• Journal of the Korea Society of Computer and Information
• /
• v.5 no.2
• /
• pp.60-70
• /
• 2000

As a result of remarkable developments in computer technology, the image retrieval system that can efficiently retrieve image data becomes a core technology of information-oriented society. In this paper, we implemented the Hierarchical Image Retrieval System for content-based image data retrieval. At the first level, to get color information, with improving the indexing method using color distribution characteristic suggested by Striker et al., i.e. the indexing method considering local color distribution characteristics, the system roughly classifies images through the improved method. At the second level, the system finally retrieves the most similar image from the image queried by the user using the shape information about the image groups classified at the first level. To extract the shape information, we use the Improved Moment Invariants(IMI) that manipulates only the pixels on the edges of objects in order to overcome two main problems of the existing Moment Invariant methods large amount of processing and rotation sensitiveness which can frequently be seen in the Directive Histogram Intersection technique suggested by Jain et al. Experiments have been conducted on 300 automobile images And we could obtain the more improved results through the comparative test with other methods.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.49 no.1
• /
• pp.1-7
• /
• 2012

This paper presents a method for 3D image construction using the hybrid (color and depth) camera system, in which the drawbacks of each camera can be compensated for. Prior to an image generation, intrinsic parameters and extrinsic parameters of each camera are extracted through experiments. The geometry between two cameras is established with theses parameters so as to match the color and depth images. After the preprocessing step, the relation between depth information and distance is derived experimentally as a simple linear function, and 3D image is constructed by coordinate transformations of the matched images. The present scheme has been realized using the Microsoft hybrid camera system named Kinect, and experimental results of 3D image and the distance measurements are given to evaluate the method.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.5
• /
• pp.9-17
• /
• 2008

This paper proposes a method of color reproduction considering the viewing conditions of a PDP TV. Because of a change of peripheral environments by a temporal and spatial location of observers, the ambient lightings should be considered in the process of color reproduction of displays. A conventional method enhances the contrast and saturation of images by controlling voltage gains in each channel, so that there is a limit to implementing the phenomenon of human adaptation. This method also faces difficulties in reproducing the perceived colors under a variety of viewing conditions. Accordingly, in order to solve this problem, we first characterize the device by considering the brightness and the flare effect. The chromatic adaptation is finally applied to reproduce colors in a PDP TV using the information from the viewing conditions by the sensor. The images reproduced by the proposed method are visually more superior to those reproduced by a conventional method in the experiments.

• Journal of Korean Academy of Oral and Maxillofacial Radiology
• /
• v.25 no.2
• /
• pp.521-533
• /
• 1995

This study was performed to develop and evaluate the method to detect Quantitatively the serial changes in the size of artificial lesion in the spongious bone by automatic color image analyzer. 15 intraoral radiograms taken before and after endodontic treatment of 5 cases were used for contour line analysis. 30 intraoral radiograms taken by geometrically standardized apparatus before and after serially the formation of artificial lesions of 0.80, 1.20, 1.75, 2.00mm in diameter at the periapical area and interdental area of spongious bone were used. The analysis of image according to the variance of lesion size by 0.25, 0.35, 0.55, and 0.85mm serially was performed by the histogram and the color enhancement with subtraction. The images inputted by CCDcamera were digitized and analyzed by NEXUS QUBE program with NEC PC-9801 computer. The obtained results were as follows: 1. There was no reliability in the analysis of lesions by contour line 2 .. The mean difference of the grey scale at each pixel was 1 step between reference image and the corrected images. 3. In the analysis by histogram of the artificial lesion in spongeous bone, the change over 0.55mm in the mesiodistal size was detectable by the change of the numbers of pixel showing the change in grey scale. 4. In the analysis by histogram of the artificial lesion in spongeous bone, the change over 0.25mm in the buccolingual size was detectable by the change in grey scale. 5. By color enbancement with- subtraction, each lesion was able to be isolated and the change in it's mesiodistal size was detectable visually , but not in it's buccolingual size.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.39 no.1
• /
• pp.67-74
• /
• 2002

A study on image searching and management techniques is actively developed by user requirements for multimedia information that are existing as images, audios, texts data from various information processing devices. If we can automatically detect and segment changing scenes from real-time image sequences, we can improve an effectiveness of image searching systems. In this paper, we propose cut detection techniques based on image color distribution and we evaluated its performance on various real-time image sequences. Results of experiments show that the proposed method are robust on various image patterns than color histogram method using statistical informations of images. Also, these methods can be used for cut detection on real-time image sequences.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.13 no.3
• /
• pp.262-266
• /
• 2020

