PFNet: an unsupervised deep network for polarization image fusion

Zhang, Junchao; Shao, Jianbo; Chen, Jianlai; Yang, Dongkai; Liang, Buge; Liang, Rongguang

doi:10.1364/ol.384189

Cited by 58 publications

(12 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…processing unstructured and complex real-world data. [29][30][31] For example, deep learning networks are capable of feature expression, can simulate the human brain to some extent, and can effectively describe highly unstructured and complexly distributed data such as images and videos.…”

Section: Effective Classification and Recognitionmentioning

confidence: 99%

“…Because deep learning networks contain multiple hidden layers and can combine low‐level features to form abstract high‐level attributes of video data, they are capable of classification, comparison, and solving problems related to complex signals. These networks are commonly used for processing unstructured and complex real‐world data 29‐31 . For example, deep learning networks are capable of feature expression, can simulate the human brain to some extent, and can effectively describe highly unstructured and complexly distributed data such as images and videos.Training with a large amount of data …”

Section: Visual Detection and Tracking Of Surgical Tools Based On Dee...mentioning

confidence: 99%

See 1 more Smart Citation

Deep learning‐based intraoperative video analysis for supporting surgery

Bai

2023

Concurrency and Computation

View full text Add to dashboard Cite

SummaryThe application of surgical tool detection and tracking and surgical phase recognition is research hotspots at present. Surgical tool detection and tracking provides accurate real‐time information to surgeons, helping them plan surgeries and get advance warnings so that procedures can be safer. Surgical phase recognition can help improve the surgery efficiency, facilitate surgeons to make correct decisions, and help evaluate surgeon skill. In this article, we described the advantages and disadvantages of applying deep learning to surgical procedures. Then, we analyzed the advantages of deep learning in visual analysis, summarized the algorithms typically used in surgical tool detection and tracking and surgical phase recognition, and reviewed the results of these algorithms. The main problems are disclosed and the development trend of deep learning‐based intraoperative video analysis for supporting surgery is briefly discussed.

show abstract

Section: Effective Classification and Recognitionmentioning

confidence: 99%

Section: Visual Detection and Tracking Of Surgical Tools Based On Dee...mentioning

confidence: 99%

Deep learning‐based intraoperative video analysis for supporting surgery

Bai

2023

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Li et al proposed the PDRDN to achieve the removing of the underwater fog effect using four angle polarization pictures (0 • , 45 , 135 • ) [42]. In addition, DL based on polarization is applied to target detection [43][44][45], underwater imaging [46], image denoising [47], and image fusion [48], etc., which can get higher detection accuracy, significant noise suppression, and effective removal of the scattered light, and can obtain more detailed target information. However, data-driven network models depend too much on the data, resulting in limited generalization capabilities, which is also a major difficulty in applying deep learning to reality.…”

Section: Introductionmentioning

confidence: 99%

High-Performance Polarization Imaging Reconstruction in Scattering System under Natural Light Conditions with an Improved U-Net

Lin

Fan

et al. 2023

Photonics

View full text Add to dashboard Cite

Imaging through scattering media faces great challenges. Object information will be seriously degraded by scattering media, and the final imaging quality will be poor. In order to improve imaging quality, we propose using the transmitting characteristics of an object’s polarization information, to achieve imaging through scattering media under natural light using an improved U-net. In this paper, we choose ground glass as the scattering medium and capture the polarization images of targets through the scattering medium by a polarization camera. Experimental results show that the proposed model can reconstruct target information from highly damaged images, and for the same material object, the trained network model has a superior generalization without considering its structural shapes. Meanwhile, we have also investigated the effect of the distance between the target and the ground glass on the reconstructing performance, in which, and although the mismatch distance between the training set and the testing sample expands to 1 cm, the modified U-net can also efficaciously reconstruct the targets.

show abstract

“…Recently, the deep learning method has shown remarkable potential for solving many computational imaging problems [15][16][17][18][19][20][21] , such as phase retrieval [22][23][24][25] , phase unwrapping 26 , image denoising 27,28 , and image super-resolution 29 . Especially, the image superresolution problem has been receiving increasing attention for decades.…”

Section: Introductionmentioning

confidence: 99%

Resolution-enhanced X-ray fluorescence microscopy via deep residual networks

Bak

Shin

et al. 2023

npj Comput Mater

View full text Add to dashboard Cite

Multimodal hard X-ray scanning probe microscopy has been extensively used to study functional materials providing multiple contrast mechanisms. For instance, combining ptychography with X-ray fluorescence (XRF) microscopy reveals structural and chemical properties simultaneously. While ptychography can achieve diffraction-limited spatial resolution, the resolution of XRF is limited by the X-ray probe size. Here, we develop a machine learning (ML) model to overcome this problem by decoupling the impact of the X-ray probe from the XRF signal. The enhanced spatial resolution was observed for both simulated and experimental XRF data, showing superior performance over the state-of-the-art scanning XRF method with different nano-sized X-ray probes. Enhanced spatial resolutions were also observed for the accompanying XRF tomography reconstructions. Using this probe profile deconvolution with the proposed ML solution to enhance the spatial resolution of XRF microscopy will be broadly applicable across both functional materials and biological imaging with XRF and other related application areas.

show abstract

PFNet: an unsupervised deep network for polarization image fusion

Cited by 58 publications

References 16 publications

Deep learning‐based intraoperative video analysis for supporting surgery

Deep learning‐based intraoperative video analysis for supporting surgery

High-Performance Polarization Imaging Reconstruction in Scattering System under Natural Light Conditions with an Improved U-Net

Resolution-enhanced X-ray fluorescence microscopy via deep residual networks

Contact Info

Product

Resources

About