A secured distributed detection system based on IPFS and blockchain for industrial image and video data security

https://doi.org/10.1016/j.jpdc.2021.02.022Get rights and content

Highlights

  • We present a platform of blockchain and IPFS to store multimedia objects as transactions.

  • The proposed framework ensures multimedia availability, immutability, and transparency.

  • The proposed approach is fully distributed where copyright information can be verified.

Abstract

Copyright infringement adversely affects the interest of copyright holders of images and videos which are uploaded to different websites and peer-to-peer image sharing systems. This paper addresses the problem of detecting copyright infringement so that copyright holders are given due credit for their work. There are several images and videos that are shared every day by millions of users with some amount of modification in images and videos originally uploaded by the copyright holders such as photographers, graphic designers, and video providers. Copyright violators, who are not the original creators of multimedia content modify them using image processing and frame modification techniques such as grayscale conversion, cropping, rotation, frame compression, and frame speed manipulation. Then, upload the tampered images and videos. To address this problem, we propose an IPFS-based (InterPlanetary File System-based) decentralized peer-to-peer image and video sharing platform built on blockchain technology. We use a perceptual hash (pHash) technique to detect copyright violations of multimedia. When multimedia is to be uploaded to the IPFS, the pHash of the same content is determined and checked against existing pHash values in the blockchain network. Similarity with existing pHash values would result in the multimedia being detected as tampered with. Blockchain technology offers the advantage of non-involvement of a third party and consequently the avoidance of a single point of failure.

Introduction

In this era of the Internet, many graphic designers, photographer, and video providers earn their income by selling their images and videos online through stock photo and video agencies. Several of them work with the royalty-free business stock model wherein the competition is tough. The main challenge they face is copyright violation. The term copyright violation is defined as: “making small changes in the image or video and getting credit for the work without being an original author of the same” [28]. Today, copyright violation encompasses not only the area of text documents but also different aspects of multimedia content such as image, video, and audio [40]. Moreover, copyright violation is an important issue in digital forensics.

There has been exponential growth in the sharing of data on various websites and information-sharing portals [22], [32]. However, this data which comes in many formats such as text, audio, video, and image, is often unprotected and prone to theft and modification on such a vast network as the Internet [20]. These types of data can be protected by certain means. However, somebody may use a copy of the data to gain financial benefit or for some other purpose [46]. Using a copy of data without the permission of the owner is a criminal offense under the digital forensic copyright protection act which exists in some form or other globally. Hence, researchers are developing methods to tackle this problem. Unfortunately, very little success has been achieved [43].

To avoid copyright violation, some websites match the uploaded multimedia content with existing versions on their database using traditional techniques of cryptographic hash matching such as SHA-1 and MD5. However, these techniques are characterized by the avalanche effect, which means that a minimal change in the original image or video results in a drastic change in the resultant hash [37].

Most of the digital forensic tools employ the date of creation, time of creation, and a unique hash (fingerprint) of an image or video to prevent copyright violation. However, due to the use of traditional hashing techniques which exhibit the avalanche effect, exact matches of the hashes of images and video keyframes are performed for detection of copyright violation [4]. Therefore, this approach cannot effectively detect minor changes made within the multimedia content. In this case, slight tampering of an image would escape copyright violation detection. Therefore, a neat solution to this problem is a perceptual hash function. A perceptual hash function has the property that the hash of the original input is correlated with the hash of the slightly-modified input [29]. Hence, a perceptual hash is useful for identifying image and video frames that have been tampered with which does not alter the appearance of an image or video frame significantly.

IPFS is a peer-to-peer distributed file-sharing system, which is gaining popularity mainly due to its distributed hash table (DHT) technique [8], [12]. In addition to this, IPFS also employs a unique hash of 46 bytes for each image or video, which helps in the reduction of the original image during access and storage. Moreover, the IPFS hash gets modified in the distributed hash table whenever the original image and video frame are modified by the author.

Blockchain technology provides features such as immutability, integrity, reliability, and availability of transactions. A blockchain consists of various blocks and each block contains various transactions [6], [7], [10], [39], [42]. In our work, we store the perceptual hash of each image or video which is shared on the IPFS distributed file-sharing system on the blockchain network.

In this paper, we propose a method to tackle the problem of copyright protection of images and videos using blockchain and IPFS-based file sharing. Blockchain-based copyright infringement provides significant advantages in images and videos which is shown in Fig. 1.

To the best of our knowledge, we are the first to provide a blockchain-based distributed structure to prevent copyright infringement for videos using the pHash algorithm. The summary of contributions of our work is given below:

  • 1.

    We present a platform of blockchain and IPFS (off-chain storage) to store multimedia objects (images and videos) as transactions.

  • 2.

    The proposed framework ensures availability, immutability, transparency, and protection of copyright of multimedia objects.

  • 3.

    The proposed approach is fully distributed where each peer can verify the information of copyright details by seeing its ledger. This is because copyright information gets disseminated among the peers of the blockchain network using a consensus algorithm (longest chain availability among the peers).

