• East Asian Economic Review
• /
• 제24권3호
• /
• pp.275-301
• /
• 2020

We analyze state-level matching efficiencies in the Indian labor market using stochastic frontier analysis. The key contribution of this research is the estimation of matching efficiencies at the state level because these can be used for a state-level measure of labor market conditions. Next, we explore the relationship between the estimated matching efficiencies and population density, labor market flexibility, and the Ease of Doing Business index, respectively. The results show that matching efficiency is heterogeneous across states with considerable variation in accordance with the regional diversity in India. However, we find that there is little relationship between the estimated matching efficiencies and the labor market conditions of interest, suggesting that other regional diversity affects matching efficiencies across states in India.

• 한국통신학회논문지
• /
• 제38A권2호
• /
• pp.174-182
• /
• 2013

Most methods for fusion-based finger vein recognition were to fuse different features or matching scores from more than one trait to improve performance. To overcome the shortcomings of "the curse of dimensionality" and additional running time in feature extraction, in this paper, we propose a finger vein recognition technology based on matching score-level fusion of a single trait. To enhance the quality of finger vein image, the contrast-limited adaptive histogram equalization (CLAHE) method is utilized and it improves the local contrast of normalized image after ROI detection. Gabor features are then extracted from eight channels based on a bank of Gabor filters. Instead of using the features for the recognition directly, we analyze the contributions of Gabor feature from each channel and apply a weighted matching score-level fusion rule to get the final matching score, which will be used for the last recognition. Experimental results demonstrate the CLAHE method is effective to enhance the finger vein image quality and the proposed matching score-level fusion shows better recognition performance.

• 방송공학회논문지
• /
• 제20권4호
• /
• pp.641-644
• /
• 2015

본 논문에서는 이진 프레임 기술자와 이를 이용한 프레임 단위 유사중복 동영상 정합 알고리즘을 제안한다. 우선 동영상으로부터 취득한 프레임을 패치(patch)단위로 나누고 패치간의 관계를 이진으로 나타낸다. 그리고 두 동영상 프레임 간의 정합비용과 보상비용으로 비용 함수를 표현한다. 초기 정합과 반복적인 정합 갱신을 통해 비용 함수를 최소화한다. 실험을 통해 제안하는 이진 프레임 기술자의 적합성과 프레임 단위 정합 알고리즘 성능이 우수함을 확인한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권5호
• /
• pp.1597-1610
• /
• 2022

At present, the main method of high-speed train chassis detection is using computer vision technology to extract keypoints from two related chassis images firstly, then matching these keypoints to find the pixel-level correspondence between these two images, finally, detection and other steps are performed. The quality and accuracy of image matching are very important for subsequent defect detection. Current traditional matching methods are difficult to meet the actual requirements for the generalization of complex scenes such as weather, illumination, and seasonal changes. Therefore, it is of great significance to study the high-speed train image matching method based on deep learning. This paper establishes a high-speed train chassis image matching dataset, including random perspective changes and optical distortion, to simulate the changes in the actual working environment of the high-speed rail system as much as possible. This work designs a convolutional neural network to intensively extract keypoints, so as to alleviate the problems of current methods. With multi-level features, on the one hand, the network restores low-level details, thereby improving the localization accuracy of keypoints, on the other hand, the network can generate robust keypoint descriptors. Detailed experiments show the huge improvement of the proposed network over traditional methods.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 1996년도 학술대회
• /
• pp.217-220
• /
• 1996

This paper describes a new hierarchial block matching algorithm especially appropriate for a large search area. The proposed algorithm consists of higher level search for an initial motion vector estimate by using a new matching criterion over the evenly subsampled search points, and lower level search for the final motion vector refinement. In the higher level matching criterion, mean absolute differences at the search points (or motion vector candidates) similar to motion vectors of causally neighboring blocks, are weighted properly so that these points can have a higher chance to being selected. The proposed algorithm outperforms existing hierarchical block matching algorithms, and its computational regularity makes hardware implementation simple.

• 전자공학회논문지S
• /
• 제34S권12호
• /
• pp.70-80
• /
• 1997

This paper proposes a hierarchical Hausdorff distance (HD) matching algorithm baased on coarse-to-fine approach. It reduces the computational complexity greatly by using the pyramidal structures consisting of distance transform (DT) and edge pyramids. Also, inthe proposed hierarchical HD matching, a thresholding method is presented to find an optimal matching position with small error, in which the threshold values are determined by using the property between adjacent level of a DT map pyramid. By computer simulation, the performance of the conventional and proposed hierarchical HD matching algorithms is compared in therms of the matching position for binary images containing uniform noise.

