Research on Method of Highlight Elimination Based on Color Space Conversion in Metal Images

摘要

In the process of three-dimensional detection with computer vision, there is a highlighted phenomenon in metal artifact surface. It leads to the interference problem in subsequent detection. To solve this question, a method of image highlight elimination is proposed which is based on color space conversion technology. Firstly, the RGB color space is converted into the Y'CbCr color space, the luminance value is adjusted, and then, the highlighted part of the image with luminance adjustment polynomial that obtained using cubic spline interpolation is disposed. Finally, the removed highlighted image in the RGB color space is obtained. Experimental results show that this method can not only effectively remove the highlighted metal in the image, but also have a better performance on preserving the texture information of the original image.