Location: Physiology and Pathology of Tree Fruits Research
Title: Detection of non-visible sun stress in ‘Granny Smith’ apples using hyperspectral imaging [abstract]Author
 |
MOGOLLON, RENE - Washington State University |
|
TORRES, CAROLINA - Washington State University |
|
Rudell, David |
|
Submitted to: American Society of Horticulture Science Meeting
Publication Type: Abstract Only Publication Acceptance Date: 4/26/2021 Publication Date: 8/9/2021 Citation: Mogollon, R., Torres, C., Rudell Jr, D.R. 2021. Detection of non-visible sun stress in ‘Granny Smith’ apples using hyperspectral imaging [abstract]. American Society of Horticulture Science Meeting. Paper No. 35134. Interpretive Summary: Technical Abstract: Climate change is triggering extreme temperatures and other climatic events in many of the world’s fruit-growing regions. These events challenge fruit production both in the orchard and during the cold chain where peel disorders such as sunscald can develop. Using hyperspectral imaging, we sought to develop spectral sorting indexes to classify ‘Granny Smith’ apples according to different sun stress levels. For this, fruit with different sun injury severity were stored in air for 6 months. Fruit were imaged monthly using a Hyperspec MV VNIR imager (400-1000 nm; Headwall Photonics, Bolton, MA). Non-sun stressed tissues showed lower reflectance between 550 to 680 nm than sun stressed ones. Reflectance at these wavelengths along with the appearance of sun-related defects. Our Vis-NIR classification model achieved an accuracy of 99% (calibration and validation) to discriminate healthy and sun-stress tissue. A cluster analysis of symptomless fruit reflectance calculated using the ratio of spectral bands (550 and 670 nm) associated with chlorophyll b and carotenoids separated fruit hemispheres and populations according to relative sun exposure. This finding could be a potential tool to identify sunscald risk in ‘Granny Smith’ apples. |
