• Journal of Internet Computing and Services
• /
• v.18 no.4
• /
• pp.61-67
• /
• 2017

Recently, CCTV has been combined with areas such as big data, artificial intelligence, and image analysis to detect various abnormal behaviors and to detect and analyze the overall situation of objects such as people. Image analysis research for this intelligent video surveillance function is progressing actively. However, CCTV images using 2D information generally have limitations such as object misrecognition due to lack of topological information. This problem can be solved by adding the depth information of the object created by using two cameras to the image. In this paper, we perform background modeling using Mixture of Gaussian technique and detect whether there are moving objects by segmenting the foreground from the modeled background. In order to perform the depth information-based segmentation using the RGB information-based segmentation results, stereo-based depth maps are generated using two cameras. Next, the RGB-based segmented region is set as a domain for extracting depth information, and depth-based segmentation is performed within the domain. In order to detect the center point of a robustly segmented object and to track the direction, the movement of the object is tracked by applying the CAMShift technique, which is the most basic object tracking method. From the experiments, we prove the efficiency of the proposed object detection and tracking method using the RGB-D model.

• Journal of Korea Multimedia Society
• /
• v.19 no.12
• /
• pp.1927-1935
• /
• 2016

This paper propose a framework for human body parts in RGB-D image. We conduct tasks of obtaining person area, finding candidate areas and local detection in order to detect hand, foot and head which have features of long accumulative geodesic distance. A person area is obtained with background subtraction and noise removal by using depth image which is robust to illumination change. Finding candidate areas performs construction of graph model which allows us to measure accumulative geodesic distance for the candidates. Instead of raw depth map, our approach constructs graph model with segmented regions by quadtree structure to improve searching time for the candidates. Local detection uses HOG based SVM for each parts, and head is detected for the first time. To minimize false detections for hand and foot parts, the candidates are classified with upper or lower body using the head position and properties of geodesic distance. Then, detect hand and foot with the local detectors. We evaluate our algorithm with datasets collected Kinect v2 sensor, and our approach shows good performance for head, hand and foot detection.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2011.07a
• /
• pp.39-41
• /
• 2011

본 논문에서는 깊이 영상의 해상도와 품질을 향상시키는 방법을 제안한다. 일반적으로 2D-plus-Depth 구조의 3D 콘텐츠에서는 깊이 영상의 품질이 매우 중요하다. 최근 들어 Time-of-Flight (TOF) 방식의 깊이 센서가 깊이 영상 획득에 많이 사용되고 있는데 TOF 깊이 센서가 제공하는 깊이 영상은 저해상도이기 때문에 고해상도 3D 콘텐츠를 제작하기 위해서는 깊이 영상의 해상도를 상향 변환하는 것이 필수적이다. 또한 고품질의 깊이 영상을 얻기 위해서는 물체 간의 경계를 정교하게 보존하는 것이 중요하다. 최근에는 깊이 영상의 해상도 상향 변환을 위해서 Joint Bilateral Upsampling(JBU) 방식이 많이 사용되고 있다. 본 논문은 깊이 영상의 해상도를 높임에 있어서 우선 보간법을 수행하여 영상의 상향 변환 시에 생긴 빈 홀들의 값을 채워준 후 Bilateral Filtering을 수행함으로써 성능을 높인다. 일반적으로 영상을 상향 변환을 할 때 다양한 방법들이 있는데 본 논문에서는 Nearest Neighborhood(NN), Gaussian과 경계 보존 보간법, 경계 보존 보간법과 Fast Curvature Based Interpolation(FCBI)를 결합한 보간법을 사용하였다. 실험 결과 제안 방법이 기존 방법보다 우수한 성능을 가짐을 보여준다. 또한 경계 보존 보간법과 FCBI를 결합한 보간법을 이용해서 상향 변환을 수행한 결과가 다른 보간법들에 의한 결과보다 우수하다는 점을 알 수 있다.

• Journal of Korea Multimedia Society
• /
• v.17 no.2
• /
• pp.152-159
• /
• 2014

Recently. utilization of remote sensing exploration and cloud computing has been extended to efficient measurement, store, and update of bathymetry map data according to cloud computing technology. In the field of real ocean, water depth measurements and measurement data management, distribution, and display equipment for the development and dissemination have generated a lot of time and cost. To improve these problems, through real-time three-dimensional display system at this location, we can determine the importance of measurement activities, and reduce the time and cost of measurement activities. Data measured from marine probe vessels and remote sensing exploration equipments and other various channels can be handled and managed. In this paper, we propose a realtime three-dimensional display system through the depth measurements from remote sensing exploration. The proposed real-time three-dimensional display system can be effectively applied in the field of measurement of the topographical survey of the land as well as bathymetry of the sea.

