Search Results (8,242)

To address the limitations of traditional cotton leaf nitrogen content estimation methods, which include low efficiency, high cost, poor portability, and challenges in vegetation index acquisition owing to environmental interference, this study focused on emerging non-destructive nutrient estimation technologies. This study proposed an innovative method that integrates multi-color space fusion with deep and machine learning to estimate cotton leaf nitrogen content using smartphone-captured digital images. A dataset comprising smartphone-acquired cotton leaf images was processed through threshold segmentation and preprocessing, then converted into RGB, HSV, and Lab color spaces. The models were developed using deep-learning architectures including AlexNet, VGGNet-11, and ResNet-50. The conclusions of this study are as follows: (1) The optimal single-color-space nitrogen estimation model achieved a validation set R² of 0.776. (2) Feature-level fusion by concatenation of multidimensional feature vectors extracted from three color spaces using the optimal model, combined with an attention learning mechanism, improved the validation R² to 0.827. (3) Decision-level fusion by concatenating nitrogen estimation values from optimal models of different color spaces into a multi-source decision dataset, followed by machine learning regression modeling, increased the final validation R² to 0.830. The dual fusion method effectively enabled rapid and accurate nitrogen estimation in cotton crops using smartphone images, achieving an accuracy 5–7% higher than that of single-color-space models. The proposed method provides scientific support for efficient cotton production and promotes sustainable development in the cotton industry. Full article

Automated quality control is critical in modern manufacturing, especially for metallic cast components, where fast and accurate surface defect detection is required. This study evaluates classical Machine Learning (ML) algorithms using extracted statistical parameters and deep learning (DL) architectures including ResNet50, Capsule Networks, and a 3D Convolutional Neural Network (CNN3D) using 3D image inputs. Using the Dataset Original, ML models with the selected parameters achieved high performance: RF reached 99.4 ± 0.2% precision and 99.4 ± 0.2% sensitivity, GB 96.0 ± 0.2% precision and 96.0 ± 0.2% sensitivity. ResNet50 trained with extracted parameters reached 98.0 ± 1.5% accuracy and 98.2 ± 1.7% F1-score. Capsule-based architectures achieved the best results, with ConvCapsuleLayer reaching 98.7 ± 0.2% accuracy and 100.0 ± 0.0% precision for the normal class, and 98.9 ± 0.2% F1-score for the affected class. CNN3D applied on 3D image inputs reached 88.61 ± 1.01% accuracy and 90.14 ± 0.95% F1-score. Using the Dataset Expanded with ML and PCA-selected features, Random Forest achieved 99.4 ± 0.2% precision and 99.4 ± 0.2% sensitivity, K-Nearest Neighbors 99.2 ± 0.0% precision and 99.2 ± 0.0% sensitivity, and SVM 99.2 ± 0.0% precision and 99.2 ± 0.0% sensitivity, demonstrating consistent high performance. All models were evaluated using repeated train-test splits to calculate averages of standard metrics (accuracy, precision, recall, F1-score), and processing times were measured, showing very low per-image execution times (as low as $3.69\times {10}^{-4}$ s/image), supporting potential real-time industrial application. These results indicate that combining statistical descriptors with ML and DL architectures provides a robust and scalable solution for automated, non-destructive surface defect detection, with high accuracy and reliability across both the original and expanded datasets. Full article

