Sonar image synthesis is crucial for advancing applications in underwater exploration, marine biology, and defence. Traditional methods often rely on extensive and expensive data collection using sonar sensors, jeopardizing data quality and diversity. To overcome these limitations, this study proposes a new sonar image synthesis framework, Synth-SONAR leveraging diffusion models and GPT prompting. The key novelties of Synth-SONAR are threefold: First, by integrating Generative AI-based style injection techniques along with publicly available real/simulated data, thereby producing one of the largest sonar data corpus for sonar research. Second, a dual text-conditioning sonar diffusion model hierarchy synthesizes coarse and fine-grained sonar images with enhanced quality and diversity. Third, high-level (coarse) and low-level (detailed) text-based sonar generation methods leverage advanced semantic information available in visual language models (VLMs) and GPT-prompting. During inference, the method generates diverse and realistic sonar images from textual prompts, bridging the gap between textual descriptions and sonar image generation. This marks the application of GPT-prompting in sonar imagery for the first time, to the best of our knowledge. Synth-SONAR achieves state-of-the-art results in producing high-quality synthetic sonar datasets, significantly enhancing their diversity and realism.

A novel multimodal Explainable AI (XAI) framework known as VALE. VALE integrates image classification with SHAP for visual explanations and utilizes the Segment Anything Model (SAM) for precise object segmentation. It enhances interpretability by employing Vision Language Models (VLMs) to generate clear textual explanations from visual inputs, addressing the semantic gap between modalities. Demonstrated on tasks like ImageNet classification and specialized applications such as underwater SONAR image analysis, LAVE underscores its adaptability and performance through prompt engineering and transfer learning for real-world applications such as defense.

This research explores the application of Explainable Artificial Intelligence (XAI) to enhance interpretability in underwater image classification, marking a pioneering effort in the field. Utilizing custom datasets including Seabed Objects KLSG, camera SONAR, mine SONAR images, and SCTD, extensive analysis of transfer learning techniques is conducted across prominent Convolutional Neural Network (CNN) architectures. The research integrates Local Interpretable Model-Agnostic Explanations (LIME) and Submodular Picks LIME (SP-LIME), leveraging algorithms like Quickshift and Simple Linear Iterative Clustering (SLIC) for image perturbation. These XAI techniques provide transparent justifications for model decisions, essential for deploying AI systems in high-security domains with confidence and reliability.