基于计算机视觉的图像采集对水果分级准确率的影响

Effest of Picture Gathering Based on Computer Vision on Fruit Grading Accuracy Rate

介绍了基于计算机视觉技术的水果分级装置和摄像系统现状与仿球形水果图像采集过程;指出了目前分级过程中水果表面信息漏采的现象;根据视场内水果、相机镜头及水果图像三者的关系和水果的直线运动与翻滚运动特性,分析了影响分级机分级准确率的水果图像拖尾和水果翻滚误差及其原因,并给出了图像拖尾和翻滚累计误差的计算公式;提出了解决因水果图像拖尾影响分级准确率的方案和水果翻滚运动跟踪的变周期图像采集系统。

This paper introduced the present situation of the fruit grading and the gathering process of the sphere-like fruit picture based on the computer vision technology; it pointed out the phenomenon of the missing gathering of the the fruit surface information in the present grading process. According to the relations of the fruit, the camera lens and the fruit picture and the characteristics of the fruit translation and the tumbling state of motion inside the field of view, we analyzed the reasons influencing sorting machine grading accuracy rate, fruit picture trailing and the fruit tumbling error, which has produced the picture trailing and the tumbling accumulative error formula. We also proposed the solution to the fruit picture trailing and the tumbling error affecting the grading accuracy rate and the changing cycle picture gathering system of the fruit tumbling movement track.