Scholarlink

Technological advancements have significantly contributed to the creation and processing of digital images through various image manipulation software available today. Consequently, the need for effective image forgery detection techniques has become paramount to distinguish between authentic and manipulated images. Image forensics is utilized in various fields, including national intelligence agencies, scientific publications, and social networks. One specific type of image manipulation, known as copy-move forgery, involves copying and pasting a portion of an image within the same image, resulting in an altered version of the original. In this study, we introduce a lightweight convolutional neural network (CNN) architecture combined with Support Vector Machine (SVM) classification for copy-move forgery detection (CMFD). This approach uses a similarity measure across feature patches to identify regions containing forgeries. We evaluated the effectiveness of our model using the challenging MICCF600 dataset commonly employed in CMFD research. The proposed model effectively reduces false positives, thereby enhancing pixel-level accuracy and reliability in CMFD applications.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Advanced Detection of Copy-Move Forgery Using Deep Convolutional Neural Network Features Integrated with Machine Learning-Based Classification Techniques

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Mohammed Y. Al-khuzaie,
  • Jasgurpreet Singh Chohan,
  • Hawraa Ali Sabah,
  • Protyay Dey,
  • Deeksha Verma,
  • Laith Hussein

摘要

Technological advancements have significantly contributed to the creation and processing of digital images through various image manipulation software available today. Consequently, the need for effective image forgery detection techniques has become paramount to distinguish between authentic and manipulated images. Image forensics is utilized in various fields, including national intelligence agencies, scientific publications, and social networks. One specific type of image manipulation, known as copy-move forgery, involves copying and pasting a portion of an image within the same image, resulting in an altered version of the original. In this study, we introduce a lightweight convolutional neural network (CNN) architecture combined with Support Vector Machine (SVM) classification for copy-move forgery detection (CMFD). This approach uses a similarity measure across feature patches to identify regions containing forgeries. We evaluated the effectiveness of our model using the challenging MICCF600 dataset commonly employed in CMFD research. The proposed model effectively reduces false positives, thereby enhancing pixel-level accuracy and reliability in CMFD applications.