• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 1998.11a
• /
• pp.181-185
• /
• 1998

Recently, research and developmental direction of communication system is concurrent adopting voice data and face image in speaking to provide more higher recognition rate then in the case of only voice data. Therefore, we present a method of lipreading in speech image sequence by using the 3-D facial shape model. The method use a feature information of the face image such as the opening-level of lip, the movement of jaw, and the projection height of lip. At first, we adjust the 3-D face model to speeching face image sequence. Then, to get a feature information we compute variance quantity from adjusted 3-D shape model of image sequence and use the variance quality of the adjusted 3-D model as recognition parameters. We use the intensity inclination values which obtaining from the variance in 3-D feature points as the separation of recognition units from the sequential image. After then, we use discrete HMM algorithm at recognition process, depending on multiple observation sequence which considers the variance of 3-D feature point fully. As a result of recognition experiment with the 8 Korean vowels and 2 Korean consonants, we have about 80% of recognition rate for the plosives and vowels.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.05a
• /
• pp.89-89
• /
• 2004

결정적인 알고리즘과 입력정보의 사용은 평가된 해석과 실제 시스템 값과의 차이로 잘못된 결론을 이끌지도 모른다. 실제 시스템은 대부분 각각의 입력 매개변수들(input parameters)과 관계된 넓은 공차 영역(tolerance band)을 가지고 있어서 입력정보로 하나의 단일한 값을 할당하는 것이 어렵다. 단일 입력에 대한 한가지 해는 변동의 이해 없이 제한된 값이라는 것을 인식할 필요가 있는데 대개 결정론적 설계는 형상과 관련된 치수변동, 항복강도나 부재의 밀도, 탄성계수와 같은 재료 물성치의 불확실성(uncertainty)과 시스템에 작용하는 하중의 변동 등을 직접 고려하지 않고 설계를 수행하기 때문에 수용할 수 있는 오차의 범위 안에서 시스템의 응답을 정확히 평가하기가 쉽지 않았다.(중략)

• Journal of Digital Contents Society
• /
• v.5 no.3
• /
• pp.176-179
• /
• 2004

Superquadrics can represent various and complex 3D object with only wome parameters(size, position, deformation etc). So if we use both superquadrics and deformed superquadrics, we can reoresent more realistic 3D object which are existed in real world. In this paper we use the Z-buffer algorithm and stencil buffer together because this is very useful when the wuperquadric primitives are combined. The fundamental ideas are illustrated with a number of tables and figures.

• Journal of Korean Institute of Industrial Engineers
• /
• v.22 no.3
• /
• pp.441-458
• /
• 1996

Many researches on feature recognition have been performed up to now, but the general solution for recognizing arbitrary features has not been developed. The most popular research area in feature recognition is automatic extraction of 2.5 dimensional features, because they are frequently used in manufacturing field. In this paper, a faster and more convenient 2.5 dimensional feature recognition algorithm is proposed using a new strategy which is quite different from the existing ones. The proposed algorithm takes process planning into consideration. The algorithm is implemented in C++. By applying the algorithm to practical complicate examples, we verify that the algorithm is working very well.

• Proceedings of the KIEE Conference
• /
• 1987.07a
• /
• pp.260-264
• /
• 1987

Algorithms to detect in real-time both position and orientation of rectangular type electronic components are developed for industrial vision. In order to conduct detection in real-time, parallel processing algorithm of image date which uses several control processor is proposed. Image processing area is divided into several regions which can be processed by each cpu. As a result, processing time is improved when two control processors are used and real-time pattern recognition of not-well-aligned components is accomplished.

• Journal of Korea Society of Industrial Information Systems
• /
• v.7 no.4
• /
• pp.29-36
• /
• 2002

In this paper, a new edge detection method for area images is presented based on the scan-line image segmentation technology. The existing algorithms are lack of precision in its detections due to the noise factors such as depth perception and illumination problems when processing the 3D image into a 2D image. The general process of applying the scan-line method is to extract straight line components to determine the shape of the objects. However, we implement this method to an arc curve for precise detections. the paper proved precise detections that from off line to on line.

• Journal of Korea Foundry Society
• /
• v.41 no.6
• /
• pp.521-527
• /
• 2021

This study presents a process management method for the detection of casting defects during in high-pressure die casting based on machine learning. The model predicts the defects of the next cycle by extracting the features appearing over the previous cycles. For design of the gearbox, the proposed model detects shrinkage defects with data from three cycles in advance with 98.9% accuracy and 96.8% recall rates.

• Journal of Sensor Science and Technology
• /
• v.32 no.5
• /
• pp.300-306
• /
• 2023

A hull-cleaning robot sticks to the surface of a vessel and moves for efficient cleaning. Precise path planning and tracking using the current position is crucial. Many robots rely on the INS algorithm, but errors accumulate. To fix this, GPS, sonar, and USBL are used, though with limitations. Selecting suitable sensors for the surface operation and accurate positioning algorithm are vital. In this study, we developed a robot position estimation algorithm using the structure of a ship. Problems that arise when expanding the 2D position estimation algorithm used in existing wall structures to 3D were evaluated and methods for solving them were proposed. In addition, we aimed to improve performance by deriving singularities that exist in the robot path and proposing an error correction algorithm based on the singularities.

• Journal of Digital Convergence
• /
• v.16 no.3
• /
• pp.303-309
• /
• 2018

Computer vision technology is a component inspection to obtain a video image from the camera to the machine to perform the capabilities of the human eye with a field of artificial intelligence, and then analyzed by the algorithm to determine to determine the good and bad of production parts It is widely applied. Shape inspection method was used as how to identify the location of the start point and the end point of the search range, measure the height to the line scan method, in such a manner as to determine the presence or absence of the bolt tabs average brightness of the inspection area in a circular scan type value And the degree of similarity was calculated. The total time it takes to test in the test performance tests of two types of bolts tab enables test 300 min, and demonstrated the accuracy and efficiency of the inspection on the production line represented a complete inspection accuracy.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.21 no.10
• /
• pp.1892-1897
• /
• 2017

This paper proposes a finger counting algorithm for a hand with stuck fingers. The proposed algorithm is based on the fact that straight line type shadows are inevitably generated between fingers. It divides the hand region into the thumb region and the four fingers region for effective shadow detection, and generates an edge image in each region. Projection curves are generated by appling a line detection and a projection technique to each edge image, and the peaks of the curves are detected as candidates for finger shadows. And then peaks due to finger shadows are extracted from them and counted. In the finger counting experiment on hand images expressing various shapes with stuck fingers, the counting success rate is from 83.3% to 100% according to the number of fingers, and 93.1% on the whole. It also shows that if hand images are generated under controlled conditions, the failure cases can be sufficiently improved.