Classification and Identification of Domestic Catsbased on Deep Learning

Abstract

In recent years, computer technology has developed very rapidly, and the hardware conditions are getting better and better. The time used to train deep neural networks has been greatly reduced. Deep learning is rapidly becoming an important hotspot of scientific research. Deep learning technology is widely used in digital recognition, speech recognition, unmanned driving, image recognition, and other fields [5]. The new generation of artificial intelligence technology represented by deep learning is gradually penetrating people's lives and promoting the development of society. As a representative of deep learning technology, the convolutional neural network has also developed rapidly in recent years. To improve the accuracy of cat breed classification and enable more people to clearly understand cat species, this paper cites and compares different deep learning models, and compares the performance of VGGNet, Inception-v3 and the optimized deep learning model in cat breed recognition. From the experimental results, the accuracy of the improved model is about 84%, which is higher than other models.