Text and Image Plagiarism Detection Using Histogram Matching and LCS
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Academic plagiarism poses a serious threat to the integrity of scholarly work, with forms ranging from direct copy-and-paste to sophisticated disguised plagiarism involving images and figures. Existing plagiarism detection tools predominantly focus on text analysis and largely ignore the role of images, which carry significant information in scientific publications. This paper presents a system for detecting both textual and image-based plagiarism in academic documents. For text plagiarism detection, the Longest Common Subsequence (LCS) algorithm is applied to measure similarity between a suspicious document and a corpus of source files. For image plagiarism detection, a Five Module Method (FMM) based on histogram comparison is employed, enabling detection of copied images including flowcharts, graphs, and photographs. The proposed system is implemented using Python and Django and tested on a publicly available flowchart image database used in CLEF-IP 2012 competitions. The system achieved an average flowchart recognition accuracy of 81.91%, demonstrating strong performance over existing methods such as CVC and INRIA. The combined approach provides a robust and scalable solution for detecting diverse forms of academic plagiarism
Fields of Study & Subjects
Academic integrity and plagiarism
Digital Media Forensic Detection
Advanced Steganography and Watermarking Techniques
Publication details
- Journal / source
- International Journal of Advanced Research in Science Communication and Technology
- Pages
- 270–270
- Publication date
- 30 March 2026
- Language
- EN