A Symbolic Contour Representation of Shape for Recognition and Retrieval

Ali Taheri Anaraki, Usman Ullah Sheikh


Image retrieval technique is one of the most important and vital part of huge image storages. A precise retrieval engine is required to meet the needs of users. In this regard, content-based image retrieval (CBIR) methods based on image contents such as shape, color and texture have been introduced as an alternative of traditional methods to improve the retrieval precision. Of all features of an image, shape holds more valuable information than others. However, due to the dynamic nature of real worlds’ objects, a shape of an object is subject to occlusion, transformation and deformation. This is the main challenge of shape-based retrieval techniques and shape descriptors. In this paper, a new shape representation and description called Symbolic Contour-based Shape Retrieval (SCSR) method is introduced to overcome drawbacks of previous methods in terms of computation time and accuracy. The performance of the proposed method has been compared to other state-of-art methods on three popular datasets, Vehicles, Brown and Mpeg7.


Image Retreival; Shape Descriptor; Content- Based; Contour- Based; Image Recognition

DOI: https://doi.org/10.11591/eei.v10i3.2051


  • There are currently no refbacks.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Bulletin of EEI Stats