• IEIE Transactions on Smart Processing and Computing
• /
• v.2 no.4
• /
• pp.189-196
• /
• 2013

This paper presents a two-stage approach to detect pedestrians in video sequences taken from a moving vehicle. The first stage is a preprocessing step, in which potential pedestrians are hypothesized. During the preprocessing step, a difference image is constructed using a global motion estimation, vertical and horizontal edge maps are extracted, and the color difference between the road and pedestrians are determined to create candidate regions where pedestrians may be present. The candidate regions are refined further using the vertical edge symmetry features of the pedestrians' legs. In the next stage, each hypothesis is verified using the integral channel features and an AdaBoost classifier. In this stage, a decision is made as to whether or not each candidate region contains a pedestrian. The proposed algorithm was tested on a range of dataset images and showed good performance.

• Journal of the Semiconductor & Display Technology
• /
• v.9 no.4
• /
• pp.63-66
• /
• 2010

Detecting marker coordinates is important in augmented reality system based on tag. If a marker is not detected, objects can't be augmented. In this paper, we propose a noise reduction method for augmented reality. Using a blue color space to HIS color transformation was performed on the binary. Erosion operator and the dilation operator of the binary images were performed. Experimental results show that proposed method produces a tag image recognizable in various light environments. And using the area of the rectangle, the labeling could be detected through the tag. Tag recognition rate is improved by removing noise.

• Journal of Information Display
• /
• v.12 no.2
• /
• pp.77-83
• /
• 2011

The memory compression algorithm CCC-LCP (color-count-controlled local color palette) reduces flicker in the liquid crystal display (LCD) overdrive. Its compression ratio is 1:5.4 for 10-bit images, with a 33 dB PSNR peak signal-to-noise ratio and with five times flicker reduction compared with the block truncation coding. The authors' two-alternative forced choice subjective tests proposed two new soundness properties, the 'CMP harmlessness' and 'OD non-lost (or OD liveliness)', to clarify the functional interaction between the overdrive functionality OD and the compression functionality CMP. The tests verified that CCC-LCP is practically applicable (at a 1.2H viewing distance threshold) for 42" 37-ppi WXGA TVs.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2004.05b
• /
• pp.463-467
• /
• 2004

본 논문에서는 컬러정보를 이용하여 외부 조명의 영향에 대응하면서 얼굴 후보영역을 추출하고, 추출된 후보 영역으로부터 다채널 스킨컬러 모델로 특정 정보를 추출하는 검출 기법을 제시한다. 외부 조명에 민감한 스킨컬러 특성을 고려해 색상정보와 광도를 분리할 수 있는 Y $C_{r}$ , $C_{b}$ 색상모델을 이용하며, Green, Blue 채널의 정보를 Gaussian 확률밀도 모델로부터 $C_{b-}$ $C_{g}$ 의 좁은 범위에 분포되어 있는 스킨컬러 영역 밀도를 모델링한다. 또한 얼굴영역에 Region Restricting과 임계값 반복 알고리즘을 사용하여 눈 영역 검출 과정을 보이고, 실시간 복합 얼굴 검출 시스템 조명방식에 의해 결과를 나타낸다.다.

• Journal of the Korean Wood Science and Technology
• /
• v.41 no.5
• /
• pp.399-405
• /
• 2013

This study demonstrates the utilization of light microscopy and Fast Fourier Transform-Peak Finding (FPF) method for microfibril angle (MFA) measurement from unstained sections of red pine (Pinus densiflora). To obtain an image with optimal contrast and resolution for MFA measurement, effects of numerical aperture (NA) of condenser lens and color filters were investigated. About 60% of NA of the maximum condenser NA produced an image with optimal contrast, but a color filter with short wavelength range (DAPI) created images with improved resolution. Manual angle measurement and the FPF method were applied to the image with optimal contrast for MFA measurement. The experimental results from the FPF method were considered to be more repeatable and less subjective than those from the manual angle measurement.

