• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.10-12
• /
• 2016

본 논문에서는 선 드로링 도면 간의 유사도 정도를 비교하여 도면으로 표현된 3차원 물체의 유사도 측정 알고리즘을 제안한다. 앞면, 뒷면, 좌측면, 우측면, 윗면, 아래면의 선 드로잉 영상으로 표현된 총 여섯 개의 영상을 한 물체의 대표 영상으로 이용한다. 데이터베이스의 3차원 물체 영상들은 전처리를 거친 후 각 영상의 여덟 방향의 그래디언트(gradient) 히스토그램을 측정하고 각 영상을 히스토그램의 기술자 벡터로서 표현하여 저장한다. 입력 영상 역시 같은 방식으로 기술자 벡터를 구하고 이를 비교될 영상의 기술자와 비교하여 유사도를 측정한다. 이와 같은 방식으로 가장 유사한 영상 집합을 가지는 N개의 물체를 탐색하여 시각적으로 제시한다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.703-705
• /
• 2005

본 논문에서는 기울기 크기 기반의 3차원 영역확장 알고리즘을 사용하여 small animal PET(Positron Emission Tomography) 영상으로부터 종양을 분할하는 연구를 수행하였다. 픽셀 값의 범위가 다양하고 저해상도의 특성을 갖는 PET영상으로부터 대상영역을 정확하게 분할하기 위해서 전처리(preprocessing)과정으로 영상 픽셀값의 분포를 펼쳐줌으로써 영상의 가시화를 높이는 히스토그램 스트레칭(histogram stretching) 기법을 적용하고 대상영역과 픽셀값이 유사한 인접영역과의 경계를 찾기 위해 가우시안의 1차 미분 함수를 사용하여 계산된 기울기 크기(gradient magnitude) 기반의 3차원 영역확장(region growing) 알고리즘을 제안한다. 제안한 알고리즘은 영역확장의 결과에 가장 큰 영향을 미치는 적절한 동질성 기준의 선택으로 대상영역의 분할을 성공적으로 수행하여 일반적인 영역확장의 단점을 보완하였다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2012.06b
• /
• pp.489-491
• /
• 2012

본 논문은 교육용 서비스로 이용 가능한 카드 인식 시스템을 제안한다. 사용자는 자유자재로 카드 등록 및 삭제가 가능하며 카드의 회전 및 크기 변화에도 강건한 인식을 보인다. 본 논문에서는 카드 템플릿의 형태 정보와 Histogram of Oriented Gradients를 특징점으로 이용한다. 또한 최종 분류기에서 계층적인 구조를 적용하여 보다 정확한 카드 검출 및 인식을 제안한다.

• Proceedings of the IEEK Conference
• /
• 2002.06d
• /
• pp.137-140
• /
• 2002

This paper presents a new variational framework for detecting and tracking moving objects in image sequence. Motion detection is performed using Level Set Model. The original frame is used to provide th moving object boundaries Then, the detection and the tracking problem are addressed in a common framework that employs a inward-outward curve evolution function. This function is minimized using a gradient decent method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.16 no.3
• /
• pp.2192-2200
• /
• 2015

In this paper, we propose an efficient vehicle image compensation algorithm based on Histogram Equalization. The proposed a vehicle image compensation algorithm was elimination to the vehicle image shake using motion compensation and motion estimation. And, algorithm was calculated the histogram of pixel values from each sub-image by dividing the image as the constant size areas in order to image enhancement. Also, it had enhancement to the image by adjusting the gradient. The proposed algorithm was evaluate the difference between of performance and time, image by applied to the IP, and were confirmed the image enhancement with removing of vehicle camera image shake. In this paper, the proposed vehicle image enhancement algorithm was demonstrated effectiveness when compared to existing vehicle image stabilization, because the elimination of shake for the vehicle images used real-time processing without using a memory. And it was obtained the reduction effect of the computation time by the calculated through block matching, and obtained the better restoration result for naturalness of the image with the lowest noise.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.20 no.1
• /
• pp.33-39
• /
• 2017

