Faster Light Detection Algorithm of Traffic Signs Based on YOLOv5s-A2

Abstract

Traffic sign recognition systems have been applied to advanced driving assistance and automatic driving systems to help drivers obtain important road information accurately. The current mainstream detection methods have high accuracy in this task, but the number of model parameters is large, and the detection speed is slow. Based on YOLOv5s as the basic framework, this paper proposes YOLOv5S-A2, which can improve the detection speed and reduce the model size at the cost of reducing the detection accuracy. Firstly, a data augmentation strategy is proposed by combining various operations to alleviate the problem of unbalanced class instances. Secondly, we proposed a path aggregation module of Feature Pyramid Network (FPN) to make new horizontal connections. It can enhance multi-scale feature representation capability and compensate for the loss of feature information. Thirdly, an attention detection head module is proposed to solve the aliasing effect in cross-scale fusion and enhance the representation of predictive features. Experiments on Tsinghua-Tencent 100K dataset (TT100K) show that our method can achieve more remarkable performance improvement and faster inference speed than other advanced technologies. Our method achieves 87.3% mean average precision (mAP) surpassing the original model's 7.9%, and the frames per second (FPS) value is maintained at 87.7. To show generality, we tested it on the German Traffic Sign Detection Benchmark (GTSDB) without tuning and obtained an average precision of 94.1%, and the FPS value is maintained at about 105.3. In addition, the number of YOLOv5s-A2 parameters is about 7.9 M.

CR-YOLOv8: Multiscale Object Detection in Traffic Sign Images

“…Wang et al [32] mitigated the loss of contextual information due to feature channel reduction by introducing an attention module and a feature enhancement module to improve the network's sensitivity to traffic signs. Yuan et al [33] incorporated a path aggregation module into the feature pyramid (FPN) structure to further enhance the encoding and decoding parts by adding horizontal connections to the spatial information, effectively enhancing the network's feature representation of traffic signs under normal weather. Liang et al [34] combined the coordinate attention module with the backbone network ResNeSt and constructed the feature pyramid for multi-scale detection [35], which enhanced the network's ability to extract shallow texture and contour information and enabled the extracted features to focus on traffic sign information, thereby improving the detection accuracy.…”

Optimizing Traffic Sign Detection and Recognition by Using Deep Learning

“…Yuan et al [19] show that systems that use autonomous driving and enhanced driving assistance depend on the correct and efficient detection of traffic signs. Although the current detection techniques have a high accuracy rate, they frequently have huge model parameters and slow detection speeds.…”

SC-YOLO: A Object Detection Model for Small Traffic Signs

“…These methods can automatically learn highlevel semantic features and have higher accuracy and robustness. [3] proposed a fast and lightweight traffic sign detection algorithm called YOLOv5s-A2. The YOLOv5s network was improved by incorporating techniques such as depth wise separable convolution, attention mechanism, and anchor box clustering.…”

“…In contrast, deep learning methods learn features directly from a large amount of training data, using a rich convolutional hierarchy that can represent complex object features. These method have made significant improvements in traffic sign detection, such as improving small object detection performance [1], improving model robustness in complex weather scenarios [2], and lightweight improvements to achieve speed-accuracy trade-offs [3], etc. Traffic signs are important facilities to ensure the safety and smooth flow of traffic, but due to various reasons such as natural factors, human factors and management factors, traffic signs may appear damaged as shown in Fig.…”