Hyperspectral imaging (HSI) has emerged as a powerful tool for non-destructive phenotyping, yet fruit crop applications remain underexplored. We propose a methodological framework to enhance the spectral characterization of apple fruits by identifying robust vegetation indices (VIs) and interpretable endmembers. We screened 284 Vis, which were evaluated using four feature selection algorithms (Boruta, MI+Lasso, RFE, and ensemble voting), generalizing across red, yellow, green, and purple apple cultivars. An ensemble criterion (≥2 algorithms) yielded 50 selected VIs from the NDSI/DSI/RSI families, preserving > 95% classification accuracy and capturing cultivar-specific variation. Pigment-sensitive wavelength bands were identified via PLS-DA VIP scores and one-vs-rest ANOVA. Using these bands, we formulated a new normalized-difference, ratio, and difference spectral indices tailored to cultivar-specific pigmentation. Several indices achieved >89% classification accuracy and showed patterns consistent with those of anthocyanin, carotenoid, and chlorophyll. A two-stage spectral unmixing pipeline (K-Means → N-FINDR) achieved the lowest reconstruction RMSE (0.043%). This multi-level strategy provides a scalable, interpretable framework for enhancing phenotypic resolution in apple hyperspectral data, contributing to fruit index development and generalized spectral analysis methods for horticultural applications. Full article

Infrared thermography non-destructive testing technology has been widely used in the defect detection of composite structures due to its advantages, including non-contact operation, rapidity, low cost, and high precision. In this study, a laser-line scanning system combined with an infrared thermography was developed, along with a corresponding dynamic sequence image reconstruction method, enabling rapid localization of surface damages. Then, high-precision quantitative characterization of defect morphology in reconstructed images was achieved by integrating an edge gradient detection algorithm. The reconstruction method was validated through finite element simulations and experimental studies. The results demonstrated that the laser-line scanning thermography effectively enables both rapid localization of surface damages and precise quantitative characterization of their morphology. Experimental measurements of ceramic materials indicate that the relative error in detecting crack width is about 6% when the crack is perpendicular to the scanning direction, and the relative error gradually increases when the angle between the crack and the scanning direction decreases. Additionally, an alumina ceramic plate with micrometer-width cracks is inspected by the continuous laser-line scanning thermography. The morphology detection results are completely consistent with the actual morphology. However, limited by the spatial resolution of the thermal imager in the experiment, the quantitative identification of the crack width cannot be carried out. Finally, the proposed method is also effective for detecting surface damage of wrinkles in ceramic matrix composites. It can localize damage and quantify its geometric features with an average relative error of less than 3%, providing a new approach for health monitoring of large-scale ceramic matrix composite structures. Full article

Spectral detection technology offers non-destructive, in situ, and high-speed capabilities, making it widely applicable for detecting biological and chemical samples and quantifying their concentrations. Water resources, essential to life on Earth, are widely distributed across the planet. The application of spectral technology to underwater environments is useful for wide-area water resource monitoring. Although spectral detection technology is well-established, its underwater application presents challenges, including waterproof housing design, power supply, and data transmission, which limit widespread application of underwater spectral detection. Furthermore, underwater spectral detection necessitates the development of compatible computational methods for sample classification or regression analysis. Focusing on underwater spectral detection, this work involved the construction of a suitable hardware system. A compact spectrometer and LEDs (400 nm–800 nm) were employed as the detection and light source modules, respectively, resulting in a compact system architecture. Extensive tests confirmed that the miniaturized design-maintained system performance. Further, this study addressed the estimation of total phosphorus (TP) concentration in water using spectral data. Samples with varying TP concentrations were prepared and calibrated against standard detection instruments. Subsequently, classification algorithms applied to the acquired spectral data enabled the in situ underwater determination of TP concentration in these samples. This work demonstrates the feasibility of underwater spectral detection for future in situ, high-speed monitoring of aquatic biochemical indicators. In the future, after adding UV LED light source, more water quality parameter information can be obtained. Full article

The main property of a steel substrate is the adhesion of its coating, which determines the quality and durability of the adhesive joint. The main objective of the research presented in this article is to evaluate the adhesion of coatings to substrates based on ultrasonic measurements and the determined reflection coefficient |r|. An experiment was carried out on disc samples, not only for ultrasonic measurements but also for the evaluation of the mechanical adhesion of coatings to substrates using the pull-off test. Three different methods of surface preparation of the samples were used: glass beading, surface treatment with P400 sandpaper, and the laser beam treatment. Based on the results, it was found that the best adhesion was obtained for samples with surfaces prepared by the glass-beading process. Reflection coefficient values in the range of 0.61–0.83 corresponded to mechanical adhesion in the range of 1.75–4.56 MPa. The results of the tests provide an important reference for the nondestructive evaluation of coating adhesion to substrates and allow for the estimation of mechanical adhesion based on the values of the reflection coefficient |r|. Full article

