전기화재 원인분석을 위한 용융흔 외형 판별 딥러닝 알고리즘 설계
A Deep learning Algorithm Design for the Classification the Shape of Molten Marks for the Cause Analysis of Electric Fires
In this paper, we proposed and verified an algorithm that compares and analyzes the performance analysis of the CNN method and Resnet for the distinguish Arc beads and Molten mark. The data for training was produced through experiments and generated as original data, and then applied to the algorithm to obtain training data and validation data. Unlike learning neural network parameters as a way to reduce CNN errors using Resnet techniques, by introducing Residual learning that utilizes the difference between input and output, it shows a structure that can be efficiently learned using a deeper network layer. As a result of the analysis, the CNN and Resnet classification accuracies were 93.54% and 96.81% respectively.
한국어
본 논문에서는, 단락흔과 열흔 판별을 위한 CNN기법과 ResNet의 성능 분석을 비교 분석하는 알고리즘 을 제안하고 검증하였다. 학습을 위한 데이터는 실험을 통해 제작하여 원본 데이터로 생성한 후, 알고리즘에 적용 시켜 학습데이터와 검증데이터로 확보하였다. Resnet 기법을 활용하여 CNN의 오차를 줄이는 방법으로 신경망의 파라미터를 학습하는 것과는 다르게 입력과 출력의 차이를 이용하는 잔차 학습(Residual learning)을 도입함으로 써, 더욱 깊은 네트워크층을 사용하여 효율적으로 학습 가능한 구조를 나타내고 있다. 분석결과 CNN과 Resnet 판별 정확도는 각각 93,54%, 96.81%로 얻었다.
목차
요약 Abstract Ⅰ. 서론 Ⅱ. 딥러닝 알고리즘을 이용한 용융흔외형 모델 구조 2.1 Machine learning 2.2 용융흔 외형 판별을 위한 알고리즘 설계 Ⅲ. 결론 REFERENCES
키워드
전기 화재단락흔열흔CNNResnet 알고리즘Electric fireArc beadsMolten markCNNResnet
저자
조장훈 [ Jang-hoon Jo | 전북대학교 IT응용시스템공학과 학생 ]
방준호 [ Junho Bang | 전북대학교 IT응용시스템공학과 교수 ]
Corresponding Author
Ever since next generation convergence technology became one of the most important industries in the nation, computing professionals have encountered a growing number of challenges. Along with scholars and colleagues in related fields, they have gathered in avariety of forums and meetings over the last few decades to share their knowledge, experiences and the outcome of their research. These exchanges have led to the founding of the International Next-generation Convergence technology (INCA) on December 1, 2015. INCA was registered as an incorporated association under the Ministry of Information and Communications. The main purpose of the organization is to improve our society by achieving the highest capability possible in next generation convergence technology.
간행물
간행물명
차세대융합기술학회논문지 [The Journal of Next-generation Convergence Technology Association]