A connection between image processing and artificial neural networks layers through a geometric model of visual perception
| Autor: | Gabriele Steidl, Thomas Batard, Marcelo Bertalmío, Eduard Ramon Maldonado |
|---|---|
| Jazyk: | angličtina |
| Rok vydání: | 2019 |
| Předmět: |
Visual perception
Artificial neural network Color constancy Computer science business.industry Color vision Vision Deep learning Color correction Image processing Pattern recognition 02 engineering and technology 01 natural sciences Neural network 010101 applied mathematics 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing Differential geometry Artificial intelligence 0101 mathematics Variational model business MNIST database |
| Zdroj: | Recercat. Dipósit de la Recerca de Catalunya instname Lecture Notes in Computer Science ISBN: 9783030223670 SSVM |
| Popis: | Comunicació presentada a: the 7th International Conference on Scale Space and Variational Methods in Computer Vision, celebrada del 30 de juny al 4 de juliol de 2019 a Hofgeismar, Alemanya. In this paper, we establish a connection between image processing, visual perception, and deep learning by introducing a mathematical model inspired by visual perception from which neural network layers and image processing models for color correction can be derived. Our model is inspired by the geometry of visual perception and couples a geometric model for the organization of some neurons in the visual cortex with a geometric model of color perception. More precisely, the model is a combination of a Wilson-Cowan equation describing the activity of neurons responding to edges and textures in the area V1 of the visual cortex and a Retinex model of color vision. For some particular activation functions, this yields a color correction model which processes simultaneously edges/textures, encoded into a Riemannian metric, and the color contrast, encoded into a nonlocal covariant derivative. Then, we show that the proposed model can be assimilated to a residual layer provided that the activation function is nonlinear and to a convolutional layer for a linear activation function. Finally, we show the accuracy of the model for deep learning by testing it on the MNIST dataset for digit classiffication. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|