Food Engineering Reviews, Год журнала: 2024, Номер unknown
Опубликована: Ноя. 20, 2024
Язык: Английский
Artificial Intelligence Review, Год журнала: 2025, Номер 58(4)
Опубликована: Янв. 25, 2025
Язык: Английский
Процитировано
4
ACS Food Science & Technology, Год журнала: 2025, Номер unknown
Опубликована: Янв. 22, 2025
Язык: Английский
Процитировано
2
Smart Agricultural Technology, Год журнала: 2024, Номер 9, С. 100533 - 100533
Опубликована: Авг. 8, 2024
As the demand for food surges and agricultural sector undergoes a transformative shift towards sustainability efficiency, need precise proactive measures to ensure health welfare of livestock becomes paramount. In egg hatchery industry, hyperspectral imaging (HSI) has emerged as cutting-edge, non-destructive technique fast accurate quality analysis, including detecting chick embryo mortality. However, high cost operational complexity compared conventional RGB are significant bottlenecks in widespread adoption HSI technology. To overcome these hurdles unlock full potential HSI, promising solution is image reconstruction from standard images. This study aims reconstruct images early prediction Initially, performance different algorithms, such HRNET, MST++, Restormer, EDSR were eggs incubation period. Later, reconstructed spectra used differentiate live dead embryos using XGBoost Random Forest classification methods. Among methods, HRNET showed impressive with MRAE 0.0955, RMSE 0.0159, PSNR 36.79 dB. motivated idea that harnessing technology integrated smart sensors data analytics improve automation, enhance biosecurity, optimize resource management sustainable agriculture 4.0.
Язык: Английский
Процитировано
10
Biomass and Bioenergy, Год журнала: 2024, Номер 186, С. 107276 - 107276
Опубликована: Июнь 19, 2024
Язык: Английский
Процитировано
8
Results in Chemistry, Год журнала: 2025, Номер 13, С. 102016 - 102016
Опубликована: Янв. 1, 2025
Язык: Английский
Процитировано
1
Animals, Год журнала: 2025, Номер 15(3), С. 292 - 292
Опубликована: Янв. 21, 2025
Avian eggs are products of consumer demand, with modern methodologies for their morphometric analysis used improving quality, productivity and marketability. Such studies open up numerous prospects the introduction artificial intelligence (AI) deep learning (DL). We first consider state art DL in poultry industry, e.g., image recognition applications detection egg cracks, content freshness. comment on how algorithms need to be properly trained ask what information can gleaned from shape. Considering geometry profiles, we revisit Preston–Biggins model, Hügelschäffer’s universal models, principles universalism “The Main Axiom”, proposing a series postulates evaluate legitimacy practical application various mathematical models. stress that different models have pros cons, using them combination may yield more useful results than individual use. classic shape index alongside other alternatives, drawing conclusions about importance indices context applying going forward. Examining weight, volume, surface area air cell calculations, might applied, storage. The value is pre-incubation sorting, optimization storage periods incubation regimes, representation dimensional characteristics. Each thus combined provide synergy threshold many scientific discoveries, technological achievements industrial successes facilitated through AI DL.
Язык: Английский
Процитировано
1
Smart Agricultural Technology, Год журнала: 2025, Номер unknown, С. 100857 - 100857
Опубликована: Фев. 1, 2025
Язык: Английский
Процитировано
1
Trends in Food Science & Technology, Год журнала: 2024, Номер 148, С. 104494 - 104494
Опубликована: Апрель 23, 2024
One of the main objectives food safety programs is assurance freshness and quality as well to prolong shelf life ingredients or products along supply value chain. It recognized that prolongation could play a key role for reducing waste but real relationships between not understood due lack data. Sensing technologies data analytics have been suggested potential tools better monitor predict fresh produces. These new produced changes in way composition, overall are analysed monitored, adding with challenges manufacturing industry. The advantages limitations on utilization sensing based vibrational spectroscopy, including mid infrared (MIR), near (NIR) hyperspectral imaging assess produces reviewed. different studies evaluated concluded use (IR) sensors allow development efficient robust systems products. IR has shown develop management decision would evaluation monitoring chains.
Язык: Английский
Процитировано
7
Software Impacts, Год журнала: 2024, Номер 21, С. 100698 - 100698
Опубликована: Авг. 23, 2024
Язык: Английский
Процитировано
7
Spectrochimica Acta Part A Molecular and Biomolecular Spectroscopy, Год журнала: 2024, Номер 322, С. 124716 - 124716
Опубликована: Июнь 26, 2024
The objective of this study was to evaluate the ability a handheld near-infrared device (900-1600 nm) predict fertility and sex (male female) traits in-ovo. NIR reflectance spectra egg samples were collected on days 0, 7, 14 18 incubation data analysed using principal component analysis (PCA), linear discriminant (LDA) support vector machines classification (SVM). overall rates for prediction fertile infertile ranged from 73 % 84 between 93 95 LDA SVM classification, respectively. highest rate obtained day 7 incubation. male female embryos achieved lower rates, 62 68 Although in-ovo sexing in are higher than those by chance (50 %), results currently not sufficient industrial chicken eggs. These demonstrated that short wavelengths range may be useful distinguish at during
Язык: Английский
Процитировано
6