• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.12
• /
• pp.2982-2988
• /
• 2014

The edge detection is a pre-processing of such as image segmentation, image recognition, etc, and many related studies are being conducted both in domestic and abroad. Representative edge detection methods are Sobel, Prewitt, Laplacian, Roberts and Canny edge detectors. Such existing methods are possible for superb detections of edges if edges are detected from videos without noises. However, for video degraded by the salt-and-pepper noise, the edge detection characteristic is shown to be insufficient due to the noise influence. Therefore, in this study, the area is separated as the top, down, left and right from the mask's center pixel first to acquire a superb edge detection characteristic from the video damaged by the salt-and-pepper noise. And the algorithm that detects the final edge by applying the directional mask on the assumed factor of mask that is obtained according to the result of determination for the noise status of representative pixel value of each area.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.42 no.6
• /
• pp.75-82
• /
• 2005

The robust feature vector selection method for the multiple frontal face detection is proposed in this paper. The proposed feature vector for the training and classification are integrated by means, amplitude projections, and its 1D Harr wavelet of the input image. And the statistical modeling is performed both for face and nonface classes. Finally, the estimated probability density functions (PDFs) are applied for the detection of multiple frontal faces in the still image. The proposed method can handle multiple faces, partially occluded faces, and slightly posed-angle faces. And also the proposed method is very effective for low quality face images. Experimental results show that detection rate of the propose method is $98.3\%$ with three false detections on the testing data, SET3 which have 227 faces in 80 images.

• Journal of the Korean Society of Propulsion Engineers
• /
• v.11 no.1
• /
• pp.34-42
• /
• 2007

Ultrasonic inspection method is more profitable than X-ray radiographic inspection in cost and effect of defect detection such as debond, damage, and it doesn't need special constructions. The method can also be a possible real time inspection with safety. This report explains 1)the experiment and analysis of ultrasonic property of solid propellant, 2)the inspection methods of propellant/liner debond by inside or outside inspection, and 3)the inspection methods of propellant microcrack by damage. From the results, it is possible to detect the defect of propellant/liner debond by inside or outside inspection. Futhermore, it can be possible to detect the propellant microcrack caused by damage using the ultrasonic attenuation.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.04c
• /
• pp.298-300
• /
• 2003

동영상이나 정지영상에서 사람의 얼굴을 검출 및 인식을 하는 여러 가지 알고리즘이 소개되고 있다. 본 논문에서는 신경망(Neural Network)과 얼굴의 기하학적 특징 중에 하나인 눈과 입을 사용하여 얼굴 영역을 추출하는 방법을 사용한다 신경망은 얼굴 인식을 비롯한 여러 분야에서 쓰이는 좋은 방법 중의 하나 이지만 신경망이 가지고 있는 특성상 많은 오차를 가질 수 있기 때문에 얼굴을 구성하고 있는 요소인 눈과 입을 사용해서 오차를 제거하는 방법을 제안한다.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.41 no.11
• /
• pp.1356-1363
• /
• 2016

In this paper, we propose a novel error detection algorithm for LZ78-compressed data. The conventional error detection method adds a certain number of parity bits in transmission, and the receiver checks the number of bits representing '1' to detect the errors. These conventional methods use additional bits resulting in increased redundancy in the compressed data which results in reduced effectiveness of the final compressed data. In this paper, we propose error detection algorithm using the structural properties of LZ78 compression without using additional bits in the compressed data. The simulation results show that the error detection ratio of the proposed algorithm is about 1.3 times better for error detection than conventional algorithms.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.24 no.1
• /
• pp.28-32
• /
• 2014

For a signal such as speech showing piece-wise linear shape in a very short time period, a nonuniform sampling method based on the inflection point detection (IPD) is proposed to reduce data rate. The method exploits the geometrical characteristics of signal further than the existing local maxima/minima detection (MMD) based sampling method. As results, the reconstructed signal by the interpolation of the IPD based sampled data resembles the original speech more. Computer simulation shows that the proposed IPD based method produces about 9~23 dB improvement over the existing MMD method. To show the usefulness of the IPD technique, it is applied to speech coding, and compared to the continuously variable slope delta modulation (CVSD). The nonuniformly sampled data is binary coded with one bit flag set "1". Noninflection samples are not sent, but only flag bits set 0 are sent. The method shows 0.3 ~ 9 dB SNR and 0.5 ~ 1.3 mean opinion score (MOS) improvements over the CVSD.

• Proceedings of the Korean Information Science Society Conference
• /
• 2003.10b
• /
• pp.523-525
• /
• 2003

방송 프로그램이나 영화와 같은 동영상을 인터넷에서 검색하는 서비스가 활성화됨에 따라 특정 인물이 등장하는 부분을 검색하는 기능은 일반 사용자나 프로그램을 만드는 제작자 모두에게 필요한 기능이 되었다. 등장인물 중심의 검색을 위해서는 해당 인물의 얼굴 검출 및 인식 기능이 필수적이며, 특히 방송영상의 특성에 적합하고 등장인물 검색 서비스에 적용 가능한 얼굴 검출 및 인식 기술이 요구된다 이를 위해 본 논문에서는 고속 얼굴 인식 시스템을 제안하고, 실시간 수행이 가능한 얼굴 검출 및 인식 알고리즘을 제안하다. 제안한 얼굴 검출 및 인식 알고리즘은 DCT 기법을 전처리 단계로 두어 계산량을 최소화하면서도 특징값의 정보량은 유지하는 방법을 사용한다. 본 논문에서는 제안하는 알고리즘이 기존 방법에 비해 우수한 성능을 보이며, 실제 방송 영상을 구현된 시스템에 적용하여 시간과 검출률/인식률 측면에서 우수한 결과를 나타냄을 보인다.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.47 no.5
• /
• pp.192-198
• /
• 2010

With the popularization of the mobile phone with a built-in camera, there are a lot of effort to provide useful information to users by detecting and recognizing the text in the video which is captured by the camera in mobile phone, and there is a need to detect the text regions in such mobile phone video. In this paper, we propose a method to detect the text regions in the mobile phone video. We employ morphological operation as a preprocessing and obtain binarized image using modified k-means clustering. After that, candidate text regions are obtained by applying connected component analysis and general text characteristic analysis. In addition, we increase the precision of the text detection by examining the frequency of the candidate regions. Experimental results show that the proposed method detects the text regions in the mobile phone video with high precision and recall.

• 한국HCI학회:학술대회논문집
• /
• 2009.02a
• /
• pp.193-196
• /
• 2009

Most of studies adopt a fixed skin color model to segment skin color region in a single image. The methods, however, result in low detection rates or high false positive error rates since the distribution of skin color is varies depending on the characteristics of input image. For the effective skin color segmentation, therefore, we need a adaptive skin color model which changes the model depending on the color distribution of input image. In this paper, we propose a novel adaptive skin color segmentation algorithm consisting of 2 stages which results in both high detection rate and low false positive error rate.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.51 no.10
• /
• pp.248-255
• /
• 2014

In this paper, we propose a pothole detection method in asphalt pavement using various features. Segmentation, candidate, and decision steps of pothole detection are processed according to the values which are derived from feature characteristics. Segmentation step, we use histogram and closing operation of morphology filter which extracts dark regions for pothole detection. Candidate step, we extract candidate regions of pothole using various features such as size, compactness, etc. Finally, decision step, candidate regions are decided whether pothole or not using comparison of pothole and background's features. Experimental results show that our proposed pothole detection method has better results than existing methods and good performance in discrimination of pothole and similar patterns.