Identification Of Seed Germination Rate And Measurement Of Radicle And Germ Of Millet Based On Image Processing

Millet is a traditional dry land crop in northern China,and the middle and upper reaches of the Yellow River are the main cultivation areas.High quality seed is the premise of high agricultural yield and the guarantee of agricultural economic growth.In order to save resources and promote the high yield of millet,the germination rate of millet seeds was taken as the starting point.The manual counting method used in the past is inefficient and consumes manpower and material resources.Therefore,in this paper,the image processing method is adopted to analyze and process the germination image of millet seeds,according to which the number of germination and the length of radicle and germ of millet are obtained,so as to realize the automatic monitoring and nondestructive measurement of the germination of millet seeds.In order to obtain the image data of millet germination,an image acquisition system was built.According to the algorithm proposed in this paper,the design of seed germination rate recognition and radicle and germ length measurement based on image processing is completed.The feasibility of solving this problem based on image processing is demonstrated through experiments.At last,the user interface of seed germination rate recognition and radicle embryo length measurement system was developed by pyqt.In the image processing part,the main research contents are as follows:(1)In order to facilitate the later correction of the collected photos and the convenience of the subsequent operation,the author uses OpenCV to calibrate the camera for collecting seed germination pictures,and obtains the camera related parameters and other information,so as to reduce the error.(2)Through image preprocessing,the image is grayed,binarized,noise eliminated and several methods are compared,and then the method suitable for this study is selected for processing.(3)On the basis of image preprocessing,the contour of each seed is extracted,and then the ROI image of each seed is obtained.(4)Using HSV color model to extract color features and other operations,the image of radicle and germ was obtained.The final image is obtained by skeleton extraction algorithm.The germination rate of seeds and the real length of radicle and germ were determined by image characteristics.