모바일 스테레오 비전 시스템을 위한 다양한 스테레오 정합기법의 오차율 비교

이주영,이광엽 한국전기전자학회 2022 전기전자학회논문지 Vol.26 No.4

In this paper, the matching error rates of modified area-based, energy-based algorithms, and learning-basedstructures were compared for stereo image matching. Census transform (CT) based on region and life propagation(BP) algorithm based on energy were selected, respectively.Existing algorithms have been improved and implementedin an embedded processor environment so that they can be used for stereo image matching in mobile systems. Even in the case of the learning base to be compared, a neural network structure that utilizes small-scale parameterswas adopted. To compare the error rates of the three matching methods, Middlebury’s Tsukuba was selected as atest image and subdivided into non-occlusion, discontinuous, and disparity error rates for accurate comparison. Asa result of the experiment, the error rate of modified CT matching improved by about 11% when compared withthe existing algorithm. BP matching was about 87% better than conventional CT in the error rate. Compared to thelearning base using neural networks, BP matching was about 31% superior. 본 논문에서는 스테레오 영상정합을 위하여 개선된 영역기반, 에너지 기반 알고리즘, 학습기반 구조의 정합 오류율을 비교하였다. 영역기반으로 census transform(CT), 에너지 기반으로 belief propagation(BP) 알고리즘을 선정하였다. 기존 알고리즘을 개선하고 모바일 시스템에서 스테레오 영상정합에 활용가능 하도록 임베디드 프로세서 환경에서 구현하였다. 비교 대상이 되는 학습기반의 경우에 도 적은 규모의 파라메터를 활용하는 신경망 구조를 채택하였다. 세 가지 정합방법의 오류율 비교를 위해 테스트 이미지로 Middlebury 데이터 세트 가운데 Tsukuba를 선정하고 정합 성능의 정확한 비교를 위해 비폐색, 불연속, 시차 오류율 등으로 세분화하였다. 실험 결과 CT 매칭의 오차율은 기존 알고리즘과 수정된 알고리즘으로 비교하였을 때 약 11% 성능 개선되었다. BP 매칭은 오류율에서 기존 CT 에 비하여 약 87% 우수하였다. 신경망을 이용한 학습기반과 비교 하였을 때 BP 매칭이 약 31% 우수함을보였다.

다해상도 가법과 AD-Census를 이용한 유전 알고리즘 기반의 스테레오 정합

홍석근,조석제,Hong, Seok-Keun,Cho, Seok-Je 한국융합신호처리학회 2012 융합신호처리학회 논문지 (JISPS) Vol.13 No.1

Stereo correspondence is the central problem of stereo vision. In this paper, we propose a stereo matching scheme based on a genetic algorithm using a multi-resolution method and AD-Census. The proposed approach considers the matching environment as an optimization problem and finds the disparity by using a genetic algorithm And adaptive chronosome structure using edge pixels and crossover mechanism are employed in this technique. A cost function is composes of certain constraints whice are commonly used in stereo matching. AD-Census measure is applied to reduce disparity error. To increase the efficiency of process, we apply image pyramid method to stereo matching and calculate the initial disparity map at the coarsest resolution. Then initial disparity map is propagated to the next finer resolution, interpolated and performed disparity refinement using local feature vector. We valid our method not only reduces the search time for correspondence compared with conventional GA-based method but also ensures the validity of matching. 스테레오 대응성은 스테레오 비전에서 중요한 문제이다. 본 논문은 다해상도 기법과 AD-Census를 이용한 유전 알고리즘 기반의 스테레오 정합 기법을 제안한다. 정합 환경을 최적화 문제로 간주하여 유전 알고리즘으로 변위를 탐색한다. 그리고 에지 픽셀을 이용한 적응적 염색체 구조와 교배 방식을 적용한다. 비용함수는 스테레오 정합에서 주로 고려할 수 있는 제약 조건으로 구성하였고, 변위오차를 줄이기 위해 AD-Census 척도를 사용하였다. 처리의 효율을 높이기 위해 영상 피라미드 방법을 적용하여 최저해상도에서 최초 변위 도를 계산한다. 그리고 최초 변위도는 다음 해상도 단계로 전파되어, 보간된 후 지역 특징 벡터를 이용하여 정제를 수행한다. 실험을 통해 제안한 방법이 다른 유전 알고리즘 기반 기법들에 비해 변위 탐색 시간을 감소시킬 뿐만 아니라 정합의 타당성을 보증함을 확인하고자 한다.

