The global food industry prioritizes the quality and safety of fish and seafood products due to their perishable nature and the increasing consumer demand for nutritious, high-quality protein sources. Traditional quality assessment methods—such as sensory evaluation, chemical analysis, physical testing, and microbiological testing—form the foundation of current practices but face notable limitations, including subjectivity, destructiveness, and labor-intensive procedures that delay results. Motivated by the need for faster, more reliable, and non-destructive quality control systems, this review investigates how emerging technologies can address these limitations. Specifically, it aims to answer the following review questions: (1) What are the limitations of traditional fish quality assessment methods? (2) How can hyperspectral imaging (HSI) and computer vision improve the accuracy and efficiency of quality assessment? (3) What roles do machine learning and deep learning techniques play in enhancing these technologies? This review explores the integration of HSI and computer vision as cutting-edge, non-invasive technologies enabling real-time, comprehensive evaluation of key fish quality attributes such as freshness, safety, nutritional content, and species verification. The fusion of HSI and computer vision with advanced learning algorithms enhances precision in quality control, reduces food waste, and supports compliance with modern standards. Finally, the review underscores the need for continued research to drive sustainable innovation and strengthen consumer confidence.
25 April 2024 in Construction and Building Materials
As materials undergo large-scale yielding or exhibit large sizes of fracture process zone in the crack tip region, multi-parameter fracture concepts should be employed to describe the complex crack-tip stress-strain fields. Fracture resistance curves (R-curves) are an established tool in characterizing the entire fracture process of such materials. However, for complex materials such as bituminous mixtures, the development of these curves is subject to experimental and computational intricacies. In this research, a framework is developed to automate the construction of R-curves for normal and rubberized asphalt concrete (AC) mixtures. AC mixtures are produced using PG58–22 and PG58–28 binders. Limestone and siliceous aggregates are used, and three binder contents are considered for the mixtures. Single-edge notched beam (SE(B)) fracture testing is conducted on AC beams with two different notch patterns. A convolutional neural network (CNN) model is developed and trained over 1260 test images with varying temperatures, notch geometries, and setups. The CNN model was used to detect the growing crack on the beam surface and each crack-detected image was sent to an image processing framework to measure the crack length. Crack extension increments were calculated and synchronized with test time and magnitude of load, load-line displacement, and cumulative fracture energy, and the R-curve could be constructed. A training accuracy of 0.91 was obtained for the model and a loss of below 0.10 as a result of a hyperparameter tuning indicating reliable classifications by the CNN architecture. The R-curves showed desirable agreement for control mixtures at temperatures of 0 °C and −15 °C. As the mixtures are rubberized, the R-curves showed favorable agreement in the crack blunting phase, transition zone, as well as the unstable propagation phase at −20 °C. Cohesive energy magnitudes were compared for the two methods with a Pearson coefficient of 0.81 while fracture rate and fracture energy magnitudes showed favorably close magnitudes with coefficients of 0.89 and 0.98 respectively.
This study develops and characterizes a patented eco-friendly engineered cementitious composite (ECC) that incorporates waste glass powder (WGP) and silica fume (SF) as supplementary cementitious materials (SCMs) and recycled glass aggregate (WGA) as an alternative aggregate. Four stages of experimental design produced 14 concrete mixtures tested at 7, 28, 56, 90, and 120 days. Fresh and hardened properties were evaluated, and the optimal mixture, S8-1, A, achieved the requirements of strength class C60/75 and workability with slump class/ consistency class S3. Microstructural analyses using X-ray diffraction and optical microscopy confirmed the formation of secondary hydration products, particularly C-S-H and A-S-H, which contributed to matrix densification and improved performance. To complement the experimental program, an artificial neural network (ANN) was developed to predict compressive strength based on mixture proportions and curing age. Each strength measurement was treated as an independent data point, resulting in 70 samples for model training and testing. A shallow feedforward ANN with three hidden layers was implemented, trained using the Adam optimizer and validated with 10-fold cross-validation. The model achieved high predictive accuracy with R² of about 0.968, mean absolute error of 1.94 MPa, and root mean square error of 2.52 MPa. The results confirm that recycled WGP and SF can be effectively incorporated into ECC while ANN modeling provides a reliable tool for predicting compressive strength and supporting sustainable concrete mix design.
This study investigates the application of artificial intelligence for predicting the compressive strength of a high-performance, eco-efficient engineered cementitious composite (ECC), designated mix S8-1, A. The composite incorporates supplementary cementitious materials and alternative aggregates derived from recycled glass waste. The binder system combines waste glass powder and silica fume, while the aggregate fraction includes recycled cobalt glass. An extensive experimental program involving 14 mixtures tested at 7, 28, 56, 90, and 120 days was performed to establish the reference mechanical and rheological properties. Mix S8-1, A achieved strength class C60/75 and workability corresponding to consistency class S4. To substantiate long-term performance, microstructural and chemical analyses were conducted on specimens preserved since 2011, using scanning electron microscopy (SEM) and X-ray fluorescence (XRF). The results confirmed a stable, densified microstructure, evidencing the long-term durability of the patented ECC formulation. For predictive modeling, a shallow feedforward artificial neural network with three hidden layers was developed and trained on 70 dataset entries representing mixture proportions and curing ages. Model performance was evaluated using cross-validation, achieving a coefficient of determination (R2) of 0.968, a mean absolute error of 1.96 MPa, and a root mean square error of 2.52 MPa. The results demonstrate that AI-based approaches can accurately predict the compressive strength of high-performance, environmentally sustainable ECCs incorporating recycled glass constituents, supporting both performance optimization and resource-efficient material design.