Design and Development of Scene Recognition and Classification Model Based on Human Pre-attentive Visual Attention

Kudus, A. R. and Teh, Chee Siong (2021) Design and Development of Scene Recognition and Classification Model Based on Human Pre-attentive Visual Attention. Journal of Physics: Conference Series, 1755 (1). pp. 1-12. ISSN 17426588

[img] PDF
abstract.pdf

Download (643kB)
Official URL: https://iopscience.iop.org/article/10.1088/1742-65...

Abstract

Recent works on scene classification still utilize the advantages of generic feature of Convolutional Neural Network while applying object-ontology technique that generates limited amount of object regions. Human can successfully recognize and classify scene effortlessly within short period of time. By utilizing this idea, we present a novel approach of scene classification model that built based on human pre-attentive visual attention. We firstly utilize saliency model to generate a set of high-quality regions that potentially contain salient objects. Then we apply a pre-trained Convolutional Neural Network model on these regions to extract deep features. Extracted features of every region are then concatenated to a final features vector and feed into one-vs-all linear Support Vector Machines. We evaluate our model on MIT Indoor 67 dataset. The result proved that saliency model used in this work is capable to generate high-quality informative salient regions that lead to good classification output. Our model achieves a better average accuracy rate than a standard approach that classifies as one whole image.

Item Type: Article
Uncontrolled Keywords: Scene Recognition and Classification, Convolutional Neural Network, pre-attentive visual attention.
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: Academic Faculties, Institutes and Centres > Faculty of Cognitive Sciences and Human Development
Depositing User: Siong
Date Deposited: 12 Apr 2021 04:28
Last Modified: 12 Apr 2021 07:09
URI: http://ir.unimas.my/id/eprint/35066

Actions (For repository members only: login required)

View Item View Item