Improve the Attention Mechanism to Realize the Clarity of Fuzzy License Plate Images

In order to avoid the impact of low‐resolution blurred images on the recognition of license plates, a spatial and channel dual attention network (RSCAN) based on residual network is proposed to recover blurred license plate images. RSCAN adds a spatial attention mechanism to the RCAN (residual channel attention) network, and introduces a new channel attention mechanism to train and test the network through a self‐made license plate data set. The RSCAN network is compared with the conventional processed pictures, SRCNN (super‐resolution) network, and RCAN network. The test license plate pictures are used for comparison experiments, and the model evaluation is carried out through the peak signal‐to‐noise ratio (PSNR) and structural similarity (SSIM), respectively, 30.312/0.889. The experimental results show that the license plate images processed by RSCAN network achieve the best results.

關鍵字

Residual network ； Attention mechanism ； License plate data set ； Evaluation index

參考文獻

Wang Z, Chen J, Hoi S C H. Deep learning for image sup-er-resolution: A survey[J]. IEEE Transactions on Pattern Ana-lysis and Machine Intelligence, 2020.

Dong C, Loy C C, He K, et al. Image super-resolution using deep convolutional networks[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 38(2): 295-307

J Kim, J.K Lee, K.M Lee. Accurate image super-resolution using very deep convolutional networks[C]// In CVPR, America: IEEE Xplore, 2016.1:1646-1654.

Lim B, Son S, Kim H, et al. Enhanced deep residual networks for single image super-resolution[C]//Proceedings of the IEEE conference on computer vision and pattern recognition workshops. 2017: 136-144.

J Kim, J.K Lee, K.M Lee. Deeply-recursive convolutional network for image super-resolution[C] //In CVPR, America: IEEE Xplore, 2016.1: 1637-1645.

國際替代計量

Improve the Attention Mechanism to Realize the Clarity of Fuzzy License Plate Images

全文下載

主題瀏覽