Improved road traffic sign recognition from feature reconstruction

Road traffic sign recognition is an essential function of modern autonomous driving systems. Accurate recognition plays an important role in ensuring vehicle safety and enabling subsequent operations. This paper introduces a Siamese architecture that is combined with feature reconstruction to overcome current method limitations. Aiming at improving accuracy and designing a flexible fusion network, this study proposes a Siamese neural network (SNN) framework that integrates an improved Mamba network with mainstream CNN architectures… named Siamese-Mamba&CNN (SM&CNN).

The proposed improved Mamba replaces the convolutional structures by introducing the Depth Separable convolution Block (DS Block), reducing computational complexity. In the fusion process of heterogeneous networks, we propose an HN-FR method to integrate heterogeneous network features and reconstruct convolution features… achieving fast and excellent performance. Taking VGG-16 network as an example, the experimental results demonstrated the method’s effectiveness.

Receive Our Newsletter

Get the latest Safety21 news sent to your inbox.