A method for segmenting regions of grape bunch stems naturally placed by machine vision

Abstract

The invention discloses a method for segmenting areas of fruit stems of grape bunches naturally placed by machine vision. Firstly, a multi-degree-of-freedom bracket is used to construct a binocular vision system, and an irradiation mode of diffuse reflection light sources is designed to obtain images of grape bunches. Then, the channel decomposition of the image was carried out under the HSI model, a circular convolution kernel with a radius of 3 pixels was designed, the median filter was performed pixel by pixel along the grape bunch image, and the edge of the grape bunch image was sharpened by using the LOG kernel function. Then, starting from the distribution characteristics of the grape edge area, the distance of the grape edge is represented by the minimum distance between the non-connected domains in the minimum neighborhood, and the segmentation point of the grape stem edge and the fruit edge is obtained according to the distribution histogram of the edge distance, and Adaptive morphological open and close convolution kernels are designed by using edge distance segmentation points, and morphological operations are performed on images with adaptive convolution kernels. Ultimately, the high-precision and high-efficiency segmentation of the area where the grape bunch stems are naturally placed by machine vision is realized.

Description

technical field [0001] The invention relates to the field of image segmentation based on machine vision, in particular to a method for segmenting fruit stem regions based on machine vision and image processing for naturally placed grape bunches, which is used for automatic sorting by robots. Background technique [0002] In recent years, my country's fruit production has grown rapidly, and traditional manual sorting methods have been difficult to meet the needs of modern agricultural production. Automatic fruit sorting based on robotic technology is of great importance to the automation, scale, and precision development of agricultural production and agricultural product processing. significance. In the robot-based automatic fruit sorting process, the accurate positioning of the grasping point is the prerequisite for the robot to achieve accurate, fast and non-destructive grasping control. Due to the advantages of non-contact, strong applicability, and high cost performance,...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the irregular distribution of fruit stalks and fruit grains of naturally placed grape bunches, and the variety of grape bunch shapes, resulting in low precision and long time-consuming segmentation of the fruit stalk area in the image, the invention proposes a machine vision based on edge distance and automatic segmentation. Segmentation method of naturally placed grape bunch stem region based on morphology

Next, aiming at the problem that the existing morphological methods with fixed convolution kernels are difficult to achieve accurate segmentation of grape stems in naturally placed areas, a stem segmentation method that can adaptively select the size of the convolution kernel is proposed

Images