Abstract
Seam carving is an excellent content-aware image resizing technology widely used, and it is also a means of image tampering. Once an image is seam carved, the distribution of magnitude levels for the pixel intensity differences in the local neighborhood will be changed, which can be considered as a clue for detection of seam carving for forensic purposes. In order to accurately describe the distribution of magnitude levels for the pixel intensity differences in the local neighborhood, local neighborhood magnitude occurrence pattern (LNMOP) is proposed in this paper. The LNMOP pattern describes the distribution of intensity difference by counting up the number of magnitude level occurrences in the local neighborhood. Based on this, a forensic approach for image seam carving is proposed in this paper. Firstly, the histogram features of LNMOP and HOG (histogram of oriented gradient) are extracted from the images for seam carving forgery detection. Then, the final features for the classifier are selected from the extracted LNMOP features. The LNMOP feature selection method based on HOG feature hierarchical matching is proposed, which determines the LNMOP features to be selected by the HOG feature level. Finally, support vector machine (SVM) is utilized as a classifier to train and test by the above selected features to distinguish tampered images from normal images. In order to create training sets and test sets, images are extracted from the UCID image database. The experimental results of a large number of test images show that the proposed approach can achieve an overall better performance than the state-of-the-art approaches.
Highlights
The histogram of oriented gradient (HOG) feature level is determined, and the final features are selected according to the HOG feature level from the extracted local neighborhood magnitude occurrence pattern (LNMOP) features
(3) In view of the selection of LNMOP features, an LNMOP feature selection algorithm based on HOG feature hierarchical matching is proposed, which reduces the dimension of LNMOP and reduces the computation of classifier effectively
We propose a novel binary pattern called local neighborhood magnitude occurrence pattern (LNMOP) for the detection of seam carving
Summary
Erefore, it is the key issue to detect image seam carving efficiently on how to accurately describe the change of the magnitude levels for local neighborhood pixels In this regard, we propose an effective and novel image pattern descriptor, we term local neighborhood magnitude occurrence pattern (LNMOP), which can be used to describe the distribution of pixel intensity differences by counting up the number of magnitude levels. An image tampering detection approach using LNMOP features for seam carving is proposed, which overcomes the insensitivity of traditional Markov feature in detecting small-scale tampering and avoids the problem that the LBP feature confuses texture region and smooth region due to noise and improves the performance as a whole.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have