학술논문
Image Copy-Paste Tampering Detection Based on Superpixel Segmentation and Feature Extraction Algorithm
Document Type
Conference
Author
Source
2019 12th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI) Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI), 2019 12th International Congress on. :1-6 Oct, 2019
Subject
Language
Abstract
With the rapid advancement of the network and the popularity of image editing software, digital image forgery has been performed increasingly. People are paying more and more attention to the authenticity of the images. In the paper, a method based on superpixel segmentation and SURF algorithm is proposed for image copy-paste tampering detection. Firstly, SLIC superpixel segmentation is used to extract SURF features and feature descriptors. K-d tree is combined with BBF algorithm when feature points are matched. RANSAC algorithm removes mismatches. Morphological corrosion expansion operation shows copy-paste tampered areas. Experiments show that this method improves the accuracy and efficiency of detection.