딥러닝 유사도 기반 현품 영상과 도면 영상의 1:1 매칭 방법
Deep Learning Similarity-based 1:1 Matching Method for Real Product Image and Drawing Image
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 논문은 주어진 현품 영상과 도면 영상의 유사도를 비교하여 1:1 검증을 위한 방법을 제시한 것으로, CNN(Convolutional Neural Network) 기반의 딥러닝 모델을 두 개로 결합하여 Siamese Net을 구성하고현품 영상과 도면 영상(정면도, 좌우 측면도, 평면도 등)을 같은 제품이면 1로 다른 제품이면 0으로학습하며, 추론은 현품 영상과 도면 영상을 쌍으로 질의하여 해당 쌍이 같은 제품인지 아닌지를 판별하는 딥러닝 모델을 제안한다. 현품 영상과 도면 영상과의 유사도가 문턱 값(Threshold: 0.5) 이상이면동일한 제품이고, 문턱 값 미만이면 다른 제품이라고 판별한다. 본 연구에서는 질의 쌍으로 동일제품의현품 영상과 도면 영상이 주어졌을 때(긍정 : 긍정) “동일제품”으로 판별할 정확도는 약 71.8%로 나타났고, 질의 쌍으로 다른 현품 영상과 도면 영상이 주어졌을 때(긍정: 부정) “다른제품”으로 판별할 정확도는 약 83..1%를 나타내었다. 향후 제안한 모델에 파라미터 최적화 연구를 접목하고 데이터 정제 등의과정을 추가하여 현품 영상과 도면 영상의 매칭 정확도를 높이는 연구를 진행할 예정이다.
Abstract
This paper presents a method for 1:1 verification by comparing the similarity between the given real product image and the drawing image. The proposed method combines two existing CNN-based deep learning models to construct a Siamese Network. After extracting the feature vector of the image through the FC (Fully Connected) Layer of each network and comparing the similarity, if the real product image and the drawing image (front view, left and right side view, top view, etc) are the same product, the similarity is set to 1 for learning and, if it is a different product, the similarity is set to 0. The test (inference) model is a deep learning model that queries the real product image and the drawing image in pairs to determine whether the pair is the same product or not. In the proposed model, through a comparison of the similarity between the real product image and the drawing image, if the similarity is greater than or equal to a threshold value (Threshold: 0.5), it is determined that the product is the same, and if it is less than or equal to, it is determined that the product is a different product. The proposed model showed an accuracy of about 71.8% for a query to a product (positive: positive) with the same drawing as the real product, and an accuracy of about 83.1% for a query to a different product (positive: negative). In the future, we plan to conduct a study to improve the matching accuracy between the real product image and the drawing image by combining the parameter optimization study with the proposed model and adding processes such as data purification.
참고문헌 (15)
- Karen Simonyan, "Very Deep Convolutional Networks for large-scale image recognition'(VGGNet)", Computer Vision and Pattern Recognition (2015)
- Deep Residual Learning for Image Recognition, "Kaiming He", Computer Vision and Pattern Recognition, 770~778 (2015)
- Gao Huang, "Densely Connected Convolutional Networks", Computer Vision and Pattern Recognition (2017)
- Mingxing Tan, "EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks", ICML 2019 (2019)
- Zhuang Liu, "A ConvNet for the 2020s", CVPR 2022 (2022)
- Andrew G. Howard, "MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applications", CVPR 2017 (2017)
- Gregory Koch, "Ruslan Salakhutdinov, ”Siamese Neural Networks for One-shot Image Recognition", Proceedings of the 32nd International Conference on Machine Learning, 37 (2015)
- Seong-Hoon Kim, "Hyperparameter optimization method based on harmony search algorithm to improve performance of 1D CNN human respiration pattern recognition system", Sensors, 20, pp.3697~ (2020)[DOI]
- Weijian Deng, "Image-Image Domain Adaptation with Preserved Self-Similarity and Domain-Dissimilarity for Person Re-identification", CVPR 2018 (2018)
- Grigori Sidorov, "Soft Similarity and Soft Cosine Measure:Similarity of Features in Vector Space Model", Computacionary Sistemas, 18(3), pp.491~504 (2014)
- Raia Hadsell, "Dimensionality Reduction by Learning an Invariant Mapping", 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'06) (2006)
- Florian Schroff, "FaceNet:A Unified Embedding for Face Recognition and Clustering", Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition (2015)
- Conghui Hu, "Feature Representation Learning for Unsupervised Cross-domain Image Retrieval", ECCV 2022 (2022)
- Qingjie Meng, "Unsupervised Cross-domain Image Classification by Distance Metric Guided Feature Alignment", Machine Learning (2020)
- Xiaoping Zhou, "Cross-domain image retrieval: methods and applications", International Journal of Multimedia Information Retrieval, 11, pp.199~218 (2022)