  • 4.

    To the best of our knowledge, we are also the first to develop a blockchain-based distributed network to store copyright details of a video.

The rest of this paper is organized as follows. Section 2 describes the related works. The proposed approach is discussed in Section 3, and its functioning is described in Section 4 along with the performance analysis, which is followed by the blockchain result analysis in Section 5. Section 6 shows the comparative analysis of the proposed model with various state-of-the-art of copyright infringement, and Section 7 concludes the paper.

Section snippets

Related work

In this section, we review concepts and previous studies (state-of-the-art), particularly those focusing on copyright infringement. The work in [40] computes the perceptual hash by combining two different approaches i.e., image-block-based and key-point-based features. They adopt Watson’s model to extract visual features. This model plays an important role to perceive image content. The underlying method can detect the local tempered regions accurately; the drawback of this model is that the

Proposed blockchain and IPFS integrated framework for copyright infringement

The main objective of this framework is to facilitate a secure distributed sharing and storage model, to detect copyright infringement for multimedia objects. The proposed framework is divided into three different modules such as (i) peers upload sections (ii) creation of addressable hash, and (iii) perceptual hash of the multimedia objects and map these hashes with appropriate peers (video provider, photographer, and graphics designer).

Copyright infringement detection of multimedia objects

This section describes the proposed IPFS blockchain-based approach to prevent copyright infringement of multimedia. First, we explain the general procedure used in the approach with the help of a flowchart. Then, we present the details of the detection process for images and videos, respectively. To meet the requirements of copyright infringement detection in peer-to-peer decentralized file storage systems, the traditional (centralized) approach needs to be extended. The objective of the

Result analysis of blockchain and IPFS

In this section, we evaluated our primary operation of IPFS and blockchain integration in the proposed framework of copyright infringement. The results of these primary operations are discussed in terms of execution time. We tested our model with 70 peers (photographer, graphics designer, and video provider) with different operations like IPFS upload, block mining, block creation, and block access.

Fig. 19 shows the computation time of varying sizes of files upload into the IPFS (off-chain)

Comparison of proposed framework with existing copyright infringement techniques

In this section, we have shown various existing works in the area of digital rights and copyright protection for multimedia objects and their limitations which is shown in Table 3. Most of the existing work is not scalable due to the centralized storage layer. This centralized storage mechanism in copyright techniques increases the chances of security and privacy threat in the field of the copyright of multimedia objects. Some of the existing work in digital right management is presented as a

Conclusion

Detection of copyright infringement is a major challenge especially in the context of multimedia objects. In this paper, we present an approach in which we employ blockchain (on-chain storage) and IPFS (off-chain storage) to share multimedia objects among peers while avoiding copyright violation. Perceptual hash (pHash) is used to detect the copyright infringement of multimedia objects. Furthermore, the IPFS hash of the original multimedia object is stored in the blockchain instead of the

CRediT authorship contribution statement

Randhir Kumar: Writing - original draft. Rakesh Tripathi: Conceptualization, Data curation, Writing - original draft. Ningrinla Marchang: Formal analysis, Writing - original draft. Gautam Srivastava: Formal analysis, Writing - original draft. Thippa Reddy Gadekallu: Funding acquisition, Investigation. Neal N. Xiong: Formal analysis, Writing - original draft.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Randhir Kumar is working toward the Ph.D. degree at the Department of Information Technology, National Institute of Technology, Raipur. He has published more than 10 research articles in the areas of Blockchain Technology and it is framework. His research interests include Blockchain Technology, Cryptography Techniques, Information Security, Web Mining, and Image Processing.

References (49)

  • DasU.K. et al.

    Digital forensic enabled image authentication using least significant bit (LSB) with tamper localization based hash function

  • DeepaN. et al.

    A survey on blockchain for big data: Approaches, opportunities, and future directions

    (2020)
  • DwivediA.D. et al.

    Optimized blockchain model for internet of things based healthcare applications

  • GadekalluT.R. et al.

    Blockchain based attack detection on machine learning algorithms for IoT based E-Health applications

    (2020)
  • A. Gangwar, E. Fidalgo, E. Alegre, V. González-Castro, PhishFingerprint: A Practical Approach for Phishing Web Page...
  • GriffinG. et al.

    Caltech-256 object category dataset

    (2007)
  • GuoJ. et al.

    Blockchain-enabled digital rights management for multimedia resources of online education

    Multimedia Tools Appl.

    (2019)
  • HamonK. et al.

    Histogram-based perceptual hashing for minimally changing video sequences

  • HavianaS.F.C. et al.

    Average hashing for perceptual image similarity in mobile phone application

    J. Telemat. Inform.

    (2016)
  • HouJ.-U. et al.

    Detection of hue modification using photo response nonuniformity

    IEEE Trans. Circuits Syst. Video Technol.

    (2016)
  • ItoK. et al.

    A critical examination of the application of blockchain technology to intellectual property management

  • IwendiC. et al.

