Copy-Move Forgery Detection of Medical Images Using Most Valuable Player Based Optimization
| Autor: | J. Sasikala, K. Thirunadana Sikamani, D. Suganya |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Copy move forgery
business.industry Computer science media_common.quotation_subject Dimensionality reduction ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Pattern recognition Health informatics Image (mathematics) Principal component analysis Quality (business) Sensitivity (control systems) Artificial intelligence Electrical and Electronic Engineering business Cluster analysis Instrumentation media_common |
| Zdroj: | Sensing and Imaging. 22 |
| ISSN: | 1557-2072 1557-2064 |
| DOI: | 10.1007/s11220-021-00367-x |
| Popis: | Present day health information systems aim to provide quality patient care through digitizing the health records that include medical images. The medical images can be tampered through copy-move forgery with a malicious target of hiding some lesions or generating several copies of lesions, causing false treatment that may cost life by attackers, thereby necessitating to perform forgery detection before performing the diagnosis of the patients. The existing methods may not provide satisfactory results as the key-points (KPs) are not uniformly distributed in the entire image region. This paper attempts to develop a new copy-move forgery detection method that spreads the KPs in the whole image region, employs SURF for evaluating the features at identified KPs, applies principle component analysis for dimensionality reduction, uses most valuable player based optimization for optimal clustering of features, and performs feature matching and false match elimination. The paper exhibits the superiority of the proposed method by studying the performances such as accuracy, sensitivity, specificity, precision and F1 on 320 medical images and comparing them with those of existing methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink