一种基于机器视觉技术的鸡蛋裂纹检测

A Detection for Cracked Eggs Based on Machine Vision

为了实现对鸡蛋壳裂纹的检测,提出了一种基于机器视觉技术的鸡蛋裂纹检测系统设计方案,该系统的硬件部分采用红色LED作为光源,选用背景光照射方式,在封闭的环境中通过CCD摄像机来获取单个静止鸡蛋的图像。软件部分采用灰度转换,中值滤波,线性锐化,阈值分割,边缘轮廓特征提取等手段实现图像的裂纹检测,最后基于图像的数学特征去除干扰因素,准确定位裂纹。实验结果表明系统对破损鸡蛋检测的准确率达到了93%。

To achieve the detection for cracked eggs,a design for a system to detect fissures on cracked eggs based on machine vi sion was presented.The system used a red LED for light source and acquired an image of a static egg by the CCD camera in the method of background light irradiation in a sealed environment.As for the software design,we utilized several specific methods such as gray-scale conversion,median filter,linear sharpening,image thresholding,selection basing on the edge-outline of char acters and so forth to achieve detection for cracked eggs by images.Finally,we excluded several interfering factors and located the fissure precisely by images’mathematical characteristics.The experimental result and data indicted that the system recognition ac curacy reached 93%.