In this paper, we proposed an effective technique for hiding the watermark in the LSB of a color image by applying spatial encryption techniques. Even if the watermark hidden in the LSB of the image is extracted, the information of the extracted watermark cannot be decrypted because the watermark is encrypted using various spatial encryption techniques. Therefore, if the watermark is concealed in the LSB using the spatial encryption techniques proposed in this paper, the security is greatly improved compared to the existing technique of embedding the watermark in the LSB. When watermarking is performed by applying the proposed technique, the image quality of the watermark-concealed image is very good, so it is impossible to distinguish it from the original image, and the watermark, which is confidential data, can be extracted from the watermarked image without loss. The performance of the proposed technique was mathematically analyzed and the superiority of the proposed technique was confirmed through experiments. When the watermark was concealed by applying the proposed technique to Lenna, airplane, Tiffany, and pepper images having a size of 512×512, the PSNR values of the watermarked images were 53.91dB, 54.10dB, 54.09dB, and 54.13dB, respectively.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.63 no.1
• /
• pp.99-107
• /
• 2014

In this paper, a landmark based localization system using a Kinect sensor is proposed and evaluated with the implemented system for precise and autonomous navigation of low cost robots. The proposed localization method finds the positions of landmark on the image plane and the depth value using color and depth images. The coordinates transforms are defined using the depth value. Using coordinate transformation, the position in the image plane is transformed to the position in the body frame. The ranges between the landmarks and the Kinect sensor are the norm of the landmark positions in body frame. The Kinect sensor position is computed using the tri-lateral whose inputs are the ranges and the known landmark positions. In addition, a new matching method using the pin hole model is proposed to reduce the mismatch between depth and color images. Furthermore, a height error compensation method using the relationship between the body frame and real world coordinates is proposed to reduce the effect of wrong leveling. The error analysis are also given to find out the effect of focal length, principal point and depth value to the range. The experiments using 2D bar code with the implemented system show that the position with less than 3cm error is obtained in enclosed space($3,500mm{\times}3,000mm{\times}2,500mm$).

• Proceedings of the Korean Vacuum Society Conference
• /
• 2000.02a
• /
• pp.211-211
• /
• 2000

Carbon nanotubes (CNTs) have been spotlighted as one of promising field emission displays(FEDs). For the first time, to authors knowledge, we have developed the 9" color CNT-FEDs with the resolution of 240x576 lines. The 9" CNT-FEDs with diode-type and triode-type structures are presented. The well-dispersed CNT paste was squeezed onto the metal-patterned cathode glass. For the anode plate, the Y2O2S:Eu, ZnS:Ag,Cl low-voltage phosphors were printed for red, green, and blue colors, respectively. The vacuum-packaged panel maintained the vacuum level of 1x10-7 Torr. The uniform moving images vacuum-packaged panel maintained the vacuum level of 1x10-7 Torr. The uniform moving images were demonstrated at 2 V/um. High brightness of 800, 200, and 150cd/m2 was observed on the green, red, and blue phosphors at V/um, respectively. Field emission characteristics of a triode-type CNT-FED were simulated using a finite element method. the resultant field strength on the cathode was modulated by gate bias and emitted electrons were focused on the anode. A relatively uniform emission image was experimentally achieved at the 800V anode. A relatively uniform emission image was experimentally achieved at the 800V anode and the 50-180 V gate biases. Energy distribution of electrons emitted from CNTs was measured using an energy analyzer. The maximum peak of energy curve corresponded to the Fermi energy level of CNTs. The whole fabrication processed of CNT-FEDs were fully scalable and reproducible. Our CNT-FEDs has demonstrated the high potential of large-area and full-color applications with very low cost fabrication and low power consumption.

• Smart Media Journal
• /
• v.9 no.4
• /
• pp.162-169
• /
• 2020

Design is a key factor that determines the competitiveness of products in the textile and fashion industry. It is very important to measure the similarity of the proposed design in order to prevent unauthorized copying and to confirm the originality. In this study, a deep learning technique was used to quantify features from images of textile designs, and similarity was measured using Spearman correlation coefficients. To verify that similar samples were actually detected, 300 images were randomly rotated and color changed. The results of Top-3 and Top-5 in the order of similarity value were measured to see if samples that rotated or changed color were detected. As a result, the VGG-16 model recorded significantly higher performance than did AlexNet. The performance of the VGG-16 model was the highest at 64% and 73.67% in the Top-3 and Top-5, where similarity results were high in the case of the rotated image. appear. In the case of color change, the highest in Top-3 and Top-5 at 86.33% and 90%, respectively.

• Journal of Fashion Business
• /
• v.28 no.2
• /
• pp.76-91
• /
• 2024

This study analyzed the colors used in the logos of infants' and children's clothing brands, comparing them based on brand type and the pursued image. A total of 34 brands with the highest recent sales rates were selected. Among these, 17 were large-sized enterprise brands found in department stores, and the other 17 were in-house production enterprise brands. All colors used in the logos of the selected brands were extracted from the brand logo images. In addition, the study classified the pursued images of the selected brands using emotional adjectives in order to assess variations in different color attributes of the logos according to the brand type and pursued image. The study found that in-house production enterprise brands used lighter and more varied colors compared to department store brands. For both brand types, reddish-to-yellowish hues were most commonly used. Brands projecting a 'cute' and 'gorgeous' image exhibited a greater array of colors in their logos, while brands with a 'neat', 'modern', and 'luxurious' image used less saturated colors. In terms of logo type, symbol logos used the most colors with a moderate level of lightness, while typographic logos used the fewest colors with a low level of lightness.