Early prediction of disease in soybeans by state-of-the-art machine vision technology

Early prediction and identification of disease in any crop is essential to prevent severe damage and enhance crop productivity. Traditional method of disease identification poses a significant challenge in terms of accuracy, time consumption and real-time detection. Computer vision including image-based analysis has been an alternative to the traditional methods for efficient, convenient, and precise disease prediction at an early stage. With the advancement in machines, technologies, camera sensors, and analysis techniques like machine learning (ML) and deep learning (DL), image-based plant disease identification has become more accurate, efficient and applicable in agriculture. The imagery data obtained from digital, spectral, and thermal images are subjected to analysis through the use of algorithms or ML and DL methods. In this review we have summarized how the integration of computer vision and artificial intelligence (ML and DL) can be used for precisely predicting the disease incidence in soybean at an early stage. Recent studies conducted regarding early prediction of soybean disease, along with the challenges and limitations and their possible solutions, have also been described. The purpose of this study is to integrate the studies on the early identification of disease in soybeans along with advancement in precision agriculture. The practical applicability of smart farming systems and their integration with sensors and the Internet of Things have also been described. This study would help the researchers understand the use of computer vision integrated with ML and DL for the early prediction of soybean disease.