라인스케너 기반 표면 정밀 검사기기의 개발

박창목,왕지남 대한설비관리학회 1998 대한설비관리학회지 Vol.3 No.1

A method of detecting the defects of floating seal is described. The proposed automatic inspection system consists of a high resolution line scan camera and a microcomputer. A 4096×4096 picture is taken from the line scan camera is 8-bit digital form and stored in the memory of the microcomputer. To reduce the processing area, the computer find the ROI(Region of interests). For high speed inspection, the computer uses a real time processor. There are two types of defects: shape defects and surface defects. Shape defects are fins, notches and etc. Surface defects are pin-halls, scratches and etc. In off-line, the computer calculates features from the ROI and learns the various defects. Basic experiments are carried out using feature extraction methods

PLC기반 차체조립라인의 안전감시를 위한 진단프로그램 생성에 관한 연구

박창목 대한안전경영과학회 2010 대한안전경영과학회지 Vol.12 No.2

In an automated industry PLC plays a central role to control the manufacturing system. Therefore, fault free operation of PLC controlled manufacturing system is essential in order to maximize a firm's productivity. On the contrary, distributed nature of manufacturing system and growing complexity of the PLC programs presented a challenging task of designing a rapid fault finding system for an uninterrupted process operation. Hence, designing an intelligent monitoring, and diagnosis system is needed for smooth functioning of the operation process. In this paper, we propose a method to continuously acquire a stream of PLC signal data from the normal operational PLC-based manufacturing system and to generate diagnosis model from the observed PLC signal data. Consequently, the generated diagnosis model is used for distinguish the possible abnormalities of manufacturing system. To verify the proposed method, we provided a suitable case study of an assembly line.

랜덤신호를 이용한 오디오 워터마킹

박창목,신승원,김종원,신동환,최종욱 한국방송·미디어공학회 2000 한국방송공학회 학술발표대회 논문집 Vol.2000 No.-

생산자동화시스템 PLC 제어프로그램의 안전성 정형검증에 관한 연구

박창목 대한안전경영과학회 2015 대한안전경영과학회지 Vol.17 No.1

In an automated industry PLC plays a central role to control the automation system. Therefore, fault free operation of PLC controlled automation system is essential in order to maximize a firm’s productivity. A prior test of control system is a practical way to check fault operations, but it is a time consuming job and can not check all possible fault operation. A formal verification of PLC program could be a best way to check all possible fault situation. Tracing the history of the study on formal verification, we found three problems, the first is that a formal representation of PLC control system is incomplete, the second is a state explosion problem and the third is that the verification result is difficult to use for the correction of control program. In this paper, we propose a transformation method to reproduce the control system correctly in formal model and efficient procedure to verify and correct the control program using verification result. To demonstrate the proposed method, we provided a suitable case study of an automation system.

컴퓨터 비젼을 이용한 다양한 볼트제품의 검사시스템 구현

박창목,왕지남 대한설비관리학회 1997 대한설비관리학회지 Vol.2 No.2

We applied computer vision for inspecting a large set of bolts. Vision system is developed and implemented to real bolt manufacturing system. The presented methodology consists of two phase inspection schemes. The first phase is to recognize the type of bolt and to set the inspection specifications. The second phase is to inspect the object For an unknown part, the inspection spectifications are added by operator in off-line. Considering the bolt types and inspection spectifications, inspection algorithm are automatically selected in off-line. Once after the inspection items are specified, an on-line inspection is performed for the next coming parts. For inspection bolt shape, the system use the edge line and corder points of two-dimensional images. Performed results are given to verify the self-setting inspection system.

영상분할과 SIFT을 이용한 작업현장 물체 자동인식에 관한 연구

박창목 대한설비관리학회 2022 대한설비관리학회지 Vol.27 No.2

An efficient SIFT-based object detection method has been described in this paper. Estimating a correct object in real plant is an important part of automated storage and retrieval system. In the conventional method, the RFID or barcode is used. This research proposes a image segmentation and sub-region matching method which can boost the accuracy of sift based image identification. The experiment shows the performance is improved for plant tool recognition and also the method shows the promising result in noisy images.

음소 음향학적 변화 정보를 이용한 한국어 음성신호의 자동 음소 분할

박창목,왕지남 한국음향학회 2001 韓國音響學會誌 Vol.20 No.8

This article is concerned with automatic segmentation for Korean speech signals. All kinds of transition cases of phonetic units are classified into 3 types and different strategies for each type are applied. The type 1 is the discrimination of silence, voiced-speech and unvoiced-speech. The histogram analysis of each indicators which consists of wavelet coefficients and SVF (Spectral Variation Function) in wavelet coefficients are used for type 1 segmentation. The type 2 is the discrimination of adjacent vowels. The vowel transition cases can be characterized by spectrogram. Given phonetic transcription and transition pattern spectrogram, the speech signal, having consecutive vowels, are automatically segmented by the template matching. The type 3 is the discrimination of vowel and voiced-consonants. The smoothed short-time RMS energy of Wavelet low pass component and SVF in cepstral coefficients are adopted for type 3 segmentation. The experiment is performed for 342 words utterance set. The speech data are gathered from 6 speakers. The result shows the validity of the method. 본 논문에서는 발음표기가 주어진 상황에서 음성 신호의 자동 음소 분할에 관한 것이며 음소의 경계를 음소 음향학적인 변화특성에 따라 3가지 형태로 분류하여 각각에 적합한 분할 알고리즘을 개발하였다. 형태 1은 묵음·유성음·무성음간의 분할이며 히스토그램분석으로 구한 문턱 값으로 초기 분할 후, 웨이블릿 계수의 SVF (Spectral Variation Function)를 이용하여 분할하였다. 형태 2는 연속적인 모음의 분할이며 각 모음변화특성을 템플릿으로 구성하여 분할에 활용하였다. 형태 3은 모음과 유성자음 혹은 유성화 자음의 분할이며 특성주파수대역의 진폭변화를 이용하여 후보구간을 정한 후, 캡스트럼 계수의 SVF를 이용하여 최종적인 분할을 수행하였다. 본 실험에서는 분할 성능을 테스트하기 위하여 한국어 PBWSpeech DB에서 342개의 단어를 자동으로 분할한 후, 수작업으로 분할한 결과와 비교하였다. 전체적인 자동 분할 성능은 20 msec내에서 81.5%의 분할성능을 보였다.

Unification of Neural Network with a Hierarchical Pattern Recognition

박창목,왕기남 대한인간공학회 1996 대한인간공학회 학술대회논문집 Vol.1996-2 No.-

Vision System for Inspection a Large Set of Bolts

박창목,박상민,왕지남 대한산업공학회 1996 대한산업공학회 추계학술대회논문집 Vol.1996 No.10

유전자 알고리즘을 이용한 롤기반 생산 절단문제 최적화

박창목 한국경영공학회 2009 한국경영공학회지 Vol.14 No.3

This paper presents an efficient approach for a one-dimensional cutting stock problem for the roll-based manufacturing industry. The problem described in this paper has been investigated by many researchers using integer programming, meta-heuristics, and hybrid method. GA(Genetic Algorithm) is a method of meta-heuristics and we improved the GA method. In the real application, demand should always be satisfied with minimum the trim loss and this means that the solution of the CSP (Cutting Stock Problem) should be guaranteed to meet the demand perfectly. The existing GA approach deals with the shortage of demand as a penalty function, so infeasible solutions degrade the performance during solution search. We suggested an efficient GA approach considering the non-generation of infeasible solutions and a heuristic to determine the production quantity in the evaluation stage without searching the production quantity. We compared our method with a existing GA method to verify the superiority of our method.