Crop Disease Classification using Deep Learning Technique

Abstract

Deep learning (DL) arises as a new development in machine learning (ML) that newlineaccomplishes the advanced outcomes of many research fields, for example, computer newlinevision,medication design, plantation health issues, and bioinformatics. The benefit of DL newlineis exploiting legitimately raw information directly without using any feature extractor. In newlinethe recent past, DL utilization gave good outcomes in both industrial and academic fields newlinebecause of two prime reasons. Primarily, the bulk of data is produced day-by-day. Consequently, this information can be utilized in order to prepare a deep model. Furthermore, the newlineintensity of computing given by the high-performance computing and graphics processing newlineunit (GPU) forms the training possible for DL models. newlineAs DL frameworks began to make advancements with time, these have been deployed to newlineperform image classification and recognition. These frameworks have also been introduced newlinein various agricultural applications, e.g. crop leaf disease classification and detection, weed newlinedetection, etc. Accurate and precise identification of crop diseases is vital for sustainable newlineagricultural productivity, along with avoiding the excessive wastage of monetary and other newlineassets. Generally, crop disease cannot be identified with the naked eyes. So, farmers need to newlinesend their infected crop samples to the pathologist who diagnoses the infected leaves of the newlineplant after doing optical observations, which is a very laborious, costly, and time consuming newlinetask to identify the disease present in the infected area of the crop. However, it is very newlineimportant to detect the early manifestations present in the crop in order to preserve the yield newlineof agriculture. Thus, automatic and intelligent systems are required that have the capability newlineto diagnose crop diseases in an accurate and efficient way. newlineVarious DL and ML techniques could be deployed to detect crop leaf diseases using the newlineImage dataset. ...