四元数小波变换优化单目图的无参考立体图像质量评价  

No-Reference Stereo Image Quality Assessment of Cyclopean Images Optimized Using Quaternion Wavelet Transform

作　　者：李一凡 李朝锋 桑庆兵[1] Li Yifan;Li Chaofeng;Sang Qingbing(College of Internet of Things Engineering, Jiangnan University, Wuxi, Jiangsu 214166, China;Institute of Logistics Science & Engineering, Shanghai Maritime University, Shanghai 200135, China)

机构地区：[1]江南大学物联网工程学院,江苏无锡214166 [2]上海海事大学物流科学与工程研究院,上海200135

出　　处：《激光与光电子学进展》2019年第18期135-143,共9页Laser & Optoelectronics Progress

基　　金：国家自然科学基金（61771223）;江苏省自然科学基金(BK20171142)

摘　　要：通过四元数小波变换分解立体图像的左右视图,获取不同尺度不同方向的幅值相位信息,并结合人眼视觉特性生成单目图;对左右视图和单目图作亮度去均值对比度归一化(MSCN)处理,获取MSCN系数图,采用广义高斯分布模型拟合MSCN系数和MSCN四方向邻域系数乘积,提取统计参数特征(联合峰度、偏度、标准偏差和能量),组成特征向量,通过XGBoost模型预测图像质量感知得分。结果表明,所提立体图像质量评价算法在LIVE3D图像库上优于其他方法,并且运行速度得到大幅度提高。First, the left and right views of stereo images are decomposed by quaternion wavelet transform to obtain the amplitude and phase information of different scales and directions, and then cyclopean images are generated by combining human visual characteristics. The left views, right views, and cyclopean images are processed via mean subtracted contrast normalization (MSCN). The MSCN coefficient map is then obtained. The MSCN coefficient and the product of MSCN four-direction neighborhood coefficients are fitted via a generalized Gauss distribution model to extract statistical parameter features. The feature vectors are formed by combining the kurtosis, skewness, standard deviation, and energy. The image quality perception score is then predicted using the XGBoost model. Experimental results show that the proposed stereo image quality assessment algorithm is superior to other reported methods in the LIVE 3D image database and it greatly improves the running speed.

关 键 词：图像处理 立体图像质量评价 四元数小波变换 单目图 去均值对比度归一化 XGBoost 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]