Sub-Nyquist computational ghost imaging with orthonormal spectrum-encoded speckle patterns

Computational ghost imaging generally requires a large number of patterns to obtain a high-quality result. It has been shown that both premodulated orthogonal patterns and postprocessing orthonormalization improve imaging quality and reduce the required pattern number. In this work, we propose and experimentally demonstrate a sub-Nyquist computational ghost imaging technique using the orthonormal spectrum-encoded speckle patterns. Our method enables the reconstruction of grayscale images at very low sampling ratios. Additionally, we show that this technique can be combined with compressive sensing to enhance image quality further. Reconstructed images are analyzed using quality indicators such as mean-square error, signal-to-noise ratio, correlation coefficient, and mean-square error of the detected edge. With our method, high-quality images can be obtained at a sampling ratio significantly lower than conventional methods.

Automated summary

In this work, a sub-Nyquist computational ghost imaging technique using orthonormal spectrum-encoded speckle patterns is proposed and experimentally demonstrated. The method enables the reconstruction of high-quality grayscale images at very low sampling ratios and can be combined with compressive sensing to further enhance image quality.