Fine-grained sedimentary rocks exhibit significant textural heterogeneity, often obscured by conventional grain size analysis techniques that require sample disaggregation. We propose a non-destructive, image-based grain size characterization workflow, utilizing stitched polarized thin-section photomicrographs, k-means clustering, and watershed segmentation algorithms. Validation against laser granulometry data indicates strong methodological reliability (absolute errors ranging from −5% to 3%), especially for particle sizes greater than 0.039 mm. The methodology reveals substantial internal heterogeneity within Es3 laminated shale samples from the Shahejie Formation (Bohai Bay Basin), distinctly identifying coarser siliceous laminae (grain size >0.039 mm, Φ < 8 based on Udden-Wentworth classification) indicative of high-energy depositional environments, and finer-grained clay-rich laminae (grain size <0.039 mm, Φ > 8) representing low-energy conditions. Conversely, massive mudstones exhibit comparatively homogeneous grain size distributions. Additionally, a multifractal analysis (Multifractal method) based on the S_50bi/S_50si ratio further quantifies spatial heterogeneity and pore-structure complexity, significantly enhancing facies differentiation and reservoir characterization capabilities. This method significantly improves facies differentiation ability, provides reliable constraints for shale oil reservoir characterization, and has important reference value for the exploration and development of the Bohai Bay Basin and similar petroliferous basins. Full article

This article presents an extensive experimental investigation of the dynamic characteristics of three-arch historical masonry bridges, using Operational Modal Analysis (OMA). The research thoroughly characterizes the dynamic behavior of four representative masonry bridges from the Apulia Region in Southern Italy through detailed experimental campaigns. These campaigns employed calibrated and optimally implemented accelerometric monitoring systems to acquire high-quality dynamic data under controlled excitation and environmental conditions. The selected bridges include the Santa Teresa Bridge in Bitonto, the Roman Bridge in Bovino, the Roman Bridge in Ascoli Satriano and a moderner road bridge on the Provincial Road SP123 in Troia; they span almost two millennia of construction history. The experimental framework incorporated several non-invasive excitation methods, including controlled vehicle passes, instrumented hammer impacts and ambient vibration tests, strategically chosen for optimal signal quality and heritage preservation. This investigation demonstrates the feasibility of capturing the dynamic behavior of these complex and specific historic structures through customized sensor configurations and various excitation methods. The resulting natural frequencies and mode shapes are accurate, robust, and reliable considering the extended data set used, and have allowed a rigorous seismic assessment. Eventually, this comprehensive data set establishes a fundamental basis for understanding and predicting the seismic response of several three-span masonry bridges to accurately identify their long-term resilience and effective conservation planning of these valuable and vulnerable heritage structures. In conclusion, the data comparison enabled the formulation of a predictive equation for the identification of the first natural frequency of bridges from geometric characteristics. Full article

Keratinous biomass, such as feathers, wool, and hair, poses environmental challenges due to its insoluble and recalcitrant nature. In this study, we identified, purified and comprehensively characterized a previously uncharacterized extracellular alkaline keratinase, KerFJ, secreted by Bacillus sp. FJ-3-16, with broad industrial application potential. KerFJ was produced at high yield (1800 U/mL) in an optimized cost-effective medium and purified to homogeneity using ion-exchange chromatography. The enzyme exhibited optimal activity at pH 9.5 and 55 °C, with remarkable alkaline and thermal stability, and high tolerance to surfactants, oxidants, and metal ions. Sequence analysis revealed that KerFJ is a member of the serine peptidase S8 family, with a molecular weight of ~27.5 kDa. It efficiently degraded native keratin substrates, achieving 70.3 ± 2.1% feather, 39.7 ± 1.8% wool, and 15.4 ± 1.2% hair degradation, and the resulting feather hydrolysates exhibited strong antioxidant activities. KerFJ also demonstrated excellent compatibility with commercial detergents and enabled effective stain removal from fabrics without damage. Moreover, both laboratory- and pilot-scale trials showed that KerFJ facilitated non-destructive dehairing of sheep, donkey, and pig skins while preserving collagen integrity. These results highlight KerFJ as a robust and multifunctional biocatalyst suitable for keratin waste valorization, eco-friendly leather processing, and detergent formulations. Full article

