Ship Identification Characters Detection Method Based on Improved TextBoxes

Zhiqiang Li ； Yingxiang Li

《International Core Journal of Engineering》 6卷5期 (2020/05) Pp. 328-332

https://doi.org/10.6919/ICJE.202005_6(5).0047

摘要

Waterway transportation is an important mode of transportation in modern society. Especially after the Silk Road, waterway transportation plays an important role in global trade transportation. However, with the expanding use of ships, ships hit the reef, illegal transportation, marine pollution and other waterway traffic accidents occur frequently. As unique identities, Ship Identification Characters(SICs) have been playing a significant role in ship traffic behavior control. In order to realize the intelligent detection of ship identification characters, this paper proposes an improved TextBoxes detection method for efficient and automatic detection of ship identification characters. Experiments show that this improved algorithm improves the detection accuracy by about 6% compared to TextBoxes themselves.

關鍵字

Textboxes ； Text detection ； Ship identification characters ； Deep learning ； Deep convolutional neural network

參考文獻

Zhang Z, Zhang C, Shen W, et al. Multi-oriented text detection with fully convolutional networks[C]//Proceedings of the IEEE Conference on Computer Vision and Patten Recognition. 2016: 4159-4167.

Lyu P, Yao C, Wu W, et al. Multi-oriented scene text detection via corner localization and region segmentation[C]//Proceeding of the IEEE Conference on Computer Vision and Patteren Recognition. 2018: 7553-7563.

Girshick R. Fast r-cnn]C]//Proceedings of the IEEE international conference on computer vision. 2015: 1440-1448.

Liao M, Shi B, Bai X. Textboxes++: A single-shot oriented scene text detector[J]. IEEE Transactions on Image Processing, 2018, 27(8): 3676-3690.

Lin Kunjie. Ship area automatic recognition system based on machine vision [D]. Nanjing University of Aeronautics and Astronautics, 2009.