Data-driven Approximate Edge Detection using Flow-based Computing on Memristor Crossbars

Jodh Singh Pannu, Sunny Raj, Steven L. Fernandes, Sumit K. Jha, Dwaipayan Chakraborty, Sarah Rafiq, Nathaniel Cady

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Scopus citations


Detection of edges in images is an elementary operation in computer vision that can greatly benefit from an implementation with a low power-delay product. In this paper, we propose a new approach for designing nanoscale memristor crossbars that can implement approximate edge-detection using flow-based computing. Instead of the traditional Boolean approach, our methodology uses a ternary logic approach with three outcomes: True representing an edge, False that representing the absence of an edge, and Don't Care that represents an ambivalent response. Our data-driven design approach uses a corpus of human-labeled edges in order to learn the concept of an edge in an image. A massively parallel simulated annealing search algorithm over 96 processes is used to obtain the design of the memristor crossbar for edge detection. We show that our approximate crossbar design is effective in computing edges of images on the BSD500 benchmark.

Original languageEnglish (US)
Title of host publication2019 IEEE Nanotechnology Symposium, ANS 2019
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781728138701
StatePublished - Nov 2019
Externally publishedYes
Event2019 IEEE Nanotechnology Symposium, ANS 2019 - Albany, United States
Duration: Nov 12 2019Nov 13 2019

Publication series

Name2019 IEEE Nanotechnology Symposium, ANS 2019


Conference2019 IEEE Nanotechnology Symposium, ANS 2019
Country/TerritoryUnited States

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture
  • Electrical and Electronic Engineering
  • Electronic, Optical and Magnetic Materials


Dive into the research topics of 'Data-driven Approximate Edge Detection using Flow-based Computing on Memristor Crossbars'. Together they form a unique fingerprint.

Cite this