The automation of concrete constructions through 3D printing (3DP) has been increasingly developed and adopted in civil engineering due to its promising advantages over traditional construction methods. However, widespread implementation is hindered by uncertainties in quality control, homogeneity, and interlayer bonding, as well as the uniqueness of each printed component. Building upon our prior work in developing 3D-printable self-sensing cementitious materials by incorporating graphite powder and carbon microfibers into a cementitious matrix to enhance its piezoresistive properties, this study aims at enabling condition assessment of cementitious 3DP by integrating the self-sensing materials as sensing nodes within conventional components. Three different 3D-printed strip patterns, consisting of one, two, and three strip lines that mimic the pattern used in fabricating foil strain gauges were investigated as conductive electrode designs to impart strain sensing capabilities, and characterized from a series of quasi-static and dynamic tests. Results demonstrate that the three-strip design yielded the highest sensitivity (${\lambda }_{\mathrm{stat}}$ of 669, ${\lambda }_{\mathrm{dyn}}$ of 630), whereas the two-strip design produced the highest signal quality (SNR_stat = 9.5 dB, SNR_dyn = 10.8 dB). These findings confirm the feasibility of integrating 3D-printed self-sensing cementitious materials through hybrid manufacturing, enabling monitoring of print quality, detection of load path changes, and identification of potential defects. Full article

The monitoring of safety and health in lithium-ion batteries (LIBs) presents a significant challenge. Ultrasonic detection techniques fulfil the requirements for high sensitivity and non-destructive evaluation in the safety assessment of these batteries. This study concentrates on the application of electromagnetic acoustic transducer (EMAT) technology for non-destructive battery testing, utilizing non-contact electromagnetic coupling to generate and receive ultrasonic waves. This method addresses the limitations associated with conventional piezoelectric ultrasonic coupling media, thereby facilitating highly reliable assessment of the internal condition of batteries. Specifically, this paper independently designs an EMAT featuring a Halbach magnet array and a butterfly coil. Based on this design, optimization is performed, and the amplitude of the received signal is increased fourfold compared to the pre-optimization configuration. The optimized transducer is employed to evaluate a set of retired batteries with a nominal capacity of 270 Ah. Experimental results demonstrate that batteries exhibiting capacities below 240 Ah produced average signal amplitudes more than 40% lower than those of batteries with higher capacities. This technology provides a non-contact, disassembly-free approach for rapid performance evaluation of batteries and demonstrates potential for effective application in sorting retired battery units. Full article

The pore structure of SCC and of all cement-based materials plays a crucial role on the mechanical and durability characteristics of the material. The pore structure is affected by mix design, water–binder ratio and the incorporation of SCM and/or nanomaterials, all of which can improve mechanical and durability characteristics by decreasing porosity. Computed tomography (CT) is a powerful, non-destructive imaging technique to investigate the internal pore structure of concrete. The main advantage compared to other investigation techniques used to assess the pore structure is in terms of sample size. More specifically, industrial CT can be used to scan large concrete samples and be able to assess the internal pore structure without damaging the specimen. CT provides accurate measurements of pore diameters, volumes and shapes and enables the assessment of the total porosity. The paper presents the results of an experimental program on the characterization of self-compacting concrete (SCC) at a very long age (7 years) in terms of static and dynamic elastic properties and compressive and splitting tensile strength, all of which are correlated with the internal pore structure assessed via the use of an industrial Nikon XTH 450 CT. The results highlight the influence of pore volume, maximum pore diameter and sphericity on the strength and elastic properties of SCC at the age of 7 years. Both the compressive strength and the static modulus of elasticity values tend to decrease with the increase in the internal total porosity, with stronger influence on the former. Full article

