[논문]Deep Learning Based Real-Time Painting Surface Inspection Algorithm for Autonomous Inspection Drone

Chang, Hyung-young; Han, Seung-ryong; Lim, Heon-young

doi:10.14773/cst.2019.18.6.253

[국내논문] Deep Learning Based Real-Time Painting Surface Inspection Algorithm for Autonomous Inspection Drone 원문보기

Corrosion science and technology, v.18 no.6, 2019년, pp.253 - 257

Chang, Hyung-young (KEPCO E&C) , Han, Seung-ryong (KEPCO E&C) , Lim, Heon-young (Department of Unmanned and Autonomous Vehicle Engineering, Kyungwoon University)

Abstract ▼ AI-Helper

A deep learning based real-time painting surface inspection algorithm is proposed herein, designed for developing an autonomous inspection drone. The painting surface inspection is usually conducted manually. However, the manual inspection has a limitation in obtaining accurate data for correct judgement on the surface because of human error and deviation of individual inspection experiences. The best method to replace manual surface inspection is the vision-based inspection method with a camera, using various image processing algorithms. Nevertheless, the visual inspection is difficult to apply to surface inspection due to diverse appearances of material, hue, and lightning effects. To overcome technical limitations, a deep learning-based pattern recognition algorithm is proposed, which is specialized for painting surface inspections. The proposed algorithm functions in real time on the embedded board mounted on an autonomous inspection drone. The inspection results data are stored in the database and used for training the deep learning algorithm to improve performance. The various experiments for pre-inspection of painting processes are performed to verify real-time performance of the proposed deep learning algorithm.

Keyword

표/그림 (5)

그림 Fig. 1 Inspection Drone System Configuration.
그림 Fig. 2 Convolutional Neural Network for Painting Surface Inspection.
그림 Fig. 3 Common Defects in Paint Works.
그림 Fig. 4. Edge Detection Results for Each Paint Defect Case
표 Table 1 Performance Comparison for Each Convolution Neural Network

AI 본문요약
AI-Helper

* AI 자동 식별 결과로 적합하지 않은 문장이 있을 수 있으니, 이용에 유의하시기 바랍니다.

문제 정의

This paper presents an autonomous painting surface inspection method based on deep learning algorithms with an inspection drone system. The inspection drone system is being developed which can inspect painting surface autonomously in real time.

제안 방법

In this paper, the drone as an unmanned mobility system and deep learning algorithms are applied to the extension of visual inspection coverage. The proposed painting in- spection drone system consists of a camera, a drone and embedded boards.
The experiment has been performed to verify the pro- posed deep learning algorithm based on data from painting surface inspection with a drone system. The captured im- ages from camera are used for input data in this system.
The deep learning algorithm operates on the image processing board in real time. The inspection results based on the deep learning algorithm and the original im- age data are transmitted to the server system through LTE (Long-Term Evolution) communication.

대상 데이터

In this paper, the drone as an unmanned mobility system and deep learning algorithms are applied to the extension of visual inspection coverage. The proposed painting in- spection drone system consists of a camera, a drone and embedded boards. The image information for inspection is obtained while the drone is flying autonomously to the inspection area with a camera and image data are trans- mitted to the embedded board.
The dataset is composed of 480 training data and 120 testing data, in which the proportion of data between training and testing is recommended to be 80:20. The image process is implemented with OpenCV library and Tensorflow that are installed in the image processing board [7,8].

성능/효과

The inspection drone system is being developed which can inspect painting surface autonomously in real time. The experiments were performed to verify the proposed method and the results of the previous method (VGGNet) and the proposed method (Edge Detection + VGGNet) were compared each other, showing the effectiveness of the proposed method. We conclude that the proposed method shows a valid improvement in detecting painting defects and applicability to painting surface inspection for an autonomous inspection drone.

참고문헌 (10)

K. Iyshwerya, B. Janani, S. Krithika, and T. Manikandan, Proc. International Conference on Smart Structures and Systems, p. 103, IEEE, Chennai, India (2013).
A. Maheshwar and G. N. N. Kumar, IJRASET, 5, 2583 (2017).
A. Leopoldo, T. Josep, H. Alvaro, and A. Jose, Proc. International Conference on Robotics and Automation, No. 5980570, IEEE, Shanghai, China (2011).
G. Mao, S. Drake, and Brian D. O. Anderson, Proc. Decision and Control Conf., p. 224, IEEE Information (2007).
A. Krizhevsky, I. Sutskever, and G. Hinton, ImageNet classificaition with deep convolutional networks, In NIPS (2012).
Y. Liu, M. M. Cheng, X. Hu, J. Bian, L. Zhang, X. Bai, and J. Tang, IEEE T. Pattern Anal., 41, 1939 (2019).

상세보기
OpenCV (Open Source Computer Vision Library), https://opencv.org/about/ (2019).
Why TensorFlow, https://www.tensorflow.org/about
R. Muthukrishnan and M. Radha, IJCSIT, 3, 259 (2011).
K. Simonyan and A. Zisserman, Proc. IC LR Conf., arXiv:1409.1556, The Hilton San Diego Resort & Spa, CA (2015).

활용도 분석정보

상세보기

다운로드

내보내기

활용도 Top5 논문

해당 논문의 주제분야에서 활용도가 높은 상위 5개 콘텐츠를 보여줍니다.
더보기 버튼을 클릭하시면 더 많은 관련자료를 살펴볼 수 있습니다.

내보내기 구분	파일저장 인쇄 메일전송
구성항목	기본정보 상세정보 관리번호, 논문명, 저널/프로시딩명, 저자 , 발행년, 권, 호, 시작페이지, 끝페이지, 발행기관 관리번호, 논문명, 대등논문명, 저자 , 저널/프로시딩명, 발행기관, 발행년, 발행언어, 권, 호, 시작페이지, 끝페이지, ISBN, ISSN, 주제분야, 키워드, 초록(한글), 초록(영문), 저자(소속기관)
저장형식	Text(ASCII format) Excel format RefWorks Direct Export RIS format (for Reference Manager, ProCite, EndNote), Scholar's Aids, Mendeley
메일정보	받는사람 (필수) @ 보내는사람 (선택) @ 제목 내용 KISTI 검색결과 이메일 서비스
안내	총 건의 자료가 검색되었습니다. 다운받으실 자료의 인덱스를 입력하세요. (1-10,000) 검색결과의 순서대로 최대 10,000건 까지 다운로드가 가능합니다. 데이타가 많을 경우 속도가 느려질 수 있습니다.(최대 2~3분 소요) 다운로드 파일은 UTF-8 형태로 저장됩니다. 파일의 내용이 제대로 보이지 않을실 때는 웹브라우저 상단의 보기 -> 인코딩 -> 자동선택 여부를 확인하십시오. ~ Text(ASCII format) Excel format

연합인증