MomentsNet: a simple learning-free method for binary image recognition

TL;DR

MomentsNet is a simple, learning-free deep network using moments, binary hashing, and histograms for binary image recognition, outperforming some learning-based methods.

Contribution

Introduces MomentsNet, a novel learning-free deep network architecture based on moments, with superior recognition performance on binary images.

Findings

MomentsNet outperforms traditional moments in recognition accuracy.

ZernikeNet achieves the best performance among MomentsNet variants.

ZernikeNet surpasses PCANet in binary image recognition.

Abstract

In this paper, we propose a new simple and learning-free deep learning network named MomentsNet, whose convolution layer, nonlinear processing layer and pooling layer are constructed by Moments kernels, binary hashing and block-wise histogram, respectively. Twelve typical moments (including geometrical moment, Zernike moment, Tchebichef moment, etc.) are used to construct the MomentsNet whose recognition performance for binary image is studied. The results reveal that MomentsNet has better recognition performance than its corresponding moments in almost all cases and ZernikeNet achieves the best recognition performance among MomentsNet constructed by twelve moments. ZernikeNet also shows better recognition performance on binary image database than that of PCANet, which is a learning-based deep learning network.