ABSTRACT
In order to solve the problem that the existing gray-scale image coloring model is not rich in color expression and the edge distinction between different objects is not obvious, this paper proposed a gray-scale image coloring method based on the generated adversarial network (GAN), which adopts convolution neural network with the same size of the feature image as the generator and the cost function, based on its "end-to-end" feature, we trained the GAN model, takes the gray image as the model input, and colorizes the color image at the output of the generator. The method proposed in this paper is applied to the remote sensing image data set UCMerced_LandUse for experimental verification. According to the experimental results, it can be observed that the gray-scale image coloring method based on GAN proposed in this paper has better evaluation index than other methods with good coloring performance, the color expression of the image after coloring by the model is more abundant.
- A. Levin, D. Lischinski, Y. Weiss. (2004) Colorization using Optimization. ACM Transactions on Graphics, 23(3), 689--694. Google ScholarDigital Library
- Y.C. Huang, Y.S. Tung, J.C. Chen, et al. (2005) An adaptive edge detection based colorization algorithm and its applications. Proceedings of the 13th annual ACM international conference on Multimedia, Singapore, November. DBLP, 2005:351--354. Google ScholarDigital Library
- T. Welsh, M. Ashikhmin, K. Mueller. (2002) Transferring color to greyscale images. ACM Transactions on Graphics, 21(3), 277--280. Google ScholarDigital Library
- G. Larsson, M. Maire, G. Shakhnarovich. (2016) Learning Representations for Automatic Colorization. 2016:577--593.Google Scholar
- B. Hariharan, P. Arbelaez, R. Girshick, et al. (2015) Hypercolumns for object segmentation and fine-grained localization. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Boston, MA, USA, June, IEEE, 2015:447--456.Google ScholarCross Ref
- Z. Cheng, Q. Yang, B. Sheng. (2015) Deep Colorization. The IEEE International Conference on Computer Vision (ICCV), Santiago, Chile, December, IEEE, 2015: 415--423. Google ScholarDigital Library
- Guang Yang, Simiao Yu, Hao Dong, et al. (2018) DAGAN: Deep De-Aliasing Generative Adversarial Networks for Fast Compressed Sensing MRI Reconstruction. IEEE Transactions on Medical Imaging, 37(6), 1310--1321.Google ScholarCross Ref
- T.M. Quan, T. Nguyen-Duc, W.K. Jeong. (2018) Compressed Sensing MRI Reconstruction using a Generative Adversarial Network with a Cyclic Loss. IEEE Transactions on Medical Imaging, 37(6), 1488--1497.Google ScholarCross Ref
- J. Zhang, Y. Peng, M. Yuan. (2018) SCH-GAN: Semi-Supervised Cross-Modal Hashing by Generative Adversarial Network. IEEE Transactions on Cybernetics, 50(2), 489--502.Google ScholarCross Ref
- Chaoyue Wang, Chang Xu, Xin Yao, et al. Evolutionary Generative Adversarial Networks. IEEE Transactions on Evolutionary Computation, 23(6), 921--934Google Scholar
Index Terms
- Gray Scale Image Coloring Method Based on GAN
Recommendations
Automatic Gray Image Coloring Method Based on Convolutional Network
Image coloring is a time-consuming and laborious work. For a work, color collocation is an important factor to determine its quality. Therefore, automatic image coloring is a topic with great research significance and application value. With the ...
Research on Colorization of Qinghai Farmer Painting Image Based on Generative Adversarial Networks
CACML '23: Proceedings of the 2023 2nd Asia Conference on Algorithms, Computing and Machine LearningAt present, deep learning method is widely used in the field of gray image colorization. Qinghai farmer painting has distinct national characteristics. The farmer painting has bright colors, high saturation, chaotic color distribution and low color ...
Multi-scale retinex-based adaptive gray-scale transformation method for underwater image enhancement
AbstractUnderwater images play an irreplaceable role as one of the carriers of underwater information acquisition. Underwater degraded images are usually affected by the color cast, noise, and blurred details, which are difficult to apply to various ...
Comments