Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection

Plums are widely consumed, both fresh and processed. During harvest, handling, or transportation, they are exposed to static and dynamic compression forces exceeding the critical stress for tissue damage. Compressionrelated damage typically develops further as the fruit ripens and softens, facilit...

Descripción completa

Detalles Bibliográficos
Autores principales: Castillo-Gironés, Salvador, Van Bellenghem, R., Wouters, N., Munera, Sandra, Blasco, José, Saeys, W.
Formato: Artículo
Lenguaje:Inglés
Publicado: Elsevier 2025
Materias:
Acceso en línea:https://hdl.handle.net/20.500.11939/9028
https://www.sciencedirect.com/science/article/abs/pii/S0925521423003769
_version_ 1855492610016149504
author Castillo-Gironés, Salvador
Van Bellenghem, R.
Wouters, N.
Munera, Sandra
Blasco, José
Saeys, W.
author_browse Blasco, José
Castillo-Gironés, Salvador
Munera, Sandra
Saeys, W.
Van Bellenghem, R.
Wouters, N.
author_facet Castillo-Gironés, Salvador
Van Bellenghem, R.
Wouters, N.
Munera, Sandra
Blasco, José
Saeys, W.
author_sort Castillo-Gironés, Salvador
collection ReDivia
description Plums are widely consumed, both fresh and processed. During harvest, handling, or transportation, they are exposed to static and dynamic compression forces exceeding the critical stress for tissue damage. Compressionrelated damage typically develops further as the fruit ripens and softens, facilitating the occurrence of rot, which might cause significant losses in the supply chain. Early detection of these damages is crucial to sorting the damaged fruit out and deviating it to processing, thus preventing food waste. However, early-stage bruises or damages on plums are not visible, especially not in dark-skin cultivars. Therefore, this study aimed to explore the potential of hyperspectral imaging in the 430 to 1 000 nm range and deep learning algorithms to detect these invisible bruises at an early stage. To this end, ’Presenta’ plums were impacted at three different levels to simulate varying degrees of damage. Images of both bruised and non-bruised plums were taken immediately after bruising and 24 and 48 h after bruise induction. Three distinct CNNs were trained to analyze the images. Two of these networks were implemented using transfer learning (ResNet and HSCNN), while the third was customdesigned for this specific purpose. The most informative wavelengths were identified as inputs for the CNNs employing PCA. F1 scores over 81% were obtained in all cases, and almost 100% accuracy was obtained in classifying the bruised plums with the highest impact energy of 0.50 J. Thus, detecting and classifying bruised plums using only three wavelengths is possible, paving the way for in-line sorting with multispectral cameras in packing houses.
format Artículo
id ReDivia9028
institution Instituto Valenciano de Investigaciones Agrarias (IVIA)
language Inglés
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Elsevier
publisherStr Elsevier
record_format dspace
spelling ReDivia90282025-04-25T14:49:49Z Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection Castillo-Gironés, Salvador Van Bellenghem, R. Wouters, N. Munera, Sandra Blasco, José Saeys, W. Spectral imaging Postharvest Invisible bruises Deep learning J10 Handling, transport, storage and protection of agricultural products Plums Plums are widely consumed, both fresh and processed. During harvest, handling, or transportation, they are exposed to static and dynamic compression forces exceeding the critical stress for tissue damage. Compressionrelated damage typically develops further as the fruit ripens and softens, facilitating the occurrence of rot, which might cause significant losses in the supply chain. Early detection of these damages is crucial to sorting the damaged fruit out and deviating it to processing, thus preventing food waste. However, early-stage bruises or damages on plums are not visible, especially not in dark-skin cultivars. Therefore, this study aimed to explore the potential of hyperspectral imaging in the 430 to 1 000 nm range and deep learning algorithms to detect these invisible bruises at an early stage. To this end, ’Presenta’ plums were impacted at three different levels to simulate varying degrees of damage. Images of both bruised and non-bruised plums were taken immediately after bruising and 24 and 48 h after bruise induction. Three distinct CNNs were trained to analyze the images. Two of these networks were implemented using transfer learning (ResNet and HSCNN), while the third was customdesigned for this specific purpose. The most informative wavelengths were identified as inputs for the CNNs employing PCA. F1 scores over 81% were obtained in all cases, and almost 100% accuracy was obtained in classifying the bruised plums with the highest impact energy of 0.50 J. Thus, detecting and classifying bruised plums using only three wavelengths is possible, paving the way for in-line sorting with multispectral cameras in packing houses. 2025-02-18T12:31:48Z 2025-02-18T12:31:48Z 2024 article publishedVersion Castillo-Girones, S., Van Belleghem, R., Wouters, N., Munera, S., Blasco, J., & Saeys, W. (2024). Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection. Postharvest Biology and Technology, 207, 112615. 0925-5214 https://hdl.handle.net/20.500.11939/9028 10.1016/j.postharvbio.2023.112615 https://www.sciencedirect.com/science/article/abs/pii/S0925521423003769 en This work was partially funded by projects GVA-PROMETEO CIPROM/2021/014. Salvador Castillo Giron´es thanks INIA for the FPIINIA grant number. PRE2020–094491, partially supported by European Union FSE funds and the COST action CA19145 for granting the research stay at KU Leuven. Remi Van Belleghem thanks the Research Foundation-Flanders (FWO, Brussels, Belgium) for his Ph.D. Fellowship strategic basic research (1S28522N). Sandra Munera thanks the postdoctoral contract Juan de la Cierva-Formaci´on (FJC2021–047786-I) cofunded by MCIN/AEI/10.13039/501100011033 and European Union NextGenerationEU/PRTR. The authors thank Mieke Thoelen from Belorta for providing the packing sheets and Wim Wouters from Fruithandel Romain Wouters & Co. nv for his flexibility in providing the plums. Attribution-NonCommercial-NoDerivatives 4.0 Internacional http://creativecommons.org/licenses/by-nc-nd/4.0/ openAccess Elsevier electronico
spellingShingle Spectral imaging
Postharvest
Invisible bruises
Deep learning
J10 Handling, transport, storage and protection of agricultural products
Plums
Castillo-Gironés, Salvador
Van Bellenghem, R.
Wouters, N.
Munera, Sandra
Blasco, José
Saeys, W.
Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
title Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
title_full Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
title_fullStr Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
title_full_unstemmed Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
title_short Detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
title_sort detection of subsurface bruises in plums using spectral imaging and deep learning with wavelength selection
topic Spectral imaging
Postharvest
Invisible bruises
Deep learning
J10 Handling, transport, storage and protection of agricultural products
Plums
url https://hdl.handle.net/20.500.11939/9028
https://www.sciencedirect.com/science/article/abs/pii/S0925521423003769
work_keys_str_mv AT castillogironessalvador detectionofsubsurfacebruisesinplumsusingspectralimaginganddeeplearningwithwavelengthselection
AT vanbellenghemr detectionofsubsurfacebruisesinplumsusingspectralimaginganddeeplearningwithwavelengthselection
AT woutersn detectionofsubsurfacebruisesinplumsusingspectralimaginganddeeplearningwithwavelengthselection
AT munerasandra detectionofsubsurfacebruisesinplumsusingspectralimaginganddeeplearningwithwavelengthselection
AT blascojose detectionofsubsurfacebruisesinplumsusingspectralimaginganddeeplearningwithwavelengthselection
AT saeysw detectionofsubsurfacebruisesinplumsusingspectralimaginganddeeplearningwithwavelengthselection