Crop disease is one of the important factors for yield. The early monitoring of crop diseases mainly rely on traditional methods of judging disease classes of plant protection, with strong subjectivity, low efficiency, delay and other defects, and crop-dust is also carried out to prevent plant diseases in advance. However, this method has the risk of excessive pesticide, which affects the ecological environment of rice growth. In recent years, with the rapid development of information technology, hyperspectral image technology has been applied in agricultural production with huge data, sensitive and reliable . It has become an effective method to detect crop diseases. This paper describes the spectral response mechanism under the plant disease stress, from 5 aspects of food crops, cash crops, vegetable crops, fruits and other crops, and summarizes the progress of hyperspectral image technology in the study of plant diseases in recent 7 years at home and abroad. On this basis, this paper puts forward the deficiencies of the technology in the field and the prospect for the future, also, provide a reference in the crop disease monitoring.