• Journal of Broadcast Engineering
• /
• v.18 no.6
• /
• pp.935-938
• /
• 2013

In this letter, we propose a method for colorization based coding, which divides the colorization matrix into smaller sub-matrices using the meanshift segmentation. Using the proposed method the computation speed becomes more than 30 times faster. Furthermore, the smearing artifact, which appears in conventional colorization based compression method, is greatly reduced.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2012.05a
• /
• pp.40-42
• /
• 2012

Colorization is a method using computer to add color to a black and white image automatically. The input is a grayscale image and some representative pixels (RPs). The RPs contain the color information for the image, and it indicates each region's color information. Colorization-based coding is a novel way for lossy image compression, it decodes a color image to get grayscale image and extracts RPs from the image. Because RPs decides the region's color and we also want small data size for image compression, form this viewpoint the paper proposes a way to get better and fewer RPs based on watershed segmentation. According to the segmentation result we also improve the original chrominance blending colorization method to save decode time and get better reconstruct image.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.38-43
• /
• 2009

The Chinese Ink Painting is an art with long history in Chinese culture. Painters can obtain various kinds of scenery by mixing water and ink properly. These papers provides a colorization technique that can transfer gray scale paintings to color paintings. Various colorization techniques for photorealistic images have good results. But these techniques are uncertainly suitable for Chinese Ink Painting. In our method, users only provide a gray scale Chinese Ink Painting and a similar color Chinese Ink Painting subjectively, system can automatically transfer the color from color painting to gray scale painting. We also provide a method for users to refine the automatically generated result.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.124-127
• /
• 2009

In this paper, we proposed automatically video colorization method with partial color sources in first frame. The input color sources are propagated to other gray pixels with the high correlation between two pixels. To robust again the errors in portion of the weak boundary, we calculate correlation between two pixels using dual-path comparison. Video colorization method should maintain the color connectivity between frames. Accordingly, we define reliability of primarily color by compare the color of neighborhood frames. We perform the color correction by blending neighboring color when the reliability of primarily color is low. We formalize this premise with energy function, and find the color to minimize the energy function. In this way, using property of video, we reduce the error caused by propagation and get result of natural changes between frames. Through simulation results, we show the proposed method derive a natural result more than previous method.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.6
• /
• pp.313-319
• /
• 2018

We propose a post-processing algorithm to improve the quality of the RGB image generated by deep learning based colorization from the gray-scale image of an infrared camera. Wavelet fusion is used to generate a new luminance component of the RGB image luminance component from the deep learning model and the luminance component of the infrared camera. PSNR is increased for all experimental images by applying the proposed algorithm to RGB images generated by two deep learning models of SegNet and DCGAN. For the SegNet model, the average PSNR is improved by 1.3906dB at level 1 of the Haar wavelet method. For the DCGAN model, PSNR is improved 0.0759dB on the average at level 5 of the Daubechies wavelet method. It is also confirmed that the edge components are emphasized by the post-processing and the visibility is improved.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.25 no.2
• /
• pp.26-38
• /
• 2021

Image coloration refers to adding plausible colors to a grayscale image or video. Image coloration has been used in many modern fields, including restoring old photographs, as well as reducing the time spent painting cartoons. In this paper, a method is proposed for colorizing grayscale images using a convolutional neural network. We propose an encoder-decoder model, adapting FusionNet to our purpose. A proper loss function is defined instead of the MSE loss function to suit the purpose of coloring. The proposed model was verified using the ImageNet dataset. We quantitatively compared several colorization models with ours, using the peak signal-to-noise ratio (PSNR) metric. In addition, to qualitatively evaluate the results, our model was applied to images in the test dataset and compared to images applied to various other models. Finally, we applied our model to a selection of old black and white photographs.

• Journal of Korea Multimedia Society
• /
• v.21 no.11
• /
• pp.1271-1278
• /
• 2018

The NDT using ultrasonic is largely divided into A-Scan and C-Scan methods. Since A-Scan method is subject to subjective judgement by trained personnel, C-Scan method has been introduced, which presents the weld area in two dimensions by placing the transducers two dimensionally used in the A-Scan method. Therefore, it is necessary to develop equipment that can provide weld quality without the help of a welding expert and the presentation of effective C-Scan images. Thus, in this paper, the algorithms that express a low resolution 2-dimensional gray image formed by C-Scan method as a high-resolution color C-Scan image and automatically determine the weld quality from the generated C-Scan color image. The high resolution color C-Scan images proposed in this paper allow the exact shape of the weld point to be expressed, and an objective algorithm to use this image to automatically determine weld quality.

• Korean Journal of Remote Sensing
• /
• v.34 no.6_3
• /
• pp.1415-1425
• /
• 2018

Kompsat-5 is the first Earth Observation Satellite which is equipped with an SAR in Korea. SAR images are generated by receiving signals reflected from an object by microwaves emitted from a SAR antenna. Because the wavelengths of microwaves are longer than the size of particles in the atmosphere, it can penetrate clouds and fog, and high-resolution images can be obtained without distinction between day and night. However, there is no color information in SAR images. To overcome these limitations of SAR images, colorization of SAR images using Cycle GAN, a deep learning model developed for domain translation, was conducted. Training of Cycle GAN is unstable due to the unsupervised learning based on unpaired dataset. Therefore, we proposed MS Cycle GAN applying multi-scale discriminator to solve the training instability of Cycle GAN and to improve the performance of colorization in this paper. To compare colorization performance of MS Cycle GAN and Cycle GAN, generated images by both models were compared qualitatively and quantitatively. Training Cycle GAN with multi-scale discriminator shows the losses of generators and discriminators are significantly reduced compared to the conventional Cycle GAN, and we identified that generated images by MS Cycle GAN are well-matched with the characteristics of regions such as leaves, rivers, and land.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.4
• /
• pp.20-27
• /
• 2010

This paper proposes a saturation compensating method by embedding pseudo-random code information in wavelet packet based colorization algorithm. In the color-to-gray process, an input RGB image is converted into YCbCr images, and a 2-level wavelet packet transform is applied to the Y image. And then, color components of CbCr are embedded into two sub-bands including minimum amount of energy on the Y image. At this time, in order to compensate the color saturations of the recovered color image during the printing and scanning process, the maximum and minimum values of CbCr components of an original image are also embedded into the diagonal-diagonal sub-band by a form of pseudo-random code. This pseudo-random code has the maximum and minimum values of an original CbCr components, and is expressed by the number of white pixels. In the gray-to-color process, saturations of the recovered color image are compensated using the ratio of the original CbCr values to the extracted CbCr values. Through the experiments, we can confirm that the proposed method improves color saturations in the recovered color images by the comparison of color difference and PSNR values.

• Journal of the Korea Computer Graphics Society
• /
• v.24 no.2
• /
• pp.1-9
• /
• 2018

In this paper, we propose a method to restore corrupted black and white facial images to color. Previous studies have shown that when coloring damaged black and white photographs, such as old ID photographs, the area around the damaged area is often incorrectly colored. To solve this problem, this paper proposes a method of restoring the damaged area of input photo first and then performing colorization based on the result. The proposed method consists of two steps: BEGAN (Boundary Equivalent Generative Adversarial Networks) model based restoration and CNN (Convolutional Neural Network) based coloring. Our method uses the BEGAN model, which enables a clearer and higher resolution image restoration than the existing methods using the DCGAN (Deep Convolutional Generative Adversarial Networks) model for image restoration, and performs colorization based on the restored black and white image. Finally, we confirmed that the experimental results of various types of facial images and masks can show realistic color restoration results in many cases compared with the previous studies.