基于深度卷积神经网络的低照度图像增强
|
摘要
针对低照度条件下图像降质严重的问题,提出了一种基于深度卷积神经网络(DCNN)的低照度图像增强算法。该算法根据Retinex模型合成训练样本,将原始低照度图像从RGB (Red Green Blue)空间转换到HSI (Hue Saturation Intensity)颜色空间,保持色度分量和饱和度分量不变,利用DCNN对亮度分量进行增强,最后将HSI颜色空间转换到RGB空间,得到最终的增强图像。实验结果表明,与现有主流的图像增强算法相比,所提算法不仅能够有效提升亮度和对比度,改善过增强现象,而且能够避免色彩失真,主观视觉和客观评价指标均得到了进一步提高。
Aiming at the problem of the severe image degradation under a low-light condition, a low-light image enhancement algorithm based on deep convolutional neural network(DCNN) is proposed. The training sample is synthesized by this algorithm according to the Retinex model. Then, the original low-light image is converted from RGB(Red Green Blue) space to HSI(Hue Saturation Intensity) color space. The luminance component is enhanced by using the DCNN while keeping the chrominance component and the saturation component unchanged. Finally, the image is turned back to the RGB space from HSI color space to get the finally enhanced image. The experimental results show that, compared with the existing excellent image enhancement algorithms, the proposed algorithm can not only effectively enhance the brightness and the contrast, but also can avoid the color distortion and the over-enhancement, and both the subjective vision and objective evaluation index are further improved.
Aiming at the problem of the severe image degradation under a low-light condition, a low-light image enhancement algorithm based on deep convolutional neural network(DCNN) is proposed. The training sample is synthesized by this algorithm according to the Retinex model. Then, the original low-light image is converted from RGB(Red Green Blue) space to HSI(Hue Saturation Intensity) color space. The luminance component is enhanced by using the DCNN while keeping the chrominance component and the saturation component unchanged. Finally, the image is turned back to the RGB space from HSI color space to get the finally enhanced image. The experimental results show that, compared with the existing excellent image enhancement algorithms, the proposed algorithm can not only effectively enhance the brightness and the contrast, but also can avoid the color distortion and the over-enhancement, and both the subjective vision and objective evaluation index are further improved.
引文
[1] Li Q Z, Liu Q. Adaptive enhancement algorithm for low illumination images based on wavelet transform[J]. Chinese Journal of Lasers, 2015, 42(2): 0209001. 李庆忠, 刘清. 基于小波变换的低照度图像自适应增强算法[J]. 中国激光, 2015, 42(2): 0209001
[2] Fei Y J, Shao F. Contrast adjustment based on image retrieval[J]. Laser & Optoelectronics Progress, 2018, 55(5): 051002. 费延佳, 邵枫. 基于图像检索的对比度调整[J]. 激光与光电子学进展, 2018, 55(5): 051002.
[3] Land E H. The retinex theory of color vision[J]. Scientific American, 1977, 237(6): 108-128.
[4] Zhang J, Zhou P C, Zhang Q. Low-light image enhancement based on iterative multi-scale guided filter Retinex[J]. Journal of Graphics, 2018, 39(1): 1-11. 张杰, 周浦城, 张谦. 基于迭代多尺度引导滤波Retinex的低照度图像增强[J]. 图学学报, 2018, 39(1): 1-11.
[5] Jobson D J, Rahman Z,Woodell G A. Properties and performance of a center/surround retinex[J]. IEEE Transactions on Image Processing, 1997, 6(3): 451-462.
[6] Lin H, Shi Z.Multi-scale retinex improvement for nighttime image enhancement[J]. Optik-International Journal for Light and Electron Optics, 2014, 125(24): 7143-7148.
[7] Chen Y, Zhan D, Liu H L. Enhancement algorithm for low-lighting images based on physical model and boundary constraint[J]. Journal of Electronics & Information Technology, 2017, 39(12): 2962-2969. 陈勇, 詹帝, 刘焕淋. 基于物理模型与边界约束的低照度图像增强算法[J]. 电子与信息学报, 2017, 39(12): 2962-2969.
[8] Fu X Y, Zeng D L, Huang Y, et al. A weighted variational model for simultaneous reflectance and illumination estimation[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 2782-2790.
[9] Guo X J, Li Y, Ling H B. LIME: low-light image enhancement via illumination map estimation[J]. IEEE Transactions on Image Processing, 2017, 26(2): 982-993.
[10] Dong X, Wang G, Pang Y, et al. Fast efficient algorithm for enhancement of low lighting video[C]. IEEE International Conference on Multimedia and Expo (ICME), 2011: 1-6.
[11] Li L, Wang R G, Wang W M, et al. A low-light image enhancement method for both denoising and contrast enlarging[C]. IEEE International Conference on Image Processing (ICIP), 2015: 3730-3734.
[12] Yang J, Jiang X W, Pan C H, et al. Enhancement of low light level images with coupled dictionary learning[C].International Conference on Pattern Recognition (ICPR), 2016: 751-756.