In this paper, an effective feature which is capable of classifying targets among the detections obtained from 2D range-bearing maps generated in active sonar environments is proposed. Most conventional approaches for target classification with the 2D maps have considered magnitude of peak and statistical features of the area surrounding the peak. To improve the classification performance, HOG(Histogram of Gradient) feature, which is popular for their robustness in the image textures analysis is applied. In order to classify the target signal, SVM(Support Vector Machine) method with reduced HOG feature by the PCA(Principal Component Analysis) algorithm is incorporated. The various simulations are conducted with the real clutter signal data and the synthesized target signal data. According to the simulated results, the proposed method considering HOG feature is claimed to be effective when classifying the active sonar target compared to the conventional methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.3
• /
• pp.621-629
• /
• 2016

This paper presents a decision method of middle ear disease which is developed in children and adults. In the proposed method, features are extracted from the middle ear disease images and normal images using HoG (histogram of oriented gradient) descriptor and the extracted features are learned by SVM (support vector machine) classifier. To obtain an input vector into SVM, an input image is resized to a predefined size and then the resized image is partitioned into 16 blocks each of which is partitioned into 4 sub-blocks (namely cell). Finally, the feature vector with 576 components is given by using HoG with 9 bins and it is used as SVM learning and classification. Input images are classified by SVM classifier based on the model of learning features. Experimental results show that the proposed method yields the precision of over 90% in decision.

• Journal of Korea Multimedia Society
• /
• v.17 no.3
• /
• pp.300-311
• /
• 2014

According to fast change of the environment, the structured study of the ecosystem by analyzing the plant leaves are needed. Expecially, the methodology that searches and classifies the leaves from captured from the smart device have received numerous concerns in the field of computer science and ecology. In this paper, we propose a plant leaf classification technique using shape descriptor by combining Scale Invarinat Feature Transform (SIFT) and Histogram of Oriented Gradient (HOG) from the image segmented from the background via Graphcut algorithm. The shape descriptor is coded in the field of Locality-constrained Linear Coding to optimize the meaningful features from a high degree of freedom. It is connected to Support Vector Machines (SVM) for efficient classification. The experimental results show that our proposed approach is very efficient to classify the leaves which have similar color, and shape.

• Journal of Broadcast Engineering
• /
• v.21 no.4
• /
• pp.609-620
• /
• 2016

Recently, there are active studies on a forward collision warning system to prevent the accidents and improve convenience of drivers. For collision evasion, the vehicle detection system is required. In general, existing learning-based vehicle detection methods use the entire appearance of the vehicles from rear-view images, so that each vehicle types should be learned separately since they have distinct rear-view appearance regarding the types. To overcome such shortcoming, we learn Haar-like features from the lower part of the vehicles which contain tail lights to detect vehicles leveraging the fact that the lower part is consistent regardless of vehicle types. As a verification procedure, we detect tail lights to distinguish actual vehicles and non-vehicles. If candidates are too small to detect the tail lights, we use HOG(Histogram Of Gradient) feature and SVM(Support Vector Machine) classifier to reduce false alarms. The proposed forward vehicle detection method shows accuracy of 95% even in the complicated images with many buildings by the road, regardless of vehicle types.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.66 no.2
• /
• pp.416-424
• /
• 2017

In this study, we propose a design of digits recognition system based on RBFNNs through a comparative study of pre-processing algorithms in order to recognize digits in handwritten. Histogram of Oriented Gradient(HOG) is used to get the features of digits in the proposed digits recognition system. In the pre-processing part, a dimensional reduction is executed by using Principal Component Analysis(PCA) and (2D)2PCA which are widely adopted methods in order to minimize a loss of the information during the reduction process of feature space. Also, The architecture of radial basis function neural networks consists of three functional modules such as condition, conclusion, and inference part. In the condition part, the input space is partitioned with the use of fuzzy clustering realized by means of the Fuzzy C-Means algorithm. Also, it is used instead of gaussian function to consider the characteristic of input data. In the conclusion part, the connection weights are used as the extended type of polynomial expression such as constant, linear, quadratic and modified quadratic. By using MNIST handwritten digit benchmarking database, experimental results show the effectiveness and efficiency of proposed digit recognition system when compared with other studies.