טכניון מכון טכנולוגי לישראל
הטכניון מכון טכנולוגי לישראל - בית הספר ללימודי מוסמכים  
M.Sc Thesis
M.Sc StudentBourchakov Marat
SubjectHyperspectral Inspection of Egg Age
DepartmentDepartment of Quality Assurance and Reliability
Supervisor Dr. Liviu Singher


Abstract

This study explores the feasibility of non-invasive non-destructive inspection of poultry eggs by spatial hyperspectral imaging. Of special interest is the estimation of the quality of an egg through spatial inspection of its interior, hypothetically subdivided into regions, each of which represents a distinct substance. The most important of these are the albumen, yolk and the air cell.

In view of their dissimilarity in terms of composition and spatial distribution, the transmittance characteristics of an electromagnetic wave vary spatially over the egg and over time. The image processing technique allows for spatial as well as temporal changes analysis from the data obtained from a set of points on the egg’s surface, and then concludes about the mapping of the absorptive capacity of the inner egg’s contents.  

Numerous experiments were performed to investigate the ability of the infrared hyperspectral imaging system to track the changes in absorption level of the poultry egg. This required not only designing the imaging system, but also the development of high quality algorithm and software interpretation of the measured results.

The optimal spectral range for monitoring the absorption level changes inside the egg was determined. “Normalized gray levels - Wavelength” diagrams were obtained for each region. Trends of increase in the values of the normalized gray level for the 710 nm wavelength were observed for all three regions, with different acceleration rates, the maximal rate being observed for the albumen, apparently due to its maximal water content. This leads to the implementation of this system to egg age evaluation.