Files
Abstract
The overall objective of this study was to develop an automated gas sensing device to detect the presence of sour skin infected onions among healthy onions based on their volatile profiles. In the first study, a metal oxide semiconductor (MOS) based gas sensing device was developed. It is comprised of mechanical, electronic and software component. The mechanical component was designed by docking MOS sensors inside a Teflon chamber attached with a gas (or volatile) delivery system. The electronic component of the device consists of a microcontroller, memory chip, clock chip, LCD, MOS sensors and other peripheral components. Software was developed at microcontroller and personal computer level. Microcontroller level software instructs the device to function on a time-based schedule resulting in automatic operation of the device. PC-based software was designed with user interface to configure the sensor, collect data and process data for analysis. The device was characterized and tested using three different concentrations of four chemicals. In the second study, the sensors potential was tested by analyzing its response to a group of healthy and infected onions. Best baseline corrected feature was selected for this application. Based on the classification results, it was evident that the gas-sensing device was capable of identifying the presence of infected onions among healthy onions.