• ETRI Journal
• /
• v.23 no.4
• /
• pp.190-198
• /
• 2001

The performance of Motion Compensated Discrete Cosine Transform (MC-DCT) video coding is improved by using the region adaptive subband image coding [18]. On the assumption that the video is acquired from the camera on a moving platform and the distance between the camera and the scene is large enough, both the motion of camera and the motion of moving objects in a frame are compensated. For the compensation of camera motion, a feature matching algorithm is employed. Several feature points extracted using a Sobel operator are used to compensate the camera motion of translation, rotation, and zoom. The illumination change between frames is also compensated. Motion compensated frame differences are divided into three regions called stationary background, moving objects, and newly emerging areas each of which is arbitrarily shaped. Different quantizers are used for different regions. Compared to the conventional MC-DCT video coding using block matching algorithm, our video coding scheme shows about 1.0-dB improvements on average for the experimental video samples.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.06a
• /
• pp.35-36
• /
• 2016

본 논문에서는 기존 Bag-of-Visual words (BoW) 접근법에서 반영하지 못한 이미지의 공간 정보를 활용하기 위해서 Spatial Pyramid Matching (SPM) 기법을 Latent Dirichlet Allocation (LDA) 모델에 결합하여 이미지를 분류하는 모델을 제안한다. BoW 접근법은 이미지 패치를 시각적 단어로 변환하여 시각적 단어의 분포로 이미지를 표현하는 기법이며, 기존의 방식이 이미지 패치의 위치정보를 활용하지 못하는 점을 극복하기 위하여 SPM 기법을 도입하는 연구가 진행되어 왔다. 또한 이미지 패치를 정확하게 표현하기 위해서 벡터 양자화 대신 희소 부호화 기법을 이용하여 이미지 패치를 시각적 단어로 변환하였다. 제안하는 모델은 BoW 접근법을 기반으로 위치정보를 활용하는 SPM 을 LDA 모델에 적용하여 시각적 단어의 토픽을 추론함과 동시에 multi-class SVM 분류기를 이용하여 이미지를 분류한다. UIUC 스포츠 데이터를 이용하여 제안하는 모델의 분류 성능을 검증하였다.

• 제어로봇시스템학회:학술대회논문집
• /
• 1996.10b
• /
• pp.920-923
• /
• 1996

This paper presents a new algorithm for the self-localization of a mobile robot using perspective invariant(Cross Ratio). Most of conventional model-based self-localization methods have some problems that data structure building, map updating and matching processes are very complex. Use of the simple cross ratio can be effective to the above problems. The algorithm is based on two basic assumptions that the ground plane is flat and two parallel walls are available. Also it is assumed that an environmental map is available for matching between the scene and the model. To extract an accurate steering angle for a mobile robot, we take advantage of geometric features such as vanishing points(V.P). Point features for computing cross ratios are extracted robustly using a vanishing point and the intersection points between floor and the vertical lines of door frames. The robustness and feasibility of our algorithms have been demonstrated through experiments in indoor environments using an indoor mobile robot, KASIRI-II(KAist SImple Roving Intelligence).

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.931-932
• /
• 2008

An implementation of modified stereo matching using efficient belief propagation(BP) algorithm is presented in this paper. Edges of the image were found using conventional edge detection algorithms. Resulting edge information is used to suppress propagation of wrong probabilistic information. Proposed method can effectively reduce errors that incurred by ambiguous scene properties.

• ETRI Journal
• /
• v.38 no.3
• /
• pp.502-509
• /
• 2016

Feature detection and description are key ingredients of common image processing and computer vision applications. Most existing algorithms focus on robust feature matching under challenging conditions, such as inplane rotations and scale changes. Consequently, they usually fail when the scene is blurred by camera shake or an object's motion. To solve this problem, we propose a new feature description algorithm that is robust to image blur and significantly improves the feature matching performance. The proposed algorithm builds a feature descriptor by considering the integral projection along four angular directions ($0^{\circ}$, $45^{\circ}$, $90^{\circ}$, and $135^{\circ}$) and by combining four projection vectors into a single highdimensional vector. Intensive experiment shows that the proposed descriptor outperforms existing descriptors for different types of blur caused by linear motion, nonlinear motion, and defocus. Furthermore, the proposed descriptor is robust to intensity changes and image rotation.

