Akademik Veri Yönetim Sistemi
IEEE Transactions on Geoscience and Remote Sensing, 2026 (SCI-Expanded, Scopus)
Accurate aerial visualization of enemy elements is essential for mission success in certain military operations, where synthetic aperture radars (SAR) are primarily employed. A reliable radar general image quality equation (GIQE) for SARs is crucial for designing operation-specific sensors or planning missions. This paper introduces a new GIQE, referred to as radar interpretability quality equation (RIQE) for SAR imaging systems to improve the prediction of image interpretability levels based on the national image interpretability rating scale (NIIRS). RIQE incorporates three critical parameters: ground sample distance, signal-to-noise ratio, and relative edge response (RER). Unlike previous approaches, this paper evaluates the combined effect of these parameters, with a particular emphasis on the contribution of RER, which is often overlooked in existing models. To develop the proposed RIQE model, a large-scale synthetic image dataset was generated to replicate a typical airborne SAR system operating in stripmap mode. Simulation results, comprising mean squared error (MSE) and confusion matrices, demonstrate that RIQE exhibits strong performance across 1 to 7 NIIRS range. We obtained the overall average MSE as 0.23 at all NIIRS levels. The confusion matrices further underscore the method’s reliability, showcasing consistent classification accuracy across different NIIRS levels. In addition, the consistency between RIQE and the NIIRS levels determined by the image analysts under more realistic conditions, such as the inclusion of land clutter scenarios, further confirms the robustness of the model. Furthermore, the practical applicability of the proposed RIQE model is validated using real SAR images. Comparative analysis with existing radar GIQE models demonstrates that RIQE provides superior NIIRS estimation accuracy in real-world scenarios, closely matching analyst evaluations. Overall, the proposed approach provides a significant improvement in SAR image quality assessment by integrating parameters that address previously overlooked aspects of SAR imaging. The inclusion of RER, along with rigorous evaluation using synthetic data, positions this model as a valuable tool for enhancing operational planning, sensor optimization, and interpretability analysis in SAR applications.