[13] Song R X, Li D, Wang X C. Low illumination image enhancement algorithm based on HSI color space[J]. Journal of Graphics, 2017, 38(2): 217-223. 宋瑞霞, 李达, 王小春. 基于HSI色彩空间的低照度图像增强算法[J]. 图学学报, 2017, 38(2): 217-223.
[14] Wu F, Kintak U. Low-light image enhancement algorithm based on HSI color space[C]. International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2017: 1-6.
[15] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[16] Xin P, Xu Y L, Tang H, et al. Fast airplane detection based on multi-layer feature fusion of fully convolutional networks[J]. Acta Optica Sinica, 2018, 38(3): 0315003. 辛鹏, 许悦雷, 唐红, 等. 全卷积网络多层特征融合的飞机快速检测[J]. 光学学报, 2018, 38(3): 0315003.
[17] Zhu M M, Xu Y L, Ma S P, et al. Airport detection method with improved region-based convolutional neural network[J]. Acta Optica Sinica, 2018, 38(7): 0728001. 朱明明, 许悦雷, 马时平, 等. 改进区域卷积神经网络的机场检测方法[J]. 光学学报, 2018, 38(7): 0728001.
[18] Lore K G,Akintayo A, Sarkar S. LLNet: a deep autoencoder approach to natural low-light image enhancement[J]. Pattern Recognition, 2017, 61: 650-662.
[19] Glorot X, Bordes A, Bengio Y. Deep sparse rectifier neural networks[C]. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, 2011: 315-323.
[20] Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]. Proceedings of the 32nd International Conference on International Conference on Machine Learning, 2015: 448-456.
[21] Chen X Y, Wang S A. Superpixel segmentation based on Delaunay triangulation[C]. International Conference on Mechatronics and Machine Vision in Practice (M2VIP), 2016: 1-6.
[22] Zhang K, Zuo W, Chen Y, et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoisin[J]. IEEE Transactions on Image Processing, 2017, 26(7): 3142-3155.
[23] He K M, Zhang X Y, Ren S Q, et al. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification[C]. IEEE International Conference on Computer Vision (ICCV), 2015: 1026-1034.
[24] Dong C,Loy C C, He K M, et al. Image super-resolution using deep convolutional networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2): 295-307.
[25] Cheng H D, Shi X J. A simple and effective histogram equalization approach to image enhancement[J]. Digital Signal Processing, 2004, 14(2): 158-170.
[26] Sheikh H R, Sabir M F,Bovik A C. A statistical evaluation of recent full reference image quality assessment algorithm[J]. IEEE Transactions on Image Processing, 2006, 15(11): 3440-3451.
[27] Wang S H, Zheng J, Hu H M, et al. Naturalness preserved enhancement algorithm for non-uniform illumination images[J]. IEEE Transactions on Image Processing, 2013, 22(9): 3538-3548.
[28] Rahman Z U, Jobson D J, Woodell G A. Investigating the relationship between image enhancement and image compression in the context of the multi-scale retinex[J]. Journal of Visual Communication and Image Representation, 2011, 22(3): 237-250.
[29] Lee C, Lee C, Kim C S. Contrast enhancement based on layered difference representation[C]. IEEE International Conference on Image Processing, 2012: 965-968.
[30] Vonikakis V, Andreadis I. Fast automatic compensation of under/over- exposured image regions[C]. IEEE Pacific Rim Symposium on Image Video and Technology (PSIVT), 2007: 510-521.
[31] Chow L S, Rajagopal H, Paramesran R. Correlation between subjective and objective assessment of magnetic resonance (MR) images[J]. Magnetic Resonance Imaging, 2016, 34(6): 820-831.
[2] Fei Y J, Shao F. Contrast adjustment based on image retrieval[J]. Laser & Optoelectronics Progress, 2018, 55(5): 051002. 费延佳, 邵枫. 基于图像检索的对比度调整[J]. 激光与光电子学进展, 2018, 55(5): 051002.
[3] Land E H. The retinex theory of color vision[J]. Scientific American, 1977, 237(6): 108-128.
[4] Zhang J, Zhou P C, Zhang Q. Low-light image enhancement based on iterative multi-scale guided filter Retinex[J]. Journal of Graphics, 2018, 39(1): 1-11. 张杰, 周浦城, 张谦. 基于迭代多尺度引导滤波Retinex的低照度图像增强[J]. 图学学报, 2018, 39(1): 1-11.
[5] Jobson D J, Rahman Z,Woodell G A. Properties and performance of a center/surround retinex[J]. IEEE Transactions on Image Processing, 1997, 6(3): 451-462.
[6] Lin H, Shi Z.Multi-scale retinex improvement for nighttime image enhancement[J]. Optik-International Journal for Light and Electron Optics, 2014, 125(24): 7143-7148.
[7] Chen Y, Zhan D, Liu H L. Enhancement algorithm for low-lighting images based on physical model and boundary constraint[J]. Journal of Electronics & Information Technology, 2017, 39(12): 2962-2969. 陈勇, 詹帝, 刘焕淋. 基于物理模型与边界约束的低照度图像增强算法[J]. 电子与信息学报, 2017, 39(12): 2962-2969.
