Human eye inspired log-polar pre-processing for neural networks

TL;DR

This paper introduces a bio-inspired log-polar pre-processing step for neural networks, inspired by the human eye, which enhances rotation and scale robustness and reduces image size without losing accuracy.

Contribution

The paper presents a novel log-polar pre-processing method inspired by the human eye, improving neural network robustness to rotation and scale variations.

Findings

Achieves rotation and scale invariance in CNNs without retraining.

Reduces image size to about 20% of original while maintaining accuracy.

Demonstrates effectiveness on image datasets inspired by human vision.

Abstract

In this paper we draw inspiration from the human visual system, and present a bio-inspired pre-processing stage for neural networks. We implement this by applying a log-polar transformation as a pre-processing step, and to demonstrate, we have used a naive convolutional neural network (CNN). We demonstrate that a bio-inspired pre-processing stage can achieve rotation and scale robustness in CNNs. A key point in this paper is that the CNN does not need to be trained to identify rotation or scaling permutations; rather it is the log-polar pre-processing step that converts the image into a format that allows the CNN to handle rotation and scaling permutations. In addition we demonstrate how adding a log-polar transformation as a pre-processing step can reduce the image size to ~20\% of the Euclidean image size, without significantly compromising classification accuracy of the CNN. The pre-processing stage presented in this paper is modelled after the retina and therefore is only tested against an image dataset. Note: This paper has been submitted for SAUPEC/RobMech/PRASA 2020.