Study of Image Classification Accuracy with Fourier Ptychography

Hongbo Zhang; Yaping Zhang; Lin Wang; Zhijuan Hu; Wenjing Zhou; Peter W. M. Tsang; Deng Cao; Ting-Chung Poon

doi:10.3390/app11104500

Study of Image Classification Accuracy with Fourier Ptychography

Hongbo Zhang^*, Yaping Zhang^*, Lin Wang, Zhijuan Hu, Wenjing Zhou, Peter W. M. Tsang, Deng Cao, Ting-Chung Poon

^*Corresponding author for this work

Department of Electrical Engineering

Research output: Journal Publications and Reviews › RGC 21 - Publication in refereed journal › peer-review

4 Citations (Scopus)

77 Downloads (CityUHK Scholars)

Abstract

In this research, the accuracy of image classification with Fourier Ptychography Microscopy (FPM) has been systematically investigated. Multiple linear regression shows a strong linear relation-ship between the results of image classification accuracy and image visual appearance quality based on PSNR and SSIM with multiple training datasets including MINST, Fashion MNIST, Cifar, Caltech 101, and customized training datasets. It is, therefore, feasible to predict the image classification accuracy only based on PSNR and SSIM. It is also found that the image classification accuracy of FPM reconstructed with higher resolution images is significantly different from using the lower resolution images under the lower numerical aperture (NA) condition. The difference is yet less pronounced under the higher NA condition.

Original language	English
Article number	4500
Journal	Applied Sciences (Switzerland)
Volume	11
Issue number	10
Online published	14 May 2021
DOIs	https://doi.org/10.3390/app11104500
Publication status	Published - May 2021

Research Keywords

Deep learning
Fourier ptychography
Image classification
Neural network

Publisher's Copyright Statement

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Access Output

10.3390/app11104500

78118941.pdfFinal Published version, 4.93 MBLicence: CC BY

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

@article{5bcda17e43074f3ea9df1deb8dd48ed6,

title = "Study of Image Classification Accuracy with Fourier Ptychography",

abstract = "In this research, the accuracy of image classification with Fourier Ptychography Microscopy (FPM) has been systematically investigated. Multiple linear regression shows a strong linear relation-ship between the results of image classification accuracy and image visual appearance quality based on PSNR and SSIM with multiple training datasets including MINST, Fashion MNIST, Cifar, Caltech 101, and customized training datasets. It is, therefore, feasible to predict the image classification accuracy only based on PSNR and SSIM. It is also found that the image classification accuracy of FPM reconstructed with higher resolution images is significantly different from using the lower resolution images under the lower numerical aperture (NA) condition. The difference is yet less pronounced under the higher NA condition.",

keywords = "Deep learning, Fourier ptychography, Image classification, Neural network",

author = "Hongbo Zhang and Yaping Zhang and Lin Wang and Zhijuan Hu and Wenjing Zhou and Tsang, {Peter W. M.} and Deng Cao and Ting-Chung Poon",

year = "2021",

month = may,

doi = "10.3390/app11104500",

language = "English",

volume = "11",

journal = "Applied Sciences (Switzerland)",

issn = "2076-3417",

publisher = "Multidisciplinary Digital Publishing Institute (MDPI)",

number = "10",

}

TY - JOUR

T1 - Study of Image Classification Accuracy with Fourier Ptychography

AU - Zhang, Hongbo

AU - Zhang, Yaping

AU - Wang, Lin

AU - Hu, Zhijuan

AU - Zhou, Wenjing

AU - Tsang, Peter W. M.

AU - Cao, Deng

AU - Poon, Ting-Chung

PY - 2021/5

Y1 - 2021/5

N2 - In this research, the accuracy of image classification with Fourier Ptychography Microscopy (FPM) has been systematically investigated. Multiple linear regression shows a strong linear relation-ship between the results of image classification accuracy and image visual appearance quality based on PSNR and SSIM with multiple training datasets including MINST, Fashion MNIST, Cifar, Caltech 101, and customized training datasets. It is, therefore, feasible to predict the image classification accuracy only based on PSNR and SSIM. It is also found that the image classification accuracy of FPM reconstructed with higher resolution images is significantly different from using the lower resolution images under the lower numerical aperture (NA) condition. The difference is yet less pronounced under the higher NA condition.

AB - In this research, the accuracy of image classification with Fourier Ptychography Microscopy (FPM) has been systematically investigated. Multiple linear regression shows a strong linear relation-ship between the results of image classification accuracy and image visual appearance quality based on PSNR and SSIM with multiple training datasets including MINST, Fashion MNIST, Cifar, Caltech 101, and customized training datasets. It is, therefore, feasible to predict the image classification accuracy only based on PSNR and SSIM. It is also found that the image classification accuracy of FPM reconstructed with higher resolution images is significantly different from using the lower resolution images under the lower numerical aperture (NA) condition. The difference is yet less pronounced under the higher NA condition.

KW - Deep learning

KW - Fourier ptychography

KW - Image classification

KW - Neural network

UR - http://www.scopus.com/inward/record.url?scp=85106575389&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85106575389&origin=recordpage

U2 - 10.3390/app11104500

DO - 10.3390/app11104500

M3 - RGC 21 - Publication in refereed journal

SN - 2076-3417

VL - 11

JO - Applied Sciences (Switzerland)

JF - Applied Sciences (Switzerland)

IS - 10

M1 - 4500

ER -

Study of Image Classification Accuracy with Fourier Ptychography

Abstract

Research Keywords

Publisher's Copyright Statement

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

GRF: Enhanced Quality Holographic Pprojection System

Cite this

Study of Image Classification Accuracy with Fourier Ptychography

Abstract

Research Keywords

Publisher's Copyright Statement

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Projects

GRF: Enhanced Quality Holographic Pprojection System

Cite this