RT info:eu-repo/semantics/article
T1 Examination of lemon bruising using different CNN-based classifiers and local spectral-spatial hyperspectral imaging
A1 Pourdarbani, Razieh
A1 Sabzi, Sajad
A1 Dehghankar, Mohsen
A1 Rohban, Mohammad H.
A1 Arribas Sánchez, Juan Ignacio
K1 Frutas
K1 Cítricos
K1 Cítricos - Cultivo
K1 Citrus
K1 Citrus cultivation
K1 Citrus fruits
K1 Classification
K1 Neural networks (Computer science)
K1 Redes neuronales (Informática)
K1 Hyperspectral imaging
K1 Fruit
K1 Fruit - Quality
K1 Food science
K1 Machine learning
K1 Aprendizaje automático
K1 Artificial intelligence
K1 Image processing - Digital techniques
K1 Procesamiento de imágenes - Técnicas digitales.
K1 Computer mathematics
K1 Ordenadores - Matemáticas
K1 Numerical analysis
K1 Análisis numérico
K1 Bruise
K1 1203.17 Informática
K1 1203.04 Inteligencia Artificial
K1 3102 Ingeniería Agrícola
AB The presence of bruises on fruits often indicates cell damage, which can lead to a decrease in the ability of the peel to keep oxygen away from the fruits, and as a result, oxygen breaks down cell walls and membranes damaging fruit content. When chemicals in the fruit are oxidized by enzymes such as polyphenol oxidase, the chemical reaction produces an undesirable and apparent brown color effect, among others. Early detection of bruising prevents low-quality fruit from entering the consumer market. Hereupon, the present paper aims at early identification of bruised lemon fruits using 3D-convolutional neural networks (3D-CNN) via a local spectral-spatial hyperspectral imaging technique, which takes into account adjacent image pixel information in both the frequency (wavelength) and spatial domains of a 3D-tensor hyperspectral image of input lemon fruits. A total of 70 sound lemons were picked up from orchards. First, all fruits were labeled and the hyperspectral images (wavelength range 400–1100 nm) were captured as belonging to the healthy (unbruised) class (class label 0). Next, bruising was applied to each lemon by freefall. Then, the hyperspectral images of all bruised samples were captured in a time gap of 8 (class label 1) and 16 h (class label 2) after bruising was induced, thus resulting in a 3-class ternary classification problem. Four well-known 3D-CNN model namely ResNet, ShuffleNet, DenseNet, and MobileNet were used to classify bruised lemons in Python. Results revealed that the highest classification accuracy (90.47%) was obtained by the ResNet model, followed by DenseNet (85.71%), ShuffleNet (80.95%) and MobileNet (73.80%); all over the test set. ResNet model had larger parameter sizes, but it was proven to be trained faster than other models with fewer number of free parameters. ShuffleNet and MobileNet were easier to train and they needed less storage, but they could not achieve a classification error as low as the other two counterparts.
PB MDPI
SN 1999-4893
YR 2023
FD 2023
LK https://uvadoc.uva.es/handle/10324/63627
UL https://uvadoc.uva.es/handle/10324/63627
LA eng
NO Algorithms, 2023, Vol. 16, Nº. 2, 113
NO Producción Científica
DS UVaDOC
RD 27-dic-2024