Smart Farming for Sustainable Rice Production:An Insight into Application,Challenge, and Future Prospect

Norhashila Hashim, Maimunah Mohd Ali, Muhammad Razif Mahadi, Ahmad Fikri Abdullah, Aimrun Wayayok, Muhamad Saufi Mohd Kassim, Askiah Jamaluddin

Review

Smart Farming for Sustainable Rice Production:An Insight into Application,Challenge, and Future Prospect

Norhashila Hashim1, Maimunah Mohd Ali2, Muhammad Razif Mahadi1, Ahmad Fikri Abdullah1, Aimrun Wayayok1, Muhamad Saufi Mohd Kassim1, Askiah Jamaluddin3

(Department of Biological and Agricultural Engineering / SMART Farming Technology Research Centre, Faculty of Engineering, Universiti Putra Malaysia, Selangor 43400, Malaysia; Department of Food Sciences, Faculty of Science and Technology, Universiti Kebangsaan Malaysia, Selangor 43600, Malaysia; Department of Resource Management & Consumer Studies, Faculty of Human Ecology, Universiti Putra Malaysia, Selangor 43400, Malaysia)

Rice has a huge impact on socio-economic growth, and ensuring its sustainability and optimal utilization is vital. This review provides an insight into the role of smart farming in enhancing rice productivity. The applications of smart farming in rice production including yield estimation, smart irrigation systems, monitoring disease and growth, and predicting rice quality and classifications are highlighted. The challenges of smart farming in sustainable rice production to enhance the understanding of researchers, policymakers, and stakeholders are discussed. Numerous efforts have been exerted to combat the issues in rice production in order to promote rice sector development. The effective implementation of smart farming in rice production has been facilitated by various technical advancements, particularly the integration of the Internet of Things and artificial intelligence. The future prospects of smart farming in transforming existing rice production practices are also elucidated. Through the utilization of smart farming, the rice industry can attain sustainable and resilient production systems that could mitigate environmental impact and safeguard food security. Thus, the rice industry holds a bright future in transforming current rice production practices into a new outlook in rice smart farming development.

rice production; smart farming; food security; agriculture sustainability

Rice (L.) is the primary food source for anestimated 3.5 billion people globally, which contributes up to 50% of the daily caloric intake for Asian populations (Li et al, 2018). More than half of the world’s population relies on rice as a staple diet, making it the second most extensively produced cereal crop (Giri et al, 2022). It is also a frontrunner in the struggle against global poverty and hunger as well as the primary staple food and a significant source of income for worldwide populations. In this regard, rice is crucial to the economy and those with lower incomes in promoting agricultural growth and reducing poverty. As the proportion of income from agriculture diminishes dueto the faster growth of other industries, the significance of rice continues to decline economically (Adjao and Staatz, 2015). The rice production must be regulated in order to meet the demand of the expanding population. As a strategic crop, rice production should be increased globally to ensure food security and sustainability.

Most countries were unprepared for the significant increase in food prices when the food crisis of 2007 to 2008 hit, particularly the price of rice (Nodin et al, 2022). Stable rice prices in most of the world’s marketplaces have been considered a sign of food security. The strategy of rice production for ensuring food security takes into consideration of rice marketing, and import and export trends. For several developing countries, rice is crucial to food security, socio-cultural practices, and government strategic initiatives. The efficient use of resources including labour, water, land, and energy along with a lower environmental footprint is needed in order to sustain food security and support economic development (Devkota et al, 2020). Conventional rice production needs to evolve to make room for additional resources, an effective and profitable systemfor future generations. It is crucial to take the sustainability of food systems into account, especially better- utilizing rice production in order to feed 10 billion people in 2050 (Saha et al, 2021).

Typically, developed countries will need to raise rice production, while least developed countries will need to maintain their rice production to meet their own needs. A sustainable approach to rice production is thus becoming even more vital in tandem with the growing population, limitation of fertile land, and insufficient natural resources. Achieving self-sufficiency in rice production towards food security has significant consequences for a stable supply of rice. The rice dominance as a food commodity guarantees the stability in the whole operation of rice production. In the early 1990s, self-sufficiency was achieved from the increased per capita availability and intensive rice production (Vaghefi et al, 2016). However, since then, there have been certain challenges with intensive rice production due to the resource shortages for modern agricultural practices. For this reason, advances in smart farming in rice production have empowered the supply in accordance with the growing population.

Smart farming is an agricultural management strategy supported by technology that monitors and assesses the requirements of specific areas and crops (Alfred et al, 2021). It is known for the use of information and data technology for the optimization of sophisticated farming systems. Instead of emphasising data storage, access, or applications, it concentrates on how the gathered information about rice production may be used intelligently. Utilizing smart farming technologies will undoubtedly help farmers with a variety of tasks to boost rice production. The ability of smart farming for rice production is one of the key issues at stake in the discussion of how to guarantee global food securityand sustainability. By utilizing natural resources, smart farming adopts the advances of artificial intelligence to meet the demands of the expanding population. Smart farming, with the integration of various technologies, offers huge potential to create automated operations withminimal human intervention. In this sense, mechanization, efficiency, and productivity are required at all rice production stages due to the global population increase and the limited availability of agricultural lands (Bacco et al, 2018). As a result, implementing precision agriculture or smart farming is essential for coping methods in the settings of food security and sustainability. Therefore, this review addresses the global production and consumption of rice. The recent applications of smart farming technologies have been highlighted with the technological advancements for sustainable rice production. The underlying challenges, trends as well as future prospects of rice production in food security and sustainability are also elucidated.

Trends of global rice production

In the past, global rice production has been relatively modest compared with crops such as wheat, maize, and soybeans. However, there was a notable expansion in rice production during the 1980s, following trade liberalization efforts by several rice-producing nations. This trend was further boosted by the implementation of the General Agreement on Tariffs and Trade in 1994 (Abera et al, 2019). During the early stages, government intervention held a pivotal position in rice production, consumption, and trade due to inadequate food supply, low consumer income, and limited foreign exchange earnings (Xie et al, 2014). However, with economic development, government policies have shifted towards market liberalization, characterized by the establishment of individual production systems, the elimination of consumption subsidies, and the abolition of export taxation. This policy evolution has generally transitioned from safeguarding consumer benefits to protecting the interests of producers. In terms of trade policy, export taxation measures are typically introduced when consumer income is low, and the government is committed to ensuring ample domestic supplies at affordable and stable prices. The absence of a coherent policy for the development and promotion of rice production, primarily due to data limitations, will inevitably impact future planning and management strategies for rice production.

In 1961, only 215 million tonnes of rice were produced worldwide, even though rice farming had been ongoing for thousands of years (Ali Chandio et al, 2022). The global harvested area of rice was recorded at 1.65 × 108hm2, whereas global rice production obtained 7.76 × 108t in 2022, respectively (FAOSTAT, 2023). During the period from 2016 to 2023, there was a significant increase in global rice production and the area of harvested rice (Fig. 1). Enhanced rice productivity is probably the main reason for the increase in global rice production and harvested area during this period. Global rice demand is expected to be equivalent to an annual growth rate of approximately 1%, ranging from 5.03 × 108to 5.44 × 108t by 2030 (FAO, 2014). Althoughshifting consumption habits have an impact, population expansion is primarily driving this demand growth. This outlook emphasizes the necessity of increasing rice production to satisfy the growing demand globally.

