摘要
针对工业产品手工测量的缺陷,提出一种基于计算机视觉技术的测量方法。根据需要,通过摄像头从不同角度定时对生产线上的产品进行拍照,首先对获得的图像进行去噪、平滑等预处理,然后对其进行基于链码表和线段表的轮廓跟踪,得到图像各个封闭区域的边界信息和区域信息,从而计算产品的面积、周长、形状因子、重心、外接矩形等几何特征参数。试验表明,该方法测试准确率为100%,适合实时检测,为产品检测的全自动化研究提供了基础。
Aiming at the drawbacks of traditional manual measurement of industrial product,this paper put forward an approach of measurement based on computer vision technology.According to the requirements,the camera timing aims at the production line to take photographs of the products from different angles.Firstly,we did some pre-treatments such as de-nos ing, smoothing and so on,and then tracked the outline of the image based on the chain codes table and line segment table to get border and region information of all the closed regions,thus computed the geometrical shape feature parameters of the image,such as area, perimeter, shape ~hetor, the center of gravity and cireum-rectangular.Experiments show that precision of this method is 100%.It is suitable for real-time detection and provides the basic for further research on the fully-automated detection of products.
出处
《电子设计工程》
2010年第3期6-8,共3页
Electronic Design Engineering
关键词
图像处理
产品检测
特征参数
链码表
线段表
image processing
products detection
feature parameters
chain codes table
line segment table
