Taxonomy of file fragments using Gray-Level Co-Occurrence Matrices

P. P. Pullaperuma, A. T. Dharmaratne

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

2 Citations (Scopus)


Researches up to data have focused on using non texture based methods in addressing the problem of classifying the data types of file fragments. In this research we considered a file fragment as a 8 bit grayscale image and the Gray Level Co-Occurrence Matrix (GLCM) based method was used to extract textural features. Texture features for fragment dimensions 8 × 8, 16 × 16, 32 × 32 and 64 × 64 and gray level quantizations from 4 to 64 with step increments of 4 were explored. The K nearest neighbor classifier was used as the classifier and the optimal GLCM features for a particular gray level and fragment dimension were determined using Sequential Forward Selection (SFS) algorithm. On the classification of 7 data types, our novel approach reached a maximum overall accuracy of 86.86% in classifying 64 × 64 sized fragments with 12 gray levels.

Original languageEnglish
Title of host publication2013 International Conference on Digital Image Computing
Subtitle of host publicationTechniques and Applications, DICTA 2013
Publication statusPublished - 2013
Externally publishedYes
EventDigital Image Computing Techniques and Applications 2013 - Wrest Point Hotel, Hobart, Australia
Duration: 26 Nov 201328 Nov 2013
Conference number: 15th (Proceedings)

Publication series

Name2013 International Conference on Digital Image Computing: Techniques and Applications, DICTA 2013


ConferenceDigital Image Computing Techniques and Applications 2013
Abbreviated titleDICTA 2013
OtherThe International Conference on Digital Image Computing: Techniques and Applications (DICTA) is the main Australian Conference on computer vision, image processing, pattern recognition, and related areas. DICTA was established as a biennial conference in 1991 and became an annual event in 2007. It is the premier conference of the Australian Pattern Recognition Society (APRS).

The conference will be held at Wrest Point Hotel, Hobart, Tasmania, Australia from 26 - 28 November 2013.
Internet address


  • File fragments
  • File visualization
  • Glcm
  • Image texture
  • Machine learning

Cite this