A custom hash algorithm for hosting secure gray scale image repository in public cloud

Nowadays, Cloud computing is an essential platform for securing resources and effectively managing files. In digital technology, many data leaks or breaches frequently occur during the storage and transmission process. Several techniques for secure image transmission have been developed by researchers worldwide. In the traditional method, data loss prevention (DLP) is the best way to protect sensitive data from breaches. The massive amount of data is not feasible in the existing storage system. However, ensuring data security remains a severe challenge. Cloud infrastructure provides a more robust, reliable, and scalable solution to overcome attacks in developing regions. The primary objective of cloud storage is to provide affordable and easy access to storage, with a vast amount of data stored across multiple cloud storage services. This paper proposed a custom block-based hash algorithm that generates a digital fingerprint from the grayscale-scale image. The pivotal contribution presented in the proposed work lies in emphasising data integrity generation and validation, tamper detection, and accurate identification of the tampered region. The entire 256 × 256 image is considered for tamper-proofing, and the hash values generated are based on the proposed work. In the integrity validation process, it compares the digest with the original digest. The cloud environment provides scalable infrastructure for securely managing and storing the digital fingerprint. User-level authentication is also incorporated into the proposed framework. Additionally, a Graphical User Interface (GUI) application has been developed for generating a hash and verifying whether the image has been tampered with or not, with the tampered region marked by a bounding box. Various benchmark metrics are analysed for validating the outfit of the proposed algorithm. The metrics, including quantitative and qualitative tests for integrity codes, collision property, and avalanche effect, were analysed, and the proposed algorithm exhibits a good ability towards integrity validation.