• 정보처리학회논문지B
• /
• 제11B권6호
• /
• pp.633-644
• /
• 2004

본 논문에서는 기존 NTSS 알고리즘을 다중해상도(MR : Multiple Resolution)기법을 이용하여 NTSS-3 레벨 알고리즘으로 제안하였다. 고속 블록정합 알고리즘은 패턴 방식에 따라 속도에 많은 영향을 미치는데 본 논문에서는 기존 NTSS의 패턴 방식과 다른 다중해상도 기법을 이용한 레벨에 따른 블록정합 알고리즘을 제안하였다. 블록 정합알고리즘에서 국부최소화 문제(Local minima problem)로 발생하는 화질 저하를 개선하기 위해 MC(Multiple Candidate)라는 다중후보를 이용하였다. 실험 결과 제안한 기법을 FS와 비교하면 16배의 탐색 속도론 나타내었고 기존 고속 블록 정합 알고리즘인 NTSS 방식에 비교 할 때 PSNR값에 있어서는 0.11-0.12(dB) 화질이 개선되었으며 속도 면에서도 0.1배 향상되었고 탐색점 대비 화질개선이 우수함을 나타내었다.

• Journal of Electrical Engineering and information Science
• /
• 제3권2호
• /
• pp.211-221
• /
• 1998

It is becoming apparent that stereo matching algorithms need much information from high level cognitive processes. Otherwise, conventional algorithms based on bottom-up control alone are susceptible to local minima. We introduce a system that consists of two levels. A lower level, using a usual matching method, is based upon the local neighborhood and a second level, that can integrate the partial information, is aimed at contextual matching. Conceptually, the introduction of bottom-up and top-down feedback loop to the usual matching algorithm improves the overall performance. For this purpose, we model the image attributes using a Markov random field (MRF) and thereupon derive a maximum a posteriori (MAP) estimate. The energy equation, corresponding to the estimate, efficiently represents the natural constraints such as occlusion and the partial informations from the other levels. In addition to recognition, we derive a training method that can determine the system informations from the other levels. In addition to recognition, we derive a training method that can determine the system parameters automatically. As an experiment, we test the algorithms using random dot stereograms (RDS) as well as natural scenes. It is proven that the overall recognition error is drastically reduced by the introduction of contextual matching.

• 반도체디스플레이기술학회지
• /
• 제19권3호
• /
• pp.43-48
• /
• 2020

The demand for stable plasma has been on the rise because of the increased delivery power amount in the chamber for improving productivity, and fast and accurate plasma impedance matching become a crucial performance measure for radio frequency (RF) power system in semiconductor manufacturing equipment. In this paper, the overall impedance matching was understood, and voltage and current values were extracted with voltage - current (VI) probe to measure plasma impedance in real-time. Actual matching data were analyzed to derive calibration coefficient for V and I measurements to understand the characteristics of VI probe, and we demonstrated the tendency of RF impedance matching according to changes in load impedance. This preliminary empirical research can contribute to fast RF matching as well as advanced equipment control for the next level of detailed investigation on embedded system based-RF matching controller.

• Management Science and Financial Engineering
• /
• 제13권1호
• /
• pp.93-113
• /
• 2007

In the previous research, we considered a logistics brokerage problem with the objective of minimizing total transportation lead time of freights in a logistics e-marketplace, in which a logistics brokerage agent intermediates empty vehicles and freights registered by car owners and shippers [7]. However, in the logistics e-marketplace, transportation due date tardiness is more important than the transportation lead time, since transportation service level is critically determined by whether the due date is met or not. Therefore, in this paper, we deal with the logistics brokerage problem with the objective of minimizing total tardiness of freights. Hungarian method based matching algorithms, real time matching(RTM), periodic matching(PM), and fixed matching(FM), are used for solving the problem considered in this paper. In order to test performance of the proposed algorithms, we perform computational experiments on a various problem instances. The results show that the waiting-and-matching algorithms, PM and FM, also give better performance than real time matching strategy, RTM, for the total tardiness minimization problem as the algorithms did for the total lead time minimization problem.