Orthogonal Moment

Orthogonal moments are mathematical tools used to represent signals, particularly images, in a way that emphasizes their inherent structure and reduces redundancy. Current research focuses on improving the computational efficiency and numerical stability of orthogonal moment calculations, particularly for high-order moments using various polynomial bases like Racah and Hahn polynomials, and integrating them into advanced architectures such as neural networks (e.g., NeRFs) for applications in image processing, computer vision, and visual servoing. These advancements enhance the robustness and accuracy of image analysis and object recognition tasks, leading to improved performance in areas like 3D reconstruction and robotic control.