稀疏表达的夜间图像采集分辨率无关处理技术
|
摘要
在图像采集和三维重建中,需要进行图像分辨率无关处理,降低光照和环境因素给分辨率带来的影响,因此提出一种基于稀疏表达的夜间图像采集分辨率无关处理技术。首先对采集的夜间图像进行小波降噪处理,对夜间图像的阴影区域和亮度区域进行曲线分割。在夜间光照的多重色差条件下,用标准化稀疏先验的正则化表达方法进行图像均衡锁光处理,采用图像像素均匀遍历方法进行图像子块连续遍历,实现夜间采集图像的阴影偏差补偿,实现夜间图像采集的分辨率无关处理,解决图像分辨率带来的图像重构问题。仿真结果表明,采用该方法进行夜间图像采集分辨率无关处理,有效提高了图像的成像质量和可分辨度,峰值信噪比和计算开销的参量指标优于传统方法。
In the image acquisition and 3 D reconstruction, independent image resolution processing is needed to reduce the influence of illumination and environmental factors to the resolution. Therefore, an independent processing technology of night image acquisition resolution is proposed based on the sparse expression. First the acquisition of night image is in wavelet denoising, conducting the curve segmentation of the shadow region and brightness region for night image. In the conditions of multiple color at night, the expression method of regularization of standardized sparse prior is used for the image equalization lock light processing, and image pixel traversal method for continuous uniform traversal of image blocks, realizing the shadow deviation compensation for acquisition image at night and independent resolution processing of that, solving the problem of image reconstruction caused by the resolution. The simulation results show that it can improve the imaging quality and resolution and the its peak signal-to-noise ratio and computing cost parameters are better than the traditional method.
In the image acquisition and 3 D reconstruction, independent image resolution processing is needed to reduce the influence of illumination and environmental factors to the resolution. Therefore, an independent processing technology of night image acquisition resolution is proposed based on the sparse expression. First the acquisition of night image is in wavelet denoising, conducting the curve segmentation of the shadow region and brightness region for night image. In the conditions of multiple color at night, the expression method of regularization of standardized sparse prior is used for the image equalization lock light processing, and image pixel traversal method for continuous uniform traversal of image blocks, realizing the shadow deviation compensation for acquisition image at night and independent resolution processing of that, solving the problem of image reconstruction caused by the resolution. The simulation results show that it can improve the imaging quality and resolution and the its peak signal-to-noise ratio and computing cost parameters are better than the traditional method.
引文
[1]赵蓉,顾国华,杨蔚.基于偏振成像的可见光图像增强[J].激光技术,2016,40(2):227-231.
[2]于涛,胡炳樑,高晓惠,等.高光谱干涉图像动态追踪补偿方法研究[J].光子学报,2016,45(7):0710003.
[3]窦慧晶,王千龙,张雪.基于小波阈值去噪和共轭模糊函数的时频差联合估计算法[J].电子与信息学报,2016,38(5):1123-1128.
[4]张伟,张合,张祥金.小型大瞬时视场光学探测系统优化设计[J].红外与激光工程,2016,45(5):518002.
[5]蒋建国,金玉龙,齐美彬,等.基于稀疏表达残差的自然场景运动目标检测[J].电子学报,2015,43(9):1738-1744.
[6]周毅敏,李光耀.多重光照色差下图像平滑美化处理算法[J].计算机科学,2016,43(10):287-291.
[7]宋涛,李鸥,刘广怡.基于空时多线索融合的超像素运动目标检测方法[J].电子与信息学报,2016,38(6):1503-1511.
[8]Evangelio R H,Patzold M,Keller I,et al.Adaptively splitted GMM with feedback improvement for the task of background subtraction[J].IEEE Transactions on Information Forensics and Security,2014,9(5):863-874.
[9]Barnich O,Van D M.Vi Be:a universal background subtraction algorithm for video sequences[J].IEEE Transactions on Image Processing,2011,20(6):1709-1724.
[10]Liu Zhi,Zhang Xiang,Luo Shuhua,et al.Superpixel-based spatiotemporal saliency detection[J].IEEE Transactions on Circuits and Systems for Video Technology,2014,24(9):1522-1540.
[11]Bae S H,Yoon K J.Robust online multiobject tracking with data association and track management[J].IEEE Transactions on Image Processing,2014,23(7):2820-2833.
[12]Jiang X,Harishan K,Thamarasa R,et al.Integrated track initialization and maintenance in heavy clutter using probabilistic data association[J].Signal Processing,2014,94:241-250.
[13]卞乐,霍冠英,李庆武.基于Curvelet变换和多目标粒子群的混合熵MRI图像多阈值分割[J].计算机应用,2016,36(11):3188-3195.
[14]李积英,党建武,王阳萍.融合量子克隆进化与二维Tsallis熵的医学图像分割算法[J].计算机辅助设计与图形学学报,2014,26(3):465-471.
[15]Ortiz A,Gorriz J M,Ramirez J,et al.Improving MR brain image segmentation using self-organising maps and entropy-gradient clustering[J].Information Sciences,2014,262(3):117-136.
[2]于涛,胡炳樑,高晓惠,等.高光谱干涉图像动态追踪补偿方法研究[J].光子学报,2016,45(7):0710003.
[3]窦慧晶,王千龙,张雪.基于小波阈值去噪和共轭模糊函数的时频差联合估计算法[J].电子与信息学报,2016,38(5):1123-1128.
[4]张伟,张合,张祥金.小型大瞬时视场光学探测系统优化设计[J].红外与激光工程,2016,45(5):518002.
[5]蒋建国,金玉龙,齐美彬,等.基于稀疏表达残差的自然场景运动目标检测[J].电子学报,2015,43(9):1738-1744.
[6]周毅敏,李光耀.多重光照色差下图像平滑美化处理算法[J].计算机科学,2016,43(10):287-291.
[7]宋涛,李鸥,刘广怡.基于空时多线索融合的超像素运动目标检测方法[J].电子与信息学报,2016,38(6):1503-1511.
[8]Evangelio R H,Patzold M,Keller I,et al.Adaptively splitted GMM with feedback improvement for the task of background subtraction[J].IEEE Transactions on Information Forensics and Security,2014,9(5):863-874.
[9]Barnich O,Van D M.Vi Be:a universal background subtraction algorithm for video sequences[J].IEEE Transactions on Image Processing,2011,20(6):1709-1724.
[10]Liu Zhi,Zhang Xiang,Luo Shuhua,et al.Superpixel-based spatiotemporal saliency detection[J].IEEE Transactions on Circuits and Systems for Video Technology,2014,24(9):1522-1540.
[11]Bae S H,Yoon K J.Robust online multiobject tracking with data association and track management[J].IEEE Transactions on Image Processing,2014,23(7):2820-2833.
[12]Jiang X,Harishan K,Thamarasa R,et al.Integrated track initialization and maintenance in heavy clutter using probabilistic data association[J].Signal Processing,2014,94:241-250.
[13]卞乐,霍冠英,李庆武.基于Curvelet变换和多目标粒子群的混合熵MRI图像多阈值分割[J].计算机应用,2016,36(11):3188-3195.
[14]李积英,党建武,王阳萍.融合量子克隆进化与二维Tsallis熵的医学图像分割算法[J].计算机辅助设计与图形学学报,2014,26(3):465-471.
[15]Ortiz A,Gorriz J M,Ramirez J,et al.Improving MR brain image segmentation using self-organising maps and entropy-gradient clustering[J].Information Sciences,2014,262(3):117-136.