4.6 Article

Enhanced polarization demosaicking network via a precise angle of polarization loss calculation method

Related references

Note: Only part of the references are listed.
Article Optics

Fast Mueller matrix microscope based on dual DoFP polarimeters

Tongyu Huang et al.

Summary: The newly developed DoFPs-MMM technology shows advantages in fast polarization imaging, and demonstrated the capability to probe the polarization properties of red blood cells smear in experiments, providing an important tool for future studies.

OPTICS LETTERS (2021)

Article Optics

Polarization image demosaicking using polarization channel difference prior

Rongyuan Wu et al.

Summary: This paper introduces a demosaicking technique for polarization images using a polarization channel difference prior, which helps recover missing polarization channels by weighted fusion of three initial predictions based on different channel differences. The method improves recovery of polarization information, particularly at edges, and is fast and parallelizable for GPU acceleration, significantly reducing processing time compared to CPU-only processing.

OPTICS EXPRESS (2021)

Article Optics

Color polarization demosaicking by a convolutional neural network

Yuanyuan Sun et al.

Summary: A color polarization demosaicking convolutional neural network (CPDCNN) is proposed for addressing color polarization demosaicking problems in polarization imaging with a color polarization camera. The CPDCNN has a two-branch structure to ensure the fidelity of polarization signatures and enhance image resolution. Experimental results demonstrate that the CPDCNN outperforms other methods significantly in contrast and resolution.

OPTICS LETTERS (2021)

Article Computer Science, Artificial Intelligence

A Sparse Representation Based Joint Demosaicing Method for Single-Chip Polarized Color Sensor

Sijia Wen et al.

Summary: The paper introduces a joint chromatic and polarimetric demosaicing model to address the demosaicing issue of single-chip polarized color sensors. Researchers also established an optical data acquisition system to collect a dataset containing various sources of polarization. Experimental results demonstrate that the method is capable of faithfully recovering full RGB information of each pixel's polarization angles.

IEEE TRANSACTIONS ON IMAGE PROCESSING (2021)

Article Computer Science, Artificial Intelligence

Binary neural networks: A survey

Haotong Qin et al.

PATTERN RECOGNITION (2020)

Article Multidisciplinary Sciences

Polarimetric image recovery method combining histogram stretching for underwater imaging

Xiaobo Li et al.

SCIENTIFIC REPORTS (2018)

Article Optics

Polarimetric 3D integral imaging in photon-starved conditions

Artur Carnicer et al.

OPTICS EXPRESS (2015)