Image Recognition With Haar Wavelet and Pseudoinverse Learning Algorithm Based Autoencoders

doi:10.1088/1742-6596/2278/1/012019

Image Recognition With Haar Wavelet and Pseudoinverse Learning Algorithm Based Autoencoders

2022-05-01

0

OA

AI

DOI：10.1088/1742-6596/2278/1/012019

原文链接

原文求助

摘要

En 中文

Abstract In this work, a recognition model with low computation and high training efficiency is proposed for images recognition. The proposed model consists of two modules: 1) haar wavelet transformation extracts features and compresses images; 2) a representation learning module using autoencoder. Because of its powerful function, Haar wavelet transformation has wide use in the feature extraction of images. The mother wavelet and scale wavelet can extract features from different scales. The representation learning module is trained with a non-gradient descent algorithm based on autoencoder structure. Two benchmark image datasets, MNIST and Fashion-MNIST, have been used to validate the proposed model. Experimental results show that our proposed model achieves its expected effect. Compared with filters, CNN, and pseudoinverse learning autoencoders, our model takes less training time, at the same time it acquires comparative recognition accuracy.

Image Recognition With Haar Wavelet and Pseudoinverse Learning Algorithm Based Autoencoders

摘要

期刊

机构

引用论文