Particle Image Recognition And Counting Methods Of Particle Image In Seed Quality To The Determination Of Species

Digital image processing techniques used more and more widely in industrial and agricultural production medical and health, etcIt has become a big hotspot both at home and abroad. Agricultural scientists also hope to have more convenient and accurate method in the process of breeding to determine the parameters of quality characteristics for seed especially corn or legume seed, to achieve purpose that selection of seeds to cultivate good. For the reason that, This issue conducted in-depth study on this problem. We worked together with HeBei Province agriculture section yard corn research institute, developed a hardware platform for Seed quality factors detecting device, we also designed and implemented a system for the determination of seed quality factors based on particle image processing techniques.This issue facing agriculture breeding research staff, it meets the actual need, Purposeful, Practical. R & D results can be generalized to the quality of crop fields of detection and classification. It has the extensive application prospect.Research content:1) Check the hardware platform of device, Take samples to seed species and camera. It requires low cost, easy to produce and easy to handle.2) Check the software platform of device. To achieve the basic treatment of the images of seed particles, segmented particle, and on this basis of particle counts and signs of extraction. The software also includes the feature data into a database. And analysis and processing. It focuses on image segmentation and image recognition.This paper explains the process of particle image processing and analysis used in some of the basics, such as image binarization,edge Detection,threshold segmentation,smooth filter image morphology method and so on foundation algorithm, based on the analysis of the key panicle image segmentation algorithm. Pellet image counting algorithm and pellet image feature extraction & pattern recognition method, and It has made some improvements to the essential algorithm. Finally, it introduces the seed quality factors in detecting the device's hardware platform and software platform solution, It uses a software engineering development methods and object-oriented VC++language. We made a system requirement analysis, system architecture, module design, algorithm design and operation test, finally we realized based on particle image processing technology seed quality parameters measurement system。...