Between the 1940s and the late 1960s, the Green Revolution or known as the Third Agricultural Revolutionled to a rise in agricultural productivity among developing countries primarily as a result of technology transfer and innovations (Muthayya et al, 2014). The global rice production is projected to keep pace with demand and consumption. Accumulation by rice-importing countries, particularly China, would maintain relative stability, whereas Thailand would focus on liquidation from the government stockpiles (Abera et al, 2019). Current population predictions indicate that Asia will require an additional 8 × 106t of rice annually despite the fact that rice production is decreasing (Mohidem et al, 2022). The demand for rice has increased along with the preference and taste of customers for rice-based cuisines, as a result of the growing Asian communities living in western regions. Therefore, to fulfill the rapidly rising demand and discerning consumer needs, rice production in the future will need to greatly enhance resource efficiency. It is anticipated that rice will continue to play a significant role in human diets in the near future.

The regional distribution of rice production has remained relatively stable over the past decades, with the highest share dominated by Asia at 89.7%, followed by Africa at 4.9%, Americas at 4.8%, Europe at 0.5%, and Oceania at 0.1%, respectively (FAOSTAT,2023). Asia’s rapid agricultural revolution has driven theregional transformation and rapid economic development (Ali Chandio et al, 2022). Americas and Africa both display similar patterns, although the changes in rice production have been less dramatic. The significant shifts in the regional production share of rice, particularly in Asia, are driven by the underlying dynamics of rapid income growth, supported by in-depth economic research and historical context. These issues in rice production should be addressed through a systematic and holistic approach. Additionally, the anticipated improvement in the customers’ standard of living and income level is expected to have an influence on rice production. In the near future, it is projected that the global rice trade will significantly increase due to the increased consumption of rice in non-Asian countries such as several regions of Africa and Europe. By deeply investing in and widely applying scientific advancements and technologies, particularly biological technologies, in rice production, there will be a substantial increase in both the productivity and quality of rice in domestic and overseas markets. Given the varying cultural and taste preferences for rice consumption across different regions of the world, it is anticipated that the long-term trend of rice consumption will continue to rise. However, marketing demands are becoming more intricate due to expanding urban societies and a growing middle-class population. This demographic shift leads to increasing demand of high-quality rice with diverse nutritional values and safety assurances, while the demand of poor-quality rice diminishes.

Fig. 1. Harvested area (A) and production (B) of rice worldwide from 2016 to 2022 (FAOSTAT, 2023).

The top rice producers worldwide in 2022 were China, India, Bangladesh, Indonesia, Vietnam, Thailand, Myanmar, the Philippines, Cambodia, and Pakistan (Table 1). China achieved the highest total rice production in 2020, with 2.08× 108t, followed by India and Bangladesh, with 1.96 × 108and 5.71 × 107t, respectively. In several countries, such as the United Republic of Tanzania, Madagascar, Peru, Colombia, Mali, and Guinea, rice production was on par with some Asian countries, such as Laos, Malaysia, Korea, and Sri Lanka. The production quantities of rice are expected to rise significantly in the coming years. This is vital as rice production quantities are important for maintaining per capita rice consumption worldwide. Asian countries will likely continue to dominate rice demand in the future due to factors including population expansion, dietary changes, and increased yields from land intensification. Given that rice remains the primarystaple in leading producing nations, the establishment ofdomestic strategic reserves holds paramount significance for ensuring food security.

Table 1. Top 30 rice producers worldwide in 2022 (FAOSTAT, 2023).

Over the past 20 years, global rice production and consumption have witnessed significant trends and patterns, which can be further analyzed by segmenting the data based on the type of rice. This segmentation allows for a more detailed understanding of the dynamics within each rice category, providing insights into the factors influencing production and consumption. Total global rice production has generally increased over the past two decades due to population growth, food demand expanding, and improvements in agricultural practices. The production growth rate may vary among different regions, with some countries experiencing substantial increases in output, driven by advancements in technology, irrigation systems, and farming techniques. It may also be influenced by factors such as weather conditions, land availability, government policies, and market demand.

Fragrant rice varieties, including basmati and jasmine, have shown a rising demand globally, driven by their unique aroma, flavour, and cultural significance. Production of fragrant rice has expanded in countries such as India, Pakistan, Thailand, and Vietnam, which are renowned for their high-quality aromatic rice. Increasing international trade and export opportunities have contributed to the growth of fragrant rice productionand consumption. Non-fragrant rice, including parboiled and glutinous varieties, remains the dominant type in global rice production and consumption. Countries, like China, India, Indonesia, and Bangladesh, are major producers and consumers of non-fragrant rice. The production of non-fragrant rice has shown steady growth, primarily driven by the increasing demand for staple food in densely populated regions. Global rice consumption has followed diverse trends over the past two decades, influenced by factors such as population growth, urbanization, dietary changes, and economic development. In some regions, there has been a shift from traditional rice-based diets to more varied food choices, resulting in changes in consumption patterns. The consumption of fragrant rice has increased in certain markets, driven by factors like globalization, cultural preferences, and rising incomes. Non-fragrant rice remains the mainstay of global consumption, particularly in countries with large populations and high rice consumption rates. Understanding these production and consumption trends, along with the segmentation by rice types, helps policymakers, researchers, and businesses make informed decisions related to agricultural policies, trade strategies, and market development. It also aids in identifying opportunities for sustainable production, addressing food security concerns, and meeting the evolving preferences of consumers worldwide.

Applications of smart farming in rice

Utilizing sustainable and smart farming technology is necessary to meet the need for rice production with higher yield and quality through mechanization and automation. Smart farming is derived from the rise of various applications of industrialization and agriculturalmodernization. Artificial intelligence, Internet of Things(IoT), cloud computing, big data, and other information technologies are specifically incorporated to deeply integrate with rice production applications for smart control and to deliver a precise management system, providing crop growth, predictive modelling in visual diagnosis, remote control, pest infestation, and disaster warning (Li et al, 2023). The fundamental element in smart farming is the extensive use of information, knowledge, and technology with the utmost goals including labour-saving, increased production scale, maximum yield capacity as well as substitution of manual operation and management. Smart farming has various applications in rice production, including yield estimation, smart irrigation systems, monitoring disease and growth as well as predicting quality and classification. Table 2 shows the applications of smart farming technologies for rice production.

Rice yield estimation

Rice yield estimation is very important due to decision- making according to the potential reduction in the crop yield. These innovative approaches provide real- time monitoring and data-driven insights, allowing farmers to optimize resource management, make informed decisions, and enhance overall productivity. The integration of smart farming techniques in rice yield estimation holds immense potential for improving crop management strategies, maximizing yields, and ensuring sustainable rice production. Accurate estimation of rice yield is crucial for effective crop management, food security, and agricultural planning. Muthusinghe et al (2019) evaluated rice harvest prediction using two different machine learning algorithms. A platform has been designed specifically for the smart farming concept in paddy cultivation, incorporating two key modules: a prediction module for forecasting paddy harvest and a prediction module for anticipating rice demand. The recurrent neural network (RNN) and long short-term memory (LSTM) models were used to successfully predict the rice harvest with a training accuracy of 78% and testing accuracy of 75%.

