The Research Of Image Capture And Recognition On Remote Montoring System Of Agricultural Pests

China is not only a large agricultural country,but also a small one.It is of great important to increase agricultural production and ensure food security.However,there are up to 13%～16%losses caused by pests every year,and it is the right time to control pests.Guided by remote monitoring system of agricultural pests,pests control can be carried out effectively,and that increaes food production and ensures national food security.Pests digital image is the main carrier of the pests information,and it is a issue to be solved that how to capture high-quality pests image efficiently,and get identified quickly and accurately.On this issue,the main work carried out in the atricle is as follows:1)For the inefficience brought by the large number of small body pests,a device for pests sampling was designed;2)Got pests image pre-procession on edge detection and image segmentation done,preparing for pests image identification;3)For low efficiency of manual identification and difficulty to promote of the machine identification,a way of self-compare identification based on profile similarity was proposed,getting pests image identified by their profile.Throughout the research on image capture and identification,pests images are captured quickly and efficiently in a high quality,and get identified by the way of self-compare identification.It ensurs the timeliness and accuracy of pests information obtainment,and lays a solid foundation for development and application of remote monitoring system of agricultural pests.