• Proceedings of the Korea Multimedia Society Conference
• /
• 2002.05c
• /
• pp.164-169
• /
• 2002

칼라 정보를 이용하여 영상을 정합하기 위해서는 적은 수의 칼라 집합으로 영상을 표현하는 영상 양자화 과정이 필요하다. 적응적 양자화를 사용하는 경우에는 균일 양자화에 비해 높은 정합 성능을 기대할 수 있지만 상이한 칼라 집합의 생성으로 인해 영상 정합 과정이 힘들게 된다. 이에 본 논문에서는 상이한 칼라 집합을 갖는 영상을 정합할 수 있는 기초적인 연구를 수행하였다. 영상 정합을 위해 우선 STR(sort-tile-recursive) 방법[1]을 응용하여 질의 영상의 각 칼라에 대한 유사 칼라를 DB 영상으로부터 빠르게 선정할 수 있는 방법을 개발하였다. 질의 칼라와 유사 칼라간의 유사도를 정의하고 이를 기반으로 영상간의 유사도를 계산함으로써 영상 정합에 이용할 수 있도록 하였다. 칼라간의 유사도는 칼라 차이가 고려되어 정의되는데 칼라 차이는 칼라 공간에서의 칼라 거리로 계산된다. 칼라 거리를 계산하기 위해 유클리디언 거리를 이용할 경우 많은 계산량이 요구되므로 기존의 시티블록 거리나 체스보드 거리에 비해 유클리디언 거리를 좀더 유사하게 근사화하면서 빠른 계산이 가능한 거리 계산 방법을 개발하였다.

• Computers and Concrete
• /
• v.7 no.6
• /
• pp.511-522
• /
• 2010

A color based segmentation procedure and a modified signature technique have been applied to the detection and analyses of complicated void shapes in cemented materials. The gray-scale segmentation and available signature methods were found to be inefficient especially for the analyses of complicated void shapes. The applicability of the developed methodology has been demonstrated on artificially prepared cemented materials made of self compacted concrete material. In order to characterize the void shapes in the investigated sample images, two new shape parameters called as coefficients of inclusion and exclusion have been proposed. When compared with the traditional use of the signature method, it was found that the methodology followed herein would better characterize complicated void shapes. The methodology followed in this study may be applied to the analysis of complicated void shapes that are often encountered in other cementitious materials such as clays and rocks.

• Journal of Surface Science and Engineering
• /
• v.54 no.2
• /
• pp.84-89
• /
• 2021

The visual inspection method is widely used for corrosion damage analysis of steel plate due to the cost-efficient, fast and reasonably accurate results. However, visual inspection of corrosion deteriorated degree has a problem that the reliability of results differs depending on the inspector's individual knowledge and experience. In this study, we evaluated the degree of corrosion from a given image by using image segmentation method based on the grabcut and HSV(Hue, Saturation, Value) color image processing techniques for the development of an automatic inspection tool. The code written in Python based OpenCV-python libraries was used to categorize the images.

• Quantitative Bio-Science
• /
• v.37 no.2
• /
• pp.81-89
• /
• 2018

It is commonplace that high false detection rates interfere with immediate vision-based fire monitoring system. To circumvent this challenge, we propose a fire detection algorithm that can accommodate color variations of RGB in temporal domain, aiming at reducing false detection rates. Despite interrupting images (e.g., background noise and sudden intervention), the proposed method is proved robust in capturing distinguishable features of fire in temporal domain. In numerical studies, we carried out extensive real data experiments related to fire detection using 24 video sequences, implicating that the propose algorithm is found outstanding as an effective decision rule for fire detection (e.g., false detection rate <10%).

• The Bulletin of The Korean Astronomical Society
• /
• v.44 no.1
• /
• pp.75.3-75.3
• /
• 2019

Understanding how ram pressure stripping (RPS) affects the star formation activity of cluster galaxies is one of the important issues in astrophysics. To examine whether we can identify any discernible trend in the star formation activity of galaxies undergoing ram pressure stripping, we study the star formation properties of galaxies in the Virgo cluster for which high-resolution HI images are available. We first classify galaxies in the Extended Virgo Cluster Catalog into different stages of RPS based on their HI morphology, HI deficiency, and location in phase space. We then examine various star formation activity indicators of these galaxies, which include starburstiness, g - r color, and WISE [3.4]-[12] color. No noticeable enhancement in star formation was identified for galaxies undergoing early or active stripping. Our results suggest that star formation activity at best seems to be enhanced locally in such galaxies, making it challenging to detect with integrated photometry. With the combination of HI deficiencies and locations in phase space, we were instead able to capture the overall quenching of star formation activity with increasing degree of ram pressure stripping, which agree with previous studies.