Elders et al (2022) discussed the application of remote sensing in predicting rice yield. In heterogeneous smallholder agricultural landscapes, the combination of moderate spatial resolution Sentinel-2 imagery and a crop cut calibrated random forest model has proven to be an effective tool for predicting crop type and estimating rice yields. Based on the findings, the random forest model was developed to predict the rice yield which obtained prediction accuracy higher than 80%. In another approach, Xu et al (2020) studied rice panicle features by counting the rice panicle yield in the field. The optimal feature learning network and the adaptive multi-scale hybrid window were selected by quantifying and analysing the correlation between the receptive field, input image size, and average panicle dimensions. This approach effectively maximizes the representation of panicle features. Additionally, a fusion algorithm is employed to eliminate duplicate counting of broken panicles, resulting in an accurate estimation of the final panicle count. A multi-scale hybrid window panicle detect model was developed to predict the rice panicle counting with an accuracy higher than 87%.

Hama et al (2020) investigated the rice yield estimation using an unmanned aerial vehicle remote sensing integrated with solar radiation datasets. The unmanned aerial vehicle remote sensing-derived normalized difference vegetation index is employed during the heading stage, as well as solar radiation data from the geostationary satellite Himawari-8 and the polar orbiting satellite Aqua/MODIS. The results obtained a coefficient of determination (2) of 0.76 and a root mean square error (RMSE) of 26.5. By combining these data sources, the study sought to improve the precision and reliability of yield estimation for paddy rice crops. Ogunti et al (2018) reported various applications of rice yield monitoring systems. A newly developed mobile application known as Connected Farm Field App was compared with existing applications which could enable farmers to conveniently record detailed information about their farming operations using a smartphone or tablet. The findings successfully demonstrated the effectiveness of a decision support system using mobile applications in order to help farmers monitor rice yield. It is worth mentioning that understanding the rice yield estimation is important for monitoring the growth of rice.

Smart rice irrigation system

Water is a vital resource for rice production. Nevertheless, its scarcity is becoming a growing concern for farmers. Irrigation, a non-natural technique of water delivering for rice production, helps boost crop yield (Selvaraj et al,2022). The efficacy and consistency of the smart irrigation system can be monitored through different data-basedframeworks by considering factors such as soil, climate,temperature, and harvest information. It is crucial to use water resources for efficient rice production, as this can open the door for an automated irrigation system, which is constantly being developed for efficient agriculture. Bamurigire et al (2021) reported the application of an IoT-Markov chain process based on expert knowledge and system datasets for fertilization and irrigation control systems. Likewise, Pham et al (2021) developed an IoT system to improve water efficiency in alternate wetting and drying irrigation. This system is feasible, in water saving 13% to 20% more water than manual practice. It consists of four elements: a cloud-based management platform, end-user applications, controller/gateway, and solar-powered water sensor. González Perea et al (2021) evaluated a pressurized irrigation system using different algorithms, including fuzzy logic, artificial neural network (ANN), and multi-objective genetic algorithms. The results yielded an2of 0.70 and an RMSE of 19.9, respectively.

ANN, Artificial neural network; CNN, Convolutional neural network; GIS, Geographic information system; IoT, Internet of Things; kNN, k-Nearest neighbour; LSTM, Long short-term memory; NDVI, Normalised difference vegetation index;2, Coefficient of determination; RNN, Recurrent neural network.

Rowshon et al (2019) determined the water demand modelling for rice irrigation by generating several hydro-climatic parameters. They used a multi-model (ensemble) projection to create a climate-smart decision support system for monitoring water resources under different climate conditions. Winter et al (2017) developed a water evaluation and planning support decision system. They denoted that rice yields appear to follow astep function, when there is sufficient water for flooding. Apart from that, Kadiyala et al (2015) established a decision support system for water-saving rice production. Their findings showed that up to 41% of water can be saved, while still producing 96% of rice yield. Xiao et al (2010) studied the moisture content and water height of rice field soil. They built a wireless moisture sensor that obtained a water-saving irrigation system of up to 65% compared to normal irrigation.

Zakzouk et al (2022) investigated rice irrigation systems based on different features of the water pump, including temperature, humidity, time, and moisture. They compared three different machine learning algorithms such as k-nearest neighbour (kNN), decision tree, and random forest to assess the classification performance of the system. Among all the algorithms, the random forest algorithm obtained the highest classification accuracy of up to 99%. González Perea et al (2018) reported the application of irrigation water demand using a hybrid combination of different machine learning algorithms. The model successfully achieved an2of 0.72 and an RMSE of 22.2. Liu et al (2021) evaluated an irrigation control system using a fuzzy-programmable intelligence device (PID) algorithm, which reduces the regulation time by 2.5 s. In addition, Rowshon and Amin (2010) developed a rice irrigation scheduling based on a geographic information system- based water management model. Regular monitoring of irrigation performance was performed based on ongoing water delivery programmes. Rashid et al (2022) assessed the feasibility of random forest model to determine consortium-treated wastewater for rice irrigation systems. They observed increasing trends in rice germination indices after irrigating with consortium- treated wastewater.

Monitoring rice disease and growth

Smart farming has demonstrated considerable effectives in identifying diseases and monitoring the growth of rice. Peng et al (2022) monitored weed detection in rice using deep learning models. The det-ResNet architecture was built to enhance the feature extraction of rice images, which obtained a mean average precision of 94%. In a similar manner, Latif et al (2022) employed a deep convolutional neural network (CNN) based on a deep learning method for rice disease detection. The results achieved the highest average accuracy of 96%, which can accurately classify five categories of rice disease, including leaf scald, leaf blast, brown spot, narrow brown spot, and bacterial leaf blight. Xia et al (2022) investigated the rice growth stage detection using a deep learning-based model. The full-resolution network model was designed with the highest accuracy of 0.89. Sai et al (2019) presented early detection of rice disease using an IoT system. Based on the findings, it was demonstrated that the IoT system increases field horizons by monitoring desired parameters in the cloud.

Sowmyalakshmi et al (2021) employed an IoT system with deep learning models for the detection of rice disease. The CNN model was used to classify based on rice disease, achieving high sensitivity of 0.91. In addition, Shahidur Harun Rumy et al (2021) proposed the detection of rice leaf disease, including brown spot, leaf blast, and hispa, using image processing techniques. A random forest model was developed for the discrimination of rice disease with an accuracy of 97%. Shrivastava et al (2021) detected six types of rice disease at the early stage, including bacterial leaf blight, brown spot, blast, sheath rot, sheath blight, and false smut. The CNN-deep learning- based model was created for rice disease detection with the highest classification accuracy of 93%. Hanif et al (2021) evaluated shelf life prediction based on rice aroma using an electronic nose coupled with the kNN algorithm. The model achieved the best aroma prediction with an2of 0.72 and an RMSE of 3.80.

Kusbandhini et al (2021) determined the shelf life prediction of rice using an electronic nose combined with a machine learning method. A support vector regression algorithm was used, which obtained2of 0.99 and RMSE of 0.36, respectively. Afzal and Kasi (2019) monitored the rice farm ontology using an IoT system. Based on the results, the IoT system aided in making phase-wise decisions for rice at an early stage of crop growth. Likewise, Ramesh et al (2022) identified rice growth using image processing techniques. The k-means clustering algorithm was developed to classify rice growth with the highest classification rate of 100%. Asif Saleem et al (2022) investigated the detection of rice disease using a deep learning model. The CNN architecture was designed for the classification of rice disease with the highest accuracy of 89%.

