Award details

Physics-based machine vision

Reference	BBS/E/S/00000452
Principal Investigator / Supervisor	Professor J Marchant
Co-Investigators / Co-Supervisors
Institution	Silsoe Research Institute
Department	Silsoe Research Institute Department
Funding type	Research
Value (£)	55,900
Status	Completed
Type	Institute Project
Start date	01/04/2001
End date	31/03/2003
Duration	24 months

Abstract

The objectives are: (a) to increase the likelihood of successful machine vision by incorporating an understanding of the underlying physics. (b) To develop image analysis methods that use this understanding. Machine vision in natural outdoor environments is difficult. In order to develop effective algorithms that can compensate for some of the variability caused by natural conditions we will exploit the physics governing the formation and capture of images as a useful source of prior knowledge. We will work on general principles where appropriate but will focus on problems from the application areas of our Robotics and Automation Group and Chemical Application Group i.e. plant sensing. In particular we will investigate on(a) Whether it is possible to use passive methods (i.e. filters on ordinary cameras) to enhance small colour differences. (b) To what extent it is possible to compensate for illumination changes. (c) How to compensate for lighting variations in hyperspectral data and how much discrimating ability is lost thereby.(d)How the details of surfaces affect their reflectance properties and hence their appearance.

Summary

unavailable

Committee	Closed Committee - Agri-food (AF)
Research Topics	X – not assigned to a current Research Topic
Research Priority	X – Research Priority information not available
Research Initiative	X - not in an Initiative
Funding Scheme	X – not Funded via a specific Funding Scheme

I accept the terms and conditions of use (opens in new window)

export PDF file

back to list new search