Research on recognition technology of Caragana korshinskii main stem used to keep stubble at a certain height after cutting

In order to meet the agronomic requirements of Caragana korshinskii for keeping stubble at a certain height after cutting it. This paper analyzes the color characteristics of Caragana korshinskii and withered grass, green grass, sky, twigs and other background noises in different color spaces. It is determined that the RGB color model with a large difference between the characteristic values of Caragana korshinskii and the background is used to process the image. The image is grayed by extracting the R-value component and G-value component that have a large difference between the characteristic values of Caragana korshinskii and the background noise, and the image is also grayed by the weighted average method. By comparison, it is determined that the graying method of extracting G value component with better segmentation effect should be adopted. The use of histogram equalization makes the image gray level evenly distributed, reducing the influence of light on the brightness of the image. The optimal threshold for image binarization is determined, and the median filter is used to eliminate salt and pepper noise. Corrosion, removal of small areas, expansion and other operations are performed on the image, and the twigs and hollows on the main stem are eliminated and filled. Using the least square method and using the polyfit function, the repair and fitting of the main stems of Caragana korshinskii are realized. Threshold segmentation algorithm was used to identify the main stem of Caragana korshinskii, and the average recognition rate was 80.97%.