• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2017.05a
• /
• pp.123-126
• /
• 2017

The information and communication technology that is being developed recently has been greatly influencing the automobile market. In recent years, devices equipped with IT technology have been installed for the safety and convenience of the driver. However, it has the advantage of increased convenience as well as the disadvantage of increasing traffic accidents due to driver 's distraction. In order to prevent such accidents, it is necessary to develop safety systems of various types and ways. In this paper, we propose a method to implement a multi-function camera driving safety system that notifies a pedestrian and lane departure warning without using a radar sensor or a stereo video image, and a study on the analysis of a lane departure alarm software result.

• ETRI Journal
• /
• v.33 no.4
• /
• pp.506-516
• /
• 2011

In this paper, we develop a registration method to eliminate the geometric inconsistency between the stereo-images and light detection and ranging (LIDAR) data obtained by an airborne multisensor system. This method consists of three steps: registration primitive extraction, correspondence establishment, and exterior orientation parameter (EOP) adjustment. As the primitives, we employ object points and linked edges from the stereo-images and planar patches and intersection edges from the LIDAR data. After extracting these primitives, we establish the correspondence between them, being classified into vertical and horizontal groups. These corresponding pairs are simultaneously incorporated as stochastic constraints into aerial triangulation based on the bundle block adjustment. Finally, the EOPs of the images are adjusted to minimize the inconsistency. The results from the application of our method to real data demonstrate that the inconsistency between both data sets is significantly reduced from the range of 0.5 m to 2 m to less than 0.05 m. Hence, the results show that the proposed method is useful for the data fusion of aerial images and LIDAR data.

• Proceedings of the Korean Society of Surveying, Geodesy, Photogrammetry, and Cartography Conference
• /
• 2007.04a
• /
• pp.201-205
• /
• 2007

In the post-Cold War era, acquisition technique of high-resolution satellite imagery (HRSI) has begun to commercialize. IKONOS-2 satellite imaging data is supplied for the first time in the 21st century. Many researchers testified mapping possibility of the HRSI data instead of aerial photography. It is easy to renew and automate a topographical map because HRSI not only can be more taken widely and periodically than aerial photography, but also can be directly supplied as digital image. In this study matching size of IKONOS Geo-level stereo image is presented lot production of digital elevation model (DEM). We applied area based matching method using correlation coefficient of pixel brightness value between the two images. After matching line (where "matching line" implies straight line that is approximated to complex non-linear epipolar geometry) is established by exterior orientation parameters (EOPs) to minimize search area, the matching is tarried out based on this line. The experiment on matching size is performed according to land cover property, which is divided off into four areas (water, urban land, forest land and agricultural land). In each of the test areas, window size for the highest correlation coefficient is selected as propel size for matching. As the results of experiment, the proper size was selected as $123{\times}123$ pixels window, $13{\times}13$ pixels window, $129{\times}129$ pixels window and $81{\times}81$ pixels window in the water area, urban land, forest land and agricultural land, respectively. Of course, determination of the matching size by the correlation coefficient may be not absolute appraisal method. Optimum matching size using the geometric accuracy therefore, will be presented by the further work.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.10
• /
• pp.35-42
• /
• 2014

It is possible to improve the obstacle avoidance capability by training and recognizing the obstacles which is in certain indoor environment. We propose the technique that use underlying intensity value along with intensity map from RGB-D image which is derived from stereo vision Kinect sensor and recognize an obstacle within constant distance. We test and experiment the accuracy and execution time of the pattern recognition algorithms like PCA, ICA, LDA, SVM to show the recognition possibility of it. From the comparison experiment between RGB-D data and intensity data, RGB-D data got 4.2% better accuracy rate than intensity data but intensity data got 29% and 31% faster than RGB-D in terms of training time and intensity data got 70% and 33% faster than RGB-D in terms of testing time for LDA and SVM, respectively. So, LDA, SVM have good accuracy and better training/testing time to use for obstacle avoidance based on intensity dataset of mobile robot.

• Journal of Broadcast Engineering
• /
• v.16 no.4
• /
• pp.596-601
• /
• 2011

In this paper, we propose a fast depth-image-based rendering method to generate a virtual view image in real-time using a graphic processor unit (GPU) for a 3D broadcasting system. Before the transmission, we encode the input 2D+depth video using the H.264 coding standard. At the receiver, we decode the received bitstream and generate a stereo video using a GPU which can compute in parallel. In this paper, we apply a simple and efficient hole filling method to reduce the decoder complexity and reduce hole filling errors. Besides, we design a vertical parallel structure for a forward mapping process to take advantage of the single instruction multiple thread structure of GPU. We also utilize high speed GPU memories to boost the computation speed. As a result, we can generate virtual view images 15 times faster than the case of CPU-based processing.

