{"title":"Monitoring corn growth with disease identification and yield prediction via advanced intelligent architecture","authors":"Mustafa Mhamed , Zhao Zhang , Man Zhang","doi":"10.1016/j.jii.2025.100922","DOIUrl":null,"url":null,"abstract":"<div><div>Corn is a vital food plant globally due to its significant financial value, beneficial wellness, and nutritional effects on humanity. The implementation of artificial intelligence (AI) in agriculture has progressed, but the vision systems, which promote machine efficiency, speed, and productivity, remain to be raised. Managing corn growth from planting to harvest and early disease diagnosis impacts yield, quality, and profitability. This research assists in anticipating growth and estimating overall damage and loss by providing New Comprehensive Corn Leaf Diseases (CCLD) data via Corn Growth Stages (CGS) with initial benchmark evaluations. Secondly, it proposed a Swin Vision Transformer with a novel Advanced Pooling Layer and PO-GELU function (PS-VT+ APL). This eliminates noise, lowers the computational burden, minimizes dimensions, captures the most beneficial local features, helps to optimize the loss functions, and optimizes efficiency. Thirdly, PS-VT+ APL successfully predicts the best time for Corn Diseased Daytime (CDDT) detection with an accuracy of 99.21%. It also efficiently identifies disease types using the Corn Leaf Diseases Types (CLST) set with an efficiency of 96.12%. In addition, it is essential to determine the best moment to identify each kind of corn disease. Finally, PS-VT+APL promptly and thoroughly recognizes and distinguishes complicated symptoms of corn diseases during growth, with the highest score (99.82%). Furthermore, it works more effectively and requires less time than the baseline methods. The recommended approach boosts the automation of corn systems while simultaneously achieving excellent effectiveness through a variety of operations. It can work in conjunction with real-time systems such as self-spraying systems and unmanned aerial vehicles (UAVs) and is flexible enough to manage a wide range of corn-related jobs. This study's findings benefit various areas, including decision-making, disease management, time-sensitive harvesting schedules, development methods, and corn robotic vision.</div></div>","PeriodicalId":55975,"journal":{"name":"Journal of Industrial Information Integration","volume":"47 ","pages":"Article 100922"},"PeriodicalIF":10.4000,"publicationDate":"2025-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Industrial Information Integration","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S2452414X25001451","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
Corn is a vital food plant globally due to its significant financial value, beneficial wellness, and nutritional effects on humanity. The implementation of artificial intelligence (AI) in agriculture has progressed, but the vision systems, which promote machine efficiency, speed, and productivity, remain to be raised. Managing corn growth from planting to harvest and early disease diagnosis impacts yield, quality, and profitability. This research assists in anticipating growth and estimating overall damage and loss by providing New Comprehensive Corn Leaf Diseases (CCLD) data via Corn Growth Stages (CGS) with initial benchmark evaluations. Secondly, it proposed a Swin Vision Transformer with a novel Advanced Pooling Layer and PO-GELU function (PS-VT+ APL). This eliminates noise, lowers the computational burden, minimizes dimensions, captures the most beneficial local features, helps to optimize the loss functions, and optimizes efficiency. Thirdly, PS-VT+ APL successfully predicts the best time for Corn Diseased Daytime (CDDT) detection with an accuracy of 99.21%. It also efficiently identifies disease types using the Corn Leaf Diseases Types (CLST) set with an efficiency of 96.12%. In addition, it is essential to determine the best moment to identify each kind of corn disease. Finally, PS-VT+APL promptly and thoroughly recognizes and distinguishes complicated symptoms of corn diseases during growth, with the highest score (99.82%). Furthermore, it works more effectively and requires less time than the baseline methods. The recommended approach boosts the automation of corn systems while simultaneously achieving excellent effectiveness through a variety of operations. It can work in conjunction with real-time systems such as self-spraying systems and unmanned aerial vehicles (UAVs) and is flexible enough to manage a wide range of corn-related jobs. This study's findings benefit various areas, including decision-making, disease management, time-sensitive harvesting schedules, development methods, and corn robotic vision.
期刊介绍:
The Journal of Industrial Information Integration focuses on the industry's transition towards industrial integration and informatization, covering not only hardware and software but also information integration. It serves as a platform for promoting advances in industrial information integration, addressing challenges, issues, and solutions in an interdisciplinary forum for researchers, practitioners, and policy makers.
The Journal of Industrial Information Integration welcomes papers on foundational, technical, and practical aspects of industrial information integration, emphasizing the complex and cross-disciplinary topics that arise in industrial integration. Techniques from mathematical science, computer science, computer engineering, electrical and electronic engineering, manufacturing engineering, and engineering management are crucial in this context.