感知特征互补的图像质量评价
|
摘要
针对图像质量客观评价方法在实际应用场景下性能退化的问题,将人眼视觉特性融入图像特征处理的多个环节,提出一种融合视觉结构显著和视觉能量显著特征互补的方法。首先,根据人眼特性对图像的灰度能量、对比度能量和梯度结构三层互补特征进行空域-频域联合变换处理;其次,分别提取前述三层视觉特征的多通道信息并进行评价;最后,基于视觉特性和图像失真度将各层视觉特征评价从内层至外层逐步自适应综合。实验表明,本方法具有较高的水平和更好的稳定性,提高了实际应用场景下的评价性能。
Aiming at the performance degradation of objective methods for image quality assessment in practical application scenar-ios, a visual saliency and complementary features method based on pooling of structure and energy by integrating human visual characteristics into many parts of image feature processing is proposed. Firstly, the three complementary features of image gray ener-gy, contrast energy and gradient structure are processed based on spatial-frequency joint transformation, according to the human eye characteristics. Secondly, multichannel information of the above three layers of visual feature is extracted and assessed, respectively.Finally, the visual feature assessment of each layer is adaptively pooled from the inner layer to the outer layer based on visual characteristics and image distortion. The experiments show that the proposed method holds higher level, better stability, and assess-ment performance is improved in practical application scenarios.
Aiming at the performance degradation of objective methods for image quality assessment in practical application scenar-ios, a visual saliency and complementary features method based on pooling of structure and energy by integrating human visual characteristics into many parts of image feature processing is proposed. Firstly, the three complementary features of image gray ener-gy, contrast energy and gradient structure are processed based on spatial-frequency joint transformation, according to the human eye characteristics. Secondly, multichannel information of the above three layers of visual feature is extracted and assessed, respectively.Finally, the visual feature assessment of each layer is adaptively pooled from the inner layer to the outer layer based on visual characteristics and image distortion. The experiments show that the proposed method holds higher level, better stability, and assess-ment performance is improved in practical application scenarios.
引文
[1]李先友,赵曙光,段永成,等.基于FPGA的实时MIPICSI-2图像采集与处理系统[J].电子技术应用,2019,45(1):97-100.
[2]吴晓元,常海涛,苟军年.Faster R-CNN定位后的工业CT图像缺陷分割算法研究[J].电子技术应用,2019,45(1):76-80.
[3]陈文艺,张龙,杨辉.HDR图像色调映射的自适应色彩调节算法[J].电子技术应用,2018,44(11):107-110.
[4]ZHANG Y C,KING N N,MA L,et al.Objective quality assessment of image retargeting by incorporating fidelity measures and inconsistency detection[J].IEEE Transactions on Image Processing,2017,26(11):5980-5993.
[5]WANG Z,BOVIK A C,SHEIKH H R,et al.Image quality assessment:from error visibility to structural similarity[J].IEEE Transactions on Image Processing,2004,13(4):600-612.
[6]GUO L,CHEN W L,LIAO Y,et al.Multi-scale structural image quality assessment based on two-stage low-level features[J].Computers and Electrical Engineering,2014,40(4):1101-1110.
[7]LIU A,LIN W,NARWARIA M.Image quality assessment based on gradient similarity[J].IEEE Transaction on Image Processing,2012,21(4):1500-1512.
[8]ZHANG L,ZHANG L,MOU X Q,et al.FSIM:a feature similarity index for image quality assessment[J].IEEETransaction on Image Processing,2011,20(8):2378-2386.
[9]ZHANG L,SHEN Y,LI H Y.VSI:a visual saliency-induced index for perceptual image quality assessment[J].IEEETransaction on Image Processing,2014,23(10):4270-4281
[10]丰明坤,王中鹏,叶绿.视觉稀疏化多通道多特征自适应的图像评价[J].仪器仪表学报,2016,37(3):667-674.
[11]丰明坤,赵生妹,邢超.基于视觉显著失真度的图像质量自适应评价方法[J].电子与信息学报,2015,37(9):2062-2068.
[12]丰明坤,赵生妹,施祥.视觉多通道梯度与低阶矩自适应图像评价[J].仪器仪表学报,2015,36(11):2531-2537.
[13]林志洁,丰明坤.深度视觉特征与策略互补融合的图像质量评价[J].模式识别与人工智能,2017,30(8):682-691.
[14]XUE W F,ZHANG L,MOU X Q,et al.Gradient magnitude similarity deviation:s highly efficient perceptual image quality index[J].IEEE Transactions on Image Processing,2014,23(2):684-695.
[15]ZHANG X D,FENG X C,WANG W W,et al.Edge strength similarity for image quality assessment[J].IEEESignal Processing Letters,2013,20(4):319-322.
[16]WANG T H,JIA H Z,SHU H Z.Full reference image quality assessment algorithm based on gradient magnitude and histogram of oriented gradient[J].Journal of Southeast University,2018,48(2):276-281.
[2]吴晓元,常海涛,苟军年.Faster R-CNN定位后的工业CT图像缺陷分割算法研究[J].电子技术应用,2019,45(1):76-80.
[3]陈文艺,张龙,杨辉.HDR图像色调映射的自适应色彩调节算法[J].电子技术应用,2018,44(11):107-110.
[4]ZHANG Y C,KING N N,MA L,et al.Objective quality assessment of image retargeting by incorporating fidelity measures and inconsistency detection[J].IEEE Transactions on Image Processing,2017,26(11):5980-5993.
[5]WANG Z,BOVIK A C,SHEIKH H R,et al.Image quality assessment:from error visibility to structural similarity[J].IEEE Transactions on Image Processing,2004,13(4):600-612.
[6]GUO L,CHEN W L,LIAO Y,et al.Multi-scale structural image quality assessment based on two-stage low-level features[J].Computers and Electrical Engineering,2014,40(4):1101-1110.
[7]LIU A,LIN W,NARWARIA M.Image quality assessment based on gradient similarity[J].IEEE Transaction on Image Processing,2012,21(4):1500-1512.
[8]ZHANG L,ZHANG L,MOU X Q,et al.FSIM:a feature similarity index for image quality assessment[J].IEEETransaction on Image Processing,2011,20(8):2378-2386.
[9]ZHANG L,SHEN Y,LI H Y.VSI:a visual saliency-induced index for perceptual image quality assessment[J].IEEETransaction on Image Processing,2014,23(10):4270-4281
[10]丰明坤,王中鹏,叶绿.视觉稀疏化多通道多特征自适应的图像评价[J].仪器仪表学报,2016,37(3):667-674.
[11]丰明坤,赵生妹,邢超.基于视觉显著失真度的图像质量自适应评价方法[J].电子与信息学报,2015,37(9):2062-2068.
[12]丰明坤,赵生妹,施祥.视觉多通道梯度与低阶矩自适应图像评价[J].仪器仪表学报,2015,36(11):2531-2537.
[13]林志洁,丰明坤.深度视觉特征与策略互补融合的图像质量评价[J].模式识别与人工智能,2017,30(8):682-691.
[14]XUE W F,ZHANG L,MOU X Q,et al.Gradient magnitude similarity deviation:s highly efficient perceptual image quality index[J].IEEE Transactions on Image Processing,2014,23(2):684-695.
[15]ZHANG X D,FENG X C,WANG W W,et al.Edge strength similarity for image quality assessment[J].IEEESignal Processing Letters,2013,20(4):319-322.
[16]WANG T H,JIA H Z,SHU H Z.Full reference image quality assessment algorithm based on gradient magnitude and histogram of oriented gradient[J].Journal of Southeast University,2018,48(2):276-281.