Venu Vasantha et al (2022) reported the application of a deep learning model in classifying eight different types of rice leaf disease. The CNN model was used for the classification of rice leaf disease with a testing accuracy of 98%. Furthermore, Lee et al (2022) evaluated rice blast detection using a feed-forward neural network model. The findings achieved the highest recall of 66% using input and blast occurrence datasets based on rice blast prediction. Singh et al (2021) studied the detection of three different rice diseases, including bacterial leaf blight, leaf smut, and brown spot. The random forest algorithm was used for rice disease classification with the highest accuracy of 100%. Hama et al (2021) investigated the monitoring of rice growth based on empirical correction using a drone, which was found to be feasible in mitigating the decrease in normalized difference vegetation index values. Tejaswini et al (2022) discussed the feasibility of deep learning method for the detection of rice leaf disease. The deep learning model obtained the highest accuracy of 78% in discriminating rice leaf disease.

Predicting rice quality and classification

Through the adoption of smart farming practices in rice production, there are tremendous prospects for monitoring quality and classification diversity to reducepostharvest losses. To improve the precision of prediction and classification tasks in practical applications, various imaging techniques can be utilized to identify different types of rice. Jin et al (2022) distinguished between rice seed varieties using a near-infrared hyperspectral imaging technique. A deep learning- based ResNet architecture was employed, achieving a classification accuracy of 86% for rice seed identification. Rahimzadeh et al (2022) identified aroma rice using an electronic nose coupled with a fuzzy clustering algorithm. The results demonstrated that fuzzy clustering is feasible for determining the number of clusters and grouping for aromatic rice based on aroma changes. Moreover, Moses et al (2022) investigated damage classification in milled rice using a machine vision system. CNN models were developed for rice damage discrimination with an overall classification accuracy of 98%. Aznan et al (2021) designed an ANN model to predict rice quality, including colour, texture, pH, and aroma. The ANN model successfully classified the rice with the highest classification rate of 98%.

Meanwhile, Ruslan et al (2022) reported the application of machine vision combined with machine learning methods to identify rice seed classification. Logistic regression was used based on the obtained rice seed images, resulting in a correct classification rate of 92%. Garnaik et al (2022) employed a machine learning algorithm to monitor soil quality based on rice productivity. The random forest model was found to be feasible for long-term fertilizer application. Arboleda and Dizon (2022) observed rice seed classification by extracting colour features from healthy and unhealthy rice images. A Coarse Tree classifier was established for rice seed identification, which successfully obtained a classification rate of 100%. Zia et al (2022) integrated the applications of computer vision and machine learning in the identification of rice kernels. Seven features of rice kernels, including width, weight, length, chalkiness, damage, yellowness, and brokenness, were analysed, with the highest classification accuracy of 98%.

Lu et al (2021) evaluated the prediction of potassium content based on the canopy hyperspectral reflectance of rice. The partial least squares model was used for potassium content prediction with an2of up to 0.76 and an RMSE of 0.37, respectively. Almaleeh et al (2022) analyzed moisture distribution detection using radio frequency tomographic imaging. Regression- based machine learning successfully detected multiple phases of moisture distribution in rice with the highest accuracy of 83%. Koklu et al (2021) investigated the classification of five different rice varieties, including Basmati, Ipsala, Jasmine, Karacadag, and Arborio. Based on the findings, the CNN model achieved the highest classification of 100% for discriminate rice varieties.

Erlangga et al (2021) assessed the detection of rice quality using an electronic nose. A neural network algorithm was developed to distinguish between expired and non-expired rice grains, achieving an accuracy score of 99%. Likewise, Aulia et al (2021) discussed the feasibility of an electronic nose in predicting the shelf life of rice. The gradient tree boosting machine learning algorithm received the highest classification accuracy of 96%. Research conducted by Wang et al (2022) reported the application of a gradient-weighed class activation mapping-based tool and CNN models in distinguishing between chalky and non-chalky rice grains. Azmi et al (2021) used radio frequency transceiversto identify the classification of moisture content in rice.The random forest model obtained the highest classification accuracy of 87% based on moisture content.

Challenges of rice production in food security and sustainability

Rice production plays a pivotal role in global food securityand socio-economic development, particularly in Asiancountries where it is a staple food for a significant portionof the population. However, ensuring the sustainabilityof rice production presents several challenges that needto be addressed from economic, social, and environmental perspectives.

From an economic point of view, the profitability of rice farming is affected by fluctuating market prices, input costs, and limited access to credit and resources for small-scale farmers. Market uncertainties, trade policies, and competition from imported rice further compound the economic challenges. Since the 1980s, farmers in developed countries had access to new tools and methods for rice production. This has led to the development of smart farming systems for precisely managing rice production based on various conditions and improving the efficiency of input applications. In developing countries, agricultural research centres and institutions are currently engaged in improving smart farming applications for rice production. However, farmers in underdeveloped countries with limited resources cannot afford the technologies and techniques employed in developed countries. The majority of rice farmers are underprivileged, yet national policies in countries where rice is the main food crop typically favour consumers by regulating market prices for rice. Rice production does not provide farmers with a significant income due to rising input costs and low rice prices. Food security for rice requires a rigorous national policy that permits appropriate investment at every stage of rice production. In other words, the correct policies must be created for better pricing, marketing, and input availability. Despite slower growth in per capita rice consumption in several countries, there is a need to meet the rising demand for rice production which is hugely influenced by the population of a country. Due to the unpredictability of trade policies of the major rice exporting countries, maintaining stability in rice prices is a significant problem in the context of shock transmission.

From the social perspective, it is crucial to identify several obstacles in terms of labour shortages, lack of advanced facilities, and conventional rice farming practices. Some farmers stop growing rice because there are other economic options available such as job opportunities in the industrial sector. Presently, the agricultural sector is experiencing a severe labour shortage due to the younger generation obtaining greatereducation levels and the industrial sector offering more enticing and promising career opportunities. Rice production is a vital source of income for about 140 million households involved in rice cultivation, such as farmers and hired labour (ESG, 2019). Due to this, the labour supply for rice farming operations has been reduced. Industrialization, urbanization, and the growth of residential areas are reducing the amount of land that can be used for rice farming activities. Approximately 79% of the total production cost per hectare in conventional rice farming is attributable to labour costs. In this sense, the mechanized systems and smart farming technologies are being implemented as there is less labour available for rice production activities.

From an environmental perspective, rice production isassociated with significant environmental impacts, includingwater scarcity, water pollution from agrochemical runoff,greenhouse gas emissions (particularly methane), loss of biodiversity, and deforestation. These environmental challenges call for sustainable farming practices that minimize resource use, promote climate resilience, and conserve natural ecosystems. Furthermore, the environment has suffered because of the intensification of rice production. The use of pesticides may also endanger human health and pollute waterways. Rice soil salinity levels in semi-arid and dry regions have increased due to intensive irrigation and poor drainage. This has led to nutrient depletion in rice soils after excessive pesticide use. Methane emissions derived from floodedrice and nitrous oxide emissions from nitrogen fertilizersare often associated with global warming. On the contrary, rising ocean water, inconsistent rainfall patterns, and rising temperatures could also potentially influence rice production. In tropical climate regions, high atmospheric temperatures may reduce rice production, whereas variations in rainfall distribution may cause more frequent and severe floods and droughts. Rice is a significant contributor to food security with different yields according to regional differences in climatic trends. Rising temperatures and unpredictable rainfall have a negative impact on rice growth and development. A considerable decline in rainfall availability as well as irregular and strong rainfall patterns during the past few decades are all related to a significant decline in rice production (Habib-ur-Rahman et al, 2022). It will be quite challenging to maintain sustainable rice production considering the fact that some underprivileged rural populations, particularly in Asia, are facing serious climate change situations. The risk of climate change may jeopardize economic productivity and food security in the least developed countries. Overcoming these challenges requires a multi-dimensional approachthat integrates economic incentives, social empowerment, and environmentally conscious farming practices. By addressing these challenges holistically, the rice industry can foster sustainable and resilient production systems that ensure economic viability, social well-being, and environmental conservation for present and future generations.

