Image Registration And Tampering Detection

Published Date: 02 Nov 2017

Abstract - The trustworthiness of photographs has an essential role in many areas, including: forensic investigation, criminal investigation, medical imaging, and journalism. But, in today’s digital age, it is possible to very easily change the information. One of the main problems is the authentication of the image received in a communication. In this paper proposed a robust alignment method which makes use of an image hash component based on the Bag of Features paradigm. Forensic hash component is a short signature attached to an image before transmission and acts as side information for analyzing the processing history and trustworthiness of the received image. The estimator is based on a voting procedure. SIFT and block-based features to detect and localize image tampering. Experiments show that the proposed approach obtaining a significant margin in terms of registration accuracy, discriminative performances and tampering detection.

Index Terms—Bag of features (BOF), forensic hash, Scale invariant feature transformation, Histogram of oriented gradient, Voting procedure, tampering.

1. INTRODUCTION AND MOTIVATIONS

The widespread use of classic and newest technologies available on Internet (e.g., emails, social networks, digital repositories) has induced a growing interest on systems able to protect the visual content against malicious manipulations that could be performed during their transmission. The validity and authenticity of a received image are needed in the context of Internet communications. The problem of tampering detection can be addressed using a watermarking-based approach. The watermark is inserted into the image, and during tampering detection, it is extracted to verify if there was a malicious manipulation on the received image. A damage into the watermark indicates a tampering of the image under consideration. A clear disadvantage in using watermarking is the need for distorting the content.

To overcome this problem signature-based approaches have been introduced. In this latter case the image hash is not embedded into the image; it is associated with the image as header information and must be small and robust against different operations. Different signature-based approaches have been recently proposed. Most of them share the same basic scheme: 1) a hash code based on the visual content is attached to the image to be sent; 2) the hash is analyzed at destination to verify the reliability of the received image. An image hash is a distinctive signature which represents the visual content of the image in a compact way (usually just few bytes). The image hash should be robust against allowed operations and at the same time it should differ from the one computed on a different tampered image.

Image hashing techniques are considered extremely useful to validate the authenticity of an image received through the Internet. Despite the importance of the binary decision task related to the image authentication, in the application context of Forensic Science is fundamental to provide scientific evidences through the history of the possible manipulations applied to the original image to obtain the one under analysis. In many cases, the source image is typically unknown, and, as in the application context of this paper, all the information about the history of the image should be recovered through the short image hash signature, making more challenging the task (blind estimation of manipulations).

The list of manipulation provides to the end user the information needed to decide whether the image can be trusted or not. In order to perform tampering localization, the receiver should be able to filter out all the geometric transformations (e.g., rotation, scaling) added to the tampered image, in order to align the received image with the one at the sender. The alignment should be done in a blind way: at destination one can use only the received image and the image hash to deal with the alignment problem since the reference image does not exist. The challenging task of blindly understanding the geometric transformations occurred on a received image motivates this paper.

To make more robust the alignment phase under manipulations such as cropping and tampering, an image hash based on robust invariant features has been proposed. The latter technique extended the idea previously proposed by employing the Bag of Visual Words (BOVW) model to represent the features to be used as image hash. The exploitation of the BOVW representation is useful to reduce the space needed for the image signature. We propose a new method to detect the geometric manipulations occurred on an image starting from the hash computed on the original one. We exploit replicated visual words and a cascade of estimators to establish the geometric parameters (rotation and scale). As pointed out by the experimental results, our approach obtains the best results with a significant margin in terms of estimation accuracy with respect to state-of-the-art methods.

The remainder of the paper is organized as follows: Section 2 presents the proposed method for the alignment component, the overall registration framework, and introduces the tampering component used by the system. Section 3 report experiments an discuss both the registration performances and the tampering localization results. Finally, Section 4 concludes the paper with avenues for further research.

2. PROPOSED METHOD

2.1 Registration Component

One of the common steps of tampering detection systems is the alignment of the received image. Image registration is crucial since all the other tasks (e.g., tampering localization) usually assume that the received image is aligned with the original one, and hence could fail if the registration is not properly done. The schema of the proposed registration component is shown in Fig. 1. we adopt a BOF-based representation to reduce the dimensionality of the descriptors to be used as hash component for the alignment

In the proposed system, a codebook is generated by clustering the set of SIFT extracted on training images. The clustering procedure points out a centroid for each cluster. The set of centroids represents the codebook to be used during the image hash generation. The computed codebook is shared between sender and receiver (Fig. 1). It should be noted that the codebook is built only once, and then used for all the communications between sender and receiver.The sender extracts SIFT features and sorts them in descending order with respect to their contrast values. Afterward, the top n SIFT are selected and associated to the id label corresponding to the closest centroid belonging to the shared codebook.

Fig. 1. Overall schema of proposed registration component.

Fig. 2 Schema of proposed block description process. First each block is described by a histogram of gradient (HOG), then it is associated to a prototype belonging to a vocabulary previously generated on training samples.

