基于改进的区域积神经网络和联合双边滤波的图像着色法

为了解决传统的卷积神经网络着色方法带来语境混淆、边缘模糊和细节信息丢失度高等问题,设计了一个改进的基于密集神经网络的区域全卷积神经网络(R-FCN)和基于局部特征网络的双分支神经网络模型。通过DenseNet可准确提取信息,产生易于训练和高参数效率的密集模型,采用全连接的条件随机场优化分割结果来提高分割的准确率。验证阶段采用联合双边滤波对图像进行处理,弥补图像边缘模糊的缺点。实验结果表明:与现有着色方法相比,该方法有效地解决了细节丢失度高、颜色不饱和及边缘模糊的问题,能够产生更真实、更合理的彩色图像,取得了优异的效果。

Abstract

In this study, we design an improved regional full convolutional neural network (R-FCN) based on dense neural networks and a two-branch neural network model based on the local feature network to solve the problems of context confusion, edge blur, and high loss of detail information associated with the traditional convolutional neural network coloring method. DenseNet is used to accurately extract information and create a dense model, which can be easily trained and exhibits high parameter efficiency, and fully connected conditional random fields is used to optimize the segmentation results for improving the segmentation accuracy. During the verification stage, joint bilateral filtering is used to process the image to solve the disadvantages associated with image edge blurring. Experiments results show that, compared with existing coloring methods, the proposed method can effectively solve the problems of high detail loss, color unsaturation, and edge blur, produce more realistic and reasonable color images, and achieve excellent results.

参考文献

[1] . Colorization using neural network ensemble[J]. IEEE Transactions on Image Processing, 2017, 26(11): 5491-5505.

[2] . Colorization using optimization[J]. ACM Transactions on Graphics, 2004, 23(3): 689-691.

[3] . Fast image and video colorization using chrominance blending[J]. IEEE Transactions on Image Processing, 2006, 15(5): 1120-1129.

[4] LuanQ, WenF, DanielC, et al. Natural image colorization[C]∥Proceedings of the 18th Eurographics conference on Rendering Techniques. [S.l.: s.n.], 2007: 309- 320.

[5] . Transferring color to greyscale images[J]. ACM Transactions on Graphics (TOG), 2002, 21(3): 277-280.

[6] Cheng ZZ, Yang QX, ShengB. Deep colorization[C]∥2015 IEEE International Conference on Computer Vision (ICCV), December 7-13, 2015, Santiago, Chile. New York: IEEE, 2015: 415- 423.

[7] . Let there be color!: joint end-to-end learning of global and local image priors for automatic image colorization with simultaneous classification[J]. ACM Transactions on Graphics, 2016, 35(4): 110.

[8] LarssonG, MaireM, ShakhnarovichG. Learning representations for automatic colorization[M] ∥Leibe B, Matas J, Sebe N, et al. Computer vision-ECCV 2016. Lecture notes in computer science. Cham: Springer, 2016, 9908: 577- 593.

[9] ZhangR, IsolaP, Efros AA. Colorful image colorization[M] ∥Leibe B, Matas J, Sebe N, et al. Computer vision-ECCV 2016. Lecture notes in computer science. Cham: Springer, 2016, 9907: 649- 666.

[10] HuangG, Liu Z, van der Maaten L, et al. Densely connected convolutional networks[C]∥2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 21-26, 2017, Honolulu, HI. New York: IEEE, 2017: 2261- 2269.

[11] DaiJ, LiY, HeK, et al. R-FCN: object detection via region-based fully convolution networks[C]∥Proceedings of the 30th International Conference on Neural Information Processing Systems, December 5-10, 2016, Barcelona, Spain. New York: Curran Associates Inc, 2016: 379- 387.

[12] . 基于区域全卷积网络结合残差网络的火焰检测方法[J]. 激光与光电子学进展, 2018, 55(4): 041011.

. Flame detection method based on regional fully convolutional networks with residual network[J]. Laser & Optoelectronics Progress, 2018, 55(4): 041011.

[13] . 基于积神经网络的图像着色方法研究[J]. 电脑知识与技术, 2019, 15(6): 198-200.

. Research on image coloring method based on convolutional neural network[J]. Computer Knowledge and Technology, 2019, 15(6): 198-200.

[14] . 基于多尺度特征提取和全连接条件随机场的图像语义分割方法[J]. 激光与光电子学进展, 2019, 56(13): 131007.

. Image semantic segmentation based on multi-scale feature extraction and fully connected conditional random fields[J]. Laser & Optoelectronics Progress, 2019, 56(13): 131007.

[15] , 等. 基于深度学习的语义分割问题研究综述[J]. 激光与光电子学进展, 2019, 56(15): 150003.

, et al. Review of deep learning-based semantic segmentation[J]. Laser & Optoelectronics Progress, 2019, 56(15): 150003.

[16] TomasiC, ManduchiR. Bilateral filtering for gray and color images[C]∥Sixth International Conference on Computer Vision (IEEE Cat. No.98CH36271), January 7, 1998, Bombay, India. New York: IEEE, 1998: 839- 846.

[17] Gonzalez RC, Woods RE. 数字图像处理[M]. 阮秋琦, 阮宇智, 译. 3版. 北京: 电子工业出版社, 2003: 222- 223.

Gonzalez RC, Woods RE. Digital image processing[M]. Ruan Q Q, Ruan Y Z, Transl. 3th ed. Beijing: Electronic Industry Press, 2003: 222- 223.

何山, 方利, 张政. 基于改进的区域积神经网络和联合双边滤波的图像着色法[J]. 激光与光电子学进展, 2020, 57(12): 121006. 何山, 方利, 张政. Image Coloring Method Based on Improved Regional Full Convolutional Neural Network and Joint Bilateral Filtering[J]. Laser & Optoelectronics Progress, 2020, 57(12): 121006.

基于改进的区域积神经网络和联合双边滤波的图像着色法

关于本站 Cookie 的使用提示

全站搜索

基于改进的区域积神经网络和联合双边滤波的图像着色法

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索