Texture-Based Image Forgery Detection Using Wavelet Transform and Adaptive SVM Optimization

  • Unique Paper ID: 183388
  • Volume: 12
  • Issue: 3
  • PageNo: 1237-1242
  • Abstract:
  • In the digital era, the integrity of visual content is often compromised due to the increasing accessibility of advanced image editing tools, making image forgery a critical issue in fields such as forensics, journalism, and legal investigations. This paper presents a texture-based image forgery detection framework that leverages Discrete Wavelet Transform (DWT) for robust texture feature extraction and employs an adaptive Support Vector Machine (SVM) optimized through a Bee Scout-inspired met heuristic algorithm. The DWT captures multi-scale texture variations caused by tampering, while the adaptive SVM model enhances classification performance by dynamically tuning hyper parameters based on selected features. Experimental evaluations on a benchmark dataset demonstrate that the proposed method achieves superior accuracy in detecting tampered regions, reflected by higher Peak Signal-to-Noise Ratio (PSNR) and lower False Rejection Rate (FRR) compared to conventional DWT-SVM approaches. This approach provides a scalable and efficient solution for authenticating digital images in real-world forensic applications.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 12
  • Issue: 3
  • PageNo: 1237-1242

Texture-Based Image Forgery Detection Using Wavelet Transform and Adaptive SVM Optimization

Related Articles