Image Piracy Detection on Social Media Platforms such as Instagram and Facebook

Abstract

Image piracy detection is a pressing issue on social media platforms such as Instagram and Facebook. This issue revolves around the unauthorised use of someone else\'s images without their permission, a practice that not only compromises the integrity of creators\' work but also leads to copyright infringement and violations of intellectual property rights. An application designed to detect image piracy could prove highly beneficial, particularly for photographers, artists, and various other creators who depend on their images for their livelihoods. The primary purpose of such an app is to safeguard the artistic and intellectual property rights of creators, ensuring that their work is not exploited without their consent. Developing a robust image piracy detection app would serve as a critical tool for creators, offering protection against the misuse of their content. In practice, this means the app scans the platform for instances of potentially plagiarised content and promptly notifies the content owner. For creators, this kind of solution is essential, as it allows them to focus on their creative endeavours with confidence, knowing that their work is less likely to be stolen or profited from without their knowledge. It represents a shield against the unlawful reproduction of images and provides a legal basis for action in case of copyright infringement. Image piracy detection app is the prevention of unlawful profits. Unauthorised use of someone else\'s images often results in financial losses for creators. These apps play a pivotal role in mitigating such losses by flagging and addressing instances of infringement. Additionally, they educate users about the proper use of content, fostering a culture of respect for creators\' rights and fair content usage. Furthermore, the presence of these tools acts as a deterrent, discouraging potential infringers from misusing images in the first place, which, in turn, reduces the occurrence of piracy.

Introduction

I. INTRODUCTION

In today's digital age, social media platforms such as Instagram and Facebook have revolutionised the way we share content, particularly images. These platforms provide creators with an exceptional opportunity to showcase their work to a global audience, offering unprecedented exposure and recognition. However, this immense digital landscape also comes with a significant challenge – the pervasive risk of image piracy. The unauthorised use of images without the creator's consent is a pressing concern, as it can result in copyright infringement and violations of intellectual property rights. Content creators invest their time, creativity, and resources into producing these images, making it vital to protect their work from being exploited without permission.

To address this issue, a proactive solution has been developed - an application designed to detect image piracy on major social media platforms such as Instagram and Facebook. This application relies on advanced image recognition techniques to meticulously scan and identify potentially infringing images. Once a suspicious image is identified, the app empowers users to take various actions, including reporting the image to platform administrators, issuing a takedown notice, or, in more severe cases, considering legal action. By providing a suite of options to content creators, this app acts as a critical tool in safeguarding their intellectual property rights and maintaining the integrity of their work in the digital realm.

The landscape of social networking sites has evolved dramatically, with their growing popularity granting users a more extensive and influential platform for content sharing. However, this also opens doors to misuse and misappropriation, allowing individuals to post content, including images that they do not have rights to use. This project's specific focus is on image plagiarism detection, especially within databases generated from content posted on social media platforms like Instagram and Facebook. While existing plagiarism detection systems are adept at identifying text-based plagiarism, flow-chart plagiarism, or similarities with content available elsewhere on the internet (excluding social media), there remains a substantial gap in the realm of identifying plagiarised images specifically within social media platforms.

II. LITERATURE REVIEW

1. The paper proposed by Lomborg, Stine; Bechmann, and Anja[3], highlighted the APIs provided by companies like Instagram and Facebook and how they set certain limitations to the collection of data and their privacy policy, along with information like what kind of data can be extracted. It was also understood that the data extracted from APIs are not 100 percent reliable because it's the developers who decide the data relevance.
2. In the paper proposed by Zubair Ahmed1, Prof. Mausumi Goswa[4], The power of Facebook API, and Facebook graph API was discussed in detail, understanding that using Facebook graph API we can access any kind of data including photos, videos, posts, etc. Graph API is a low-level HTTP-based API to get or post data through different techniques asks than normally an application would. Most Graph APIs an application can request by getting an access token by implementing Facebook Logins. CCEW, Department of Computer Engineering 2022-23 17
3. The system in the paper proposed by S, Akshay & B N, Chaitanya & Kumar[5], and Rishabh constitutes 80 images that are extracted from research papers through a third-party website. The images have been pre-processed using Compression, Re-sizing, and Binarization. The results obtained through comparison using subtraction are quite accurate. The F-measure value obtained is 70%. The drawback of this system is that it cannot detect plagiarism on images that are cropped to vital content.
4. The system in the paper proposed by S Sowmya Mithra, and K P Supreethi[6] works for images that are of different sizes, brightness, background, and grey-scaling (doesn’t work for A Hash computation). This system uses the Hash value computation technique to find similar images. Images are retrieved along with the hash value differences as well as a graphical representation of the same. The drawback of this system is that it works only for assignmentsubmitted images and has no mention of the numeric accuracy of the system.
5. The system proposed by Prajakta Ovhal [7] used the CBIR technique. This system consisted of 3 stages which included the pre- processing stage, feature extraction stage, and comparison stage. The feature extraction stage was then divided into 2 main parts which are color feature extraction(HSV color space) and shape feature extraction. The system used the Euclidean distance formula to calculate the similarity between any 2 images. After the comparison between the images, the results displayed were sorted using merge sort in decreasing order of similarity index.
6. In this paper by Ibrahin,[8] an improved system for the detection of plagiarism of images was proposed. The feature extracted were color features, texture features (Tamura), and Shape features (Canny Edge). The results showed were ascending order of the similarity index and whether it is true or false. The accuracy was 100% for unedited images and differed for different variations in images.
7. This paper by P. Hurtik and P. Hodakova[1] focused on searching for plagiarized images in databases. It proposed the searching algorithm FTIP based on the F^1 -transform technique(Fuzzy Transform) which was mainly used in the preprocessing step speeding up CCEW, Department of Computer Engineering 2022-23 18 the whole process of image searching. The system showed that the proposed algorithm is very fast (approximately 100ms) with a 100% success rate for the uncropped images. Whereas they also demonstrated that a 100% success rate can be achieved also for the cropped images with a slight improvement of the original algorithm.