Technical advancements and future prospects

Global rice production has begun to show signs that may not remain constant as the 21st century progresses. Despite a slower pace of rice production, there are also fewer resources available to grow rice. There are numerous limitations and challenges to diminish rural hunger and poverty within rice-based production systems. Fortunately, there are technical ways to deal with these challenges as more than half of the world’s population depends on rice as their primary food source. The most viable path to sustainable rice production is constant productivity growth enabled by yield-enhancing smart farming technologies. The increased yields in environmentally sound and fertile cultivation areas will lessen the need to use fragile land for rice production,which will stop soil erosion and deforestation. In addition to reducing farmers’ reliance on pesticides, various research projects have been conducted to create rice varieties that are resistant to major pests and diseases as well as to control natural biological agents. Global industrialization has led to a severe manpower shortage in the agricultural sector. It is possible to reduce reliance on labour and increase worker productivity by rationalizing land ownership with an emphasis on the adoption of innovative smart farming technologies. The majority of rice production tasks have been substituted by automated machinery systems and operations. The neglect of rice fields has also been associated with lack of resources and infrastructure. By establishing a corporation with supply and distribution chains, it is possible to improve the situation of farmers in current rice production schemes by assisting them in becoming self-sufficient.

To preserve rice quality, a systematic approach relatedto rice control and management practices can boost the effectiveness of water and fertilizer usage. This will reduce the risk of environmental deterioration while increasing farmer profitability and reducing poverty. In this case, sustainable rice production is promoted to provide a sufficient supply of rice at affordable prices while utilizing environmental, water, and energy resources. Sustainable rice production strategies should take into account the economic, political, environmental, and social goals of each nation. The top priorities for sustainable rice production are international support for workable policies and political decision-making fromnational leaders. In order to achieve sustainability in rice production, smart farming technologies incorporated with the joint force from private sectors and farmer reorganizations have emerged as a new catalyst.

Several attempts have been made to modernize rice production through the design of policies in order to address the issues related to rice industry. For instance, main improvements were rapid infrastructure expansion,enlargement of cultivation area through land consolidation, adoption of mechanization, subsidies, and the development of rice plantations. The commercialization of rice production has been a great success, which has shown to be profitable and attracted potential investors. A policy change for export and import in rice production is quite reasonable and has obvious implications for the shifting role of rice international markets in ensuring food security. It is denoted that the rice international market is recognized as an institutional framework for allocating resources for the efficient and rapid processing of information. In Malaysia, the national policy is to maintain a reasonable level of self-sufficiency of at least 65% because rice is regarded as a security commodity (Vaghefi et al, 2016). Income protection of rice farmersreduced reliance on external resources, and appropriate rice pricing for consumers are the goals of policy initiatives and interventions. In addition, another goal for maintaining sustainable rice production is ensuring a stable and fair rice supply for customers. To make rice farming economically viable and to promote the rice industry, the government offers subsidies by subsidizing fertilizers to reduce production costs. The commercial markets are allowed to determine the rice price based on the quality grade in order to safeguard the interests of low-income consumers. Policy tools targeted at controlling rice trade flows need to be managed carefully to avoid negative price volatility that can influence food security and nutrition.

In the late 1980s, the primary focus of agronomic research was the introduction and verification of individual technologies such as herbicides, fertilizers, and green manures across three distinct rice growing environments: irrigated lowland, mangrove rice, and floating rice (Rodenburg and Saito, 2022). Researchers from various disciplines, including agronomists, crop modellers, physiologists, geographers, and hydrologists, conducted numerous studies to characterize these rice-growing environments, as well as inland valleys. Additionally, efforts were made to quantify yield gaps through comprehensive investigations. It is evident that policy shocks can significantly contribute to price volatility. This volatility can have adverse effects on farming communities, as higher yields and increased trade volumes may potentially lead to lower prices (Abera et al, 2019). Such price fluctuations can have negative implications for global food security and nutrition, particularly considering the substantial population reliant on rice as a major staple. Therefore, policy instrumentsaimed at regulating rice trade flows should be carefully managed to prevent detrimental price volatility that could impact food security and nutrition. Rice production programs should incorporate nutrition objectives to ensure that achieved gains positively influence nutrition and health outcomes, recognizing the crucial role that rice plays not only as a significant food source but also as a vital component of the livelihoods of many impoverished individuals worldwide.

The 1980 National Agricultural Policy placed a strong emphasis on new land development and the consolidation of economically unviable land holdings (Najim et al, 2007). This policy aimed to enhance agricultural productivity while ensuring the long-term sustainability of the sector. In line with these policy objectives, Malaysia implemented a series of measures to modernize rice farming, beginning in the late 1960s. These transformative changes were introduced gradually and encompassed various aspects of rice production. The key initiatives include improving infrastructure, increasing the size of land parcels through consolidation, reorganizing individual land lots to optimize their size and shape, introducing mechanization, providing subsidies, and establishing rice estates. The estatization of rice farming, which transformed rice cultivation into a commercial enterprise, yielded remarkable successes. The profitability of collective farming under this model prompted the private sector to also invest in rice estates, further contributing to the sector’s growth and development. These policy-driven transformations in Malaysia’s rice sector were undertaken to address the challenges faced by the industry. By modernizing farming practices and promoting collective farming, the government aims to enhance productivity, improve the livelihoods of rice farmers, and ensure a sustainable future for the rice sector. The success of these initiatives demonstrated the viability and profitability of the commercial approach to rice farming, attracting private sector investments, and further fostering the growth of the rice industry in Malaysia.

As a fundamental strategic item, investing in rural infrastructure, especially irrigation and easy access to fertilizers for rice production, is important since rice has a significant impact on food security. Given the rising demand for rice, much work needs to be done to address potential trade-offs for long-term sustainability. Furthermore, successful postharvest handling and management that reduces losses and preserves nutritional rice quality should be developed. New breeding techniques, such as various forms of biofortification, are suggested to develop new rice varieties with improvedspecific nutrient quality. There are interesting breeding techniques that can be used to improve nutrient content and reduce illness risks brought on by climatic variabilities, such as for golden rice and zinc rice. Wide-ranging rice research and breeding programmes are available from various institutions and organizations with the goal of creating climate-ready rice varieties that are more resilient to submersion, drought, heat, salinity, and biotic stresses. Additionally, research work on rice varieties resistant to salinity and drought has been developed to find ways to increase rice yields under these circumstances. Through genetic manipulation, genetic diversity can be enhanced to improve rice quality and yields. Nevertheless, these breeding methods should prevent the loss of other beneficial agronomic traits linked to productivity and disease resistance. Efforts have been exerted to effectively utilize rice genetic resources for long-term food security and nutrition.

