• Journal of Broadcast Engineering
• /
• v.9 no.3
• /
• pp.185-195
• /
• 2004

This paper presents a multi-depth map fusion method for the 3D scene reconstruction. It fuses depth maps obtained from the stereo matching technique and the depth camera. Traditional stereo matching techniques that estimate disparities between two images often produce inaccurate depth map because of occlusion and homogeneous area. Depth map obtained from the depth camera is globally accurate but noisy and provide a limited depth range. In order to get better depth estimates than these two conventional techniques, we propose a depth map fusion method that fuses the multi-depth maps from stereo matching and the depth camera. We first obtain two depth maps generated from the stereo matching of 3-view images. Moreover, a depth map is obtained from the depth camera for the center-view image. After preprocessing each depth map, we select a depth value for each pixel among them. Simulation results showed a few improvements in some background legions by proposed fusion technique.

• Journal of Korean Society for Geospatial Information Science
• /
• v.21 no.2
• /
• pp.11-17
• /
• 2013

Nowadays 3D models are used in diverse sectors. The 3D maps provide better reality than existing plane maps as well as diverse pieces of information that cannot be expected from the limited plane maps. A process proposed in this paper enables easy and quick production by replacing the expensive laser scanners for modeling by an improved digital camera stereo matching algorithm. The algorithm used in this study was a SURF algorithm contained in the OpenCV library. The unconformity points of the algorithm were eliminated using the homography conversion and epipolar lines. In addition, the improved algorithm was compared with the commercial program, and it showed a better performance than the commercial program. It is expected that the proposed method can contribute to the digital maps and 3D virtual reality because it enables easy and quick 3D modeling provided that the stereo matching conditions are met.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2013.05a
• /
• pp.635-637
• /
• 2013

This paper proposed a new hardware architecture for stereo matching in real time. We minimized the amount of calculation and the number of memory accesses through analyzing calculation of stereo matching. From this, we proposed a new stereo matching calculating cell and a new hardware architecture by expanding it in parallel, which concurrently calculates cost function for all pixels in a search range. After expanding it, we proposed a new hardware architecture to calculate cost function for 2-dimensional region. The implemented hardware can be operated with minimum 250Mhz clock frequence in FPGA environment, and has the performance of 813fps in case of the search range of 64 pixels and the image size of $640{\times}480$.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.2
• /
• pp.99-105
• /
• 2017

Generally, stereo matching methods are used to estimate depth information based on color and spatial similarity. However, most depth estimation methods suffer from the occlusion region because occlusion regions cause inaccurate depth information. Moreover, they do not consider the temporal dimension when estimating the disparity. In this paper, we propose a temporal stereo matching method, considering occlusion and disregarding inaccurate temporal depth information. First, we apply a global stereo matching algorithm to estimate the depth information, we segment the image to occlusion and non-occlusion regions. After occlusion detection, we fill the occluded region with a reasonable disparity value that are obtained from neighboring pixels of the current pixel. Then, we apply a temporal disparity estimation method using the reliable information. Experimental results show that our method detects more accurate occlusion regions, compared to a conventional method. The proposed method increases the temporal consistency of estimated disparity maps and outperforms per-frame methods in noisy images.

• Journal of IKEEE
• /
• v.28 no.1
• /
• pp.65-71
• /
• 2024

This paper proposes a neural network that can reduce the number of parameters while reducing matching errors in occluded regions to increase the accuracy of depth maps in stereo matching. Stereo matching-based object recognition is utilized in many fields to more accurately recognize situations using images. When there are many objects in a complex image, an occluded area is generated due to overlap between objects and occlusion by background, thereby lowering the accuracy of the depth map. To solve this problem, existing research methods that create context information and combine it with the cost volume or RoIselect in the occluded area increase the complexity of neural networks, making it difficult to learn and expensive to implement. In this paper, we create a depthwise seperable neural network that enhances regional feature extraction before cost volume generation, reducing the number of parameters and proposing a neural network that is robust to occlusion errors. Compared to PSMNet, the proposed neural network reduced the number of parameters by 30%, improving 5.3% in color error and 3.6% in test loss.

