摘要
为了解决工业带材生产过程中带材速度实时精确检测这一问题,设计了一种基于纹理图像功率谱的非接触式工业带材速度检测系统。首先利用线阵CCD采集带材图像的纹理图;然后对纹理图像进行滤波、二值化、腐蚀等预处理;最后对图像进行功率谱分析,利用霍夫变换求得带材图像功率谱分布方向,从而间接得到带材的速度。实验结果表明该方法检测工业带材速度的误差可以缩减到±0.03 m/s之内,能够有效、实时地测得工业带材的运动速度。
In order to realize real-time accurate detection on velocity of strip during production of the industrial strip, a new non-contact detection system based on the texture image power spectrum is designed in this paper. The system chooses linear array CCD to get the strip image and then puts pretreatments on the texture image such as image denoising, binarization and corrosion. Finally, the power spectrum of the image is analyzed and the direction of the power spectral distribution of the strip image is obtained by the use of Hough transform. Thus, the velocity of the strip can be worked out indirectly. The experimental results show that the proposed system can effectively measure the real-time strip velocity and make the deviation within ±0.03m/s.
出处
《现代电子技术》
2013年第13期31-35,共5页
Modern Electronics Technique
基金
陕西省教育厅产业化培育项目(2012JC19)
西安市技术转移促进工程重大项目资助(CX12166)
关键词
纹理图像功率谱
带材速度
霍夫变换
工业带材
texture image power spectrum
strip velocity
Hough transform
industrial strip
