Rotation-invariant image classification using a novel 1D CNN and Multichannel Accurate Bessel-Fourier moments
César Camacho-Bello, Lucia Gutiérrez-Lazcano, Rosa María Ortega-Mendoza · PÄDI Boletín Científico de Ciencias Básicas e Ingenierías del ICBI · 2022
This work presents a proposal to use Bessel-Fourier moments as inputs to 1D convolutional neural networks in such a way that they take advantage of the inherent characteristics of moment type descriptors such as rotational invariance and minimal information redundancy. The results presented show that the proposal has a better performance than the deep neural network with rotation invariance.