基于机器视觉的鲜枣群体大小检测算法

目前,鲜枣的分级一般都是由人工测量分级的,效率低,且需要大量劳动力。该文通过建立鲜枣群体大小检测软硬件平台,利用黑白校正获取较清晰的鲜枣目标图像,采用HSV加权法灰度化得到轮廓鲜明的鲜枣图像,采取图像处理技术对鲜枣的轮廓进行弧段组合及弧段分割,获取单个鲜枣的整体边缘,最后通过最小二乘法椭圆拟合得到鲜枣的直径大小。实验表明,该方法能很好地检测每个鲜枣的大小,检测的准确率达到98.91%。

Fresh jujubes group size detection algorithm based on machine vision

Nowadays, the method of fresh jujubes grading is usually manual measurement classifica-tion, which is slowly, and require a lot of labor. This study detected the fresh jujubes group size through CCD camera and some image processing method. The results showed that the CCD camera and the image processing method can detect the fresh jujubes group size. And the accuracy of the system reaches 98. 91%, while the system is low cost, fast efficiency and other characteristics.