A Novel Approach in Adopting Finite State Automata for Image Processing Applications

A Novel Approach in Adopting Finite State Automata for Image Processing Applications

R. Obulakonda Reddy, Kashyap D. Dhruve, R. Nagarjuna Reddy, M. Radha, N. Sree Vani
Copyright: © 2018 |Pages: 16
DOI: 10.4018/IJCVIP.2018010104
(Individual Articles)
No Current Special Offers


This article describes how robust image processing application rely heavily on image descriptors extracted. Limited work is carried out in adopting probabilistic finite state automata (PFSA) models for image processing. A finite state automata for image processing (FSAFIP) method is presented here. Texture classification and content based image retrieval (CBIR) is considered. In FSAFIP, foreground and background regions of an image are identified and later split into patches. Using a tristate PFSA model, feature descriptors corresponding to background/foreground regions are constructed. A distance based large margin nearest neighbor (LMNN) classifier is considered in FSAFIP to impart intelligence. A performance and experimental study to evaluate performance of FSAFIP for CBIR and texture classification is presented. Comparison results in CBIR obtained prove superior performance of FSAFIP over existing methods on Corel-1K dataset. High texture classification accuracy of 99.2% is reported using FSAFIP on KHT-TIPS dataset. An improved texture classification accuracy is achieved using FSAFIP in comparison to former methods.
Article Preview

1. Introduction

Expedited growth of world-wide web has to lead to easy access to huge volume of digital multimedia data especially image data. Unfortunately, this image data in most cases are scatted and unorganized, making searching, analysis and retrieval of such data difficult. Extensive work is carried out by researchers towards image feature detectors that are used to establish feature descriptors. Image processing applications like image classification (Fang, 2017), content based image retrieval (CBIR). (Guo, 2015), (Yang, 2017), (Elalami, 2014), image representation (Yap, 2010), image classification (Liu, 2012), motion tracking and crowd analysis (Zerdi, 2014) [7], texture analysis and classification (Song, 2017), medical image processing (Satheesha, 2017), to name a few rely on accurate feature detectors and feature descriptors for robust operations. Significance of image features in image processing applications is clearly highlighted in (Satheesha, 2017), (Godtliebsen, 2004), (Hassaballah, 2016). In this paper discusses about CBIR and texture classification applications.

In CBIR systems a set of visually similar images are obtained from a large collection of images in a database. To retrieve visually similar images, it is essential to understand the content present in images. Researchers have proposed various features to describe content. In (Guo, 2015), color co-occurrence feature and bit pattern features are considered to understand content. Numerous features like gray, color co-occurrence matrix, difference observed between pixels of scan patterns, histogram of oriented gradient and local binary patterns features is considered in (Yang, 2017). Though researchers have considered numerous feature combinations, the existing CBIR systems in place neglect to describe correlation that exists between low-level features and high-level concepts observed in images effecting performance (Zhao, 2016).

Complete Article List

Search this Journal:
Volume 13: 1 Issue (2023): Forthcoming, Available for Pre-Order
Volume 12: 4 Issues (2022): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2021)
Volume 10: 4 Issues (2020)
Volume 9: 4 Issues (2019)
Volume 8: 4 Issues (2018)
Volume 7: 4 Issues (2017)
Volume 6: 2 Issues (2016)
Volume 5: 2 Issues (2015)
Volume 4: 2 Issues (2014)
Volume 3: 4 Issues (2013)
Volume 2: 4 Issues (2012)
Volume 1: 4 Issues (2011)
View Complete Journal Contents Listing