• Journal of Broadcast Engineering
• /
• v.19 no.2
• /
• pp.166-173
• /
• 2014

Recently, human action recognition have been developed for various broadcasting and video process. Since a video can consist of various scenes, keypoint approaches have been more attracted than template based methods for real application. Keypoint approahces tried to find regions having motion in video, and made 3-dimensional patches. Then, descriptors using histograms were computed from the patches, and a classifier based on machine learning method was applied to detect actions in video. However, a single classifier was difficult to handle various human actions. In order to improve this problem, approaches using multi classifiers were used to detect and to recognize objects. Thus, we propose a new human action recognition using decision-level fusion with support vector machine and sparse representation. The proposed method extracted descriptors based on keypoint approach from a video, and acquired results from each classifier for human action recognition. Then, we applied weights which were acquired by training stage to fuse each results from two classifiers. The experiment results in this paper show better result than a previous fusion method.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.1
• /
• pp.23-30
• /
• 2014

Feature matching using image descriptors is robust method used recently. However, mismatches occur in 3D transformed images, illumination-changed images and repetitive-pattern images. In this paper, we observe that there are a lot of mismatches in the images which have repetitive patterns. We analyze it and propose a method to eliminate these mismatches. MDMF(Multiple Descriptors-based Mismatch Filtering) eliminates mismatches by using descriptors of nearest several features of one specific feature point. In experiments, for geometrical transformation like scale, rotation, affine, we compare the match ratio among SIFT, ASIFT and MDMF, and we show that MDMF can eliminate mismatches successfully.

• Journal of Space Technology and Applications
• /
• v.1 no.1
• /
• pp.7-22
• /
• 2021

In this paper, we describe the results of the docking phase test in the ground environment of the rendezvous/docking technology verification satellite under development for the first time in Korea. rendezvous/docking technology is a high-level technology in space technology, which is also very important for accessing and performing tasks on relative objects in space orbit. In this paper, we describe the ground test results that the chaser finally docks the fixed target using an air bearing device. Based on the thrust control algorithm in the docking phase and the relative object recognition and relative distance estimation algorithm using visual-based sensors validated in this paper, we intend to use them for later expansion to rendezvous/docking algorithms in three-dimensional space for testing in space.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.10
• /
• pp.67-75
• /
• 1997

A design for a system to perform rapid recognition of three dimensional objects is presented, focusing on efficient indexing. In order to retrieve the best matched models without exploring all possible object matches, we have employed a bayesian framework. A decision-theoretic measure of the discriminatory power of a feature for a model object is defined in terms of posterior probability. Detectability of a featrue defined as a function of the feature itselt, viewpoint, sensor charcteristics, nd the feature detection algorithm(s) is also considered in the computation of discribminatory power. In order to speed up the indexing or selection of correct objects, we generate and verify the object hypotheses for rfeatures detected in a scene in the order of the discriminatory power of these features for model objects.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.209-212
• /
• 2005

모션 캡쳐 기술은 영화나 게임 혹은 애니메이션과 같은 다양한 분야에서 새롭게 시도되고 있다. 기존의 모션 캡쳐 방식은 마커라는 불편한 인식장치로 인하여 움직임에 적지 않은 제약을 받을 뿐만 아니라 엄청난 고가 장비를 필요로 한다. 갈수록 좀 더 편리한 모션 캡쳐 기술이 요구됨에 따라 실시간으로 물체의 위치를 트래킹(위치추적)하고, 스테레오 비전을 이용하여 3 차원 재 구축을 수행해 입체적인 가상모델을 생성하여 보았다. 본 논문에서는 효율적인 트래킹 연산에 의한 움직임 제약을 최소화한 개선된 알고리즘을 설계, 구현하였고, 저렴한 웹 캠을 이용하여 스테레오 비전방식(Stereo Vision Based)을 접목시켜 기존의 고가 장비들과 유사한 환경을 이 시스템으로도 가능하다는 것을 확인하였다.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.33B no.7
• /
• pp.138-146
• /
• 1996

