一种彩色图像质量客观评价方法

为了研究图像质量客观评价结果与主观质量评价结果不一致的问题，依据图像的灰度直方图作为统计特征可以应用于图像质量的客观评价理论，采用数据拟合方法，研究了彩色图像的基于分量理论和色彩量化理论的颜色直方图的图像质量客观评价方法，并分别给出了拟合直方图曲线的算法步骤，根据失真图像和参考图像的拟合系数之差的权值是否在可失真范围内给出图像质量的二元评价．实验中计算基于灰度直方图拟合的灰度图像的每像素的数据差值后，可知在小于等于5×10。时失真图像质量为优，该方法和数据也适用于失真彩色图像质量的二元评价．

A Kind of Method of Color Image Quality Assessment

In order to study the inconsistent problem of the objective image quality evaluation result and subjec- tive quality evaluation result, according to the theory of the image histogram as the statistical characteristics applied to image quality assessment, the data fitting approach is adopted and the color image quality objective assessment approaches of the color histogram based on the color components and the color quantization. The algorithm steps of fitting the histogram curve are differently listed. Then the two element evaluation is given according to the distortion range after the difference weight data of the fitting coefficients about distorted image and the reference image is cal- culated. The experiments show that the distorted image quality assessment is excellent when the difference weight data of the fitting coefficients about distorted image and the reference image is less than or equal to 5 x 10-5. Ac- cording to the approach and the data, the two element evaluation results of the distorted color images are also effec- tively given.