Climate change is projected to have an impact on the welfare and food security status of rice production because of limited land and scarce resources. Increased food insecurity in rice production is caused by various factors, including inadequate storage, limited rice operation facilities, and unfavourable environmental circumstances such as erosion and flooding (Ojo et al, 2022). Despite recent attention to the potentially major influence of climate change on rice production, the links between climate change and other components of food security have not yet been well investigated. For this reason, rice availability and nutritional value are both important factors that affect food security. Numerous aspects of food security must be carefully studied and action plans must be made in order to create an effective framework for sustainable rice production. Therefore, it has been suggested that in orderto ensure the economic, social, and environmental systems working well, planned and proactive rice production strategies should be developed.

With the global demand for rice on the rise, the need for sustainable rice production has become paramount. To address this challenge, integrating advanced technologies and data-driven approaches holds significant promise for the future. Precision agriculture, smart farming, and IoT system present opportunities to optimize the management of resources, refine yield estimation, enhance pest and disease control measures, and minimize environmental impacts. Leveraging remote sensing, drones, and satellite imagery can provide valuable information for monitoring and effectively managing rice fields, enabling more efficient and sustainable agricultural practices. By embracing these innovative technologies, the rice industry can strive towards achieving higher productivity, improved resource efficiency, and reduced environmental footprints, thereby meeting the increasing global demand for rice while ensuring long-term sustainability.

A significant contribution to the quest for sustainablerice production involves the development and adoption of climate-smart rice varieties. These specially bred varieties are designed to withstand various climate stresses, including drought, heat, and flooding while maintaining high yields and nutritional qualities. Moreover, the exploration of genetic engineering and biotechnology opens up new possibilities for enhancing desirable traits in rice, such as disease resistance and nutrient content. These advancements play a crucial role in promoting sustainable rice production. The implications of sustainable rice production extend beyond the agricultural realm. By embracing environmentally friendly practices like water-efficient irrigation systems and reduced use of agrochemicals, rice farming can actively contribute to conserving water resources, preserving soil health, and safeguarding biodiversity. Furthermore, sustainable rice production can have positive socio-economic impacts, particularly in the lives of small-scale farmers. It empowers women in agriculture, promotes inclusive and equitable value chains, and ultimately improves the livelihoods of those involved in rice farming. By adopting a holistic approach that combines scientific advancements, eco-friendly practices, and social empowerment, the pursuit of sustainable rice production paves the way for a more resilient, environmentally conscious, and socially inclusive future. Therefore, the path towards sustainable rice production hinges on embracing technological progress, fostering the development of climate-smart varieties, and adopting eco-friendly practices. These strategies hold great potential for boosting productivity, mitigating environmental consequences, and fostering socio-economic prosperity. However, the realization of these prospects necessitates collaborativeendeavours involving researchers, policymakers, farmers, and other stakeholders. By uniting our efforts, we can establish sustainable rice production systems that safeguard food security, preserve the environment, and ensure the enduring socio-economic well-being of current and future generations.

In recent decades, rice has been the most important crop for ensuring food security worldwide. Rice is a strategic agricultural commodity in terms of global trade in addition to being a significant staple food. Apart from that, rice farming activities have provided job opportunities for millions of people across the world. However, numerous challenges need to be addressed in the rice industry. Conventional farming techniques and low-quality rice varieties are the main bottlenecks, which also have a detrimental impact on global export production and trade share. Research and development priorities as well as the deployment of smart farming technologies, should time-saving, rapid, and labour-saving methods in response to the manual workforce and tedious operations in rice production. Authorities, industries, universities, and research institutions should also collaborate to acceleratethe development of smart farming technologies and their contributions to agriculture. The review systematically illustrated that the technological advancements for sustainable rice production. It is noted that integrating smart farming approaches has a significant positive impact on rice productivity.

The applications of smart farming in rice production have grown rapidly including rice yield estimation, smart rice irrigation systems, monitoring rice disease and growth, and predicting rice quality and classifications. To evaluate the potential use of smart farming in various operations in rice production, intensive work has been initiated. Various studies are being conducted to transfer this technology to the rice industry based on the prospective uses of smart farming. More studies are required to develop effective and robust algorithms since a large amount of time-series data is needed for solving problems in rice production. This review has successfully shown that advanced applications in rice production benefit from smart farming. In order to reap the maximum benefits of smart farming, this technology can play a major role in establishing sustainable rice production as well as providing social and environmental benefits. Thus, the technical revolution of smart farming could potentially be used in a real-time monitoring system for a variety of applications in rice production.

Acknowledgement

The authors wish to acknowledge the Ministry of Higher Education,Malaysia for financial support via the Transdisciplinary Research Grant Scheme Project (Grant No. TRGS/1/2020/UPM/02/7).

Abera B B, Terefe B, Baye K, Covic N. 2019. Rice contribution to food and nutrition security and leveraging opportunities for sustainability, nutrition and health outcomes.: Ferranti P, Berry E M, Anderson J R. Encyclopedia of Food Security and Sustainability. Amsterdam, the Netherland: Elsevier: 257–263.

Adjao R T, Staatz J M. 2015. Asian rice economy changes and implications for sub-Saharan Africa., 5: 50–55.

Afzal H, Kasi M K. 2019. Ontology-based knowledge modeling for rice crop production.: 7th International Conference on Future Internet of Things and Cloud (FiCloud). August 26–28, 2019. Istanbul, Turkey: 343–350.

Alfred R, Obit J H, Chin C P Y, Haviluddin H, Lim Y. 2021. Towards paddy rice smart farming: A review on big data, machine learning, and rice production tasks., 9: 50358–50380.

Ali Chandio A, Gokmenoglu K K, Ahmad M, Jiang Y S. 2022. Towards sustainable rice production in Asia: The role of climatic factors., 6(1): 1–14.

Almaleeh A A, Zakaria A, Kamarudin L M, Rahiman M H F, Ndzi D L, Ismail I. 2022. Inline 3D volumetric measurement of moisture content in rice using regression-based ML of RF tomographic imaging., 22(1): 405.

Arboleda E R, Dizon A A. 2022. Seed of rice plant classification using coarse tree classifier., 11(2): 727–735.

Asif Saleem M, Aamir M, Ibrahim R, Senan N, Alyas T. 2022. An optimized convolution neural network architecture for paddy disease classification.-, 71(3): 6053–6067.

Aulia I, Wijaya D R, Hidayat W. 2021. Rice quality detection using gradient tree boosting based on electronic nose dataset.: 2021 International Conference on Artificial Intelligence and Mechatronics Systems (AIMS). April 28–30, 2021. Bandung, Indonesia: 1–5.

Azmi N, Kamarudin L M, Zakaria A, Ndzi D L, Rahiman M H F, Zakaria S M M S, Mohamed L. 2021. RF-based moisture content determination in rice using machine learning techniques., 21(5): 1875.

Aznan A, Gonzalez Viejo C, Pang A, Fuentes S. 2022. Rapid assessment of rice quality traits using low-cost digital technologies., 11(9): 1181.

Bacco M, Berton A, Ferro E, Gennaro C, Gotta A, Matteoli S, Paonessa F, Ruggeri M, Virone G, Zanella A. 2018. Smart farming: Opportunities, challenges and technology enablers.: 2018 IoT Vertical and Topical Summit on Agriculture-Tuscany (IOT Tuscany). May 8–9, 2018. Tuscany, Italy: 1–6.

