• Journal of the Korea Convergence Society
• /
• v.9 no.9
• /
• pp.31-39
• /
• 2018

In this paper, we proposed a license plate detection method based on contour extraction that adapts to environmental changes. The proposed method extracts contour lines using DoG (Difference of Gaussian) to remove unnecessary noise parts in the contour extraction process. Binarization was applied in ugly outline images, and erosion and dilation operations were used to emphasize the contour of the character part. Then, only the outline of the ratio of the characters of the plate was extracted through the ratio of the width and height of the characters. And the case where the outline is the longest is estimated by estimating the characters of the license plate. For the experiment, we applied 130 image data to license plate on the front of the vehicle, oblique environment, and environment images with various backgrounds. I also experimented with motorcycle images of different license plate patterns. Experimental results showed that the detection rate of the oblique image was 93% and that of the various background environment was 70% in the motorcycle image but 98% in the front image.

• Journal of Korea Game Society
• /
• v.1 no.1
• /
• pp.49-54
• /
• 2001

본 논문에서는 복잡한 랜덤 배경 하에서 위치하고 있는 게임 플레이어의 얼굴 영상을 스테레오 매칭을 이용하여 배경과 분리하여 추출할 수 있는 방법에 대하여 기술한다. 사람과의 상호 작용이 필요한 게임일수록 사람의 동작이나 각 부위에 대한 인식이 필요하다. 이 방법은 게임 이외에도 보안 시스템, 의류 시뮬레이션, 3D 모델링 그리고 로보틱스와 같은 분야에 적용될 수 있다. 스테레오 매칭에 관해서는 많은 연구가 있어왔으며, 기본적으로 영역기반 방법과 특징기반 방법으로 분류될 수 있다. 본 논문의 제안 방법 에서는 영역기반 방법으로 처리를 시작하고, 다단계 크기의 윈도우를 적용하여 물체의 경계선을 찾는 작업을 진행한다. 각 윈도우 크기에 대하여 유사성 커브가 생성되며, 이 값은 물체의 경계선을 판별하는 특징으로 사용된다. 전단계에서 생성된 코어스(coarse) 영역은 유사성 커브 방식에 의하여 머지 작업을 거치며, 최종적으로 대상 물체의 영상을 추출하게 된다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.4 s.310
• /
• pp.58-66
• /
• 2006

In this paper, we apply DCM(Dilation of Color and Motion information) mask and Active Contour Models(Snakes) to extract facial outline in moving pictures with complex background. First, we propose DCM mask which is made by applying morphology dilation and AND operation to combine facial color and motion information, and use this mask to detect facial region without complex background and to remove noise in image energy. Also, initial curves are automatically set according to rotational degree estimated with geometric ratio of facial elements to overcome the demerit of Active Contour Models which is sensitive to initial curves. And edge intensity and brightness are both used as image energy of snakes to extract contour at parts with weak edges. For experiments, we acquired total 480 frames with various head-poses of sixteen persons with both eyes shown by taking pictures in inner space and also by capturing broadcasting images. As a result, it showed that more elaborate facial contour is extracted at average processing time of 0.28 seconds when using interpolated initial curves according to facial rotation degree and using combined image energy of edge intensity and brightness.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2015.10a
• /
• pp.492-494
• /
• 2015

본 논문에서는 철도선상에서 발생할 수 있는 자살 사고를 예방하기 위한 전단계로서 열차의 철도를 추출하는 방법을 제안한다. 본 논문에서 제안된 방법은 철도 영상에서 각각의 RGB 채널 값을 추출한다. 추출된 각각의 RGB 채널 값을 삼각형 타입 의 소속 함수에 적용하여 상한 값과 하한 값을 퍼지 스트레칭 기법으로 철도 영상의 명암 대비를 강조시킨다. 퍼지 스트레칭 기법이 적용된 영상에서 각각의RGB 채널 값을 이용하여 배경을 제거한 후에 그레이 영상으로 변환한다. 변환된 그레이 영상에서 캐니 마스크를 적용하여 철도선의 에지를 검출하고 에지가 검출된 영상에서 허프 변환 기법과 유클리디안 거리를 적용하여 철도를 검출한다. 제안된 방법의 성능을 확인하기 위해서 다양한 각도의 철도 영상을 대상으로 실험한 결과, 제안된 방법이 철도 검출에 가능성 있는 방법인 것을 확인할 수 있었다.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• v.9 no.1
• /
• pp.145-148
• /
• 2005