In this paper, w epropose a new mehtod to represent the shape and to recognize the object. The shape description and the matching is implemented by using the distance transform on the morphological skeleton. The employed distance transform is the chamfer (3,4) distance transform, because the chamfer distance transform (CDT) has an approximate value to the euclidean distance. The 2-D object can be represented by means of the distribution of the distance transform on the morphological skeleton, the number of skeletons, the sum of the CDT, and the other features are employed as the mtching parameters. The matching method has the invariant features (rotation, translation, and scaling), and then the method is used effectively for recognizing the differently-posed objects and/or marks of the different shape and size.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2019.05a
• /
• pp.657-660
• /
• 2019

최근 모바일 디바이스의 발전으로 기기에 탑재된 카메라를 가지고도 눈으로 인식하지 못할 정도로 빠르게 움직이는 물체들의 고화질 촬영이 가능하게 되었다. 보급형 하드웨어의 고성능화는 생활스포츠 환경 전반에서 거쳐 활용도가 높아지고 있으며, 고가의 초고속 카메라 대신 개인용 모바일 디바이스에 탑재된 초고속 카메라를 이용하여 경기장 내 목표물의 움직임의 촬영이 가능하게 되었다. 본 논문에서는 모바일 디바이스를 이용한 경기장의 라인 인/아웃 판정 지원과 3차원 재구성을 위한 판정 분석 도구 개발과 관련된 모바일 디바이스 카메라 캘리브레이션 방법에 대하여 연구하였다.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.50 no.6
• /
• pp.201-211
• /
• 2013

In this paper, we propose a hand tracking and gesture recognition system. Our system employs a depth capture device to obtain 3D geometric information of user's bare hand. In particular, we build a flexible tracking volume and restrict the hand tracking area, so that we can avoid diverse problems caused by conventional object detection/tracking systems. The proposed system computes running average of the hand position, and tracking volume is actively adjusted according to the statistical information that is computed on the basis of uncertainty of the user's hand motion in the 3D space. Once the position of user's hand is obtained, then the system attempts to detect stretched fingers to recognize finger gesture of the user's hand. In order to test the proposed framework, we built a NUI system using the proposed technique, and verified that our system presents very stable performance even in the case that multiple objects exist simultaneously in the crowded environment, as well as in the situation that the scene is occluded temporarily. We also verified that our system ensures running speed of 24-30 frames per second throughout the experiments.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.10
• /
• pp.1853-1859
• /
• 2008

In this paper, we propose a novel method to improve the visual quality of reconstructed images for 3D pattern recognition based on the computational integral imaging reconstruction (CIIR). The proposed CIIR method provides improved 3D reconstructed images by superimposing magnified elemental images by a combination of a round mapping model and image interpolation algorithms. To objectively evaluate the proposed method, we introduce an experimental framework for a computational pickup process and a CIIR process using a Gaussian function and evaluate the proposed method. We also carry out experiments on 3D objects and present their results.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.255-263
• /
• 2022

3D object detection pipelines often incorporate RGB-based object detection methods such as YOLO, which detects the object classes and bounding boxes from the RGB image. However, in complex environments where objects are heavily cluttered, bounding box approaches may show degraded performance due to the overlapping bounding boxes. Mask based methods such as Mask R-CNN can handle such situation better thanks to their detailed object masks, but they require much longer time for data preparation compared to bounding box-based approaches. In this paper, we present a 3D object recognition pipeline which uses either the YOLO or Mask R-CNN real-time object detection algorithm, K-nearest clustering algorithm, mask reduction algorithm and finally Principal Component Analysis (PCA) alg orithm to efficiently detect 3D poses of objects in a complex environment. Furthermore, we also present an improved YOLO based 3D object detection algorithm that uses a prioritized heightmap clustering algorithm to handle overlapping bounding boxes. The suggested algorithms have successfully been used at the Artificial-Intelligence Robot Challenge (ARC) 2021 competition with excellent results.