Textile Retrieval Based on Image Content from CDC and Webcam Cameras in Indoor Environments.

Oscar García-Olalla,Surajit Saikia,Laura Fernández-Robles,Enrique Alegre,Eduardo Fidalgo

doi:10.3390/s18051329

Abstract

Textile based image retrieval for indoor environments can be used to retrieve images that contain the same textile, which may indicate that scenes are related. This makes up a useful approach for law enforcement agencies who want to find evidence based on matching between textiles. In this paper, we propose a novel pipeline that allows searching and retrieving textiles that appear in pictures of real scenes. Our approach is based on first obtaining regions containing textiles by using MSER on high pass filtered images of the RGB, HSV and Hue channels of the original photo. To describe the textile regions, we demonstrated that the combination of HOG and HCLOSIB is the best option for our proposal when using the correlation distance to match the query textile patch with the candidate regions. Furthermore, we introduce a new dataset, TextilTube, which comprises a total of 1913 textile regions labelled within 67 classes. We yielded 84.94% of success in the 40 nearest coincidences and 37.44% of precision taking into account just the first coincidence, which outperforms the current deep learning methods evaluated. Experimental results show that this pipeline can be used to set up an effective textile based image retrieval system in indoor environments.

Highlights

The process of automatically finding objects, textiles, faces, or other patterns in images and videos is one of the most studied topics in computer vision
For n ≤ 11, Histogram of Oriented Gradients (HOG) + Half CLOSIB (HCLOSIB) descriptor outperformed the rest with a precision of 37.17% for n = 1
We presented a new application for textile based image retrieval in indoor environments

Summary

Introduction

The process of automatically finding objects, textiles, faces, or other patterns in images and videos is one of the most studied topics in computer vision. Most works related to CBIR aim at finding objects in datasets of images. Research groups face this problem using different approaches such as invariant local features (SIFT [11,12], SURF [13]), color description [14,15], template matching [16,17] or, more recently, deep learning techniques [18,19,20]

Methods

Results

Conclusion