• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.712-714
• /
• 2004

본 논문에서는 다양한 변화에서 얼굴을 효과적으로 검출할 수 있는 방법론을 제안한다. 우리는 복잡한 배경에서 보다 효과적으로 얼굴 영역을 검출하기 위해 영역 분할 알고리즘인 JSEG를 이용하여 영역을 분할을 하게 된다. 그리고 조명 변화에 따른 간섭이 비교적 작은 YCrCb 칼라 모델을 이용하여 분할된 영역에서 후보 얼굴 영역을 찾는다. 마지막으로 보다 정확한 결과를 위하여 검출된 얼굴 후보 영역에서 눈과 눈썹을 검출하고 눈과 눈썹의 기하학적 정보를 이용해서 최종 얼굴 영역을 결정한다. 영역 분할을 이용함으로써 복잡한 배경과 다양한 조명 변화를 지닌 환경에서 다양한 얼굴 영상들을 실험한 결과 높은 정확도를 보여주었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2012.04a
• /
• pp.359-361
• /
• 2012

움직임 검출은 비디오 감시 시스템의 작업부하를 줄여주는 주요한 이슈가 되고 있다. 본 논문에서는 다양한 움직임 검출 알고리즘의 작업부하를 분석하고, 움직임 검출의 정확도와 작업부하를 고려한 경우의 최적 알고리즘을 도출한다. 비디오 감시 시스템에서 획득된 실제 데이터를 이용한 실험 결과, 움직임 프레임 비율이 낮은 환경의 비디오 감시 시스템에서는 차 프레임과 GMM을 이용하는 알고리즘이, 움직임 프레임 비율이 높은 환경에서는 GMM만을 이용하는 알고리즘이 정확도와 수행시간을 통합한 성능지수 관점에서 가장 효과적인 움직임 검출 솔루션이 될 수 있음을 확인하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2019.07a
• /
• pp.313-314
• /
• 2019

본 논문에서는 특징점 검출 알고리즘을 적용하여 ultrasound image에서 특징점을 검출하는 것과 object dectection을 위한 keypoints가 object에 올바르게 위치하는지를 검증하는 실험을 진행한다. 특징점 검출을 위한 알고리즘으로는 Scale Invariant Feature Transform(SIFT)과 Harris corner detection 을 적용하여 검증한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2011.11a
• /
• pp.442-444
• /
• 2011

본 논문에서는 접사에서의 주요 객체 검출과 검출된 주요객체의 가장 최적화된 구도를 사용자에게 안내하는 방법을 제안한다. 대부분의 접사는 주요객체에 초점을 맞추고 배경이 되는 영역은 아웃 포커싱 기법을 사용하여 촬영한다는 점에서 착안하여 주요 객체를 검출하고 검출된 주요객체와 사진 구도의 3등분할점과 구도점의 상관관계에 대하여 계산하여 최적의 구도라고 판단되는 화면으로 사용자를 유도한다. 제안하는 방법으로의 실험했을 때 좋은 결과를 얻는 것을 확인할 수 있었다.

• The KIPS Transactions:PartB
• /
• v.12B no.5 s.101
• /
• pp.617-630
• /
• 2005

General scene change detection determines the changes of a scene by using feature comparison of two continuous images that are above the fixed threshold. But existing algerian detects scene change that was used in comparing the features of two images continuously, it usually takes a lot of time in decrypting the image data and false-detection problem occurs when there is an object motion or a change of illumination. In this paper, macroblock were used to extract the information directly from the MPEG compression area and suggests algorithm that will detect scene changes more effectively. Existing algorithm have shown numerous arithmetic problems that were improved in the proposed algorithm. The existing algorithm cannot detect the changes of a scene after analyzing the relationship of the previousand futureimages while the algorithm being proposed can detect the changes of a scene continuously and resolves the problem of false-detection. To this end, the data used in general were tested to prove that this algerian would be able to detect the scene changes faster and more correctly than the existing ones. The performance of the suggested algorithm was analyzed basedontheresultsoftheexperiment. .

