Analysis of activation maps through global pooling measurements for texture classification

Author(s)

Condori, Rayner H. M. ^{[1, 2]} ; Bruno, Odemir M. ^{[1, 2]}

Affiliation

^[1] Univ Sao Paulo, Inst Math & Comp Sci, BR-13560970 Sao Carlos, SP - Brazil

^[2] Univ Sao Paulo, Sao Carlos Inst Phys, Sci Comp Grp, POB 369, BR-13560970 Sao Carlos, SP - Brazil

• Document type: Journal article
• Source: INFORMATION SCIENCES; v. 555, p. 260-279, MAY 2021.
• Web of Science Citations: 1

Abstract

We analyzed the effects of global pooling measurements on extracting relevant texture information from a given set of activation maps. Initially, using a layer-by-layer approach (GP-CNN), we experimentally demonstrated that layers at various depth levels could provide high-quality texture information. Based on this finding, we developed RANKGP-CNN, a method that performs multi-layer feature extraction. More specifically, RANKGP-CNN treats every CNN model as a vast collection of deep composite functions, where each function computes a 2D activation map for every input image. A feature ranking approach then assigns a score to each deep composite function by processing the activation maps generated for a particular dataset bank. Eventually, RauxGP-CNN uses the top-ranked deep composite functions to compute feature vectors for different texture datasets. Experiments on a dedicated classifier showed that RauxGP-CNN achieves good results and can adapt to different texture problems. Finally, we present RauxGP-3M-CNN as the version of RANKGP-CNN that considers multiple CNN models. Overall, RaNKGP-3M-CNN achieves promising results with the advantage of only using the default scale of the input images. (C) 2020 Published by Elsevier Inc. (AU)

• FAPESP's process: 16/18809-9 - Deep learning and complex networks applied to computer vision
• Grantee: Odemir Martinez Bruno
• Support Opportunities: Research Grants - Research Partnership for Technological Innovation - PITE
• FAPESP's process: 14/08026-1 - Artificial vision and pattern recognition applied to vegetal plasticity
• Grantee: Odemir Martinez Bruno
• Support Opportunities: Regular Research Grants