III. ACKNOWLEDGEMENT

We would like to thank our seminar guide Prof. Y.A.Handge for timely support and guidance for the entire process of proposing this work. We would also like to thank the entire staff of the computer engineering department of Pune Institute of Computer Technology for supporting us.

Conclusion

The implementation of this project has significant implications for copyright protection on social media platforms. By automating the process of detecting image piracy, it reduces the burden on copyright holders and improves the efficiency of identifying copyright infringements. This system not only benefits individual creators and copyright owners but also contributes to maintaining a fair and ethical online environment.

References

[1] Hurtik, Petr, and Petra Hodakova. \"FTIP: A tool for an image plagiarism detection.\" In 2015 7th International Conference of Soft Computing and Pattern Recognition (SoCPaR), pp. 42-47. IEEE, 2015. [2] R. Keerthana, S. Divakaran and S. Nisha, \"A robust deep learning method for radiation induced pulmonary fibrosis disease classification in lung CT-a review,\" 2022 International Conference on Power, Energy, Control and Transmission Systems (ICPECTS), Chennai, India, 2022, pp. 1-4, doi: 10.1109/ICPECTS56089.2022.10047728. [3] Lomborg, Stine, and Anja Bechmann. \"Using APIs for data collection on social media.\" The Information Society 30, no. 4 (2014): 256-265. [4] Zubair Ahmed1, Prof. Mausumi Goswami, Prof. K. Balachandran,” The power of Facebook API”,2014. [5] Akshay, S., B. N. Chaitanya, and Rishabh Kumar. \"Image plagiarism detection using compressed images.\" IJITEE 8 (2019): 1423- 1426. [6] \"Online Plagiarism Detection for Images\", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.8, Issue 5, page no.e391-e395, May2021, [7] Available:http://www.jetir.org/papers/JETIR2105593.pdf [8] Ovhal, Prajakta. \"Detecting plagiarism in images.\" In 2015 International Conference on Information Processing (ICIP), pp. 85-89. IEEE, 2015. [9] Ibrahin, Amirul S. Bin, Othman O. Khalifa, and Diaa Eldein M. Ahmed. \"Plagiarism Detection of Images.\" In 2020 IEEE Student Conference on Research and Development (SCOReD), pp. 183- 188. IEEE, 2020. CCEW, Department of Computer Engineering 2022- 23 64 [10] Hodáková, Petra. \"Fuzzy (F-) transform of functions of two variables and its applications in image processing.\" PhD diss., Ph. D. dissertation, University of Ostrava, 2014. [11] Holcapek, Michal, and Tomáš Tichý. \"Discrete fuzzy transform of higher degree.\" In 2014 IEEE International Conference on Fuzzy Systems (FUZZ-IEEE), pp. 604-611. IEEE, 2014. [12] Zhang, Dengsheng, and Guojun Lu. \"Evaluation of similarity measurement for image retrieval.\" In International Conference on Neural Networks and Signal Processing, 2003. Proceedings of 2003, vol. 2, pp. 928-931. IEEE, 2003. [13] Chaudhari, Reshma, and A. M. Patil. \"Content-based image retrieval using color and shape features.\" International Journal of Advanced Research in Electrical, Electronics and Instrumentation Engineering 1, no. 5 (2012): 386-392. [14] Sakhare, Swati V., and Vrushali G. Nasre. \"Design of feature extraction in content-based image retrieval (CBIR) using color and texture.\" International Journal of Computer Science & Informatics 1, no. 2 (2011): 57-61.

Copyright

Copyright © 2023 Yogesh Handge, Kshitija Gondhalekar, Tanmay Thanvi, Piyusha Khandare, Dhiraj Darakhe. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.