University of Worcester Worcester Research and Publications
 
  USER PANEL:
  ABOUT THE COLLECTION:
  CONTACT DETAILS:

Automatically Evolving Rotation-invariant Texture Image Descriptors by Genetic Programming

Al-Sahaf, H., Al-Sahaf, A., Xue, B., Johnston, Mark and Zhang, M. (2017) Automatically Evolving Rotation-invariant Texture Image Descriptors by Genetic Programming. IEEE Transactions on Evolutionary Computation, 21 (1). pp. 83-101. ISSN Print 1089-778X Online 1941-0026

[img] Text
TEVC_Harith_RotationInvariant_2016June.pdf - Accepted Version
Restricted to Repository staff only

Download (1MB) | Request a copy

Abstract

In computer vision, training a model that performs classification effectively is highly dependent on the extracted features, and the number of training instances. Conventionally, feature detection and extraction are performed by a domain-expert who, in many cases, is expensive to employ and hard to find. Therefore, image descriptors have emerged to automate these tasks. However, designing an image descriptor still requires domain-expert intervention. Moreover, the majority of machine learning algorithms require a large number of training examples to perform well. However, labelled data is not always available or easy to acquire, and dealing with a large dataset can dramatically slow down the training process. In this paper, we propose a novel Genetic Programming based method that automatically synthesises a descriptor using only two training instances per class. The proposed method combines arithmetic operators to evolve a model that takes an image and generates a feature vector. The performance of the proposed method is assessed using six datasets for texture classification with different degrees of rotation, and is compared with seven domain-expert designed descriptors. The results show that the proposed method is robust to rotation, and has significantly outperformed, or achieved a comparable performance to, the baseline methods.

Item Type: Article
Additional Information:

Staff and students at the University of Worcester can access the full-text of the online published article via the official URL. External users should check availability with their local library or Interlibrary Requests Service.

Uncontrolled Discrete Keywords: genetic programming, classification, image descriptor, keypoint detection, feature extraction
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Divisions: College of Health, Life and Environmental Sciences > School of Science and the Environment
Related URLs:
Depositing User: Mark Johnston
Date Deposited: 02 Aug 2016 09:09
Last Modified: 17 Jun 2020 17:12
URI: https://eprints.worc.ac.uk/id/eprint/4716

Actions (login required)

View Item View Item
 
     
Worcester Research and Publications is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.