Bamurigire P, Vodacek A, Jayavel K, Ngoga S R. 2021. A decision-making module for fertilization and irrigation control system in rice farming using markov chain process and SARSA algorithms.: 2021 11th International Workshop on Computer Science and Engineering, WCSE. June 19–21, 2021. DOI: 10.18178/wcse.2021.06.052.

Devkota K P, Khanda C M, Beebout S J, Mohapatra B K, Singleton G R, Puskur R. 2020. Assessing alternative crop establishment methods with a sustainability lens in rice production systems of Eastern India., 244: 118835.

Elders A, Carroll M L, Neigh C S R, D’Agostino A L, Ksoll C, Wooten M R, Brown M E. 2022. Estimating crop type and yield of small holder fields in Burkina Faso using multi-day Sentinel-2., 27: 100820.

Erlangga F, Wijaya D R, Wikusna W. 2021. Electronic nose dataset for classifying rice quality using neural network.: 2021 9th International Conference on Information and Communication Technology (ICoICT). August 3–5, 2021. Yogyakarta, Indonesia: IEEE: 462–466.

ESG (Earth Security Group). 2019. Financing Sustainable Rice for a Secure Future: Innovative Finance Partnerships for Climate Mitigation and Adaptation. Essex, United Kingdom: © Earth Security Partnerships 2019: 1–56.

FAO. 2014. A regional rice strategy for sustainable food security in Asia and the Pacific. Rome, Italy: Food and Agriculture Organization of the United Nations Regional Office for Asia and the Pacific Bangkok: 1–73.

FAOSTAT. 2023. Production/yield quantities of rice in World (2016–2022). (2023-12-27) [2024-01-17]. https://www.fao.org/ faostat/en/#data/QCL/visualize.

Garnaik S, Samant P K, Mandal M, Mohanty T R, Dwibedi S K, Patra R K, Mohapatra K K, Wanjari R H, Sethi D, Sena D R, Sapkota T B, Nayak J, Patra S, Parihar C M, Nayak H S. 2022. Untangling the effect of soil quality on rice productivity under a 16-years long-term fertilizer experiment using conditional random forest., 197: 106965.

Geollegue K W V, Arboleda E R, Dizon A A. 2022. Seed of rice plant classification using coarse tree classifier., 11(2): 727–735.

Giri D, Dhital M, Chaudhary B, Pandey R, Bastakoti B, Shrestha S. 2022. Effect of different nitrogen levels on yield and yield attributes of different rice varieties in DDSR condition at Kanchanpur, Nepal., 7(3): 310–317.

González Perea R, Camacho Poyato E, Montesinos P, Rodríguez Díaz J A. 2018. Prediction of applied irrigation depths at farm level using artificial intelligence techniques., 206: 229–240.

González Perea R, Camacho Poyato E, Rodríguez Díaz J A. 2021. Forecasting of applied irrigation depths at farm level for energy tariff periods using Coactive neuro-genetic fuzzy system., 256: 107068.

Habib-ur-Rahman M, Ahmad A, Raza A, Hasnain M U, Alharby H F, Alzahrani Y M, Bamagoos A A, Hakeem K R, Ahmad S, Nasim W, Ali S, Mansour F, EI Sabagh A. 2022. Impact of climate change on agricultural production: Issues, challenges, and opportunities in Asia., 13: 925548.

Hama A, Tanaka K, Mochizuki A, Tsuruoka Y, Kondoh A. 2020. Improving the UAV-based yield estimation of paddy rice by using the solar radiation of geostationary satellite Himawari-8., 14(1): 56–61.

Hama A, Tanaka K, Chen B, Kondoh A. 2021. Examination of appropriate observation time and correction of vegetation index for drone-based crop monitoring., 77(3): 200–209.

Hanif S, Wijaya D R, Wikusna W. 2021. K-nearest neighbors algorithm for prediction shelf life of rice based on electronic nose dataset.: 2021 IEEE Asia Pacific Conference on Wireless and Mobile (APWiMob). April 8–10, 2021. Bandung, Indonesia: IEEE: 116–120.

Jeyaraj P R, Asokan S P, Samuel Nadar E R. 2022. Computer-assisted real-time rice variety learning using deep learning network., 29(5): 489–498.

Jin B C, Zhang C, Jia L Q, Tang Q Z, Gao L, Zhao G W, Qi H N. 2022. Identification of rice seed varieties based on near-infrared hyperspectral imaging technology combined with deep learning., 7(6): 4735–4749.

Kadiyala M D M, Jones J W, Mylavarapu R S, Li Y C, Reddy M D. 2015. Identifying irrigation and nitrogen best management practices for aerobic rice-maize cropping system for semi-arid tropics using CERES-rice and maize models., 149: 23–32.

Koklu M, Cinar I, Taspinar Y S. 2021. Classification of rice varietieswith deep learning methods., 187: 106285.

Kusbandhini O J, Wijaya D R, Hidayat W. 2021. Rice shelf-life prediction using support vector regression algorithm based on electronic nose dataset.: International Conference on Artificial Intelligence and Mechatronics Systems (AIMS). April 28–30, 2021. Bandung, Indonesia: 1–5.

Latif G, Abdelhamid S E, Mallouhy R E, Alghazo J, Kazimi Z A. 2022. Deep learning utilization in agriculture: Detection of rice plant diseases using an improved CNN model., 11(17): 2230.

Lee K T, Han J, Kim K H. 2022. Optimizing artificial neural network-based models to predict rice blast epidemics in Korea., 38(4): 395–402.

Li D P, Nanseki T, Chomei Y, Kuang J S. 2023. A review of smart agriculture and production practices in Japanese large-scale rice farming., 103(4): 1609–1620.

Li Y, Xiao J H, Chen L L, Huang X H, Cheng Z K, Han B, Zhang Q F, Wu C Y. 2018. Rice functional genomics research: Past decade and future., 11(3): 359–380.

Lin S D, Yao Y, Li J Y, Li X B, Ma J, Weng H Y, Cheng Z X, Ye D P. 2023.Application of UAV-based imaging and deep learning in assessment of rice blast resistance., 30(6): 652–660.

Liu S J, Hu W W, Xu S P, Wu Y J, Liu Y, Wang Y S. 2021. Research and design of intelligent irrigation control system based on fuzzy.: 4th International Conference on Computer and Communication Engineering Technology (CCET). August 13–15, 2021. Beijing, China: 396–401.

Lu J S, Eitel J U H, Jennewein J S, Zhu J E, Zheng H B, Yao X A, Cheng T, Zhu Y, Cao W X, Tian Y C. 2021. Combining remote sensing and meteorological data for improved rice plant potassium content estimation., 13(17): 3502.

Mohidem N A, Hashim N, Shamsudin R, Che Man H. 2022. Rice for food security: Revisiting its production, diversity, rice milling process and nutrient content., 12(6): 741.

Moses K, Miglani A, Kumar Kankar P. 2022. Deep CNN-based damage classification of milled rice grains using a high-magnification image dataset., 195: 106811.

Muthayya S, Sugimoto J D, Montgomery S, Maberly G F. 2014. An overview of global rice production, supply, trade, and consumption., 1324: 7–14.

Muthusinghe M R S, Palliyaguru S T, Weerakkody W A N D, Hashini Saranga A M, Rankothge W H. 2019. Towards smart farming: Accurate prediction of paddy harvest and rice demand.: IEEE Region 10 Humanitarian Technology Conference (R10-HTC). December 6–8, 2018. Malambe, Sri Lanka: 1–6.

