Coded Aperture

Coded aperture imaging uses strategically designed masks to encode scene information onto a sensor, enabling the reconstruction of high-dimensional data like light fields or hyperspectral images from fewer measurements than traditional methods. Current research emphasizes optimizing both the coded aperture design and the reconstruction algorithms, often employing deep learning architectures like neural networks and unrolling algorithms inspired by optimization methods such as ADMM, to achieve improved accuracy and efficiency. This approach is driving advancements in compact, fast, and low-power imaging systems with applications spanning diverse fields including microscopy, medical imaging, and autonomous driving.