• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.2
• /
• pp.124-132
• /
• 2014

In this paper, we present a novel backlit region detection and contrast enhancement method using fuzzy C-means clustering and adaptively partitioned block based contrast stretching. The proposed method separates an image into both dark backlit and bright background regions using adaptively partitioned blocks based on the optimal threshold value computed by fuzzy logic. The detected block-wise backlit region is refined using the guided filter for removing block artifacts. Contrast stretching algorithm is then applied to adaptively enhance the detected backlit region. Experimental results show that the proposed method can successfully detect the backlit region without a complicated segmentation algorithm and enhance the object information in the backlit region.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.37 no.5
• /
• pp.31-41
• /
• 2000

This paper proposes a block-based motion detection algorithm for digital video surveillance system which adaptively decides the threshold according to the kinds of images We first compute the features of a block after dividing each Image into small sub-block regions, and analyze performance of the motion detection algorithm based on statistic features by using the proposed threshold-decision method. Motion vectors are used to analyze motion degree and adaptively determine the threshold The simulation results show the performances of motion detection algorithms according to sub-block size, statistic features, noise, and threshold.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.45 no.2
• /
• pp.95-102
• /
• 2008

This paper proposes an adaptive speech enhancement system based on an auditory system to enhance speech that is degraded by various background noises. As such, the proposed system detects voiced and unvoiced sections, adaptively adjusts the coefficients for both the lateral inhibition and the amplitude component according to the detected sections for each input fame, then reduces the noise signal using a time-delay neural network. Based on measuring the signal-to-noise ratio, experiments confirm that the proposed system is effective for speech degraded by various noises.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.46 no.4
• /
• pp.102-110
• /
• 2009

In this paper, we propose an object detection method for real-time in infrared (IR) images and PowerPC (PPC) and H/W design based on field programmable gate array (FPGA). An open H/W architecture has the advantages, such as easy transplantation of HW and S/W, support of compatibility and scalability for specification of current and previous versions, common module design using standardized design, and convenience of management and maintenance. Proposed background modeling for an open H/W architecture design decreases size of search area to construct a sparse block template of search area in IR images. We also apply to compensate for motion compensation when image moves in previous and current frames of IR sensor. Separation method of background and objects apply to adaptive values through time analysis of pixel intensity. Method of clutter reduction to appear near separated objects applies to median filter. Methods of background modeling, object detection, median filter, labeling, merge in the design embedded system execute in PFC processor. Based on experimental results, proposed method showed real-time object detection through global motion compensation and background modeling in the proposed embedded system.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.05a
• /
• pp.579-582
• /
• 2015

The Mean-Shift is a robustness algorithm that can be used for tracking the object using the similarity of histogram distributions of target model and target candidate. However, Mean-shift using hue information has disadvantage of tracking a wrong target when the target and background has similar hue distributions. We then propose a robust Mean-Shift tracking algorithm using new image that combined upper 4bit-planes in hue and saturation, respectively.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2005.11a
• /
• pp.550-554
• /
• 2005

본 논문에서는 블럽을 사용해서 다수의 자동차 후면의 번호판을 추출하는 방법을 제안한다. 입력 영상에서 번호판의 문자와 배경사이의 명암도 차이를 이용하여, 입력 영상의 모든 블럽을 찾고, 찾아낸 블럽을 둘러싸는 최소의 사각형들을 구한다. 이 사각형들 중에서 일련의 경향성을 갖는 블럽 그룹을 찾는다. 찾아난 블럽 그룹이 자동차 번호판인지 아닌지를 SVM을 이용하여 확인한다. 적응적 이진화를 제외한 전처리작업을 하지 않았음에도 불구하고 번호판 검출률은 매우 높았으며, 번호판을 검출하는데 걸리는 시간도 길지 않았다.

• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.413-416
• /
• 2003

This paper proposes an adaptive background image generation method based on the frame difference for traffic monitoring. The performance of the conventional method is limited when there are more vehicles due to traffic Jam. To improve on this, we use frame differencing to separate vehicles from background in frame differencing, we adopt selective approach by using part of the image not considered as vehicle fer extraction of background. The proposed method generates background more efficiently than conventional methods even in the presence of heavy traffic.

• Proceedings of the Korean Information Science Society Conference
• /
• 2008.06c
• /
• pp.468-471
• /
• 2008

이 논문은 응급상황을 인식하기 위하여 어안렌즈를 통해 획득된 영상을 이용하여 기절 동작을 인식하는 방법을 제안한다. 거실의 천장 중앙에 위치한 어안렌즈(fish-eye lens)를 장착한 카메라로부터 화각이 170인 RGB 컬러 모델의 어안 영상을 입력 받은 뒤, 가우시안 혼합 모델 기반의 적응적 배경 모델링 방법을 이용하여 동적으로 배경 영상을 갱신한다. 입력 영상의 평균 밝기를 구하고 평균 밝기가 급격하게 변화하지 않도록 영상 픽셀을 보정한 뒤, 입력 영상과 배경 영상과 차이가 큰 픽셀을 찾음으로써 움직이는 객체를 추출하였다. 그리고 연결되어 있는 전경 픽셀 영역들의 외곽점들을 추적하여 타원으로 매핑하고 움직이는 객체 영역의 형태를 단순화하였다. 이 타원을 추적하면서 어안 렌즈 영상을 투시 영상으로 변환한 다음 타원의 크기 변화, 위치 변화, 이동 속도 정보를 추출하여 이동과 정지 및 움직임이 기절동작과 유사한지를 판단하도록 하였다. 본 논문에서는 실험자로 하여금 기절동작, 걷기 동작, 앉기 동작 등 여러 동작을 취하게 하고 기절 동작 인식을 실험하였다. 실험 결과 어안 렌즈 영상을 그대로 사용하는 것보다 투시 영상으로 변환하여 타원의 크기변화, 위치변화, 이동속도 정보를 이용하는 것이 높은 인식률을 보였다.

• The Journal of the Korea Contents Association
• /
• v.7 no.2
• /
• pp.40-50
• /
• 2007

In this paper, we propose an efficient method for detecting and tracking multiple moving objects based on morphological region merging from real-time video sequences. The proposed approach consists of adaptive threshold extraction, morphological region merging and detecting and tracking of objects. Firstly, input frame is separated into moving regions and static regions using the difference of images between two consecutive frames. Secondly, objects are segmented with a reference background image and adaptive threshold values, then, the segmentation result is refined by morphological region merge algorithm. Lastly, each object segmented in a previous step is assigned a consistent identification over time, based on its spatio-temporal information. The experimental results show that a proposed method is efficient and useful in terms of real-time multiple objects detecting and tracking.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.42 no.1
• /
• pp.101-108
• /
• 2005

Adaptive constant false alarm rate(CFAR) algorithm is used for good detection probability as well as constant false alarm rate in clutter background. Especially, filtering technique adaptive to spatial variation is necessary for improving detection quality in non stationary clutter environment which has spatial correlation and large magnitude deviation. In this paper, we propose a two-dimensional block interpolation(TBI) adaptive CFAR algorithm that calculates the node estimate in the fred two dimensional region and subsequently determines the final estimate for each resolution cell by two-dimensional interpolation. The proposed method is efficient for filtering abnormal ejection by adopting distribution median in fixed region and also has advantage of reducing required memory space by using estimation method which gets final values after calculating the block node values. Through simulations, we show that the proposed method is superior to the traditional adaptive CFAR algorithms which are transversal or recursive in aspect of the detection performance and required memory space.