Real-time object detection and classification for high-speed asymmetric-detection time-stretch optical microscopy on FPGA

Maolin Wang, Ho-Cheung Ng, Bob MF Chung, B Sharat Chandra Varma, Manish Kumar Jaiswal, Kevin K Tsia, Ho Cheung Shum, Hayden Kwok-Hay So

Research output: Contribution to conferencePaper

2 Citations (Scopus)

Abstract

A real-time object detection and classification system using FPGA developed for high-speed asymmetric time-stretched optical microscopy (ATOM) framework is presented. Due to the massive amount of data generated by optical frontend, storing the raw data for offline post-processing is slow and impractical for the targeted single cell analysis applications. The proposed FPGA solution eliminates the need to transfer and persist the entire raw data by processing low-level signals and forming high-level images in real-time. Objects of interest are detected and segmented from the image stream and a classifier subsequently performs high-level analysis on the segmented images. When compared with existing software-based post-processing workflow, this FPGA-based approach will improve both the number of objects captured per experiment and the overall end-to-end object classification performance. The system also allows co-optimization between optical system, low-level signal processing and image analytic in a unified environment that enables new scientific discoveries previously unachievable.
LanguageEnglish
Pages261-264
Number of pages4
DOIs
Publication statusPublished - 18 May 2017

Fingerprint

Optical microscopy
Field programmable gate arrays (FPGA)
Processing
Optical systems
Signal processing
Classifiers
Object detection
Experiments

Keywords

  • Field programmable gate arrays
  • Image segmentation
  • Imaging
  • Real-time systems
  • Atom optics
  • Detectors
  • Hardward

Cite this

Wang, Maolin ; Ng, Ho-Cheung ; Chung, Bob MF ; Varma, B Sharat Chandra ; Jaiswal, Manish Kumar ; Tsia, Kevin K ; Shum, Ho Cheung ; So, Hayden Kwok-Hay. / Real-time object detection and classification for high-speed asymmetric-detection time-stretch optical microscopy on FPGA. 4 p.
@conference{ca5212aa60ef489a8aa1df162bd3135a,
title = "Real-time object detection and classification for high-speed asymmetric-detection time-stretch optical microscopy on FPGA",
abstract = "A real-time object detection and classification system using FPGA developed for high-speed asymmetric time-stretched optical microscopy (ATOM) framework is presented. Due to the massive amount of data generated by optical frontend, storing the raw data for offline post-processing is slow and impractical for the targeted single cell analysis applications. The proposed FPGA solution eliminates the need to transfer and persist the entire raw data by processing low-level signals and forming high-level images in real-time. Objects of interest are detected and segmented from the image stream and a classifier subsequently performs high-level analysis on the segmented images. When compared with existing software-based post-processing workflow, this FPGA-based approach will improve both the number of objects captured per experiment and the overall end-to-end object classification performance. The system also allows co-optimization between optical system, low-level signal processing and image analytic in a unified environment that enables new scientific discoveries previously unachievable.",
keywords = "Field programmable gate arrays, Image segmentation, Imaging, Real-time systems, Atom optics, Detectors, Hardward",
author = "Maolin Wang and Ho-Cheung Ng and Chung, {Bob MF} and Varma, {B Sharat Chandra} and Jaiswal, {Manish Kumar} and Tsia, {Kevin K} and Shum, {Ho Cheung} and So, {Hayden Kwok-Hay}",
year = "2017",
month = "5",
day = "18",
doi = "10.1109/FPT.2016.7929548",
language = "English",
pages = "261--264",

}

Real-time object detection and classification for high-speed asymmetric-detection time-stretch optical microscopy on FPGA. / Wang, Maolin; Ng, Ho-Cheung; Chung, Bob MF; Varma, B Sharat Chandra; Jaiswal, Manish Kumar; Tsia, Kevin K; Shum, Ho Cheung; So, Hayden Kwok-Hay.

2017. 261-264.

Research output: Contribution to conferencePaper

TY - CONF

T1 - Real-time object detection and classification for high-speed asymmetric-detection time-stretch optical microscopy on FPGA

AU - Wang, Maolin

AU - Ng, Ho-Cheung

AU - Chung, Bob MF

AU - Varma, B Sharat Chandra

AU - Jaiswal, Manish Kumar

AU - Tsia, Kevin K

AU - Shum, Ho Cheung

AU - So, Hayden Kwok-Hay

PY - 2017/5/18

Y1 - 2017/5/18

N2 - A real-time object detection and classification system using FPGA developed for high-speed asymmetric time-stretched optical microscopy (ATOM) framework is presented. Due to the massive amount of data generated by optical frontend, storing the raw data for offline post-processing is slow and impractical for the targeted single cell analysis applications. The proposed FPGA solution eliminates the need to transfer and persist the entire raw data by processing low-level signals and forming high-level images in real-time. Objects of interest are detected and segmented from the image stream and a classifier subsequently performs high-level analysis on the segmented images. When compared with existing software-based post-processing workflow, this FPGA-based approach will improve both the number of objects captured per experiment and the overall end-to-end object classification performance. The system also allows co-optimization between optical system, low-level signal processing and image analytic in a unified environment that enables new scientific discoveries previously unachievable.

AB - A real-time object detection and classification system using FPGA developed for high-speed asymmetric time-stretched optical microscopy (ATOM) framework is presented. Due to the massive amount of data generated by optical frontend, storing the raw data for offline post-processing is slow and impractical for the targeted single cell analysis applications. The proposed FPGA solution eliminates the need to transfer and persist the entire raw data by processing low-level signals and forming high-level images in real-time. Objects of interest are detected and segmented from the image stream and a classifier subsequently performs high-level analysis on the segmented images. When compared with existing software-based post-processing workflow, this FPGA-based approach will improve both the number of objects captured per experiment and the overall end-to-end object classification performance. The system also allows co-optimization between optical system, low-level signal processing and image analytic in a unified environment that enables new scientific discoveries previously unachievable.

KW - Field programmable gate arrays

KW - Image segmentation

KW - Imaging

KW - Real-time systems

KW - Atom optics

KW - Detectors

KW - Hardward

U2 - 10.1109/FPT.2016.7929548

DO - 10.1109/FPT.2016.7929548

M3 - Paper

SP - 261

EP - 264

ER -