[8] Fu X Y, Zeng D L, Huang Y, et al. A weighted variational model for simultaneous reflectance and illumination estimation[C]. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016: 2782-2790.
[9] Guo X J, Li Y, Ling H B. LIME: low-light image enhancement via illumination map estimation[J]. IEEE Transactions on Image Processing, 2017, 26(2): 982-993.
[10] Dong X, Wang G, Pang Y, et al. Fast efficient algorithm for enhancement of low lighting video[C]. IEEE International Conference on Multimedia and Expo (ICME), 2011: 1-6.
[11] Li L, Wang R G, Wang W M, et al. A low-light image enhancement method for both denoising and contrast enlarging[C]. IEEE International Conference on Image Processing (ICIP), 2015: 3730-3734.
[12] Yang J, Jiang X W, Pan C H, et al. Enhancement of low light level images with coupled dictionary learning[C].International Conference on Pattern Recognition (ICPR), 2016: 751-756.
[13] Song R X, Li D, Wang X C. Low illumination image enhancement algorithm based on HSI color space[J]. Journal of Graphics, 2017, 38(2): 217-223. 宋瑞霞, 李达, 王小春. 基于HSI色彩空间的低照度图像增强算法[J]. 图学学报, 2017, 38(2): 217-223.
[14] Wu F, Kintak U. Low-light image enhancement algorithm based on HSI color space[C]. International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2017: 1-6.
[15] Krizhevsky A, Sutskever I, Hinton G E. ImageNet classification with deep convolutional neural networks[J]. Communications of the ACM, 2017, 60(6): 84-90.
[16] Xin P, Xu Y L, Tang H, et al. Fast airplane detection based on multi-layer feature fusion of fully convolutional networks[J]. Acta Optica Sinica, 2018, 38(3): 0315003. 辛鹏, 许悦雷, 唐红, 等. 全卷积网络多层特征融合的飞机快速检测[J]. 光学学报, 2018, 38(3): 0315003.
[17] Zhu M M, Xu Y L, Ma S P, et al. Airport detection method with improved region-based convolutional neural network[J]. Acta Optica Sinica, 2018, 38(7): 0728001. 朱明明, 许悦雷, 马时平, 等. 改进区域卷积神经网络的机场检测方法[J]. 光学学报, 2018, 38(7): 0728001.
[18] Lore K G,Akintayo A, Sarkar S. LLNet: a deep autoencoder approach to natural low-light image enhancement[J]. Pattern Recognition, 2017, 61: 650-662.
[19] Glorot X, Bordes A, Bengio Y. Deep sparse rectifier neural networks[C]. Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, 2011: 315-323.
[20] Ioffe S, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift[C]. Proceedings of the 32nd International Conference on International Conference on Machine Learning, 2015: 448-456.
[21] Chen X Y, Wang S A. Superpixel segmentation based on Delaunay triangulation[C]. International Conference on Mechatronics and Machine Vision in Practice (M2VIP), 2016: 1-6.
[22] Zhang K, Zuo W, Chen Y, et al. Beyond a Gaussian denoiser: residual learning of deep CNN for image denoisin[J]. IEEE Transactions on Image Processing, 2017, 26(7): 3142-3155.
[23] He K M, Zhang X Y, Ren S Q, et al. Delving deep into rectifiers: surpassing human-level performance on ImageNet classification[C]. IEEE International Conference on Computer Vision (ICCV), 2015: 1026-1034.
[24] Dong C,Loy C C, He K M, et al. Image super-resolution using deep convolutional networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2016, 38(2): 295-307.
[25] Cheng H D, Shi X J. A simple and effective histogram equalization approach to image enhancement[J]. Digital Signal Processing, 2004, 14(2): 158-170.
[26] Sheikh H R, Sabir M F,Bovik A C. A statistical evaluation of recent full reference image quality assessment algorithm[J]. IEEE Transactions on Image Processing, 2006, 15(11): 3440-3451.
[27] Wang S H, Zheng J, Hu H M, et al. Naturalness preserved enhancement algorithm for non-uniform illumination images[J]. IEEE Transactions on Image Processing, 2013, 22(9): 3538-3548.
[28] Rahman Z U, Jobson D J, Woodell G A. Investigating the relationship between image enhancement and image compression in the context of the multi-scale retinex[J]. Journal of Visual Communication and Image Representation, 2011, 22(3): 237-250.
[29] Lee C, Lee C, Kim C S. Contrast enhancement based on layered difference representation[C]. IEEE International Conference on Image Processing, 2012: 965-968.
[30] Vonikakis V, Andreadis I. Fast automatic compensation of under/over- exposured image regions[C]. IEEE Pacific Rim Symposium on Image Video and Technology (PSIVT), 2007: 510-521.
[31] Chow L S, Rajagopal H, Paramesran R. Correlation between subjective and objective assessment of magnetic resonance (MR) images[J]. Magnetic Resonance Imaging, 2016, 34(6): 820-831.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |