关键词: 农业工程, 机器视觉技术, 表面纹理检测, 梯度算法, 苹果

Abstract: A detection and classification technique for the apple surface texture based on the machine vision was proposed.The color images of apple were extracted and greyed using different algorithms.According to the feature that the surface textures locate in the sudden change areas of image grey value,the double grade sharpening was performed for the image using the typical grad algorithm.To make the texture more obvious,the data filtering based on the neighbor information was used to remove the image noise.The average grad of the image was calculated by the absolute difference algorithm.With the threshold of 55 selected,apples were divided into grad a with sharpcut texture and grad B with blurred texture.An automatic grading system composed of a PC computer,a programmable logic controller,a camera,and an image grabbing card was developed.41 crystal Fuji apples were graded automatically by the system with the grading error 4.9%, satisfying the requirement of application.

Key words: agricultural engineering, machine vision, surface texture detecting, grad algorithm, apple