Hence, the final signature for the alignment component is created by considering the id label, the dominant direction , and the keypoint coordinates(x,y) for each selected SIFT (Fig. 1). The source image and the corresponding hash component for the alignment(hs) are sent to the destination. The system assumes that the image is sent over a network consisting of possibly untrusted nodes, whereas the signature is sent upon request through a trusted authentication server which encrypts the hash in order to guarantee its integrity. During the untrusted communication the image could be manipulated for malicious purposes.

Once the image reaches the destination, the receiver generates the related hash signature for registration(hr) by using the same procedure employed by the sender. Then, the entries of the hashes hs and hr are matched by considering the id values (see Fig. 1).Note that an entry of hs may have more than one association with entries of hr (and vice versa) due to possible replicated elements in the hash signatures.After matchings are obtained, the alignment is hence performed by estimating two transformation parameters: the rotation angle and the scaling factor. To this aim, we consider the differences between dominant directions of the signature entries and the ratio of scales of the matchings between hs and hr. The estimation of the geometric parameters proceeds with a cascade of simple estimators.

The approach described in this paper deals with the problem of wrong matchings combining in cascade a filtering strategy based on the SIFT dominant direction with a robust estimator based on a voting strategy on the parameter space of a geometric transformation model.

2.2. Tampering Localization Component

Once the alignment has been performed, the image is analyzed to detect tampered regions. Tampering localization is the process of localizing the regions of the image that have been manipulated for malicious purposes to change the semantic meaning of the visual message. The tampering manipulation typically changes the properties (e.g., edges distributions, colors, textures, etc.) of some image regions. To deal with this problem the image is usually divided into non-overlapping blocks which are represented through feature vectors computed taking into account their content. The feature vectors computed at the source are then sent to a destination where these are used as forensic hash for the tampering detection component of the system. The check to localize tampered blocks is performed by the receiver taking into account the received signature and the one computed on the received image. The comparison of the signatures is performed block-wise after the alignment.

In the proposed approach, the prototypes (i.e., centroids) of the produced clusters are retained to form the vocabulary. Images at sender and receiver are first split into blocks and then each block is associated to the closest histogram prototype belonging to the shared vocabulary. Comparison between signatures is made by simply comparing the IDs of corresponding blocks after the alignment. The overall scheme related to the generation of the block representation is reported in Fig.2. Experimental results confirm the effectiveness of the proposed non-uniform quantization in terms of both compactness of the final hash signature and tampering detection accuracy.

3. EXPERIMENTAL RESULTS

This section reports a number of experiments on which the proposed approach has been tested and compared. This dataset is made up of 4485 images (average size of 244 _ 272 pixels) belonging to fifteen different scene categories at basic level of description: bedroom, suburb, industrial, kitchen, living room, coast, forest, highway, inside city, mountain, open country, street, tall building, office, store. The training set used in our experiments is built through a random selection of 150 images from the aforementioned dataset. Specifically, ten images have been randomly sampled from each scene category. The test set consists of 5250 images generated through the application of different transformations on the training images. The following image manipulations have been applied: cropping, rotation, scaling, tampering, JPEG compression. Tampering has been performed through the swapping of blocks (50 _ 50) between two images randomly selected from the training set. Various combinations of the basic transformations have been also used to make more challenging the task to be addressed.Finally, a codebook with 1000 visual words has been used to compare the approaches. The codebook has been learned through k-means clustering on the overall SIFT extracted on the training set. The results have been obtained considering sixty SIFT to build the signature of training and test images. As shown in fig.3. Additional experiments have been performed in order to examine the dependence of the average rotational and scaling error with respect to the rotation and scale transformation parameters respectively. Hence, it is better to retrieve the scale information only at the end on more reliable information filtered through the rotational estimation step.

The final step of the proposed framework is the tampering detection, i.e., the localization of the image regions that have been modified for malicious purposes. We adopt an image representation based on histogram of gradients to properly describe image blocks (see Fig. 2). Two kinds of quantization have been considered to test the proposed framework: uniform and non-uniform. Since the performance of the uniform quantization depends on the selected threshold, to properly compare uniform and non-uniform quantization the threshold has been fixed to obtain similar true positive values, the non-uniform quantization describes a single block making use of only log2k instead of 12 bits used by uniform quantization.

(b) (c) (d)

Fig. 4. Example of proposed tampering detection workflow. (a) Original image. (b) Tampered image. (c) Image registration.(d) Tampering localization.

(b)

Fig. 4. SIFT point selection: (a) SIFT selected by original image (b) tampered image

4. CONCLUSION AND FUTURE WORKS

The assessment of the reliability of an image received through the Internet is an important issue in nowadays society. In this paper the task of building a robust image signature to be used for the alignment component of distributed forensic systems has been addressed. The image registration step is crucial to perform further analysis which are commonly used to establish if a received image is trustable or not. The proposed solution exploits the Bag of Features paradigm representation, and a cascade of estimators to establish the rotational and scaling factors. The proposed hash encodes the spatial distribution of features to better deal with highly texturized and contrasted tampering patches. Experiments have been performed on a representative dataset of scenes. The proposed framework outperforms a recently appeared technique by obtaining a significant margin in terms of registration accuracy, discriminative performances and tampering detection. Future works should concern a more in-depth analysis to establish the minimal number of SIFT needed to guarantee an accurate estimation of the geometric transformations and a study in terms of bits needed to represent the overall image signature.