Dynamic Scanning Augmentation

Dynamic scanning augmentation involves optimizing the order and selection of data acquisition points (e.g., pixels in an image, points in a scan) to improve the efficiency and quality of various imaging and sensing techniques. Current research focuses on applying this concept to diverse fields, including multi-modal image fusion, computed tomography, scanning probe microscopy, and remote sensing, often employing state-space models (like Mamba) or deep learning architectures (like Vision Transformers) to process the resulting data. These advancements aim to reduce data acquisition time, improve image quality and robustness to noise or artifacts, and enable fully automated operation in applications ranging from materials science to autonomous robotics.