• Journal of the Korean Society for Nondestructive Testing
• /
• v.36 no.6
• /
• pp.451-459
• /
• 2016

Crack detection on panels during manufacturing process is an important step for ensuring the product quality. The accuracy and efficiency of traditional crack detection methods, which are performed by eye inspection, are dependent on human inspectors. Therefore, implementation of an on-line and precise crack detection is required during the panel pressing process. In this paper, a regular CCTV camera system is utilized to obtain images of panel products and an image process based crack detection technique is developed. This technique uses a comparison between the base image and a test image using an amplitude mapping of the local image. Experiments are performed in the laboratory and in the actual manufacturing lines to evaluate the performance of the developed technique. Experimental results indicate that the proposed technique could be used to effectively detect a crack on panels with high speed.

• Journal of Digital Convergence
• /
• v.10 no.7
• /
• pp.123-128
• /
• 2012

In this paper, we propose a landmark-detection system of object for more accurate object recognition. The landmark-detection system of object becomes divided into a learning stage and a detection stage. A learning stage is created an interest-region model to set up a search region of each landmark as pre-information necessary for a detection stage and is created a detector by each landmark to detect a landmark in a search region. A detection stage sets up a search region of each landmark in an input image with an interest-region model created in the learning stage. The proposed system uses Fast Fourier Transform to detect landmark, because the landmark-detection is fast. In addition, the system fails to track objects less likely. After we developed the proposed method was applied to environment video. As a result, the system that you want to track objects moving at an irregular rate, even if it was found that stable tracking. The experimental results show that the proposed approach can achieve superior performance using various data sets to previously methods.

• Fire Science and Engineering
• /
• v.26 no.6
• /
• pp.24-30
• /
• 2012

This study is on the method of the detection for flame and series arc which can be happened at poor contact point added a vibration in part of contact point of low voltage line. In general, the causes of electric fire are over current, short circuit, poor contact, ect. The over-current or short circuit among those causes is detected by measuring a instant current value, but poor contact is difficult to detect by measuring a excessive value of the voltage and current and a distortion of waveforms. And therefore, in this paper, it is studied on the optimal technique of the arc judgement using fuzzy logic and MDET (Multi Dimension Estimation Technique). And it carries out the simulation for arc detection and the experiment for controller and load test. In result, the controller and detection algoristhm, is classified with normal wave and abnormal arc wave without relation with each loads and so the controller can detect a series arc successfully.

• Journal of Broadcast Engineering
• /
• v.18 no.3
• /
• pp.401-408
• /
• 2013

In this paper, we propose a method that automatically detects organs having similar appearances in medical images by learning both context and appearance features. Since only the appearance feature is used to learn the classifier in most existing detection methods, detection errors occur when the medical images include multiple organs having similar appearances. In the proposed method, based on the probabilities acquired by the appearance-based classifier, new classifier containing the context feature is created by iteratively learning the characteristics of probability distribution around the interest voxel. Furthermore, both the efficiency and the accuracy are improved through 'region based voting scheme' in test stage. To evaluate the performance of the proposed method, we detect femur and tibia which have similar appearance from SKI10 knee joint dataset. The proposed method outperformed the detection method only using appearance feature in aspect of overall detection performance.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.20 no.5
• /
• pp.564-570
• /
• 2014

This paper propose a method to detect and track a human face using depth information as well as color images for detection of drowsy driving. It consists of a face detection procedure and a face tracking procedure. The face detection procedure basically uses the Adaboost method which shows the best performance so far. But it restricts the area to be searched as the region where the face is highly possible to exist. The face detected in the detection procedure is used as the template to start the face tracking procedure. The experimental results showed that the proposed detection method takes only about 23 % of the execution time of the existing method. In all the cases except a special one, the tracking error ratio is as low as about 1 %.