Scholarlink

This paper proposes a comprehensive solution to combat the growing threat of Deepfake technology, employing Convolutional Neural Networks (CNNs) and Blockchain. CNNs analyze video frames for anomalies indicative of Deepfake manipulation, while Blockchain ensures content integrity through timestamping and IPFS hashes. Leveraging transfer learning with EfficientNet-B1 architecture and dropout layers to prevent overfitting, the CNN model attains a 98.53% accuracy on testing data. Smart contracts are utilized to store social media content and AI verification results. The methodology integrates private storage on social media platforms, AI-based CNN verification, Blockchain timestamping, smart contract verification, and continuous improvement. By synergizing AI and Blockchain technologies, our approach aims to bolster defenses against deceptive content proliferation, enhancing the reliability of online media ecosystems.

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

Securing Digital Integrity: Proposed Comprehensive Framework for Deepfake Detection and Blockchain Validation

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Anant Jain,
  • Adamya Gaur,
  • Gauranshi Gupta,
  • Shubhangi Mishra,
  • Rahul Johari,
  • Deo Prakash Vidyarthi

摘要

This paper proposes a comprehensive solution to combat the growing threat of Deepfake technology, employing Convolutional Neural Networks (CNNs) and Blockchain. CNNs analyze video frames for anomalies indicative of Deepfake manipulation, while Blockchain ensures content integrity through timestamping and IPFS hashes. Leveraging transfer learning with EfficientNet-B1 architecture and dropout layers to prevent overfitting, the CNN model attains a 98.53% accuracy on testing data. Smart contracts are utilized to store social media content and AI verification results. The methodology integrates private storage on social media platforms, AI-based CNN verification, Blockchain timestamping, smart contract verification, and continuous improvement. By synergizing AI and Blockchain technologies, our approach aims to bolster defenses against deceptive content proliferation, enhancing the reliability of online media ecosystems.