Assessment of Internal and External Quality of Fruits and Vegetables

Hernández-Sánchez, N.; Moreda, G.P.; Herrero-Langreo, Ana; Herreros, Ángela Melado

doi:10.1007/978-3-319-24735-9_9

Cited by 12 publications

(5 citation statements)

References 96 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These objects were chosen to reflect different shapes and tissue textures. Their CT attenuation was in the range of tissues typically found in diagnostic imaging in humans at 120 kV, with values resembling fat (apple, mean -179 HU; onion, mean -29 HU), water (orange, mean -5 HU), and soft tissue (kiwifruit, mean 45 HU) [ 27 ]. Objects were placed on a radiolucent cushion used in daily routine scanning ( Figure 1 ).…”

Section: Methodsmentioning

confidence: 99%

Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics

Euler

Laqua

Cester

et al. 2021

Cancers

View full text Add to dashboard Cite

The purpose of this study was to (i) evaluate the test–retest repeatability and reproducibility of radiomic features in virtual monoenergetic images (VMI) from dual-energy CT (DECT) depending on VMI energy (40, 50, 75, 120, 190 keV), radiation dose (5 and 15 mGy), and DECT approach (dual-source and split-filter DECT) in a phantom (ex vivo), and (ii) to assess the impact of VMI energy and feature repeatability on machine-learning-based classification in vivo in 72 patients with 72 hypodense liver lesions. Feature repeatability and reproducibility were determined by concordance–correlation–coefficient (CCC) and dynamic range (DR) ≥0.9. Test–retest repeatability was high within the same VMI energies and scan conditions (percentage of repeatable features ranging from 74% for SFDE mode at 40 keV and 15 mGy to 86% for DSDE at 190 keV and 15 mGy), while reproducibility varied substantially across different VMI energies and DECTs (percentage of reproducible features ranging from 32.8% for SFDE at 5 mGy comparing 40 with 190 keV to 99.2% for DSDE at 15 mGy comparing 40 with 50 keV). No major differences were observed between the two radiation doses (<10%) in all pair-wise comparisons. In vivo, machine learning classification using penalized regression and random forests resulted in the best discrimination of hemangiomas and metastases at low-energy VMI (40 keV), and for cysts at high-energy VMI (120 keV). Feature selection based on feature repeatability did not improve classification performance. Our results demonstrate the high repeatability of radiomics features when keeping scan and reconstruction conditions constant. Reproducibility diminished when using different VMI energies or DECT approaches. The choice of optimal VMI energy improved lesion classification in vivo and should hence be adapted to the specific task.

show abstract

Section: Methodsmentioning

confidence: 99%

Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics

Euler

Laqua

Cester

et al. 2021

Cancers

View full text Add to dashboard Cite

show abstract

“…The spatial distribution of the spectra can be used to identify objects and their attributes such as shape, size, texture, etc. (de Juan et al, 2009;Amigo et al, 2013Amigo et al, , 2015Hernández-Sánchez et al, 2016). As a consequence, HSI is inherently linked to multivariate data analysis (Amigo et al, 2013) and the particular spectral analysis to be applied depends on the objective as well as on the chemical properties of the sample (de Juan et al, 2009).…”

Section: Spectral Analysismentioning

confidence: 99%

Close range hyperspectral imaging of plants: A review

Mishra

Asaari

Herrero-Langreo³

et al. 2017

Biosystems Engineering

232

120

View full text Add to dashboard Cite

“…Additionally, vision transformers provide the ability to process high-resolution images and handle large-scale datasets, making them ideal for the task of citrus fruit classification Li et al (2020). The combination of vision transformers and machine learning in citrus fruit classification allows for a more accurate and efficient analysis of fruit attributes, leading to quality in the citrus fruit industry Hernández-Sánchez et al (2016). Considerable contributions of the study include:…”

Section: Introductionmentioning

confidence: 99%

Improving Citrus Fruit Classification Accuracy in Xray Images Using Features Enhanced Vision Transformer Architecture

Raza,

Babeker

et al. 2024

Preprint

View full text Add to dashboard Cite

Quality assessment is a cornerstone of fruit production and distribution, particularly regarding storage conditions and duration. Citrus fruits, a staple in global consumption patterns, serve as aultimate example. This study employs a non-destructive analytical technique, X-ray Computed Tomography (CT) scanning to meticulously analyze a substantial sample of 300 citrus fruits, specifically satsuma, subjected to both ambient (20-22°C, 50-60% humidity) and refrigeration conditions (6-8°C, 65-75% humidity). Through a methodologically rigorous approach, we conduct Stratified Dataset Splitting, allocating 60% of the X-ray datasets for training, with 20% dedicated to validation and testing, respectively. Our research introduces a pioneering methodology termed Feature Enhancement Vision Transformers (FEViT), meticulously designed to augment precision in 1 citrus fruit classification via X-ray image analysis. By leveraging the transforma-tive potential of Vision Transformer architecture, FEViT integrates convolutional layers and expanded input dimensions to enhance classification performance substantially. Our empirical findings unequivocally demonstrate the superior efficacy of FEViT models vis-à-vis conventional ViT counterparts across diverse datasets. Particularly noteworthy are the marked performance gains exhibited by FEViT-Large variants, evidenced by notable increases in precision (5.08%), accuracy (5.47%), recall (4.55%), and F1 scores (5.28%). This underscores the distinguish-able enhanced discriminatory prowess of FEViT models. Extensive validation through rigorous experimentation ratifies FEViT’s supremacy over traditional deep learning architectures, affirming heightened accuracy, precision, recall, and F1 scores. Our study heralds the advent of FEViT architecture as a milepost in citrus fruit image classification, promising augmented accuracy and robustness vis-à-vis extant methodologies. This research holds profound implications for the agricultural sector, especially in domains such as citrus and broader fruit classification , where nuanced image analysis is indispensable for quality assessment and informed decision-making.

show abstract

Assessment of Internal and External Quality of Fruits and Vegetables

Cited by 12 publications

References 96 publications

Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics

Virtual Monoenergetic Images of Dual-Energy CT—Impact on Repeatability, Reproducibility, and Classification in Radiomics

Close range hyperspectral imaging of plants: A review

Improving Citrus Fruit Classification Accuracy in Xray Images Using Features Enhanced Vision Transformer Architecture

Contact Info

Product

Resources

About