• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2014.11a
• /
• pp.218-221
• /
• 2014

기존의 지역기반 스테레오 정합 방법은 정합에 사용하는 정합창 크기에 따라 다양한 결과를 갖게 된다. 특히 사용하는 정합창의 크기가 커질수록 영상의 잡음에 강인하지만 객체의 경계부분이 모호해지는 단점을 갖고 있다. 본 논문은 고정된 크기의 정합창을 사용하는 지역기반 스테레오 정합 방법과 다른 방법을 제안한다. 제안한 방법은 영상의 경계를 검출하는 알고리즘을 이용하여 경계부분에는 작은 크기의 정합창을 사용한 변이값을 적용하고 경계가 없는 부분은 큰 크기의 정합창을 사용하여 얻은 변이값을 적용하도록 하였다. 경계를 검출하는 과정에서 본 논문은 AND 연산을 사용하여 최대한 객체의 테두리만을 나타내는 경계값을 구하도록 하였다. 또한 두 가지 크기의 정합창을 이용함으로써 발생하는 복잡도 증가를 감소시키기 위해 기존의 SAD 연산 알고리즘을 수정하여 복잡도를 감소시켰다. 본 논문에서 사용한 정합창의 크기는 $5{\times}5$와 $15{\times}15$이며 실험결과 제안한 방법은 $15{\times}15$ 정합창을 사용한 결과와 비교하여 변이지도에서 객체의 경계부분은 더 잘 살리면서 수행시간을 줄여 효율적인 정합결과를 얻어냈다.

• The KIPS Transactions:PartB
• /
• v.19B no.2
• /
• pp.113-122
• /
• 2012

This paper presents a new stereo matching technique which is based on the matching of feature windows and feature links. The proposed method uses the FAST feature detector to find image features in stereo images and determines the correspondences of the detected features in the stereo images. We define a feature window which is an image region containing several image features. The proposed technique consists of two matching steps. First, a feature window is defined in a standard image and its correspondence is found in a reference image. Second, the corresponding features between the matched windows are determined by using the feature link technique. If there is no correspondence for an image feature in the standard image, it's disparity is interpolated by neighboring feature sets. We evaluate the accuracy of the proposed technique by comparing our results with the ground truth of in a stereo image database. We also compare the matching accuracy and computation time with two conventional feature-based stereo matching techniques.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.7
• /
• pp.1437-1446
• /
• 2012

In this paper we propose a new stereo matching algorithm which is suitable for application to obtain depth information with high-speed in stereoscopic camera environment. For satisfying these condition we propose a new adaptive stereo matching technique using frequency information in discrete wavelet (DWT) domain and variable matching window. The size of the matching window is selected by analysis of the local property of the image in spatial domain and the feature and scaling factor of the matching window is selected by the frequency property in the frequency domain. For using frequency information we use local DWT and global DWT. We identified that the proposed technique has better peak noise to signal ratio (PSNR) than the fixed matching techniques with similar complexity.

• Broadcasting and Media Magazine
• /
• v.2 no.1
• /
• pp.45-56
• /
• 1997

이 논문에서는 여러대의 카메라를 이용하여, 화면내의 모든 화소에 대해 정확한 깊이정보를 추출하는 알고리듬을 제안한다. 제안방법은 스테레오정합에 기반하고 있다. 스트레오정압에 의해 깊이정보를 추정하는데 있어서는 , 오클루전(occlusion)이 추정에 미치는 악영향을 어떻게 극복하는가가매우 중요한 문제이다. 오클루전문제는 2대의 카메라로는 근본적으로 해결이 불가능하다. 따라서 이논문에서는 5대의 카메라를 사용하여, 각 카메라로부터의 정보를 선택적으로 사용함으로써, 이 문제를 상당히 개선한다. 한편, 스테레오 정합법에서는 정합창의 크기가 추정성능에 영향을 미치는데, 큰 창을 이요하면 잡음내성은 우수하나 깊이가 불연속인 곳에서 오차가 발생하고, 작은창을 이용하면 잡음성능은 저하되나 불연속부에서의 창크기에 의한 오차가 줄어드는, 대차관계가 존재한다. 이 논문에서는 처음에 큰 창으로 추정하고 차츰 창의 크기를 줄여나가는 계층적 방법을 제안하여, 잡음내성을 강화하고 불연속부의 오차를 줄이고 있다. 실험경과를 통하여 제안방법의 성능을 제시하고 타당성을 확인한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2010.11a
• /
• pp.206-208
• /
• 2010

