Design of a Universal Image Classification and Recognition System Based on Tensorflow

DOI: https://doi.org/10.62517/jes.202402116

Author(s)

Ding Lilei

Affiliation(s)

College of Electronic and Information Engineering, Ankang University, Ankang, Shaanxi, China

Abstract

With the continuous development of computer technology and big data applications, research on image classification and recognition has become an important content in the fields of science and engineering, with wide applications in multiple industries. On the basis of studying the process of image classification and recognition, the deep learning framework TensorFlow and Convolutional Neural Network (CNN), this system divides the entire recognition and classification process into multiple universal module units, and designs a visual and universal image classification and recognition system that can automatically adapt to the accurate classification and recognition of various types of images. The experimental results show that the system has a high recognition accuracy (over 95%) and good universality.

Keywords

Tensorflow; Image Classification; CNN

References

[1] Zhao Liping, Yuan Xiao, Zhu Cheng, et al. Research on the Progress of Residual Networks for Image Classification [J]. Computer Engineering and Applications, 2020, 56(20):9-19. [2] Feng Guier. GUI (Tkinter) instance development based on Python [J]. Information and Computer (Theoretical Edition), 2023, 35(01):175-178. [3] Xing Yanfang, Duan Hongxiu, He Guangwei the Application of TensorFlow in Image Recognition Systems [J]. Computer Technology and Development, 2019, 29(05):192-196. [4] Yuan Wenchui, Kong Xue. Research on Convolutional Neural Networks Based on TensorFlow Deep Learning Framework [J]. Microcomputer Applications, 2018, Vol. 34, Issue 2. [5] Zhao Wenchao, Wang Yanxin, Yong Junfei, et al. Underwater pipeline inspection robot based on TensorFlow image classification [J]. Modern Computer, 2022, 28(11):92-95. [6] Liu Mengya, Mao Jianlin. Research on the Impact of an Improved Pooling Model on the Performance of Convolutional Neural Networks [J]. Electronic Measurement Technology, 2019, Volume 42, Issue 5.