An Approach for Detection of Copy-Move Forgery of Images Using Histogram of Orientated Gradients and Prewitt Filter
Paper ID : 1125-ICTCK (R3)
1مطهره بختیارپاکزاد *, 2علی اکبر خزاعی
1دانشگاه آزاد بجنورد
2دانشگاه آزاد مشهد
With the emergence of power image editing tools, it is now simply possible to manipulate the image, change the content, and edit or remove important features of an image. The operation of identifying manipulated images is of high importance since several millions of images are sent to the internet every day. The most common type of forgery in digital images is the copy-move forgery. The most common type of image forgery in the digital images is copy-move forgery. In this research, histogram of orientated gradients imposed to each block and statistical features extracted. Here, features are reduced which has improved results relative to previous works. The results of the experiment show that the proposed method is an effective and useful scheme with a degree of diagnostic accuracy of 77% for images obtained from the "comofod" database and the ability to detect forged images using copy-move method, even when the image is distorted via transformations such as changes in brightness, color reduction , blurring or JPEG compression.
Keywords: Digital image forgery detection, HOG, Copy-move image forgery, Feature extraction
Status : Paper Accepted