DEM 작성을 위한 항공사진의 컬러입체정합

이효성 ( Hyo-seong Lee ),안기원 ( Ki-weon Ahn ) 한국환경기술학회 2013 한국환경기술학회지 Vol.14 No.1

입체정합 관련 연구에 있어서 컬러영상보다는 주로 흑백영상이 사용되어 왔다. 본 연구에서는 수치컬러영상으로부터 수치표고모델을 작성을 위하여 영역기반의 표준상관기법을 사용하는 입체정합기법을 연구하였다. 컬러영상을 사용하는 입체정합기법의 유효성을 제시하기 위하여 두 가지 실험을 수행하였는데 첫째는 정합결과의 평가에 사용될 참조영상이 공개되어 있는 Middlebury 데이터세트로 실험을 수행하였고 둘째는 항공사진측량용 UltraCamX 카메라로 촬영된 공간해상도 0.125m의 입체수치컬러영상으로 실험하였다. 첫 번째 실험 결과, 컬러영상을 사용하는 입체정합기법이 흑백영상을 사용하는 경우보다 평균상관계수는 커지지만 정합정확도는 향상되지 않았다. 따라서 실제의 컬러항공영상을 사용하는 두 번째 실험에서는 공간교차이론의 적용시에 발생하는 공간교차오차를 사용하여 정합결과를 필터링 하는 기법을 제안하고 제안된 기법으로 수치표고모델을 작성하였다. Most of stereo matching methods have been limited to panchromatic or mono-spectral image while few papers have dealt with color image matching. To generate a digital elevation model from digital color aerial images, we tried to use a color stereo matching technique based on area-based correlation coefficients. In order to show effectiveness of the color stereo matching technique, the first test was performed using the well-known Middlebury stereo dataset whose ground truth disparity map is available. The second test was then carried out for the large format digital color aerial images acquired from UltraCamX whose ground sampling distance is about 0.125 meters. The test results indicated that while the color matching technique could increase the average correlation coefficients than conventional mono-spectral matching, the matching precision based on the similarity measure or average absolute residual parallax was not improved. Therefore, we proposed a filtering method for reliable matching points having smallest absolute value of residual parallax and we could generate a DEM using the filtered points.

Stereo Disparity Measurement for Binocular Stereo Video Systems

Kunio Takaya 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8

This paper reports the development of a stereo disparity measurement system by the binocular video camera. The system uses a simple optical configuration such that raster lines of the right and left cameras share the same vertical position. This allows 1D search alongthe rasters can line to match local patterns in the RGB color space. In order to deal with the occlusion, a strategy to minimize the number of line crossings among the lines connecting the pixel points in the right and left rasterlines. Small disparity for alarge distance to the object necessitates subpixel stereo matching. A method based on the phase-only cross correlation was in corporated. The system was developed for the Windows platform using VisualC++ and video OCXlibrary. The system is capable of displaying dense disparity maps continuously at a lower video rate.

A Study on Edge Block Matching for Stereo Image Coding

Kim, Sungwook,Lee, Youngwook,Kim, ShinHwan 대구대학교 정보통신연구소 2003 情報通信硏究 Vol.1 No.2

Many compression techniques for stereo pair use block based schemes to remove the redundancy between the left and right images efficiently. General block matching methods remove the redundancy of the stereo pair by obtaining the disparity map of the stereo. In this paper we propose a new compression method using the edge of the stereo pair for edge block matching. Right image can be reconstructed from the left image by using edge block matching, because the edge distributions of the left image and the right image are similar. As the result of the edge block matching interpolation map and decimation map are made and transmitted. It is possible to compress the stereo pair efficiently since these maps have less entropy than that of the disparity map derived from the general block matching.

Post Processing to Reduce Wrong Matches in Stereo Matching

Park, Hee-Ju,Lee, Suk-Bae Korean Society of Surveying 2001 Korean journal of geomatics Vol.1 No.1

Although many kinds of stereo matching method have been developed in the field of computer vision and photogrammetry, wrong matches are not easy to avoid. This paper presents a new method to reduce wrong matches after matching, and experimental results are reported. The main idea is to analyze the histogram of the image attribute differences between each pair of image patches matched. Typical image attributes of image patch are the mean and the standard deviation of gray value for each image patch, but there could be other kinds of image attributes. Another idea is to check relative position among potential matches. This paper proposes to use Gaussian blunder filter to detect the suspicious pair of candidate match in relative position among neighboring candidate matches. If the suspicious candidate matches in image attribute difference or relative position are suppressed, then many wrong matches are removed, but minimizing the suppression of good matches. The proposed method is easy to implement, and also has potential to be applied as post processing after image matching for many kinds of matching methods such as area based matching, feature matching, relaxation matching, dynamic programming, and multi-channel image matching. Results show that the proposed method produces fewer wrong matches than before.

Stereo Matching Based on Least Square

Yu Haihao,Fang Yurun,Kong Leilei,Wang Xin 보안공학연구지원센터 2015 International Journal of Multimedia and Ubiquitous Vol.10 No.2

Least Square method is widely adopted in stereo matching owing to it high precision, but the fact that transformation parameters is obtained by solving linear equations leads to the instability of its solutions and the process of matching oscillates and decreases convergence speed. To overcome this disadvantage, improve convergence speed and keep high precision, this paper provides gradient method to resolve stereo matching. The experiments show that the algorithm is valid and practical.

Local Stereo Matching Using an VariableWindow, Census Transform and an Edge-preserving Filter

Vinh Quang Dinh,Dung Duc Nguyen,Vinh Dinh Nguyen,Jae Wook Jeon 제어로봇시스템학회 2012 제어로봇시스템학회 국제학술대회 논문집 Vol.2012 No.10

In this paper, we propose an alternative blocking-matching approach to the correspondence problem in stereo matching. In blocking-matching algorithms, a local window is used to measure the similarity (or dissimilarity) between pixels of a stereo pair. Although some area-based stereo matching methods have been developed and work well in many kinds of regions such as textureless or object boundary regions, their performance can degrade when working in some types of radiometric conditions. Our proposed algorithm is an improved method that uses a non-parametric transform in the pre-processing step and an edge-preserving filter in the post-processing step. Input images are first pre-processed by the census transform, which makes the proposed method more robust when the image pair is captured in different light sources or camera exposure conditions. The window cost in our approach is computed from the transformed images using the Hamming distance, and the correspondence is finally chosen by a Winner-Takes-All strategy. The experimental results for the Middleburry images show that the proposed method outperforms test local stereo methods in radiometrically different images.

종합적 정합방법을 이용한 광량변화에 강건한 스레테오 정합

이대성(Dae-Sung Lee),이상근(Sang-keun Lee) 대한전자공학회 2017 전자공학회논문지 Vol.54 No.11

실제 스테레오 영상획득과정에서 광량이 변화된 상황에서 변위정보를 추정함에 있어서 어려움을 주는 요소이다. 따라서 본 논문에서는 CIELab와 BDIP(Block difference of inverse probabilities)를 이용하여 6차원의 기술자를 구성하고 Census matching과 Gradient matching이용하여 질감정보를 계산한뒤 앞의 3가지의 종합 정합정보를 이용하여 정합비용 추정 후 WTA(Winner Takes All)를 통해 깊이영상을 만든다. 종합적인 깊이정보를 추정하면서 생긴 오정합률을 줄이기 위해 좌우 교차검사를 실시하고 폐색영역 수정을 통하여 광량변화에도 강건한 스테레오 정합을 제안한다. 실험 결과, 제안하는 방법은 비교하는 방법들보다 강건하게 영상을 획득 하는 것을 확인 했으며 평균 5.02%의 오정합률을 나타냈다. It is a difficult factor to estimate displacement information in a situation where radiometric changed in the actual stereo image acquisition process. In this paper, we construct a six-dimensional descriptor using CIELab and BDIP (Block Difference of Inverse Probabilities), calculate texture information using census matching and gradient matching, After estimation, depth image is made through WTA(Winner Takes All). In order to reduce the misalignment rate caused by the estimation of comprehensive depth information, we perform left and right crossing tests and propose a robust stereo matching to the variation of light intensity by correcting the occlusion region. Experimental results show that the proposed method obtains images more robustly than the comparison methods and shows an average error rate of 5.02%.

Genetic Stereo Matching Algorithm with Locally Adaptive Chromosomes

Han, Kyu-Phil,Lee, Ky-Soo 金烏工科大學校 2000 論文集 Vol.21 No.-

A new stereo matching scheme using a genetic algorithm is presented to improve the depth reconstruction method of stereo vision systems. Genetic algorithms are efficient search methods based on principles of population genetics, i.e. mating, chromosome crossover, gene mutation, and natural selection. The proposed approach considers the matching environment as an optimization problem and finds the optimal solution by using an evolutionary strategy. Accordingly, genetic operators are adapted for the circumstances of stereo matching: 1) an individual is a disparity set. 2) a chromosome has a 2D structure for handling image signals efficiently, and 3) a fitness function is composed of certain constraints which are commonly used in stereo matching. Since the fitness function consists of intensity similarity and disparity smoothness, the matching and relaxation processes are considered at the same time in each generation. In order to acquire a disparity map consistent with the image appearance, a region of the input image, divided by zero-crossing points, is extracted and used in the determination of the chromosome shape. As a result, all chromosomes contain the external image form, and the disparity output coincides with the input image without any modification of the matching algorithm. In addition, an informed gene generation based on intensity difference is applied to reduce the searching space of the genetic operations.