基于注意力机制的改进VGG车辆类型识别研究

Research on Vehicle Type Recognition of Improved VGG Based on Attention Mechanism

提出一种基于改进VGG11的车辆类型识别算法,用于处理不同类别的车辆型号的识别问题。为了解决一些车辆型号之间非常相似而导致的误检问题,应用注意力机制来增加有效特征图权重,减小无效或效果小的特征图权重,来使得训练模型得到更好的效果,从而提升算法的准确率。为了验证改进的VGG11的性能,将改进模型与经典模型AlexNet和VGG11原模型进行了实验对比。实验结果表明,改进的VGG11模型的收敛速度和精度都要远高于Alexnet和VGG11原模型,在训练50epochs时,就达到了96%的识别精度。

A vehicle type recognition algorithm based on improved VGG11 is proposed to deal with the problem of vehicle type recognition of different categories.In order to solve the error detection problem caused by the similarity between some vehicle models,the attention mechanism is applied to increase the weight of effective feature images and reduce the weight of ineffective or smalleffect feature images,so as to make the training model get a better effect and improve the accuracy of the algorithm.In order to verify the performance of the improved VGG11,the improved model was compared with the classical model AlexNet and the original VGG11 model.The experimental results show that the convergence speed and accuracy of the improved VGG11 model are much higher than that of the AlexNet and the original VGG11 model,and the recognition accuracy can reach 96%when training 50epochs.