Share Email Print

Proceedings Paper

VAPI: low-cost, rapid automated visual inspection system for Petri plate analysis
Author(s): L. T. Chatburn; B. C. Kirkup; M. F. Polz
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Most culture-based microbiology tasks utilize a petri plate during processing, but rarely do the scientists capture the full information available from the plate. In particular, visual analysis of plates is an under-developed rich source of data that can be rapid and non-invasive. However, collecting this data has been limited by the difficulties of standardizing and quantifying human observations, by the limits of a scientists' fatigue, and by the cost of automating the process. The availability of specialized counting equipment and intelligent camera systems has not changed this - they are prohibitively expensive for many laboratories, only process a limited number of plate types, are often destructive to the sample, and have limited accuracy. This paper describes an automated visual inspection solution, VAPI, that employs inexpensive consumer computing hardware and digital cameras along with custom cross-platform open-source software written in C++, combining Trolltech's Qt GUI toolkit with Intel's OpenCV computer vision library. The system is more accurate than common commercial systems costing many times as much, while being flexible in use and offering comparable responsiveness. VAPI not only counts colonies but also sorts and enumerates colonies by morphology, tracks colony growth by time series analysis, and provides other analytical resources. Output to XML files or directly to a database provides data that can be easily maintained and manipulated by the end user, offering ready access for system enhancement, interaction with other software systems, and rapid development of advanced analysis applications.

Paper Details

Date Published: 25 September 2007
PDF: 11 pages
Proc. SPIE 6755, Advanced Environmental, Chemical, and Biological Sensing Technologies V, 67550S (25 September 2007); doi: 10.1117/12.752526
Show Author Affiliations
L. T. Chatburn, Massachusetts Institute of Technology (United States)
B. C. Kirkup, Massachusetts Institute of Technology (United States)
M. F. Polz, Massachusetts Institute of Technology (United States)

Published in SPIE Proceedings Vol. 6755:
Advanced Environmental, Chemical, and Biological Sensing Technologies V
Tuan Vo-Dinh; Robert A. Lieberman; Günter Gauglitz, Editor(s)

© SPIE. Terms of Use
Back to Top