본 논문에서는 최종 변이영상의 정확도를 높이기 위해 영상의 특징점을 이용한 적응적 가변 정합창 방법과 교차 일치성 검사의 신뢰도를 높이는 방법을 제안한다. 제안한 적응적 가변 정합창 방법은 색상정보를 이용하여 영상을 분할하고 분할된 각 영상의 특징점을 찾아 그 특징점들의 유무에 따라 정합창의 크기를 적응적으로 가변시키는 방법이다. 또한 제안한 알고리즘을 GPU를 기반으로 구현하여 연산속도가 평균 128배 빨라졌다.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.14 no.7
• /
• pp.1667-1679
• /
• 2010

This thesis proposes an adaptive variable-sized matching window method using the characteristic points of the image and a method to increase the reliability of the cross-consistency check to raise the correctness of the final disparity image. The proposed adaptive variable-sized window method segments the image with the color information, finds the characteristic points in each segmented image, and varies the size of the matching window according to the existence of the characteristic points inside the window. Also the proposed cross-consistency check method processes the two cases with the cost values corresponding to the best disparity and the second-best disparity: when the cost values themselves are too large and when the difference between the two cost values are too small. The two proposed methods were experimented with the four test images provided by the Middleburry site. As the results from the experiments, the proposed adaptive variable-sized matching window method decreased up to 18.2% of error ratio and the proposed cross-consistency check method increased up to 7.4% of reliability.

• Journal of Broadcast Engineering
• /
• v.2 no.2
• /
• pp.216-224
• /
• 1997

In this paper, we propose a depth information extraction method for intermediate image synthesis. As stereo matching method, MBS(Multiple-Baseline Stereo) method has been proposed, in which the matching accuracy increases by using the multiple camera, but there are some inherent problems such as computational complexity, boundary overreach(BO) in depth map, and occlusion. So, we propose the modified version of MBS so called Multi-Resolution MBS(MR-MBS). Moreover, we also propose an adaptive occlusion area processing technique to improve the accuracy of the depth information in occlusion area.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.15 no.3
• /
• pp.583-588
• /
• 2011

In this paper, we propose an adaptive variable-sized matching window method using the characteristic points of the image and a method to increase the reliability of the cross-consistency check to raise the correctness of the final disparity image. The proposed adaptive variable-sized window method segments the image with the color information, finds the characteristic points inside the window. Also the proposed algorithm implement using a graphic processing unit(GPU). The GPU, we used in this paper is GeForce GTX296 (NVIDIA) and we can use programming based on CUDA. The calculation speed realizes a speed approximately 128 times faster than that of a CPU.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.6A
• /
• pp.610-617
• /
• 2002

In this paper, a new intermediate view reconstruction technique by using a disparity estimation method based-on the adaptive matching window size is proposed. In the proposed method, once the feature values are extracted from the input stereo image, then the matching window size for the intermediate view reconstruction is adaptively selected in accordance with the magnitude of this feature values. That is, coarse matching is performed in the region having smaller feature values while accurate matching is carried out in the region having larger feature values by comparing with the predetermined threshold value. Accordingly, this new approach is not only able to reduce the mismatching probability of the disparity vector mostly happened in the accurate disparity estimation with a small matching window size, but is also able to reduce the blocking effect occurred in the disparity estimation with a large matching window size. Some experimental results on the 'Parts' and 'Piano' images show that the proposed method improves the PSNR about 2.32∼4.16dB and reduces the execution time to about 39.34∼65.58% than those of the conventional matching methods.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.34 no.3C
• /
• pp.279-287
• /
• 2009

In this paper, a hierarchical stereo matching with color information is proposed. To generate an initial disparity map, feature based stereo matching is carried out and to generate a final disparity map, hierarchical stereo matching is carried out. The boundary (edge) region is obtained by segmenting a given image into R, G, B and White components. From the obtained boundary, disparity is extracted. The initial disparity map is generated when the extracted disparity is spread to the surrounding regions by evaluating autocorrelation from each color region. The initial disparity map is used as an initial value for generating the final disparity map. The final disparity map is generated from each color region by changing the size of a block and the search range. 4 test images that are provided by Middlebury stereo vision are used to evaluate the performance of the proposed algorithm objectively. The experiment results show better performance compared to the Graph-cuts and Dynamic Programming methods. In the final disparity map, about 11% of the disparities for the entire image were inaccurate. It was verified that the boundary for the non-contiguous point was clear in the disparity map.