Machine-harvested seed cotton impurity rate detection method and system

The invention provides a machine-harvested seed cotton impurity rate detection method. The method comprises the following steps: acquiring a saturation channel image of a machine-harvested seed cotton impurity image; carrying out image segmentation by using a truncation segmentation method and then carrying out edge detection by using an improved Canny operator; performing morphological closing operation on the image after edge detection to form a closed area, and performing hole filling on the formed closed area; for the image with impurities intersecting with the edge of the image, performing gray value addition on the image after noise reduction and the image after hole filling, determining a segmentation threshold through a gray histogram, and then performing binarization operation on the gray image; and outputting the processed image to a data processing thread for solving the real-time impurity rate and the average impurity rate, and outputting the image. By means of the method, the condition of the imp