• Journal of Broadcast Engineering
• /
• v.18 no.1
• /
• pp.31-42
• /
• 2013

This paper proposes a photogrammetric rectification method based on Bayesian approach as a method that eliminates vertical parallax between stereo images to minimize visual fatigue of 3D contents. The image rectification consists of two phases; geometry estimation and epipolar transformation. For geometry estimation, coplanarity-based relative orientation algorithm was used in this paper. To ensure robustness for mismatch and localization error occurred by automation of tie point extraction, Bayesian approach was applied by introducing several prior constraints. As epipolar transformation perspective transformation was used based on condition of collinearity to minimize distortion of result images and modification for input images. Other algorithms were compared to evaluate performance. For geometry estimation, traditional relative orientation algorithm, 8-points algorithm and stereo calibration algorithm were employed. For epipolar transformation, Hartley algorithm and Bouguet algorithm were employed. The evaluation results showed that the proposed algorithm produced results with high accuracy, robustness about error sources and minimum image modification.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.32 no.1
• /
• pp.39-47
• /
• 2014

In this study, the epipolar images were created by both methods of resolution adjustment and piecewise approach using RPC(Rational Polynomial coefficients) and ancillary data of IKONOS-2 and SPOT-5 satellite images whose resolutions are different from each other. The stereo geometry of these two satellite images was analyzed and the RPC block modelling was accomplished for generating epipolar images. In order to evaluate the accuracy of created epipolar images, the y-parallaxes were analyzed for the specific points which were apparently identified in mountainous, plain and urban area. Also the RMSEs of the specific points were calculated using the coordinates from the epipolar stereo images and the coordinates from the block triangulation. Y-parallaxes were within one pixel and the RMSEs were within two meters for X, Y and Z each.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.29 no.5C
• /
• pp.716-725
• /
• 2004

In this paper, a new multiview stereoscopic 3D display system using a VHLS(volume holographic lenticular sheet) is suggested. The VHLS, which acts just like an optical direction modulator, can be implemented by recording the diffraction gratings corresponding each directional vector of the multiview stereoscopic images in the holographic recording material by using the angularly multiplexed recording property of the conventional volume hologram. Then, this fabricated VHLS is attached to the panel of a LCD spatial light modulator and used to diffract each of the multiview image loaded in a SLM to the corresponding spatial direction for making a 3D stereo view-zone. Accordingly, in this paper, the operational principle and characteristics of the VHLS are analyzed and an optimized 4-view VHLS is fabricated by using a commercial photopolymer. Then, a new VHLS-based 4-view stereoscopic 3D display system is implemented. Through some experimental results using a 4-view image synthesized with adaptive disparity estimation algorithm, it is suggested that implementation of a new VHLS-based multiview stereoscopic 3D display system can be possible.

• Journal of Broadcast Engineering
• /
• v.17 no.5
• /
• pp.789-807
• /
• 2012

The purpose of this paper is to find various disparity factors in conjunction with situations and contexts of the contents that cause viewer's discomfort, which is to make a content production guideline for viewer's safety, eventually. As a methodology, we extract the experimental data by letting various viewers push a pre-defined key as long as they feel discomfort during watching stereo 3D contents. For each contents, we extract the disparities and disparity changes for the focal point, vicinity of the focal point (ROI, region of interest), and the whole image. For each contents, each point or part of contents is analyzed to find the factors causing discomfort with the two prepared data with watching the contents in 3D. Then, all the analyzed data are re-analyzed to find the strength of each factor. The results from the analyses are explained from the factors having relatively high strength first. It includes the factors that are not known as well as the ones that we can already guess.

• Journal of Internet Computing and Services
• /
• v.21 no.2
• /
• pp.9-18
• /
• 2020

In the field of image processing, a method of detecting and counting passengers as moving objects when getting on and off the bus has been studied. Among these technologies, one of the artificial intelligence techniques, the deep learning technique is used. As another method, a method of detecting an object using a stereo vision camera is also used. However, these techniques require expensive hardware equipment because of the computational complexity of used to detect objects. However, most video equipments have a significant decrease in computational processing power, and thus, in order to detect passengers on the bus, there is a need for an image processing technology suitable for various equipment using a relatively low computational technique. Therefore, in this paper, we propose a technique that can efficiently obtain the number of passengers on the bus by detecting the contour of the object through the background subtraction suitable for low-cost equipment. Experiments have shown that passengers were counted with approximately 70% accuracy on lower-end machines than those equipped with stereo vision camera.