通过研究非对称失真立体图像人眼视觉感知特性,提出了一种基于小波图像融合的非对称失真立体图像质量客观评价方法。首先,采用固定尺寸块匹配方法得到原始立体图像块视差值,接着,根据该视差值的大小,同时结合非对称失真立体图像视觉感知特性,对匹配块进行小波融合得到原始融合图像和失真融合图像。最后,通过计算原始融合图像和失真融合图像之间的方向投影变化距离来度量立体图像的失真程度。实验结果表明该模型与各类非对称失真立体图像的主观评价值具有较好的一致性,尤其在评价交叉失真立体图像时取得了较好的评价效果。
Through studying visual perception of asymmetric distortion of stereoscopic image, a quality assessment method for asymmetric distortion of stereoscopic image based on Wavelet Image Fusion (WIF) is proposed. Firstly, the block disparity value is computed by fixed block marching method. Then, the original and distorted fusion images are obtained by WIF according to original block disparity value and stereoscopic visual perception. Finally, the stereoscopic image quality is objectively assessed by calculating the distances between directional projection of original and distorted fusion images. Experimental results show that the proposed objective method achieves consistent stereoscopic image quality evaluation results with subjective assessment for various types of distortions, especially shows its effectiveness for assessing stereoscopic image with cross-distortion.