• Proceedings of the KSRS Conference
• /
• 1998.09a
• /
• pp.274-279
• /
• 1998

For the accurate generation of DEMs, the determination of match quality in adaptive least square matching algorithm is significantly important. Traditionally, only the degree of convergence of a solution matrix in least squares estimation has been considered for the determination of match quality. It is, however, not enough to determine the true match quality. This paper reports two approaches of match quality determination based on adaptive least square correlation : the conventional if-then logic approaches with scene geometry and correlation as additional quality measures; and, the fuzzy logic approaches. Through these, accurate decision of match quality will minimize the number of blunder and maximize the number of exact match. The proposed methods have been tested on JERS and SPOT images and the results show good performance.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.37 no.2
• /
• pp.50-59
• /
• 2000

Stereo vision is a technique extracting depth information from stereo images, which are two images that view an object or a scene from different locations. The most important procedure in stereo vision, which is called stereo matching, is to find the same points in stereo images. It is difficult to match stereo images in real time because stereo matching requires heavy calculation. In this Paper we design a digital VLSI to Process stereo matching in real time, which we call stereo image match processor (SIMP). For implementation of real time stereo matching, sliding memory and minimum selection tree are presented. SIMP is designed with pipeline architecture and parallel processing. SIMP takes 64 gray level 64$\times$64 stereo images and yields 8 level 64 $\times$64 disparity map by 3 bit disparity and 12 bit address outputs. SIMP can process stereo images with process speed of 240 frames/sec.

• Proceedings of the KIEE Conference
• /
• 2004.11c
• /
• pp.144-146
• /
• 2004

Stereo image analysis has been an important tool for reconstructing 3D terrain. By In its nature, occlusion is one of difficulties we cannot avoid in stereo matching. This paper presents a study on occlusion detection by employing LRC(Left-Right Check) and OCC(Occlusion Constraint). Experimental results show that these method can effectively detect occluded regions and those regions are usually occurred around object contours and scene discontinuity.

• Proceedings of the KSRS Conference
• /
• 2008.10a
• /
• pp.301-304
• /
• 2008

Since longer wavelength microwave radiation can penetrate clouds, satellite passive microwave sensors can observe sea ice of the entire polar region on a daily basis. Thus, it is becoming popular to derive sea ice motion vectors from a pair of satellite passive microwave sensor images observed at one or few day interval. Usually, the accuracies of derived vectors are validated by comparing with the position data of drifting buoys. However, the number of buoys for validation is always quite limited compared to a large number of vectors derived from satellite images. In this study, the sea ice motion vectors automatically derived from pairs of AMSR-E 89GHz images (IFOV = 3.5 ${\times}$ 5.9km) by an image-to-image cross correlation were validated by comparing with sea ice motion vectors manually derived from pairs of cloudless MODIS images (IFOV=250 ${\times}$ 250m). Since AMSR-E and MODIS are both on the same Aqua satellite of NASA, the observation time of both sensors are the same. The relative errors of AMSR-E vectors against MODIS vectors were calculated. The accuracy validation has been conducted for 5 scenes. If we accept relative error of less than 30% as correct vectors, 75% to 92% of AMSR-E vectors derived from one scene were correct. On the other hand, the percentage of correct sea ice vectors derived from a pair of SSM/I 85GHz images (IFOV = 15 ${\times}$ 13km) observed nearly simultaneously with one of the AMSR-E images was 46%. The difference of the accuracy between AMSR-E and SSM/I is reflecting the difference of IFOV. The accuracies of H and V polarization were different from scene to scene, which may reflect the difference of sea ice distributions and their snow cover of each scene.

• ETRI Journal
• /
• v.26 no.6
• /
• pp.597-604
• /
• 2004

In stereoscopic television, there is a trade-off between visual comfort and 3-dimensional (3D) impact with respect to the baseline-stretch of a 3DTV camera. It is necessary to adjust the baseline-stretch at an appropriate the distance depending on the contents of a scene if we want to obtain a subjectively optimal quality of an image. However, it is very hard to obtain a small baseline-stretch using commercially available cameras of broadcasting quality where the sizes of the lens and CCD module are large. In order to overcome this limitation, we attempt to freely control the baseline-stretch of a stereoscopic camera by synthesizing the virtual views at the desired location of interval between two cameras. This proposed technique is based on the stereo matching and view synthesis techniques. We first obtain a dense disparity map using a hierarchical stereo matching with the edge-adaptive multiple shifted windows. Then, we synthesize the virtual views using the disparity map. Simulation results with various stereoscopic images demonstrate the effectiveness of the proposed technique.