본 논문에서는 동작 검출 기법을 소형 화상 카메라에 적용하여 감시 영상을 검출했으며, 감시영상은 모바일 환경에서 실시간 모니터링을 통해 유무선 연동의 감시 시스템을 구축하였다. 기존에 사용되던 화소 값을 이용하여 동작 검출을 수행하는 방법은 배경 영상을 저장할 메모리가 필요하고 화소단위의 데이터 처리로 인하여 수행 시간의 증가와 노이즈에 민감한 단점을 감수해야만 한다. 이러한 단점을 해결하기 위해 본 연구에서는 블록 단위의 특징 값을 비교하는 기법을 사용하였다. 검출된 영상을 모바일 클라이언트로 전송하기 위해서 WAP 풀 기반 영상전송 방법을 사용하여 WIPI SDK로 구현함으로써 감시 영상의 변환과 전송을 실시간으로 구현할 수 있었다.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.9-18
• /
• 2020

In the field of image processing, a method of detecting and counting passengers as moving objects when getting on and off the bus has been studied. Among these technologies, one of the artificial intelligence techniques, the deep learning technique is used. As another method, a method of detecting an object using a stereo vision camera is also used. However, these techniques require expensive hardware equipment because of the computational complexity of used to detect objects. However, most video equipments have a significant decrease in computational processing power, and thus, in order to detect passengers on the bus, there is a need for an image processing technology suitable for various equipment using a relatively low computational technique. Therefore, in this paper, we propose a technique that can efficiently obtain the number of passengers on the bus by detecting the contour of the object through the background subtraction suitable for low-cost equipment. Experiments have shown that passengers were counted with approximately 70% accuracy on lower-end machines than those equipped with stereo vision camera.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.2
• /
• pp.31-37
• /
• 2014

In conventional methods for car navigational path detection using Hough transform, navigational path deviation of a car is decided in car navigational images with simple background. But in case of car navigational images having complex background with obstacles on the road, shadows, other cars, and so on, it is very difficult to detect navigational path because these obstacles obstruct correct detection of car navigational path. In this paper, I proposed an effective navigational path detection method having better performance than conventional navigational path detection methods using Hough transform only, and fuzzy binarization method and Canny mask are applied in the proposed method for the better performance. In order to evaluate the performance of the proposed method, I experimented with 20 car navigational images and verified the proposed method is more effective for detection of navigational path.

• Journal of Korea Society of Industrial Information Systems
• /
• v.24 no.5
• /
• pp.129-137
• /
• 2019

Ubiquitous computing technology is emerging as information technology advances. In line with this, a number of studies are being carried out to increase device miniaturization and user convenience. Some of the proposed devices are user-friendly and uncomfortable with hand-held operation. To address these inconveniences, this paper proposed a virtual button that could be used in watching television. When watching a video on television, a camera is installed at the top of the TV, using the fact that the user watches the video from the front, so that the camera takes a picture of the top of the head. Extract the background and hand area separately from the filmed image, extract the outline to the extracted hand area, and detect the tip point of the finger. Detection of the end point of the finger produces a virtual button interface at the top of the image being filmed in front, and the button activates when the end point of the detected finger becomes a pointer and is located inside the button.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.7
• /
• pp.838-846
• /
• 2004

In the case where both a camera and a target are moving at the same time, the image background is successively changed, and the overlap with other moving objects is apt to be generated. The snake algorithms have been variously used in tracking the object, but it is difficult to be applied in the excessive overlap with other objects and the large bias between the snake and the target. To solve this problem, this paper presents an extended snake model. It includes an additional energy function which considers the temporal variation rate of the snake's area and a SSD algorithm which generates the template adaptive to the snake detected in the previous frame. The new energy function prevents the snake from over-shrinking or stretching and the SSD algorithm with adaptively changing template allows the prediction of the target's position in the next frame. The experimental results have shown that the proposed algorithm successfully tracks the target even when the target is temporarily occluded by other objects.

• Journal of Digital Contents Society
• /
• v.16 no.5
• /
• pp.717-725
• /
• 2015

Prostate images have been used in the diagnosis of prostate using TRUS images being relatively cheap. Ultrasound images are recorded with 3 dimension and one diagnostic exam is made with a number of the images. A doctor can see 2 dimensional images on the monitor sequentially and 3 dimensional ones to diagnose a disease. To display the images, 2-d images are used with raw 2-d ones, but 3-d images need to be segmented by the prostates and their backgrounds to be seen from different angles and with cut images of inner side. Especially on detecting the boundary, the ones in the middle of all images are easy to find the boundary but the base and apex of the images are hard to do it since there are lots of uncertain boundary. So, in this paper we propose the method that applies an average shape model and detects the boundary, and shows its superiority compared to the existing methods with experiments.