Cyperus esculentus is a very destructive perennial weed, rapidly propagating and spreading through large amounts of daughter tubers. Successful control relies on depleting the soil tuber bank. This study investigated the effect of different control measures, applied across several cropping systems, on tuber bank dynamics over time. Therefore, 52 infested fields were monitored over 3 consecutive years, with annual quantification of the C. esculentus tuber bank. In maize monocropping systems, substantial 3-year tuber bank reductions (>90%) are achievable with preplant incorporation of dimethenamid-P or S-metolachlor, followed by a post-emergence application of mesotrione and pyridate at the 4–5 leaf stage, combined with delayed sowing (after 20 May) or mechanical measures (e.g., hoeing, harrowing). On non-maize fields, effective strategies (median tuber bank reductions of 57–70%) include intensive black fallow with at least four control timings or winter cereal cropping followed by intensive control (at least three measures) during the stubble phase. Established, fertilized grasslands also offer moderate reductions (17–67%) via intensive grazing or mowing. These results demonstrate that significant C. esculentus reductions are possible across different crops, but control remains challenging, requiring intensive, repeated strategies over multiple years. Less intensive approaches may undermine previous efforts. Full article

Dendrobium officinale is an important medicinal and edible plant in China, widely used in the dietary health industry and pharmaceutical field. Due to the different geographical origins and cultivation methods, the nutritional value, medicinal quality, and price of Dendrobium are significantly different, and accurate identification of the origin is crucial. Current origin identification relies on expert judgment or requires costly instruments, lacking an efficient solution. This study proposes a Variational Inference-enabled Data-Efficient Learning (VIDE) model for high-precision, non-destructive origin identification using a small number of image samples. VIDE integrates dual probabilistic networks: a prior network generating latent feature prototypes and a posterior network employing variational inference to model feature distributions via mean and variance estimators. This synergistic design enhances intra-class feature diversity while maximizing inter-class separability, achieving robust classification with limited samples. Experiments on a self-built dataset of Dendrobium officinale samples from six major Chinese regions show the VIDE model achieves 91.51% precision, 92.63% recall, and 92.07% F1-score, outperforming state-of-the-art models. The study offers a practical solution for geographical origin identification and advances intelligent quality assessment in Dendrobium officinale. Full article

Leaf area (LA) and SPAD value (a proxy for chlorophyll content) are two key determinants of seedling quality. This study aimed to develop and validate approaches for the efficient retrieval of these features in order to facilitate both management and screening practices. In cucumber, different models were developed and tested for the accurate estimation of LA at the scale of the individual organ (cotyledon, leaf) by using its linear dimensions (length (L) and width (W)), and of the whole seedling by using the 2D image-extracted projected area (from three different angles: 0°, 45°, and 90°). At either scale, the SPAD value was computed by using image (90°)-based colorimetric features. The estimation of individual organ area was more accurate when using L alone, compared with W alone. By using the two dimensions and specific colorimetric features, the individual organ area (R² ≥ 0.92) and SPAD value (R² of 0.77) were accurately predicted. When considering a single view, the top one (90°) was associated with the highest accuracy in whole-seedling LA estimation, and the side view (0°) with the lowest (R² of 0.88 and 0.73, respectively). Using any combination of two angles, the whole-seedling LA was accurately retrieved (R² ≥ 0.88). When using colorimetric features, a poor estimation of the whole-seedling SPAD value was noted (R² ≤ 0.43). The deployment of artificial neural networks (ANNs) further allowed the estimation of specific organ shape traits, and improved the accuracy of all the aforementioned predictions, including the whole-seedling SPAD value (R² of 0.597). In conclusion, the findings of this study highlight that features readily retrieved from 2D images hold promising potential for improving screening routines within the nursery industry. Full article