• Journal of Broadcast Engineering
• /
• v.15 no.2
• /
• pp.163-172
• /
• 2010

The free viewpoint video delivers an active contents where users can see the images rendered from the viewpoints chosen by them. Its applications are found in broad areas, especially museum tour, entertainment and so forth. As a new free-viewpoint application, this paper presents a stereoscopic free-viewpoint TIP (Tour Into Picture) where users can navigate the inside of a single image controlling a virtual camera and utilizing depth data. Unlike conventional TIP methods providing 2D image or video, our proposed method can provide users with 3D stereoscopic and free-viewpoint contents. Navigating a picture with stereoscopic viewing can deliver more realistic and immersive perception. The method uses semi-automatic processing to make foreground mask, background image, and depth map. The second step is to navigate the single picture and to obtain rendered images by perspective projection. For the free-viewpoint viewing, a virtual camera whose operations include translation, rotation, look-around, and zooming is operated. In experiments, the proposed method was tested eth 'Danopungjun' that is one of famous paintings made in Chosun Dynasty. The free-viewpoint software is developed based on MFC Visual C++ and OpenGL libraries.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.07a
• /
• pp.116-118
• /
• 2010

본 논문에서는 영역 분할과 영상의 움직임 정보를 이용한 깊이맵 생성에 관한 기법을 제안하였다. 2D/3D 변환 알고리즘에서 2차원 영상에서 얻은 깊이 정보는 2차원 영상을 3차원 영상으로 변환 가능하게 하는 핵심 기술이 된다. 영역을 분할하고 계산되어진 움직임 값 (intensity)을 분할된 각 영역에 부여함으로서 깊이맵을 얻을 수 있다. 본 논문에서는 초기 단계에서 영역을 분할한 뒤, 입력 영상을 그룹화 하여 양방향 탐색을 통한 움직임 추정 연산을 수행토록 하여 보다 정확한 깊이 정보를 획득하고, 최종적으로 얻은 결과에 각 화소에 해당 되는 확률적 통계에 의한 후처리 기법을 사용하였다. 보다 정확한 깊이정보를 영역별로 지정하고, 후처리 기법을 사용함에 따라 보다 신뢰도 높은 깊이맵 영상을 생성할 수 있었다.

• Smart Media Journal
• /
• v.2 no.4
• /
• pp.27-33
• /
• 2013

Multi-view video system provides both realistic 3D feelings and free-view navigation. But it is hard to transmit too huge data, so we send only two or three view images and generate intermediate view image using depth information. In this paper, we propose high quality multi-view image generation method from stereoscopic images. Since the stereo matching method does not provide accurate disparity values for all the pixels, especially at the occlusion area, we propose an occlusion handling method using the background pixels at first. We also apply a joint bilateral filtering to enhance the disparity map at the object boundary since it can affect the quality of synthesized images significantly. Finally, we can generate virtual view images at intermediate view positions using confidence map to reduce bad pixel and hole's error. Experimental results show the proposed method performs better than the conventional method.

• Korean Journal of Optics and Photonics
• /
• v.20 no.1
• /
• pp.16-22
• /
• 2009

We propose a method to convert a two-dimensional movie to a three-dimensional movie using normalized cut and optical flow. In this paper, we segment an image of a two-dimensional movie to objects first, and then estimate the depth of each object. Normalized cut is one of the image segmentation algorithms. For improving speed and accuracy of normalized cut, we used a watershed algorithm and a weight function using optical flow. We estimate the depth of objects which are segmented by improved normalized cut using optical flow. Ordinal depth is estimated by the change of the segmented object label in an occluded region which is the difference of absolute values of optical flow. For compensating ordinal depth, we generate the relational depth which is the absolute value of optical flow as motion parallax. A final depth map is determined by multiplying ordinal depth by relational depth, then dividing by average optical flow. In this research, we propose the two-dimensional/three-dimensional movie conversion method which is applicable to all three-dimensional display devices and all two-dimensional movie formats. We present experimental results using sample two-dimensional movies.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.12
• /
• pp.91-97
• /
• 2013

Image and it's processing techniques are widely applied and very important in the recent IT environment. In this paper, we try to reconstruct original BMP(Bitmap) image into indexed image and codebook utilizing vector quantization and represent high quality image only with same pixel depth of previous indexed image like JPEG etc. That is, By adopting common map method onto index image with $2^n$ color codebook, image can be represented as high quality as $2^{n+1}$ color codebook. When proposed output image is compared with original BMP image, it provides as much around 2dB as higher PSNR than conventional 8 bit index image(normal JPEG). Furthermore, this improvement(2 dB higher PSNR) could be provided when using the 9 bit indexed image.

• Proceedings of the Korean Geotechical Society Conference
• /
• 2010.03a
• /
• pp.687-698
• /
• 2010

Borehole drilled depend on the point is bound to be limited to obtain the 2-D or 3-D layer information for entire targer area. On the other hand, SASW and MASW provide the sectional form of layer information through the shear wave velocity($V_s$). Therefore the useful information of the target area can be derived from SASW, MASW and borehole data. In this research, the correlation reflected locality and nationwide between sectional geo-layer and $V_s$ was investigated and analyzed. The target areas are westside of Pyeongtaek and Incheon. The shear wave velocity($V_s$) obtained from SASW, MASW and borehole data conducted within the scope of crossline for survey was utilized in each region. In the 2D distribution of $V_s$ from SASW, MASW, $V_s$ tend to continually increase deeper and deeper. By the target area, the depth of each representative geo-layer was nested on the sectional distribution map of $V_s$ to suggest the range of $V_s$ in accordance of strata by using borehole data. The 2D sectional geo-layer distribution map is presented based on the range of $V_s$. In addition the correlation between measured and calculated $V_s$ according to the empirical equation was analyzed.