밝기 변화에 강인한 적대적 음영 생성 및 훈련 글자 인식 알고리즘
Adversarial Shade Generation and Training Text Recognition Algorithm that is Robust to Text in Brightness
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
The system for recognizing text in natural scenes has been applied in various industries. However, due to the change in brightness that occurs in nature such as light reflection and shadow, the text recognition performance significantly decreases. To solve this problem, we propose an adversarial shadow generation and training algorithm that is robust to shadow changes. The adversarial shadow generation and training algorithm divides the entire image into a total of 9 grids, and adjusts the brightness with 4 trainable parameters for each grid. Finally, training is conducted in a adversarial relationship between the text recognition model and the shaded image generator. As the training progresses, more and more difficult shaded grid combinations occur. When training with this curriculum-learning attitude, we not only showed a performance improvement of more than 3% in the ICDAR2015 public benchmark dataset, but also confirmed that the performance improved when applied to our’s android application text recognition dataset.
Abstract
The system for recognizing text in natural scenes has been applied in various industries. However, due to the change in brightness that occurs in nature such as light reflection and shadow, the text recognition performance significantly decreases. To solve this problem, we propose an adversarial shadow generation and training algorithm that is robust to shadow changes. The adversarial shadow generation and training algorithm divides the entire image into a total of 9 grids, and adjusts the brightness with 4 trainable parameters for each grid. Finally, training is conducted in a adversarial relationship between the text recognition model and the shaded image generator. As the training progresses, more and more difficult shaded grid combinations occur. When training with this curriculum-learning attitude, we not only showed a performance improvement of more than 3% in the ICDAR2015 public benchmark dataset, but also confirmed that the performance improved when applied to our’s android application text recognition dataset.
참고문헌 (20)
- J. Baek, "What Is Wrong With Scene Text Recognition Model Comparisons? Dataset and Model Analysis", 2019 IEEE, 2019, ~
- Y. Lecun, "Gradient-based learning applied to document recognition", Proceedings of the IEEE, 86(11), pp.2278~2324 (1998), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- Baoguang Shi, "ASTER: An Attentional Scene Text Recognizer with Flexible Rectification", IEEE Transactions on Pattern Analysis and Machine Intelligence, 41(9), pp.2035~2048 (2019), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- F. F. Borisyuk, "Rosetta : Large Scale System for Text Detection and Recognition in Images", 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (2018)
- Baoguang Shi, "An End-to-End Trainable Neural Network for Image-Based Sequence Recognition and Its Application to Scene Text Recognition", IEEE Transactions on Pattern Analysis and Machine Intelligence, 39(11), pp.2298~2304 (2017), Institute of Electrical and Electronics Engineers (IEEE)[DOI]
- 김정환, "딥러닝을 이용한 번호판 검출과 인식 알고리즘", 전기전자학회논문지, 23(2), pp.642~651 (2019), 한국전기전자학회[DOI]
- 서민석, "행동 인식을 위한 시공간 앙상블 기법", 로봇학회 논문지, 15(4), pp.385~391 (2020), 한국로봇학회[DOI]
- D. Karatzas, "ICDAR 2015competition on Robust Reading", 2015 13th International Conference on Document Analysis and Recognition(ICDAR) (2015)
- A. Gupta, "Synthetic Data for Text Localisation in Natural Images", 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)
- M. Jaderberg, "Synthetic data and artificial neural networks for natural scene text recognition", NIPS DLW (2014)
- D. Hendrycks, "Benchmarking neural network robustness to common corruptions and perturbations", International Conference on Learning Representations (ICLR) (2019)
- E. Rusak, "A Simple Way to Make Neural Networks Robust Against Diverse Image Corruptions", Lecture Notes in Computer Science, pp.53~69 (2020)
- K. Wang, "End-to-end scene text recognition", 2011 International Conference on Computer Vision (2011)
- B. Shi, "Robust Scene Text Recognition with Automatic Rectification", 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2016)
- M. Jaderberg, "Spatial transformer networks", NIPS (2015)
- W. Liu, "STAR-Net: A SpaTial Attention Residue Network for Scene Text Recognition", British Machine Vision Conference 2016 (2016)
- Y. Mou, "PlugNet : Degradation Aware Scene Text Recognition Supervised by a Pluggable Super-Resolution Unit", European Conference on Computer Vision, pp.158~174 (2020)
- 김진호, "스마트폰을 이용한 은행 보안카드 자동 인식", 한국콘텐츠학회 논문지, 16(12), pp.19~26 (2016), 한국콘텐츠학회[DOI]
- 이승주, "가상 데이터를 활용한 번호판 문자 인식 및 차종 인식 시스템 제안", 방송공학회 논문지, 25(5), pp.776~788 (2020), 한국방송∙미디어공학회[DOI]
- C. -Y. Lee, "Recursive recurrent nets w ith attention modeling for ocr in the wild", Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.2231~2239 (2016)