机器视觉油炸鱼豆腐外观缺陷识别与剔除系统

摘要

设计了外观缺陷识别与剔除试验系统,利用产品图像的颜色特征和形状特征,可实现油炸鱼豆腐外观质量的在线检测、识别与自动剔除.在产品外观颜色检测中利用RGB颜色空间中的B分量阈值作为特征;在产品外形缺陷检测中采用目标面积和边缘形状矩形度实现特征识别.经排序机构将产品有序排列后,根据各列缺陷识别结果,由单片机控制剔除装置完成相应动作,剔除不合格品.实验表明:该试验系统可有效完成油炸鱼豆腐产品的在线外观检测及残次品的剔除,提升生产线成套装备的技术水平.%A detecting and eliminating system to deal with appearance defects of fired-fish-bean-curd has been developed, which makes use of color and shape features to detect,recognize and eliminate products of bad appearance on line.It takes the B component threshold in RGB color space as the criterion for the color appearance of the product;the area and degree of similarity of the image with rectangle are used to distinguish defects in shape appearance.The defect recognition results of each line can be translated into special information code automatically after the products are sorted into order by the sorting mechanism,and then the elimination device will use SCM to reject unqualified products automatically.Experimental results show that this system can be applied to fried-fish-bean-curd appearance detection and elimination,and thus improve the technology and quality of the production line.