    Keysplitwatermark: Zero watermarking algorithm for software protection against cyber-attacks

    IEEE Access

    (2020)
  • JananiT. et al.

    Secure similar image search and copyright protection over encrypted medical image databases

    IRBM

    (2020)
  • KamalM. et al.

    Blockchain-based lightweight and secured V2V communication in the internet of vehicles

    IEEE Trans. Intell. Transp. Syst.

    (2020)
  • Cited by (76)

    • A novel and secured email classification and emotion detection using hybrid deep neural network

      2024, International Journal of Cognitive Computing in Engineering
    • Zooming Into Video Conferencing Privacy

      2024, IEEE Transactions on Computational Social Systems
    View all citing articles on Scopus

    Randhir Kumar is working toward the Ph.D. degree at the Department of Information Technology, National Institute of Technology, Raipur. He has published more than 10 research articles in the areas of Blockchain Technology and it is framework. His research interests include Blockchain Technology, Cryptography Techniques, Information Security, Web Mining, and Image Processing.

    Rakesh Tripathi received his Ph.D. degree in Computer Science and Engineering from the Indian Institute of Technology Guwahati, India. He is an Assistant Professor with the Department of Information Technology, National Institute of Technology, Raipur, India. He has over ten years of experience in academics. He has published over 20 referred articles and served as a Reviewer of several journals. His research interests include Mobile-Ad hoc Networks, Sensor Networks, Data Center Networks, Distributed Systems, Network Security, Blockchain and Game Theory in Networks.

    Ningrinla Marchang (M’13) received the B.Tech. degree in Computer Science and Engineering from the North Eastern Regional Institute of Science and Technology, Itanagar, India, in 1993, the M.Tech. degree in Computer Science and Engineering from the Indian Institute of Technology (IIT) Delhi, India, in 1995, and the Ph.D. degree in Computer Science and Engineering from NERIST in 2010. From 1995 to 1996, she was a Research Engineer with the Department of Computer Science and Engineering, IIT Delhi. From 1996 to 2001, she taught at the Department of Computer Applications, Sathyabama Engineering College, Chennai, India. Since 2001, she has been a Faculty Member of NERIST, where she is an Associate Professor with the Department of Computer Science and Engineering. She is currently a Team Member involved with an Information Technology Research Academy project in cognitive radio network funded by the Government. Her research interests include mobile ad hoc networks, cognitive radio networks, and IoT.

    Dr. Gautam Srivastava was awarded his B.Sc. degree from Briar Cliff University in the U.S.A. in the year 2004, followed by his M.Sc. and Ph.D. degrees from the University of Victoria in Victoria, British Columbia, Canada in the years 2006 and 2012, respectively. He then taught for 3 years at the University of Victoria in the Department of Computer Science, where he was regarded as one of the top undergraduate professors in the Computer Science Course Instruction at the University. From there in the year 2014, he joined a tenure-track position at Brandon University in Brandon, Manitoba, Canada, where he currently is active in various professional and scholarly activities. He was promoted to the rank Associate Professor in January 2018. Dr. G, as he is popularly known, is active in research in the field of Data Mining and Big Data. In his 8-year academic career, he has published a total of 160 papers in high-impact conferences in many countries and in high-status journals (SCI, SCIE) and has also delivered invited guest lectures on Big Data, Cloud Computing, Internet of Things, and Cryptography at many Taiwanese and Czech universities. He is an Editor of several international scientific research journals. He currently has active research projects with other academics in Taiwan, Singapore, Canada, Czech Republic, Poland and U.S.A. He is constantly looking for collaboration opportunities with foreign professors and students. Assoc. Prof. Gautam Srivastava is funded at the national level in Canada by the Natural Sciences and Engineering Research Council of Canada (NSERC) and Mathematics of Information Technology and Complex Systems (MITACS). He is also an IEEE Senior Member.

    Dr. Thippa Reddy G is currently working as Assistant Professor (Senior) at School of Information Technology and Engineering, VIT, Vellore, Tamil Nadu, India. He obtained his B.Tech. in CSE from Nagarjuna University, A.P., M.Tech. in CSE from Anna University, Chennai, Tamil Nadu, India and completed his Ph.D. at VIT, Vellore, Tamil Nadu, India. He has 14 years of experience in teaching. He produced more than 25 international/national publications. Currently, he is working in the area of Machine Learning, Internet of Things, Deep Neural Networks, Blockchain.

    Neal N. Xiong (S’05–M’08–SM’12) is currently an Associate Professor (5rd year) at Department of Mathematics and Computer Science, Northeastern State University, OK, USA. He received both his Ph.D. degrees from Wuhan University (2007, on sensor system engineering), and Japan Advanced Institute of Science and Technology (2008, on dependable communication networks), respectively. Before he attended Northeastern State University, he worked in Georgia State University, Wentworth Technology Institution, and Colorado Technical University (full professor for about 5 years) for about 10 years. His research interests include Cloud Computing, Security and Dependability, Parallel and Distributed Computing, Networks, and Optimization Theory.

    View full text