Earticle

초록

영어

The development of powerful image editing software has made it easy to create visually convincing digital image forgeries. Recently some research works based on general low-level visual features in the field of digital image forensics have been conducted to address this problem. However, there has been little work by analyzing the high-level semantic content of the image. This paper discusses the forgery detection problem of digital image from the point of high-level semantic, and proposes the framework of image forgery detection based on semantics. The framework consists of three components including image recognition, semantic logic reasoning engine and generation of semantic rule. A new fuzzy logic semantic reasoning model for image forgery detection is proposed in this paper. Our reasoning model based on fuzzy logic can give the deterministic and uncertainty reasoning results. Based on the proposed framework of image forgery detection, the prototype system of image forgery detection based on semantics is completed using MATLAB, MySQL database software and ontology software protégé and has the ability to detect abnormal splicing from the foreground and background, avoiding detection using the traditional methods based on low-level image features. Experimental results show that the framework and logic semantic reasoning model for image forgery detection are effective.

목차

Abstract
 1. Introduction
 2. Related Work
 3. The Proposed Framework of Image Forgery Detection Based on Semantic
  3.1. Image Recognition
  3.2. Image Forgery Detection Based on Semantic Logic Reasoning Model
  3.3. Generation of Semantic Rule
 4. Semantic Logic Reasoning Model for Image Forgery Detection
  4.1. Semantic Reasoning Rule
  4.2. Ontology Storage Structure
 5. Experimental Procedure and Results
  5.1. Experimental Environment and Procedure
  5.2 Detection Results and Discussion
 6. Conclusions and Future Work
 References

키워드

저자

• Yongzhen Ke [ School of Computer Science and Software Engineering, Tianjin Polytechnic University, China ]
• Weidong Min [ School of Computer Science and Software Engineering, Tianjin Polytechnic University, China ]
• Fan Qin [ Department of Logistics Management, Nankai University, China ]
• Junjun Shang [ School of Computer Science and Software Engineering, Tianjin Polytechnic University, China ]

참고문헌

발행기관

간행물

표지보기 관심저널 등록

• 간행물명

  International Journal of Hybrid Information Technology

• 간기

  격월간

• pISSN

  1738-9968

• 수록기간

  2008~2016

• 십진분류

  KDC 505 DDC 605