• Proceedings of the IEEK Conference
• /
• 2003.11a
• /
• pp.171-174
• /
• 2003

In this paper, we propose the method of stereo images composition using adaptive dense disparity estimation. For the correct composition of stereo image and 3D virtual object, we need correct marker position and depth information. The existing algorithms use position information of markers in stereo images for calculating depth of calibration object. But this depth information may be wrong in case of inaccurate marker tracking. Moreover in occlusion region, we can't know depth of 3D object, so we can't composite stereo images and 3D virtual object. In these reasons, the proposed algorithm uses adaptive dense disparity estimation for calculation of depth. The adaptive dense disparity estimation is the algorithm that use pixel-based disparity estimation and the search range is limited around calibration object.

• The Journal of the Acoustical Society of Korea
• /
• v.14 no.6
• /
• pp.74-79
• /
• 1995

This paper proposes the stereo LVFC-Robot which Imitates eyes and arms of man. we derived linear approximation equation between visual space and joint space by minimum square method and then applied it to the proposed stereo LVFC-Robot. As a result of a simulation, its efficency is verified. Compared with the stereo VFC, the stereo LVFC Robot don't need the Image Jacobian and the Robot Jacobian. Thus it is possible to control Robot in real time.

• Proceedings of the KOSOMBE Conference
• /
• v.1998 no.11
• /
• pp.265-266
• /
• 1998

This paper describes a depth extraction algorithm in the stereo endoscopic images using adaptive window. First, The relation between the 3D coordinates in the world and the 2D coordinates in the image plane is estimated using camera calibration. Next, stereo matching is performed to find the conjugate pairs in the left and right images. To improve the precision of the matching result, adaptive window which can be varied on the shape as well as on the size according to the area characteristics is used. Finally, the result from the stereo matching and that of camera modeling are combined to extract the real depth information.

• Transactions of the Korean Society of Machine Tool Engineers
• /
• v.17 no.3
• /
• pp.71-77
• /
• 2008

This paper presents a stereo vision based autonomous inspection system for welding quality control of a RV(Recreational Vehicle) sinking seat. The three dimensional geometry of the welding bead, which is the welding quality criteria, is measured by using the captured stereo images with a median filter applied on it. The image processing software for the system was developed using the NI LabVTEW software with NI vision system. In the manufacturing process of a RV sinking seat, the developed system can be used for overcoming the precision error that arises from a visible inspection by an operator. The welding quality inspection system for RV sinking seat was verified using experimentation.

• 한국정보디스플레이학회:학술대회논문집
• /
• 2000.01a
• /
• pp.79-80
• /
• 2000

General stereo vision system shows things in 3D, using two visions of left and right side. When the viewpoints of left/right sides are not in accord with each other, it gives fatigue to human eyes and prevents them from having the 3-D feeling. Also, it would be difficult to track mobile objects that are not in the middle of a screen. Therefore, the object tracking function of stereo vision system is to control tracking objects to always be in the middle of a screen while controlling convergence angles of mobile objects in the input image of the left/right cameras. In this paper, object-tracker in stereo vision system is presented which would track mobile objects by using block matching algorithm of preprocessing and JTC.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.6 no.3
• /
• pp.349-354
• /
• 2011

In this paper we propose a simple efficient stereo matching algorithm to recover sharp object boundaries and to obtain dense disparity map using new multiple line shape windows. To this end, we consider left-right consistency and unique constraint. From the experimental results it is found that the proposed algorithm is very good for obtaining sharp and dense disparity maps for stereo image pairs.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.28.2-28
• /
• 2001

Three dimensional shape measurement techniques are widely needed in industries for product quality monitoring and control. X-ray imaging method is a promising technology to achieve three-dimensional Information, both the surface and inner structure of an object, since it can overcome the limitations of conventional visual or optical methods such as an occlusion problem or surface reflection properties. In this paper, we propose three dimensional volume reconstruction method based on x-ray stereo imaging technology. Here, the stereo images of an object from two different views are taken by changing the object pose rather than moving imaging plane as in conventional stereo vision method. We propose a series of image processing techniques to extract the features efficiently from x-ray images, where the occluded features in case of normal camera vision could be found ...

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.5
• /
• pp.179-186
• /
• 2021

For stereo matching based on deep learning, the design of network structure is crucial to the calculation of matching cost, and the time-consuming problem of convolutional neural network in image processing also needs to be solved urgently. In this paper, a method of stereo matching using sparse loss volume in parallax dimension is proposed. A sparse 3D loss volume is constructed by using a wide step length translation of the right view feature map, which reduces the video memory and computing resources required by the 3D convolution module by several times. In order to improve the accuracy of the algorithm, the nonlinear up-sampling of the matching loss in the parallax dimension is carried out by using the method of multi-category output, and the training model is combined with two kinds of loss functions. Compared with the benchmark algorithm, the proposed algorithm not only improves the accuracy but also shortens the running time by about 30%.

• Journal of Sensor Science and Technology
• /
• v.33 no.2
• /
• pp.98-104
• /
• 2024

This study proposes a method to overcome the limited detection range of short-baseline stereo cameras (SBSCs). The proposed method includes two steps: (1) predicting an unscaled initial depth using monocular depth estimation (MDE) and (2) adjusting the unscaled initial depth by a scale factor. The scale factor is computed by triangulating the sparse visual keypoints extracted from the left and right images of the SBSC. The proposed method allows the use of any pre-trained MDE model without the need for additional training or data collection, making it efficient even when considering the computational constraints of small platforms. Using an open dataset, the performance of the proposed method was demonstrated by comparing it with other conventional stereo-based depth estimation methods.

• Journal of Korean Ophthalmic Optics Society
• /
• v.21 no.1
• /
• pp.61-68
• /
• 2016

Purpose: This study investigated differences in stereoscopic fusional area between those with good and poor stereo acuity in viewing stereoscopic displays. Methods: Stereo acuity of 39 participants (18 males and 21 females, $23.6{\pm}3.15years$) was measured with the random dot stereo butterfly method. Participants with stereo-blindness were not included. Stereoscopic fusional area was measured using stereoscopic stimulus by varying the amount of horizontal disparity in a stereoscopic 3D TV. Participants were divided into two groups of good and poor stereo acuity. Criterion for good stereo acuity was determined as less than 60 arc seconds. Measurements arising from the participants were statistically analyzed. Results: 26 participants were measured to have good stereo acuity and 13 participants poor stereo acuity. In case of the stereoscopic stimulus farther than the fixation point, threshold of horizontal disparity for those with poor stereo acuity were measured to be smaller than the threshold for those with good stereo acuity, with a statistically significant difference. On the other hand, there was no statistically significant difference between the two groups, in case of the stereoscopic stimulus nearer to the fixation point. Conclusions: In viewing stereoscopic displays, the boundary of stereoscopic fusional area for the poor stereo acuity group was smaller than the boundary of good stereo acuity group only for the range behind the display. Hence, in viewing stereoscopic displays, participants with poor stereo acuity would have more difficulty perceiving the fused image at farther distances compared to participants with good stereo acuity.