PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.
This work tries to demonstrate the potential of Near Infrared Spectroscopy combined with class-modelling and discriminant methods, for improvement of EVOO authentication. To cover that goal, 209 olive oil samples from the three mentioned categories (68 EVOO, 93 VOO and 48 LOO) were analyzed in a FT-NIR instrument coupled to an in-line fiber optic probe. The best models developed allow to classify correctly 82% of samples as EVOO and 84.93% as VOO. These results show that NIRS technology can be a great instrumental method to replace/complement the Panel Test.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
In this study, the optimal distance between the light source and the sensor by each apple size was investigated for soluble solid content (SSC) measurement, and 1D-Convolutional Neural Network (CNN) SSC models were developed at that distance. The visible/near-infrared transmittance spectra of apple in the range of 400 to 1100 nm were measured using a 100W halogen light source. The distance between the light source and the sensor was set at three levels, which had less impact on the size of the apple investigated in the previous study. The transmission spectra of the fruit were measured at the distance of each level by size, and the SSC was also measured. 1D- CNN was used to develop SSC estimation models. The results of this study showed that 1D-CNN technology could improve the SSC measurement performance of apples. In the future, these deep learning results can be applied to a high-performance online non-destructive fruit sorter.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The objective of this study is to measure the growth status of garlic crops in open-field using VIS/NIR hyperspectral imaging system. A hyperspectral imaging system capable of acquiring a wavelength of 400 nm to 1000 nm was used, and the hyperspectral image data were analyzed by PLSR (Partial Least Square Regression), LS-SVM (Least Square Support Vector Machine), CNN (Convolutional Neural). Networks) and Spatial-Spectral Residual network (SSRN). The optimal model was able to classify the difference by fertilization levels with an accuracy of 80 to 99%, and the difference by soil covering with an accuracy of 93-99. These results show that the Vis/NIR hyperspectral imaging system and data can be utilized to predict the growth status of garlic.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
Deep-learning models were used to evaluate egg quality based on the surface condition of the eggs. Three different deep learning image classification models (EfficientNet, Swin-transformer, YOLO v5) were used for the training, and EfficientNet showed the highest accuracy with 99.7% for assessing egg quality based on the reference with 8 conditions, such as chicken manure, yolk, spot, sandy, calcium, swelling, deformed, and normal. The result demonstrates that deep learning image classification technique can be used for automated evaluation of egg quality with good accuracy.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
The food service industry must keep premises clean and free of foodborne pathogens that can be harbored in biofilms and organic residues. These may cause foodborne infections, endangering consumers and service providers. New fluorescence technology with advanced artificial intelligence algorithms can be a solution for detecting invisible contamination problems. However, improving algorithms requires access to data, raising concerns about data privacy and potential leaks of sensitive data. We present federated learning, a decentralized privacy-preserving method, to train algorithms for precisely detecting contamination in food preparation facilities and improving cleanliness while providing data privacy assurance for clients in the food-service industry.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This conference presentation was prepared for the Defense + Commercial Sensing Conference, 2023.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
This conference presentation was prepared for the Defense + Commercial Sensing Conference, 2023.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.