Kiwifruits can easily suffer damage during picking, harvest handling, transport, or other forms of mechanical stress. This issue is inconsistent with marketing requirements and may result in significant economic and reputational losses (Hou 2012). In this contest, the study aims to assess the feasibility of rapidly detecting occurrence and progression of mechanical damage on kiwifruit by a non-destructive technique: Near Infrared Hyperspectral Imaging (HSI). A representative sample of kiwifruits from the green cultivar produced in different countries (Spain and Portugal) was collected from local sellers in Bilbao, Spain. The samples were stored at room temperature (25 ± 1 °C). Various types of fruit damage were investigated through impact tests, compression tests, and their combination. Reference techniques were employed to assess destructive parameters such as pH, soluble solid contents, and titratable acidity. Hyperspectral data were acquired using a SWIR camera (Specim, Spectral Imaging Ltd., Oulu, Finland) (Figure 1a) at different sampling time after inducing the damage. The acquired spectral data were analyzed using HYPER-Tools 3 software (Mobaraki 2018). Interesting challenges arise from the physico-chemical characteristics of kiwi, including the brown colour of the peel, its opacity, and the high-water content, which may complicate HSI analysis (Liu 2023). Thus, several steps were considered to identify similarities between chemical and physical changes in the fruit images and the progression of mechanical damage. First of all, preprocessing techniques such as derivatives and scattering removal were tested. Then, principal component analysis (PCA) and K-means clustering models were developed. Finally, the relationship between moisture and the reference parameters with time was investigated (Figure 1b). The developed method could aid in preventing damaged fruits from entering the fresh market while facilitating alternative uses such smoothies’ preparation.
The story of kiwifruit trauma: seeing beyond the surface with Near Infrared Hyperspectral Imaging / S. Grassi, G. Gorla, A. Grassi, I. Locatelli, J. Manuel Amigo - In: IASIM 2024[s.l] : International Association of Spectral Imaging, 2024 Jun. - pp. 236-236 (( IASIM Bilbao 2024.
The story of kiwifruit trauma: seeing beyond the surface with Near Infrared Hyperspectral Imaging
S. Grassi
Primo
;I. LocatelliPenultimo
;
2024
Abstract
Kiwifruits can easily suffer damage during picking, harvest handling, transport, or other forms of mechanical stress. This issue is inconsistent with marketing requirements and may result in significant economic and reputational losses (Hou 2012). In this contest, the study aims to assess the feasibility of rapidly detecting occurrence and progression of mechanical damage on kiwifruit by a non-destructive technique: Near Infrared Hyperspectral Imaging (HSI). A representative sample of kiwifruits from the green cultivar produced in different countries (Spain and Portugal) was collected from local sellers in Bilbao, Spain. The samples were stored at room temperature (25 ± 1 °C). Various types of fruit damage were investigated through impact tests, compression tests, and their combination. Reference techniques were employed to assess destructive parameters such as pH, soluble solid contents, and titratable acidity. Hyperspectral data were acquired using a SWIR camera (Specim, Spectral Imaging Ltd., Oulu, Finland) (Figure 1a) at different sampling time after inducing the damage. The acquired spectral data were analyzed using HYPER-Tools 3 software (Mobaraki 2018). Interesting challenges arise from the physico-chemical characteristics of kiwi, including the brown colour of the peel, its opacity, and the high-water content, which may complicate HSI analysis (Liu 2023). Thus, several steps were considered to identify similarities between chemical and physical changes in the fruit images and the progression of mechanical damage. First of all, preprocessing techniques such as derivatives and scattering removal were tested. Then, principal component analysis (PCA) and K-means clustering models were developed. Finally, the relationship between moisture and the reference parameters with time was investigated (Figure 1b). The developed method could aid in preventing damaged fruits from entering the fresh market while facilitating alternative uses such smoothies’ preparation.| File | Dimensione | Formato | |
|---|---|---|---|
|
Abstract_IASIM_Kiwi_DEF.pdf
accesso aperto
Tipologia:
Post-print, accepted manuscript ecc. (versione accettata dall'editore)
Licenza:
Creative commons
Dimensione
192.12 kB
Formato
Adobe PDF
|
192.12 kB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.