Najim M M M, Lee T S, Haque M A, Esham M. 2007. Sustainability of rice production: A Malaysian perspective., 3(1): 1–12.

Nodin M N, Mustafa Z, Hussain S I. 2022. Assessing rice production efficiency for food security policy planning in Malaysia: A non-parametric bootstrap data envelopment analysis approach., 107: 102208.

Ogunti E O, Akingbade F K, Segun A, Oladimeji O. 2018. Decision support system using mobile applications in the provision of day to day information about farm status to improve crop yield., 6(2): 89–99.

Ojo T O, Ogundeji A A, Emenike C U. 2022. Does adoption of climate change adaptation strategy improve food security? A case of rice farmers in ogun state,, 11(11): 1875.

Peng H X, Li Z H, Zhou Z Y, Shao Y Y. 2022. Weed detection in paddy field using an improved RetinaNet network., 199: 107179.

Pham V B, Diep T T, Fock K, Nguyen T S. 2021. Using the Internet of Things to promote alternate wetting and drying irrigation for rice in Vietnam’s Mekong Delta., 41(3): 43.

Rahimzadeh H, Sadeghi M, Ahmad Mireei S, Ghasemi-Varnamkhasti M. 2022. Unsupervised modelling of rice aroma change during ageing based on electronic nose coupled with bio-inspired algorithms., 216: 132–146.

Ramesh S, Tamilselvi M, Ramkumar G, Anitha G, Nirmala P. 2022. Comparison and analysis of rice blast disease identification in greenhouse controlled environment and field environment using ML Algorithms.: International Conference on Advances in Computing, Communication and Applied Informatics (ACCAI). January 28–29, 2022. Chennai, India: 1–5.

Rashid A, Mirza S A, Keating C, Ijaz U Z, Ali S, Campos L C. 2022. Machine learning approach to predict quality parameters for bacterialconsortium-treated hospital wastewater and phytotoxicity assessment on radish, cauliflower, hot pepper, rice and wheat crops., 14(1): 116.

Rodenburg J, Saito K. 2022. Towards sustainable productivity enhancement of rice-based farming systems in sub-Saharan Africa., 287: 108670.

Rowshon M K, Amin M S M. 2010. GIS-based irrigation water management for precision farming of rice., 3(3): 27–35.

Rowshon M K, Dlamini N S, Mojid M A, Adib M N M, Amin M S M, Lai S H. 2019. Modeling climate-smart decision support system (CSDSS) for analyzing water demand of a large-scale rice irrigation scheme., 216: 138–152.

Ruslan R, Khairunniza-Bejo S, Jahari M, Ibrahim M F. 2022. Weedy rice classification using image processing and a machine learning approach., 12(5): 645.

Saha I, Durand-Morat A, Nalley L L, Alam M J, Nayga R. 2021. Rice quality and its impacts on food security and sustainability in Bangladesh., 16(12): e0261118.

Sai K K, Satyanarayana P, Hussain M A, Suman M. 2019. A real time precision monitoring and detection of rice plant diseases by using internet of things (IoT) based robotics approach., 8(7): 403–408.

Selvaraj S K, Sanjeev Kumar R, Balamurugan K, Joany R M, Dorothy R, Nguyen T A, Rajendran S. 2022. Wireless nanosensor network for irrigation control.: Nanosensors for Smart Agriculture. Amsterdam, the Netherland: Elsevier: 463–478.

Shahidur Harun Rumy S M, Arefin Hossain M I, Jahan F, Tanvin T. 2021. An IoT based system with edge intelligence for rice leaf disease detection using machine learning.:IEEE International IOT, Electronics and Mechatronics Conference (IEMTRONICS). April 21–24, 2021. Ontario, Canada: 1–6.

Shrivastava V K, Pradhan M K, Thakur M P. 2021. Application of pre-trained deep convolutional neural networks for rice plant disease classification.: International Conference on Artificial Intelligence and Smart Systems (ICAIS). March 25–27, 2021. Coimbatore, India: 1023–1030.

Singh A K, Chourasia B, Raghuwanshi N, Raju K. 2021. BPSO based feature selection for rice plant leaf disease detection with random forest classifier., 69(4): 34–43.

Sowmyalakshmi R, Jayasankar T, Ayyem PiIllai V, Subramaniyan K, Pustokhina I V, Pustokhin D A, Shankar K. 2021. An optimal classification model for rice plant disease detection.-, 68(2): 1751–1767.

Tejaswini P, Singh P, Ramchandani M, Rathore Y K, Janghel R R. 2022. Rice leaf disease classification using CNN.:, 1032(1): 012017.

Vaghefi N, Shamsudin M N, Radam A, Rahim K A. 2016. Impact of climate change on food security in Malaysia: Economic and policy adjustments for rice industry., 13(1): 19–35.

Venu Vasantha S, Samreen S, Lakshmi Aparna Y. 2022. Rice disease diagnosis system (RDDS).-, 73(1): 1895–1914.

Wang C X, Caragea D, Kodadinne Narayana N, Hein N T, Bheemanahalli R, Somayanda I M, Krishna Jagadish S V. 2022. Deep learning based high-throughput phenotyping of chalkiness in rice exposed to high night temperature., 18(1): 9.

Winter J M, Young C A, Mehta V K, Ruane A C, Azarderakhsh M, Davitt A, McDonald K, Haden V R, Rosenzweig C. 2017. Integrating water supply constraints into irrigated agricultural simulations of California., 96: 335–346.

Xia L, Zhao F, Chen J, Yu L, Lu M, Yu Q Y, Liang S F, Fan L L, Sun X, Wu S R, Wu W B, Yang P. 2022. A full resolution deep learning network for paddy rice mapping using Landsat data., 194: 91–107.

Xiao K H, Xiao D Q, Luo X W. 2010. Smart water-saving irrigation system in precision agriculture based on wireless sensor network., 26(11): 170–175. (in Chinese with English abstract)

Xie S N, Artachinda O N, Yang J, Liu H G. 2014. Rice policy reviews in China, Thailand and Vietnam: Policy instruments, targets and impacts.: Proceedings of Selected Articles of 2013 World Agricultural Outlook Conference. Berlin, Heidelberg: Springer: 117–134.

Xu C, Jiang H Y, Yuen P, Zaki Ahmad K, Chen Y. 2020. MHW-PD: A robust rice panicles counting algorithm based on deep learning and multi-scale hybrid window., 173: 105375.

Zakzouk S, Ehab M, Atef S, Yousri R, Tawfik R M, Darweesh M S. 2022. A comparative study of machine learning techniques for automatic rice crop irrigation.: 17th International Computer Engineering Conference (ICENCO). December 29–30, 2021. Cairo, Egypt: 46–50.

Zia H M, Fatima H S, Khurram M, Hassan I U, Ghazal M. 2022. Rapid testing system for rice quality control through comprehensive feature and kernel-type detection., 11(18): 2723.

1 May 2023;

4 August 2023

Norhashila Hashim(norhashila@upm.edu.my); Maimunah Mohd Ali (maimunahmma@ukm.edu.my)

Copyright © 2024, China National Rice Research Institute. Hosting by Elsevier B V

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)

Peer review under responsibility of China National Rice Research Institute

http://doi.org/10.1016//j.rsci.2023.08.004

(Managing Editor: Wang Caihong)