|M.Sc Student||Clara Shenderey|
|Subject||Diagnosis of Fruit-Core-Seed Disease Using NIR Spectroscopy|
|Department||Department of Agricultural Engineering||Supervisors||Professor Emeritus Shmulevich Itzhak|
|Dr. Ze'ev Schmilovich|
|Full Thesis text - in Hebrew|
Export markets show increasing demands for high-quality sorted fruits and vegetables, and the revenue from such high-quality products is much higher than the average income. However, in many cases mechanical and manual sorting is based on external indices and criteria, and lacks the ability to examine essential internal quality attributes. Nondestructive, rapid determination methods are increasingly demanded by the growers and packers.
Moldy core of apples is undetectable until the fruit is cut open or bitten into, and, therefore, can pose serious problems to both producer and consumer. One of the seed core diseases is caused by the Alternaria-alternata fungus. Infection occurs through the style of the flower, and the disease develops in the orchard within the fruits of open-calyx cultivars of the ‘Red Delicious group’. The decayed tissue spreads out from the seed cavity during fruit development at harvest and during storage.
The technology suggested in the present work is based on Near Infrared Transmittance Radiation (NIR), because the Alternaria fungus starts developing from the seed core in the majority of cases. The method is non-destructive, fast, relatively cheap, precise and simple to use.
The goal of the present research is to develop a non-destructive NIR spectroscopy based method to examine for Alternaria infection in the apple seed core.
The experiments were carried out using two devices based on NIR transmittance radiation: a stationary system and rotating system with rate of 11 apples per minute. During measuring the fruit was laid down horizontally, and measurements were carried out at three points along the cross-section perimeter. Destructive tests were conducted included cutting each apple into two equal parts and measuring the infection ratio. The experiment results were processed using the following statistical methods: Partial Least Square (PLS), Cluster Analysis and Two-Way ANOVA.
According to the experiment results the high correlation (80%-92%) was obtained when the sample contained more than 30% apples with infection ratio more than 10%. It is possible to identify Alternaria-infected apples with the infection ratio more than 10% according to the intensity at a particular wavelength and according to the gravity center of spectra at confidence level 95%.
Further research should be concentrated on extending the present method to the damage percentages of less than 10% using more sophisticated hardware and examining additional spectrum characteristics.