http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
P&ID의 파이프라인 인식 향상을 위한 라인 검출 개선에 관한 연구
오상진,채명훈,이현,이영환,정은경,이현식,Oh, Sangjin,Chae, Myeonghoon,Lee, Hyun,Lee, Younghwan,Jeong, Eunkyung,Lee, Hyunsik 한국플랜트학회 2020 플랜트 저널 Vol.16 No.4
For several decades, productivity in construction industry has been regressed and it is inevitable to improve productivity for major EPC players. One of challenges to achieve this goal is automatically extracting information from imaged drawings. Although computer vision technique has been advanced rapidly, it is still a problem to detect pipe lines in a drawing. Earlier works for line detection have problems that detected line elements be broken into small pieces and accuracy of detection is not enough for engineers. Thus, we adopted Contour and Hough Transform algorithm and reinforced these to improve detection results. First, Contour algorithm is used with Ramer Douglas Peucker algorithm(RDP). Weakness of contour algorithm is that some blank spaces are occasionally found in the middle of lines and RDP covers them around 17%. Second, HEC Hough Transform algorithm, we propose on this paper, is improved version of Hough Transform. It adopted iteration of Hough Transform and merged detected lines by conventional Hough Transform based on Euclidean Distance. As a result, performance of Our proposed method improved by 30% than previous.
시맨틱 세그멘테이션 기반의 P&ID 심볼 탐지에 대한 연구
오상진(Sangjin Oh),이현(Hyun Lee),이정규(Jeongkyu Lee),이영환(Younghwan Lee),정은경(Eunkyung Jeong),이현식(Hyunsik Lee) (사)한국CDE학회 2020 한국CDE학회 논문집 Vol.25 No.4
Although the Fourth Industrial Revolution comes to reality rapidly, construction industry is going through a difficult time to adopt new technologies. Also, improving productivity is one of the most urgent issues for major construction companies. However, reading information and digitizing them from imaged drawings takes much time and it becomes a reason for low productivity. Thus, in this paper, we propose a method to recognize symbols in P&ID (piping and instrumentation diagram) using neural networks for Semantic Segmentation. First, crop a drawing into small patches and label on them 8 classes of symbol. Then, Train U-net and FCN with 2,500 patches with annotation. After training, results of recognition are displayed with color code on imaged drawings. Finally, we run tests with 5 new P&ID drawings and scored the performance of our recognition models.