Abstract
In terms of small objects in traffic scenes, general object detection algorithms have low detection accuracy, high model complexity, and slow detection speed. To solve the above problems, an improved algorithm (named YOLO-MXANet) is proposed in this paper. Complete-Intersection over Union (CIoU) is utilized to improve loss function for promoting the positioning accuracy of the small object. In order to reduce the complexity of the model, we present a lightweight yet powerful backbone network (named SA-MobileNeXt) that incorporates channel and spatial attention. Our approach can extract expressive features more effectively by applying the Shuffle Channel and Spatial Attention (SCSA) module into the SandGlass Block (SGBlock) module while increasing the parameters by a small number. In addition, the data enhancement method combining Mosaic and Mixup is employed to improve the robustness of the training model. The Multi-scale Feature Enhancement Fusion (MFEF) network is proposed to fuse the extracted features better. In addition, the SiLU activation function is utilized to optimize the Convolution-Batchnorm-Leaky ReLU (CBL) module and the SGBlock module to accelerate the convergence of the model. The ablation experiments on the KITTI dataset show that each improved method is effective. The improved algorithm reduces the complexity and detection speed of the model while improving the object detection accuracy. The comparative experiments on the KITTY dataset and CCTSDB dataset with other algorithms show that our algorithm also has certain advantages.
Highlights
Object detection is an essential field of computer vision, and its task is to locate and classify objects with the variable number in an image
Based on general one-stage object detection algorithms, we propose a small object detection algorithm in traffic scenes, which solves the problem that original algorithm is not high in detecting small-scale objects and reduces the number of parameters from 61.5 M to 13.8 M and improves the detection speed
To further enhance the feature extraction capability of MobileNeXt, we present SA-MobileNeXt based on the Shuffle Channel and Spatial Attention module as the backbone network
Summary
Object detection is an essential field of computer vision, and its task is to locate and classify objects with the variable number in an image. Object detection algorithms based on anchor-free [9,10,11,12] are developing rapidly in the one-stage algorithms. Two-stage algorithms depend on the proposals, and their detection speed is generally slow, in other words, their real-time performance cannot meet the demand of traffic scenes, even though its detection accuracy is constantly improving. The speed of onestage algorithms based on regression is fast enough to satisfy the requirements of most tasks. Many scholars have applied general object detection algorithms to the traffic field.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
