New Advances of Privacy Protection and Multimedia Content Security for Big Data and Cloud Computing

Multimedia signal processing and Internet technology are popular in our daily lives. However, a lot of security problems have emerged correspondingly, including covert communication using multimedia files, copy-move forgery in digital images and videos, and biometric attacks. To address these problems, many techniques about multimedia information security were proposed by using the methods such as steganography, watermarking, encryption, privacy protection, and deep learning. Nevertheless, there are still some obvious shortcomings in research methodologies. Current studies for multimedia content security are usually based on strict conditions that are nearly impossible to be satisfied in real applications. Furthermore, high computation complexity of the currentmethodsmakes it difficult to handle the big data in the cloud computing environment. All these drawbacks limit the real applications of these techniques for multimedia security, and deserve additional in-depth investigations.

Multimedia signal processing and Internet technology are popular in our daily lives. However, a lot of security problems have emerged correspondingly, including covert communication using multimedia files, copy-move forgery in digital images and videos, and biometric attacks. To address these problems, many techniques about multimedia information security were proposed by using the methods such as steganography, watermarking, encryption, privacy protection, and deep learning. Nevertheless, there are still some obvious shortcomings in research methodologies. Current studies for multimedia content security are usually based on strict conditions that are nearly impossible to be satisfied in real applications. Furthermore, high computation complexity of the current methods makes it difficult to handle the big data in the cloud computing environment. All these drawbacks limit the real applications of these techniques for multimedia security, and deserve additional in-depth investigations.
This special issue will focus on the latest research on the topics of privacy protection and multimedia content security, with a particular emphasis on novel and highly efficient methods that have the potentials to be applied for big data. According to the rigorous review procedure, each manuscript submitted to this special issue was technically reviewed by at least three anonymous experts in several rounds. Finally, twelve papers were selected to be included in this special issue, which fall into four main categories: (1) Steganography (six papers), (2) Privacy Protection (two papers), (3) Encryption (two papers), and (4) Attack (two papers). In the following paragraphs, we introduce these papers briefly.

STEGANOGRAPHY
The paper, entitled "A Robust Image Steganography on Resisting JPEG Compression with No Side Information", first analyzes the error diffusion of steganographic embedding encoder, and then proposes a robust steganography with no side information based on the error diffusion analysis result and the relationship between DCT coefficients. The results based on BOSSbase 1.01 image library show that this method can resist JPEG compression effectively with acceptable robustness to steganalysis statistical detection. Different with conventional works, the paper, entitled "Constructive Steganography Using Texture Synthesis", propose to construct an intermediate texture before data hiding, thus, more candidate patches can be utilized to carry secret bits during texture synthesis with better quality of synthesized image. The paper, entitled "A Coverless Information Hiding Algorithm Based on Grayscale Gradient Co-occurrence Matrix", uses the grayscale gradient co-occurrence matrix to encode images and utilizes the mapping relationship between images and random numbers to express payload information. Experimental results show that this algorithm can be tolerant to JPEG compression and low-pass filtering. The paper, entitled "Data hiding in Iris Image for Privacy Protection", restricts the modifications of the embedding within the regions that seldom affects iris recognition with the assist of syndrome trellis coding (STC), and a new distortion function is presented to measure the impact of data embedding, where the regions with important iris features are assigned with high embedding cost. The paper, entitled "Anti-HEVC Recompression Video Watermarking Algorithm Based on the All Phase Biorthogonal Transform and SVD", embeds watermark in the quantized coefficients based on high efficiency video coding (HEVC) compression domain, which can decrease cumulative errors. In order to resist recompression attack, based on the all phase biorthogonal transform (APBT) and SVD, an anti-HEVC recompression video watermarking scheme is proposed. The paper, entitled "RTP Timestamp Steganography Detection Method", realizes timestamp-based least significant bit (LSB) steganography detection, and a clustering analysis of the area between the 5th degree polynomial best-fit curves with different message windows is conducted with high accuracy in the 2nd round of clustering. aims to solve the risk problem of location privacy disclosure in social network. Based on RCCAM access control model, a location privacy preservation scheme is proposed to assign accessing users of access permission and visibility level through the combined strategies of conflict resolution, permission allocation and location generalization. The paper, entitled "A Privacy-Preserving Image Retrieval Method Based on Improved BoVW Model in Cloud Environment", proposes a privacy-preserving image retrieval method, which uses an improved bag of visual words (BoVW) model based on Hamming embedding and orthogonal transformation to provide binary signatures refining the matching of visual words with privacy preserving at the same time.

ENCRYPTION
The paper, entitled "Efficient Fully Homomorphic Encryption with Large Plaintext Space", optimizes the rounding function of decryption procedure in Ducas and Micciancio's scheme, called as FHEW, and then designs an algorithm for extracting the most significant digit of external ciphertext by employing the homomorphic accumulator, which can enlarge the plaintext space effectively. The paper, entitled "An IBE Scheme with Verifiable Outsourced Key Generation Based on a Single Server", proposes an identity-based encryption scheme where the private key generation (PKG) outsources the task to a single server, and the results can be verified effectively. The PKG only requires the execution of one modular exponentiation, therefore, the total computing time of PKG and the server in the outsourcing algorithm is far less than that of direct computation.

ATTACK
The paper, entitled "An Attack on Hollow CAPTCHA Using Accurate Filling and Nonredundant Merging", proposes an attack method on hollow CAPTCHA through accurate filling and non-redundant merging, in which the operations of thinning, inner-outer contour filling, segmentation, minimum-nearest neighbor merging and convolutional neural network (CNN) are involved. Results show that this method has higher success rate and superior efficiency compared with some existing attack methods on hollow CAPTCHA. The paper, entitled "Exploration for Software Mitigation to Spectre Attacks of Poisoning Indirect Branches", claims that the pure software fix does impact on performance to existing software but varies depending on how the applications interact with the kernel, and the more a program relies on the kernel, the greater regression it shows. In order to alleviate this impact, a method that uses user space network stack is proposed in this paper, which is also evaluated by the fast packet I/O framework.
The publication of this special issue was achieved based on sincere cooperation. The Guest Editorial team would like to thank all the authors for the submission of their excellent works and also appreciate the technical reviewers for their precious time and hard work in reviewing. We also extend our appreciations to the IETE and Taylor & Francis staffs, especially Mrs Sandeep Kaur Mangat, for their kind and careful work. We do hope that the twelve papers published in this special issue of IETETR could benefit the readers especially the young researchers with wider range of information and better understanding in the field. Specially, we would like to deeply thank Prof M Jagadesh Kumar, Editor-in-Chief of IETE Technical Review, for his great support and guidance from the beginning to the eventual publication of this special issue. His research interests include multimedia security, image processing, and digital forensics. He has published more than 200 papers in these areas.