• Proceedings of the KIEE Conference
• /
• 1999.07g
• /
• pp.2941-2943
• /
• 1999

The application of the aerial images are to find the 3-D elevations. Image matching techniques such as Multi-resolution techniques, WCC (Weighted Cross-Correlation), NSSR (Narrow Search Sub-pixel Registration) that we know robustly apply to images which have enough features. But the method is not adaptive in images which have not enough features due to increasing of disparity errors. In this paper, we propose Disparity Interpolation that decrease disparity errors occurring in the area where images have not enough features. By using real aerial images we compare the result from existing image matching techniques to the result from proposed method.

• Journal of KIISE
• /
• v.44 no.9
• /
• pp.932-945
• /
• 2017

In this paper, a new hardware and software architecture for a stereo vision processing system including rectification, disparity estimation, and visualization was developed. The developed method, named LArge scale stereo matching method using Support Point Interpolation (LASPI), shows excellence in real-time processing for obtaining dense disparity maps from high quality image regions that contain high density support points. In the real-time processing of high definition (HD) images, LASPI does not degrade the quality level of disparity maps compared to existing stereo-matching methods such as Efficient LArge-scale Stereo matching (ELAS). LASPI has been designed to meet a high frame-rate, accurate distance resolution performance, and a low resource usage even in a limited resource environment. These characteristics enable LASPI to be deployed to safety-critical applications such as an obstacle recognition system and distance detection system for autonomous vehicles. A Field Programmable Gate Array (FPGA) for the LASPI algorithm has been implemented in order to support parallel processing and 4-stage pipelining. From various experiments, it was verified that the developed FPGA system (Xilinx Virtex-7 FPGA, 148.5MHz Clock) is capable of processing 30 HD ($1280{\times}720pixels$) frames per second in real-time while it generates disparity maps that are applicable to real vehicles.

• Journal of Institute of Control, Robotics and Systems
• /
• v.3 no.3
• /
• pp.297-304
• /
• 1997

In stereo vision, camera modeling is very important because the accuracy of the three dimensional locations depends considerably on it. In the existing stereo camera models, two camera planes are located in the same plane or on the optical axis. These camera models cannot be used in the active vision system where it is necessary to obtain two stereo images simultaneously. In this paper, we propose four kinds of stereo camera models for active stereo vision system where focal lengths of the two cameras are different and each camera is able to rotate independently. A single closed form solution is obtained for all models. The influence of the stereo camera model to the field of view, occlusion, and search area used for matching is shown in this paper. And errors due to inaccurate focal length are analyzed and simulation results are shown. It is expected that the three dimensional locations of objects are determined in real time by applying proposed stereo camera models to the active stereo vision system, such as a mobile robot.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.28 no.4
• /
• pp.463-469
• /
• 2010

Since MOLA(Mars Orbiter Laser Altimeter) data, which provides altimetry data for Mars, does not cover the whole Mars area, image matching with MOC imagery should be implemented for the generation of DEM. However, automatic image matching is difficult because of insufficient features and low contrast. In this paper, we present the area based semi-automatic image matching algorithm with MOC-NA(Mars Orbiter Camera ? Narrow Angle) imagery. To accomplish this, seed points describing conjugate points are manually added for the stereo imagery, and interesting points are automatically produced by using such seed points. Produced interesting points being used as initial conjugate points, area based image matching is implemented. For the points which fail to match, the locations of initial conjugate points are recalculated by using matched six points and image matching process is re-implemented. The quality assessment by reversing the role of target and search image shows 97.5 % of points were laid within one pixel absolute difference.

• Korean Journal of Remote Sensing
• /
• v.11 no.3
• /
• pp.61-70
• /
• 1995

In this paper, the algorithm to extract the height of flat-rooftop structures in stereo aerial image is suggested with an assumption that location, orientation, focal length, and field of view of a camera are known. It can be adapted to stereo aerial or satellite images. For performing feature-based stereo matching, the line segments suitable to describe the shape of general buildings are chosen as the feature. This paper is composed of three categories;the first step is to extract edges of structures with the polygon extraction algorithm which utilizes the edge following method, the second step is to perform the line segment matching with the camera information, and the last step is to calculate the location of each matched line and to estimate heights. The stereo images used in experiments are not real but synthetic ones. The experiment shows good results.