부산대학교 도서관

This chapter presents a discursive literature survey on the applications of artificial intelligence (AI) techniques in plant disease detection. The agriculture field faces many problems from cultivating to harvesting. Major concerns are various disease infections. This leads to severe yield loss with environmental hazards due to extreme usage of insecticides. With insane expansion of human population, the demand for food is incessantly surging. Conventional techniques used by farmers are not only adequate to satisfy the augmenting demand but also hamper the soil by intense use of hazardous pesticides. Besides, conventional techniques, AI gives many advantages in disease detection. In 1983, computer application was used to solve a problem in agriculture for the first time. Since then, numerous approaches have been designed to figure out a large number of problems in the field of agriculture. Furthermore, many databases and decision support systems have been developed. Out of these, AI techniques have been conveyed to deliver results with better accuracy and robustness. In addition, it enabled researchers to detect the complicated details of each condition and offer a solution that could be a perfect fit for the respective problem. Different AI techniques like convolutional neural network, artificial neural network, and deep learning have been successfully used for disease detection in rice, wheat, maize, cotton, tomato, peas, potato, cucumber, cassava, berries, peach, grapes, olives, mango, banana, apple, sweet paper, tea, and so on. This chapter discusses various AI techniques that were developed and used in agriculture for plant disease detection and discourses in its future to achieve precision in farming.