木门生产线中运动状态下条码定位与复原

Position and recovery of motion-state bar code in wooden door production line

针对木门生产线中对于运动状态下条码定位不准以及由于相对运动所产生的图像退化,提出一种解决方法。对于大范围中复杂木门背景条件下条码的定位,采用Canny算法对图像进行边缘提取,并对边缘检测结果通过Hough变换检测图中竖直直线。再将图像分条,并通过DBSCAN聚类算法对条码特征进行分条判断,将符合特征的区域提取出即为目标条码区域。对于生产线中采用CMOS相机拍摄运动物体产生的运动模糊的图像退化,通过结合生产线运行速度及方向与相机参数计算出系统的退化模型,再采用维纳滤波对运动退化条码图像进行复原。通过图像处理能使生产线在不改变硬件条件下达到对于运动状态下木门条码的有效定位与复原。能准确定位木门背景下小占比的多个条码区域,并能对定位出的条码区域高效、快速的复原。经过实验与生产实践证明该算法经济、简单、有效。

In this paper, a solution is proposed for the bar code positioning in wooden door production line and the image degradation caused by relative motion.For the positioning of the barcode under the background of complex wooden door in a large range, the Canny algorithm is used to extract the edge of the image, and the edge detection results are detected by Hough transform to detect the vertical straight line.Then, the image is segmented, and the bar code characteristics are judged by the DBSCAN clustering algorithm, which is the target bar code region.For resulting in the production line adopts CMOS cameras moving object motion blur image degradation, by combining the production line running speed and direction and the camera parameters to calculate the degradation model of the system, then using wiener filtering for sports degradation barcode image restoration.Through image processing, the production line can effectively locate and recover the wooden door barcode under the condition of no change of hardware.It can accurately locate multiple barcode areas of small proportion in the wooden door background, and can effectively and quickly recover the barcode area located.The experiment and production practice prove that the algorithm is economical, simple and effective.