• Journal of the Semiconductor & Display Technology
• /
• v.23 no.2
• /
• pp.23-27
• /
• 2024

Underwater images are typically degraded due to color distortion, light absorption, scattering, and noise from artificial light sources. Restoration of these images is an essential task in many underwater applications. In this paper, we propose a two-phase deep learning-based method, Underwater Deep Curve Estimation (UWDCE), designed to effectively enhance the quality of underwater images. The first phase involves a white balancing and color correction technique to compensate for color imbalances. The second phase introduces a novel deep learning model, UWDCE, to learn the mapping between the color-corrected image and its best-fitting curve parameter maps. The model operates iteratively, applying light-enhancement curves to achieve better contrast and maintain pixel values within a normalized range. The results demonstrate the effectiveness of our method, producing higher-quality images compared to state-of-the-art methods.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.34 no.10
• /
• pp.1701-1716
• /
• 2010

This is a follow-up study of "A study on the color image of baby's wear brands". This study was performed to classify the style images and analyze the style and color images of 7 baby wear brands that suggests how their harmony and consistency identify their clothing images. In order to classify the style image, questionnaires were organized with 40 pieces of the style board of 7 brands and 20 baby wear images adjectives. The total 324 copies were used to survey students who majored in fashion design. Questionnaires were analyzed by factor analysis from the SPSS 12.0 package program. The results of this study are as follows: First, the style image of baby wear brands was classified by 4 factors, 'loveliness', 'chic', 'liveliness', and 'pureness'. This was similar with the results of the color images that were surveyed in advance of this study. There were different characteristics for the style image of 7 baby wear brands. Second, gender (boy, girl, and new-borns) affected the style image and not season. Third, the harmony and consistency of the style and color image of 7 baby wear brands were different. It was possible to identify the clothing image of 5 baby wear brands by consistency and harmony in style and color images. However, it was evaluated that 2 baby wear brands were needed to plan the style and color image closely as well as harmoniously to make their clothing image definitive.

• Korean Institute of Interior Design Journal
• /
• no.7
• /
• pp.31-37
• /
• 1996

The purpose of this study is to grasp the characteristics by image type in interior color. This experiment is carried out by the evaluation method of color simulation with the color image processor. And the result of this experimental evaluation is analyzed quantitatively by HAYASI 1 Program. The results of this analysis are as follows : 1) In casual and clear images, the most main colors are GY, PB, Y , and N, Casual image has high chroma and the most arrangement of colors is hue-contrast or contrast -harmony with white. Also the main colors of clear image are in identical or similar harmony with the hues of floor. 2) In romantic, elegant, pretty, and gorgeous images, the most main colors are GY, RP, R , YR and Y and the most arrangement of colors is identical or similar harmony. The romantic image of pastel tone is wholly lighter than the pretty image of bright tone. And elegant image is lower in chroma than romantic images, so generally dark. Also gorgeous image is the vivid tone with high chroma. 3) In chic and modern images, the main colors are the hues of B, PB, high value and low chroma with bright tone. Also, the main colors are in identical or similar harmony with the hues of floor : BG , B, PB and P. 4) In natural and semiclassic images, the main color is the warm color of Y, YR and the most arrangement of colors is identical or similar harmony. Also Semiclassic images is the dull tone with middle value and middle chroma and darker tone than natural image. 5) In dynamic image, the main color is the hue of N, Y, PB and GY and most of color is high chroma. And the most arrangement of colors is value-contrast.

• Journal of the Korea Fashion and Costume Design Association
• /
• v.23 no.1
• /
• pp.47-59
• /
• 2021

This study aims to provide inspiration and methods for menswear design by analyzing elements for men's fashion using flower images. The results are as follows: Men's fashion types with flower images were categorized as classic tailored, casual tailored, casual wear, sports-outdoor. The order of frquency was casual tailored, casual, classic tailored, and sports outdoor. For the classic tailored type, the flower images are related with an X-line silhouette, and the arrangement methods, such as a scattered patterns, one-point patterns, and surface techniques, such as printing and embroidery were used, and similar color or monochromatic schemes appeared sequentially. For the casual tailored type, the flower images are related to an H-line silhouette, arrangement methods such as a scattered pattern, panel pattern, and surface techniques, such as print, embroidery, and jacquard were used, similar color and accent color schemes appeared sequentially. For the casual type, the flower images are related to H-line and Y-line silhouettes, and arrangement methods, such as a scattered pattern, all-round continuous pattern, and panel pattern, and surface techniques, such as print, jacquard, embroidery, and patchwork were used, similar color and contrast color schemes appeared sequentially. For the sports outdoor type, the flower image were related to A-line and H-line silhouettes, arrangement methods, such as a scattered pattern and all-round continuous pattern, and surface techniques, such as print and jacquard were used, monochromatic scheme and contrast color schemes appeared sequentially. Therefore, the flower images in men's fashion were applied to various design elements, and displayed an interesting result, different from conventional design approach.

