平面视觉图像舒适色度范围测量算法研究

Research on Measurement Algorithm of Comfortable Chroma Range of Plane Vision Image

针对传统方法处理下图像舒适度范围测量结果不够精准,不符合人眼视觉需求的问题,提出平面视觉图像舒适色度范围测量算法。设计图像采集模块提取图像信息,在图像信息中将图像颜色的亮度分量与色度分量分离,根据对比敏感度对亮度分量和色度分量的结构相似度进行加权,得出图像结构相似度。通过约束条件获取全局匹配最小误差,满足平滑性假设,从而提高误差估计精度。利用Mean Shift算法分割图像,对分割区域实施全局匹配,得到新的视差图。根据色度分割提高弱纹理区域视差估计质量,去除图像边缘的模糊效应,获得视差序列,实现最佳色泽感观范围测量。研究结果表明,所提算法视差估计结果更加准确,测量范围误差小,舒适色度的量化范围更符合人眼视觉需求。

In order to solve the problem that the measurement results of the comfort range of the image processed by traditional methods are not accurate enough and can not meet the needs of human vision,an algorithm for measuring the comfort chroma range of plane vision images is proposed.The image acquisition module was designed to extract the image information,and the luminance component and chroma component of the image color were separated in the image information,and the structural similarity of the luminance component and the chrominance component was weighted according to the contrast sensitivity to obtain the image structure similarity.The minimum error of global matching was obtained by constraint conditions,which satisfied the assumption of smoothness,thus improving the accuracy of error estimation.The mean shift algorithm was used to segment the image,and a new disparity map was obtained by global matching.According to the color segmentation,the quality of disparity estimation in weak texture region was improved,the fuzzy effect of image edge was removed,and the disparity sequence was obtained to realize the measurement of optimal color perception range.The results show that the disparity estimation results of the proposed algorithm are more accurate,the measurement range error is small,and the quantization range of comfortable chroma is more in line with human visual requirements.