It uses convolutional neural networks in the paper to provide an effective solution for dog recog-nition problems.Firstly,the dataset is sourced from the standard dataset of 120 dog breeds at Stanford U-niversity.Secondly,it constructs models of Alexnet and VGG16 convolutional neural networks.After train-ing and testing,both convolutional neural network models can recognize dogs in the dataset.Finally,through experimental comparison on the validation set,it is found that the VGG16 convolutional neural network model constructed in this paper has better recognition performance than the Alexnet convolutional neural network model,with a recognition rate of 89.17％ on the validation set.
关键词
犬类识别/卷积神经网络/Alexnet模型/VGG16模型
Key words
dog recognition/convolutional neural network/Alexnet model/VGG16 model
维普
万方数据