Photon-limited imaging through scattering medium based on deep learning

Speckle noise Ghost Imaging SIGNAL (programming language)
DOI: 10.1364/oe.27.033120 Publication Date: 2019-10-28T20:49:55Z
ABSTRACT
Imaging under ultra-weak light conditions is affected by Poisson noise heavily. The problem becomes worse if a scattering media present in the optical path. Speckle patterns detected condition carry very little information which makes it difficult to reconstruct image. Off-the-shelf methods are no longer available this condition. In paper, we experimentally demonstrate use of deep learning network images through illumination. weak limitation method analyzed. random detection obtains partial object. Based on property, demonstrated better performance our enlarging training dataset with multiple detections speckle patterns. Our results that approach can from close 1 signal photon per pixel (PPP)
SUPPLEMENTAL MATERIAL
Coming soon ....
REFERENCES (34)
CITATIONS (32)
EXTERNAL LINKS
PlumX Metrics
RECOMMENDATIONS
FAIR ASSESSMENT
Coming soon ....
JUPYTER LAB
Coming soon ....