• Journal of Korea Multimedia Society
• /
• v.15 no.12
• /
• pp.1442-1448
• /
• 2012

The problem of Metal Area Segmentation (MAS) in X-ray CT images is a very hard task because of metal artifacts. This research features a practical yet effective method for MAS in X-ray CT images that exploits both projection image and reconstructed image spaces. We employ the Relevant Neighbor Area (RNA) idea [1] originally developed for projection image inpainting in order to create a novel feature in the projection image space that distinctively represents metal and near-metal pixels with opposite signs. In the reconstructed result of the feature image, application of a simple thresholding technique provides accurate segmentation of metal areas due to nice separation of near-metal areas from metal areas in its histogram.

• Publications of The Korean Astronomical Society
• /
• v.32 no.1
• /
• pp.33-35
• /
• 2017

We have been working on data processing and calibration of AKARI/IRC images from pointed observations. As of September 2014, a data package for each pointing only contains raw data and quick-look data, so that users have to process them using the toolkit by themselves. We plan to change this situation and to provide science-ready data sets, which are easy-to-use for non-AKARI experts. For Phase 1&2, we have updated dark and flat calibrations, and also the toolkit itself to produce images more reliable and easier to use. A new data package includes fully calibrated images with WCS information. We released it for about 4000 pointings at the end of March 2015.

• Journal of Korea Multimedia Society
• /
• v.4 no.1
• /
• pp.67-74
• /
• 2001

In this paper, we propose a face detection algorithm using the color and shape information in color still images. The proposed algorithm is only applied to chrominance components(Cb and Cr) in order to reduce the variations of lighting condition in YCbCr color space. Input image is segmented by pixels with skin-tone color and then the segmented mage follows the morphological filtering an geometric correction to eliminate noise and simplify the segmented regions in facial candidate regions. Multiple facial regions in input images can be isolated by connected component labeling. Moreover tilting facial regions can be detected by extraction of second moment-based ellipse features.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.3
• /
• pp.91-102
• /
• 1994

For the analysis of textured image, it requires large storage space and computation time to calculate the matrix features such as SGLDM(Spatial Gray Level Dependence Matrix). NGLDM(Neighboring Gray Level Dependence Matrix). NSGLDM(Neighboring Spatial Gray Level Dependence Matrix) and GLRLM(Gray Level Run Length Matrix). In spite of a large amount of information that each matrix contains, a set of several correlated scalar features calculated from the matrix is not sufficient to approximate it. In this paper, we propose a new classifier for textured images based on these matrices in which the projected vectors of each matrix on the meaningful directions are used as features. In the proposed method, an unknown image is classified to the class of a known image that gives the maximum similarity between the projected model vector from the known image and the vector from the unknown image. In the experiment to classify images of agricultural products, the proposed method shows good performance as much as 85-95% of correct classification ratio.

• Journal of Korean Institute of Industrial Engineers
• /
• v.28 no.2
• /
• pp.162-170
• /
• 2002

Detecting straight lines in an image is frequently required for various machine vision applications such as restoring CAD drawings from scanned images and object recognition. The standard Hough transform has been dominantly used to that purpose. However, massive storage requirement and low precision in estimating line parameters due to the quantization of parameter space are the major drawbacks of the Hough transform technique. In this paper, to overcome the drawbacks, an iterative algorithm based on a self-organizing map is presented. The self-organizing map can be adaptively learned such that image points are clustered by prominent lines. Through the procedure of the algorithm, a set of lines are sequentially detected one at a time. The algorithm can produce highly precised estimates of line parameters using very small amount of storage memory. Computational results for synthetically generated images are given. The promise of the algorithm is also demonstrated with its application to two natural images of inserts.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.268-272
• /
• 2003

Large medical images in PACS are compressed for saving storage space and improving network speed. The integrated compression environment was designed and developed for uniting of various compression methods. Various compression algorithm-RLE compression, lossless JEPG, JPEG, was built into it, complying with DICOM. A image compression using DWT was also implemented in it. And a unified algorithm of lossless compression and lossy compression was designed to improve images quality and to make compression ratios high. And integrated compression environment was operating together with a database program for efficient and user-friendly management.

• Journal of applied mathematics & informatics
• /
• v.9 no.2
• /
• pp.793-806
• /
• 2002

Feature-based similarity retrieval become an important research issue in image database systems. The features of image data are useful to discrimination of images. In this paper, we propose the highspeed k-Nearest Neighbor search algorithm based on Self-Organizing Maps. Self-Organizing Maps (SOM) provides a mapping from high dimensional feature vectors onto a two-dimensional space. The mapping preserves the topology of the feature vectors. The map is called topological feature map. A topological feature map preserves the mutual relations (similarity) in feature spaces of input data. and clusters mutually similar feature vectors in a neighboring nodes. Each node of the topological feature map holds a node vector and similar images that is closest to each node vector. In topological feature map, there are empty nodes in which no image is classified. We experiment on the performance of our algorithm using color feature vectors extracted from images. Promising results have been obtained in experiments.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.62 no.9
• /
• pp.1276-1282
• /
• 2013

This paper presents a demosaicking method based on high-order interpolation with parameters. Demosaicking is an essential process in capturing color images through a single sensor-array. Thus, a lot of methods including the Hamilton-Adams(HA) method has been studied in this literature. However, the image quality depends on various factors such as contrast and correlation in color space; existing algorithms depend on test images in use. Consequently, a new test image set was suggested to develop demosaicking algorithms properly. According to previous studies, the HA method shows high performances with the new test data set. In this paper, we improve the HA method using high-order interpolation with parameters. Also, we provide an analysis and formulations for the proposed method. To evaluate our method, we compare our method with the existing methods both objectively and subjectively. The experimental results indicate that the proposed method is superior to the existing methods.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2012.05a
• /
• pp.284-286
• /
• 2012

There has been increasing attention about the prevention and counter-measure of disasters. Particularly, for the case of fire disaster, early detection reduces the damage caused by fire significantly and effective detection method is important. Since most existing detectors need to be located at a close distance to fire, analyzing camera images to find fire becomes active research topic. In this paper, we analyze the color characteristics of fire images in various color spaces and report the experimental detection results. The best result is 77.8% success rate in YIQ space.

• International Journal of Contents
• /
• v.14 no.3
• /
• pp.1-6
• /
• 2018

For 3D display applications, auto-stereoscopic display methods that can provide 3D images without glasses have been actively developed. This paper is concerned with developing a display system for elemental images of real space using integral imaging. Unlike the conventional method, which reduces a color image to the level as much as a generated depth image does, we have minimized original color image data loss by generating an enlarged depth image with interpolation methods. Our method was efficiently implemented by applying a GPU parallel processing technique with OpenCL to rapidly generate a large amount of elemental image data. We also obtained experimental results for displaying higher quality integral imaging rather than one generated by previous methods.