论文部分内容阅读
与现有的带材速度检测方法不同，本文提出了一种新的基于纹理图像功率谱的带材速度检测方法，该方法利用霍夫变换求得带材图像功率谱的分布方向，从而间接得到带材的速度。该方法首先对采集到的图像进行二值化，以此根据被测物体运动速度的不同而显现出具有不同方向趋势的纹理图案；然后对二值化后的图像进行去噪、腐蚀，以此得到更高纹理质量的带材图像；最后对带材图像进行功率谱分析，并使用霍夫变换求得移动角度，从而计算出物体的运动速度。本文对多个被测物体进行测试，实验结果表明该方法能够有效地测得带材的运动速度，并且在带材表面横向纹理周期性较强的情况下，该方法的效果要明显优于相关测速法。“,”In the proposed method,different from other measure velocity of the strip method,this paper puts forward a new approach to measure velocity of the strip based on the power spectrum of texture image. The method uses hough transform to obtain the direction of the power spectral distribution of the strip image,thus obtains the velocity of the strip. Firstly, the strip image is processed by binarization and shows the different direction trends of texture image;Secondly,the binary image is used by denoising and corrosion,in order to obtain the higher quality of texture image;Finally,analysed power spectrum of the strip image , though the Hough transform obtained moving angle, in order to calculate the velocity of the object. The experimental results for different measured objects show the proposed method can be effectively measured the velocity of strip, and in the situation of horizontal strip surface texture cyclical, the effect of the method is significantly better than the relevant velocimetry.