Статья

MobileNetV2 Model for Image Classification

Ke DongSchool of Computer Science and Information, Hefei University of Technology, Hefei, ChinaChengjie ZhouCollege of Letters and Science, University of California, Los Angeles Los Angeles, CAYihan RuanThe Grainger College of Engineering, University of Illinois at Urbana Champaign, Champaign, ILYuzhi LiFaculty ofElectronic and Information Engineering, Xi’an Jiaotong University, Xi’an, China

2020en

ABI

Аннотация

Machine learning has been increasingly prevailing all over the world, especially in the computer vision field. This paper mainly focused on the performance of MobileNetV2 model for image classification. To verify the advanced performance of MobileNetV2 model better, this paper adopted MobileNetVl model as the control group and introduced an experiment of identifying images in a variety of datasets extracted from TensorFlow. With the T-SNE visualization tool, the conclusion can be generated by comparing the accuracy and effectiveness of these two models. The experimental results demonstrated that the proficiency of MobileNetV2 model achieved higher accuracy rates compared to MobileNetVl model. In order to enhance the performance of MobileNetV2, extensive experiments are performed.

Перевод пока недоступен

Идентификаторы

DOI: 10.1109/itca52113.2020.00106

Цитирования и источники

Цитирований: 2Использованных источников: 0

Показатели — AkademScholar