Scholarlink

<p>IoT security technologies are increasingly used in real-time applications, including face recognition for surveillance systems. Various methods have been developed, but contemporary methods focus on face data. Video face recognition methods have recently been implemented to enhance protection systems. As deepfake technology evolves, detection methods also evolve. Researchers are developing more sophisticated approaches to keep up with AI-generated content. Combining multiple detection techniques and ongoing research collaboration is essential for effective deepfake detection and mitigation. To overcome these issues, a Secure Guard—DeepFake Deep Learning-based Deep Q Network is developed to improve the learning process. The proposed Q-Learning approach is used to start with face detection and tracking. The proposed Secure Guard framework enhanced with Blockchain techniques like Elliptic Curve Cryptography (ECC) for efficient public key cryptography, Advanced Encryption Standard (AES-256) for data encryption, and Secure Hash Algorithm (SHA-256) for data integrity, and the model is able to achieve 98% accuracy.</p>

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

Blockchain-enhanced secure guard: a deep Q network framework for robust IoT surveillance person detection

  • 期刊论文

  • 发表日期：

    2025年2月25日

查看全文
  • Srikanth Bethu,
  • Suresh Babu Erukala

摘要

IoT security technologies are increasingly used in real-time applications, including face recognition for surveillance systems. Various methods have been developed, but contemporary methods focus on face data. Video face recognition methods have recently been implemented to enhance protection systems. As deepfake technology evolves, detection methods also evolve. Researchers are developing more sophisticated approaches to keep up with AI-generated content. Combining multiple detection techniques and ongoing research collaboration is essential for effective deepfake detection and mitigation. To overcome these issues, a Secure Guard—DeepFake Deep Learning-based Deep Q Network is developed to improve the learning process. The proposed Q-Learning approach is used to start with face detection and tracking. The proposed Secure Guard framework enhanced with Blockchain techniques like Elliptic Curve Cryptography (ECC) for efficient public key cryptography, Advanced Encryption Standard (AES-256) for data encryption, and Secure Hash Algorithm (SHA-256) for data integrity, and the model is able to achieve 98% accuracy.