Detection of pests and diseases for vegetable and fruit plants using machine vision: A review

Shafik  Kiraga1   Ali Mohammad2   Reza Md Nasim1,2   Chowdhury Milon1,2   Chung Sun-Ok1   Hong Soon Jung3,*   

1Department of Smart Agricultural System, Graduate School, Chungnam National University, Daejeon 34134, Republic of Korea
2Department of Agricultural Machinery Engineering, Graduate School, Chungnam National University, Daejeon 34134, Republic of Korea
3Korea National College of Agriculture and Fisheries, Jeonju 54874, Republic of Korea


Early detection of pests and diseases can significantly increase crop yield in agricultural fields. Farmers normally use visible characteristics to identify and characterize different pests and diseases. In some cases, human experts are hired to identify the resulting anomalies. All these measures are time-consuming, expensive, and impractical in some cases. The objective of this paper was to review the previous research related to pests and diseases detection in fruit and vegetable plants using machine vision techniques, focusing on visible RGB images. . Based on captured images of infected leaves or pests, machine vision techniques were applied to give a general idea about disease/pest infestation. Machine vision applications in agriculture have been encouraging for several decades, and the steady development of useful vision algorithms has been well matched by modern computers' ability to implement them at sufficiently high speeds to make them viable. The development of these vision systems has involved moving from academic research to agricultural application, which presents more problems to solve to ensure mature application in the real natural environment.

Figures & Tables

Fig. 1. Common diseases and pest infestation to tomato plants. a) Leaf mold b) Gray mold c) Canker d)Canker e) Miners f) Whitefly (Modified from Fuentes et al., 2017).