The modern society generates vast amounts of digital content, whose credibility plays a pivotal role in shaping public opinion and decision-making processes. The rapid development of social networks and generative technologies, such as deepfakes, significantly increases the risk of disinformation through image manipulation. This article aims to review methods for verifying images’ integrity, particularly through deep learning techniques, addressing both passive and active approaches. Their effectiveness in various scenarios has been analyzed, highlighting their advantages and limitations. This study reviews the scientific literature and research findings, focusing on techniques that detect image manipulations and localize areas of tampering, utilizing both statistical properties of images and embedded hidden watermarks. Passive methods, based on analyzing the image itself, are versatile and can be applied across a broad range of cases; however, their effectiveness depends on the complexity of the modifications and the characteristics of the image. Active methods, which involve embedding additional information into the image, offer precise detection and localization of changes but require complete control over creating and distributing visual materials. Both approaches have their applications depending on the context and available resources. In the future, a key challenge remains the development of methods resistant to advanced manipulations generated by diffusion models and further leveraging innovations in deep learning to protect the integrity of visual content.
This paper presents a method for actively safeguarding image integrity based on embedding a hidden signature generated by a neural network. The proposed solution utilizes two cooperating networks: the first one is responsible for generating a signature pattern that is difficult to detect, while the second one embeds this signature into the original image in a way that ensures high visual transparency. This automatically generated signature enables the detection of manipulations in the image and allows for localizing areas where integrity violations have occurred. The main innovation of the proposed approach lies in its ability to handle images of virtually any resolution, including widely used standards in modern communication and publications, such as HD (1280 × 720), Full HD (1920 × 1080) and even 4K (3840 × 2160). This represents a significant improvement over most existing methods, which are typically limited to the small, square images (e.g., 256 × 256 or 512 × 512 pixels) commonly found in popular object classification datasets, such as ImageNet (approximately 224 × 224 pixels). As a result, the proposed method opens new possibilities for actively securing the integrity of large and non-standard image formats while maintaining reasonable computational requirements. It surpasses previous limitations in terms of scalability and image proportions.
We present DualSecure, a resolution-agnostic semi-fragile watermarking framework that combines semantic tamper detection with provenance attribution. The proposed method embeds an imperceptible watermark that remains robust to benign transformations while degrading under semantic image alterations, enabling reliable detection, localization, and interpretation of manipulations. In addition to verifying integrity, the presented solution can identify the likely source of an image, such as a specific social media or communication platform, based on characteristic alterations introduced during its processing pipeline. The system maintains high visual transparency and is scalable to very high-resolution images, including HD, Full HD, and 2K. Experimental evaluation demonstrates strong performance in both manipulation detection and source attribution across a diverse set of image resolutions and processing conditions.
In response to the growing need for practical and scalable methods of protecting digital imagery, we propose a novel end-to-end approach based on transparent, zero-bit watermarking. Our method enables not only the detection and localization of visual manipulations but also the classification of attack types, with explicit sup-port for both signal-level perturbations and semantically meaningful edits. The system architecture comprises four specialized mod-ules: a semi-fragile encoder, a decoder for tamper localization, a binary manipulation detector, and a multi-label attack classifier. Designed with modularity and resolution-agnostic processing in mind, the pipeline supports input images of arbitrary size, including ultra high-resolution formats up to 8K. Experimental results on a custom dataset of over 304,000 images show that the proposed solution achieves high visual transparency(PSNR around 43 dB and SSIM close to 0.994), strong localization accuracy, and precise classification performance across a wide range of attacks. The system remains computationally efficient and can operate in reasonable time on standard CPUs without requiring dedicated GPU acceleration. The proposed framework demonstrates a balanced trade-off between robustness, interpretability, and deployment practicality, offering a viable solution for real-world image integrity verification.