Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jul 5;21(13):4612.
doi: 10.3390/s21134612.

An Improved Character Recognition Framework for Containers Based on DETR Algorithm

Xiaofang Zhao  1 , Peng Zhou  2 , Ke Xu  3 , Liyun Xiao  1
Affiliations
Free PMC article

An Improved Character Recognition Framework for Containers Based on DETR Algorithm

Xiaofang Zhao et al. Sensors (Basel). .
Free PMC article

Abstract

An improved DETR (detection with transformers) object detection framework is proposed to realize accurate detection and recognition of characters on shipping containers. ResneSt is used as a backbone network with split attention to extract features of different dimensions by multi-channel weight convolution operation, thus increasing the overall feature acquisition ability of the backbone. In addition, multi-scale location encoding is introduced on the basis of the original sinusoidal position encoding model, improving the sensitivity of input position information for the transformer structure. Compared with the original DETR framework, our model has higher confidence regarding accurate detection, with detection accuracy being improved by 2.6%. In a test of character detection and recognition with a self-built dataset, the overall accuracy can reach 98.6%, which meets the requirements of logistics information identification acquisition.

Keywords: DETR (detection with transformers); character recognition; multi-scale location coding; split-attention.

PubMed Disclaimer

Conflict of interest statement

The authors declare no conflict of interest.

Figures

Figure A1
Figure A1
Some images of the self-built dataset in this paper.
Figure A2
Figure A2
Some small sample character images after data reconstruction.
Figure A3
Figure A3
Comparison of category detection results after amplification of small sample characters.
Figure 1
Figure 1
Container characters recognition framework based on the improved DETR framework.
Figure 2
Figure 2
The structure of transformer.
Figure 3
Figure 3
Multi-head attention.
Figure 4
Figure 4
Data enhancement results of the sample set: (a) original drawing, (b) simulating foggy days, (c) simulating rainy days, (d) simulating snow days, (e) foreign bodies blocked 1, (f) foreign bodies blocked 2, (g) simulating smoke, (h) simulating strong light.
Figure 5
Figure 5
The mAP@0.5 of training 100 epochs.
Figure 6
Figure 6
The detection results of the two models respectively on the characters.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Druzhkov P.N., Kustikova V.D. A survey of deep learning methods and software tools for image classification and object detection. Pattern Recognit. Image Anal. 2016;26:9–15. doi: 10.1134/S1054661816010065. - DOI
    1. Liu X., Meng G., Pan C. Scene text detection and recognition with advances in deep learning: A survey. Int. J. Doc. Anal. Recognit. 2019;22:143–162. doi: 10.1007/s10032-019-00320-5. - DOI
    1. Ren S., He K., Girshick R., Sun J. Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks. IEEE Trans. Pattern Anal. Mach. Intell. 2017;39:1137–1149. doi: 10.1109/TPAMI.2016.2577031. - DOI - PubMed
    1. Redmon J., Farhadi A. YOLOv3: An Incremental Improvement. [(accessed on 4 July 2021)];arXiv. 2018 Available online: https://arxiv.org/abs/1804.02767.1804.02767
    1. Bochkovskiy A., Wang C.Y., Liao H. YOLOv4: Optimal Speed and Accuracy of Object Detection. arXiv. 20202004.10934

LinkOut - more resources