• Journal of the Korean Society of Clothing and Textiles
• /
• v.43 no.3
• /
• pp.373-383
• /
• 2019

This research investigated the effect of individual yarn colors on the perception of overall colors of yarn-dyed woven fabrics. The way the colors of yarn-dyed woven fabrics are perceived is known as visual color mixing: when the different colored yarns juxtaposed on the fabric surface are observed from some distance away, they are visually mixed in our eyes and perceived as a solid color. However, we can still see individual yarn colors that make the fabrics look obviously different from actual solid colors. To quantify this visual color mixing effect, twenty-one sateen fabrics were produced in a wide range of colors using cyan, magenta, and yellow yarns, and the colors were measured instrumentally. The obtained colorimetric values were converted into solid color images on a CRT monitor. Then, the physical fabrics were scanned, and the scanned images were displayed on the monitor with solid color images in pairs for visual color difference evaluation. The woven and solid colors in each pair were of physically identical color; however, the visual color difference was as large as $4.81{\Delta}{E^*}_{ab}$ on average. A visual color difference model was proposed by considering this parametric effect of individual yarn colors.

• Journal of Korea Multimedia Society
• /
• v.12 no.7
• /
• pp.947-952
• /
• 2009

In this paper, we propose a method of generating color sketch images using the DIP operator. In the proposed method, an input RGB color image is first transformed into an HSV color image. A sketch image of the V component image is then extracted by applying the DIP operator to the V component image, which is the brightness component of the input image. For the visual convenience, the extracted sketch image of the V component image is next inverted and contrast-stretched. The S component image is also enhanced to deepen the color of output sketch image while maintaining its color. Finally, the V and S component images along with the original H component image are transformed into an output RGB color sketch image. Experimental results show that the proposed method yields output color sketch images similar to hand-drawn sketch pictures whose colors are the same as those of input color images.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2003.09a
• /
• pp.567-570
• /
• 2003

A color restoration algorithm for dynamic images under multiple luminance conditions is proposed by using correction vectors, defined for sub regions that the original target is divided into and calculated from color information given in well-illuminated regions. These vectors restore chromatic information of the restored image obtained by the color restoration algorithm in a low luminance condition. Under the condition that the size of dynamic color images in multiple luminance conditions is $320\times240$, experimental results show that the restored image by the proposed algorithm decreases the color-difference about 30％ than that of the restoration algorithm with color change vectors in a low luminance condition. The proposed algorithm aims to construct the surveillance system with a low cost CCD camera in the real world.

• Proceedings of the Korea Society of Information Technology Applications Conference
• /
• 2005.11a
• /
• pp.153-156
• /
• 2005

Skin region detection in images is an important process in many computer vision applications targeting humans such as hand gesture recognition and face identification. It usually starts at a pixel-level, and involves a pre-process of color spae transformation followed by a classification process. A color space transformation is assumed to increase separability between skin classes and other classes, to increase similarity among different skin tones, and to bring a robust performance under varying imaging conditions, without any complicated analysis. In this paper, we examine if the color space transformation actually brings those benefits to the problem of skin region detection on a set of human hand images with different postures, backgrounds, people, and illuminations. Our experimental results indicate that color space transfomation affects the skin detection performance. Although the performance depends on camera and surround conditions, normalized [R, G, B] color space may be a good choice in general.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2009.01a
• /
• pp.1-6
• /
• 2009

Due to the different camera properties of the multi-view camera system, the color properties of captured images can be inconsistent. This inconsistency makes post-processing such as depth estimation, view synthesis and compression difficult. In this paper, the method to correct the different color properties of multi-view images is proposed. We utilize a gray gradient bar on a display device to extract the color sensitivity property of the camera and calculate a look-up table based on the sensitivity property. The colors in the target image are converted by mapping technique referring to the look-up table. Proposed algorithm shows the good subjective results and reduces the mean absolute error among the color values of multi-view images by 72% on average in experimental results.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.470-473
• /
• 2000

Digital watermarking is a technical solution to the copyright problem and is a necessary technique to protect copyright of multimedia content. Recently, there are many digital watermarking methods that deal in grey scale still images. However, only a few researchers are interested in digital watermarking fro video and color images. In this paper, we focus on digital watermarking for color images. At first, in order to embed the watermark signal in color image, we converted RGB color space to YCbCr color space which is a world-wide digital component video standard. In addition, we adopted the acceptable degree of color difference in order to keep the invisibility.