{"title":"Combining satellite data and artificial intelligence with a crop growth model to enhance rice yield estimation and crop management practices","authors":"Nguyen-Thanh Son, Chi-Farn Chen, Youg-Sin Cheng, Cheng-Ru Chen, Chien-Hui Syu, Yi-Ting Zhang, Shu-Ling Chen, Shih-Hsiang Chen","doi":"10.1007/s12518-024-00575-6","DOIUrl":null,"url":null,"abstract":"<div><p>Rice is the staple food of more than half of the world’s population, especially in Asia, where rice provides more than 50% of the caloric supply for at least 520 million people, most of them are either extremely impoverished or poor. Information on rice production is thus essential for agricultural management and the formulation of food security policies. The objective of this research is to develop an approach combining remote sensing and artificial intelligence (AI) techniques with a crop growth model for enhancing yield estimation and crop management in Taiwan. The data processing involves three main steps: (1) data pre-processing to generate model inputs, (2) crop yield modeling through assimilating satellite-derived leaf area index (LAI) into a crop growth model using the AI particle swarm optimization (PSO) algorithm, and (3) model validation. The assimilation process was performed using a cost function based on the difference between remotely-sensed and simulated LAI values. The optimization process began with an initial parameterization and appropriately adjusted input parameters in the model. The fitness value derived from a cost function was determined using the PSO. The results of yield estimates obtained from the crop growth model based on optimized inputs were evaluated using the government’s yield statistics, revealing close agreement between these two datasets. The root mean square percentage error (RMSPE) and the mean absolute percentage error (MAPE) for the first crop were 19.8% and 17.1%, and the values for the second crop were 8.4% and 6.3%, respectively. The relative percentage error (RPE) values of 18.5% and − 5.1%, respectively, showed a slight overestimate and underestimate for the first and second crops.</p></div>","PeriodicalId":46286,"journal":{"name":"Applied Geomatics","volume":null,"pages":null},"PeriodicalIF":2.3000,"publicationDate":"2024-07-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Applied Geomatics","FirstCategoryId":"1085","ListUrlMain":"https://link.springer.com/article/10.1007/s12518-024-00575-6","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"REMOTE SENSING","Score":null,"Total":0}
引用次数: 0
Abstract
Rice is the staple food of more than half of the world’s population, especially in Asia, where rice provides more than 50% of the caloric supply for at least 520 million people, most of them are either extremely impoverished or poor. Information on rice production is thus essential for agricultural management and the formulation of food security policies. The objective of this research is to develop an approach combining remote sensing and artificial intelligence (AI) techniques with a crop growth model for enhancing yield estimation and crop management in Taiwan. The data processing involves three main steps: (1) data pre-processing to generate model inputs, (2) crop yield modeling through assimilating satellite-derived leaf area index (LAI) into a crop growth model using the AI particle swarm optimization (PSO) algorithm, and (3) model validation. The assimilation process was performed using a cost function based on the difference between remotely-sensed and simulated LAI values. The optimization process began with an initial parameterization and appropriately adjusted input parameters in the model. The fitness value derived from a cost function was determined using the PSO. The results of yield estimates obtained from the crop growth model based on optimized inputs were evaluated using the government’s yield statistics, revealing close agreement between these two datasets. The root mean square percentage error (RMSPE) and the mean absolute percentage error (MAPE) for the first crop were 19.8% and 17.1%, and the values for the second crop were 8.4% and 6.3%, respectively. The relative percentage error (RPE) values of 18.5% and − 5.1%, respectively, showed a slight overestimate and underestimate for the first and second crops.
期刊介绍:
Applied Geomatics (AGMJ) is the official journal of SIFET the Italian Society of Photogrammetry and Topography and covers all aspects and information on scientific and technical advances in the geomatics sciences. The Journal publishes innovative contributions in geomatics applications ranging from the integration of instruments, methodologies and technologies and their use in the environmental sciences, engineering and other natural sciences.
The areas of interest include many research fields such as: remote sensing, close range and videometric photogrammetry, image analysis, digital mapping, land and geographic information systems, geographic information science, integrated geodesy, spatial data analysis, heritage recording; network adjustment and numerical processes. Furthermore, Applied Geomatics is open to articles from all areas of deformation measurements and analysis, structural engineering, mechanical engineering and all trends in earth and planetary survey science and space technology. The Journal also contains notices of conferences and international workshops, industry news, and information on new products. It provides a useful forum for professional and academic scientists involved in geomatics science and technology.
Information on Open Research Funding and Support may be found here: https://www.springernature.com/gp/open-research/institutional-agreements