Search Results (3,245)

This study presents an advanced framework for fabric texture classification by leveraging macro- and micro-texture extraction techniques integrated with deep learning architectures. Co-occurrence histograms, local binary patterns (LBPs), and albedo-dependent feature maps were employed to comprehensively capture the surface properties of fabrics. A late fusion approach was applied using four state-of-the-art convolutional neural networks (CNNs): InceptionV3, ResNet50_V2, DenseNet, and VGG-19. Excellent results were obtained, with the ResNet50_V2 achieving a precision of 0.929, recall of 0.914, and F1 score of 0.913. Notably, the integration of multimodal inputs allowed the models to effectively distinguish challenging fabric types, such as cotton–polyester and satin–silk pairs, which exhibit overlapping texture characteristics. This research not only enhances the accuracy of textile classification but also provides a robust methodology for material analysis, with significant implications for industrial applications in fashion, quality control, and robotics. Full article

Remote sensing images are often degraded by atmospheric haze, which not only reduces image quality but also complicates information extraction, particularly in high-level visual analysis tasks such as object detection and scene classification. State-space models (SSMs) have recently emerged as a powerful paradigm for vision tasks, showing great promise due to their computational efficiency and robust capacity to model global dependencies. However, most existing learning-based dehazing methods lack physical interpretability, leading to weak generalization. Furthermore, they typically rely on spatial features while neglecting crucial frequency domain information, resulting in incomplete feature representation. To address these challenges, we propose ScaleViM-PDD, a novel network that enhances an SSM backbone with two key innovations: a Multi-scale EfficientViM with Physical Decoupling (ScaleViM-P) module and a Dual-Domain Fusion (DD Fusion) module. The ScaleViM-P module synergistically integrates a Physical Decoupling block within a Multi-scale EfficientViM architecture. This design enables the network to mitigate haze interference in a physically grounded manner at each representational scale while simultaneously capturing global contextual information to adaptively handle complex haze distributions. To further address detail loss, the DD Fusion module replaces conventional skip connections by incorporating a novel Frequency Domain Module (FDM) alongside channel and position attention. This allows for a more effective fusion of spatial and frequency features, significantly improving the recovery of fine-grained details, including color and texture information. Extensive experiments on nine publicly available remote sensing datasets demonstrate that ScaleViM-PDD consistently surpasses state-of-the-art baselines in both qualitative and quantitative evaluations, highlighting its strong generalization ability. Full article

In the era of smart agriculture, accurate identification of plant species is critical for effective crop management, biodiversity monitoring, and the sustainable use of medicinal resources. However, existing deep learning approaches often underperform when applied to fine-grained plant classification tasks due to the lack of domain-specific, high-quality datasets and the limited representational capacity of traditional architectures. This study addresses these challenges by introducing a novel, well-curated leaf image dataset consisting of 39 classes of medicinal and aromatic plants collected from the Aromatic and Medicinal Plant Research Station in Odakkali, Kerala, India. To overcome performance bottlenecks observed with a baseline Convolutional Neural Network (CNN) that achieved only 44.94% accuracy, we progressively enhanced model performance through a series of architectural innovations. These included the use of a pre-trained VGG16 network, data augmentation techniques, and fine-tuning of deeper convolutional layers, followed by the integration of Squeeze-and-Excitation (SE) attention blocks. Ultimately, we propose a hybrid deep learning architecture that combines VGG16 with Batch Normalization, Gated Recurrent Units (GRUs), Transformer modules, and Dilated Convolutions. This final model achieved a peak validation accuracy of 95.24%, significantly outperforming several baseline models, such as custom CNN (44.94%), VGG-19 (59.49%), VGG-16 before augmentation (71.52%), Xception (85.44%), Inception v3 (87.97%), VGG-16 after data augumentation (89.24%), VGG-16 after fine-tuning (90.51%), MobileNetV2 (93.67), and VGG16 with SE block (94.94%). These results demonstrate superior capability in capturing both local textures and global morphological features. The proposed solution not only advances the state of the art in plant classification but also contributes a valuable dataset to the research community. Its real-world applicability spans field-based plant identification, biodiversity conservation, and precision agriculture, offering a scalable tool for automated plant recognition in complex ecological and agricultural environments. Full article

Stainless steel is recognized for its excellent durability and anti-corrosion properties, which are essential qualities across various industrial applications. The machining of stainless steel, particularly under a dry environment to attain sustainability, poses several challenges. The poor heat conductivity and high ductility of stainless steel results in poor heat distribution, accelerating tool wear and problematic chip formation. To mitigate these challenges, the implementation of surface texturing has been identified as a beneficial strategy. This study investigates the impact of wave-type texturing patterns, developed on the flank surface of tungsten carbide ceramic tool inserts, on the machinability of AISI 316 stainless steel under dry cutting conditions. In this investigation, chip morphology and surface roughness were used as key indicators of machinability. Analysis of Variance (ANOVA) was conducted for chip thickness, chip thickness ratio, and surface roughness, while Taguchi mono-objective optimization was applied to chip thickness. The ANOVA results showed that linear models accounted for 71.92%, 83.13%, and 82.86% of the variability in chip thickness, chip thickness ratio, and surface roughness, respectively, indicating a strong fit to the experimental data. Microscopic analysis confirmed a substantial reduction in chip thickness, with a minimum observed value of 457.64 µm. The corresponding average surface roughness Ra value 1.645 µm represented the best finish across all experimental runs, highlighting the relationship between thinner chips and enhanced surface quality. In conclusion, wave textures on the cutting tool’s flank face have the potential to facilitate the dry machining of AISI 316 stainless steel to obtain favorable machinability. Full article

Legumes are vital sources of protein, dietary fibre and nutrients, making them crucial for global food security and sustainable agriculture. However, their widespread acceptance and consumption are often limited by undesirable sensory characteristics, such as “a beany flavour”, bitterness or variable textures. Addressing these challenges requires a comprehensive understanding of the complex molecular mechanisms governing appearance, aroma, taste, flavour, texture and palatability in legumes, aiming to enhance their sensory appeal. This review highlights the transformative power of multi-omics approaches in dissecting these intricate biological pathways and facilitating the targeted enhancement of legume sensory qualities. By integrating data from genomics, transcriptomics, proteomics and metabolomics, the genetic and biochemical networks that directly dictate sensory perception can be comprehensively unveiled. The insights gained from these integrated multi-omics studies are proving instrumental in developing strategies for sensory enhancement. They enable the identification of key biomarkers for desirable traits, facilitating more efficient marker-assisted selection (MAS) and genomic selection (GS) in breeding programs. Furthermore, a molecular understanding of sensory pathways opens avenues for precise gene editing (e.g., using CRISPR-Cas9) to modify specific genes, reduce off-flavour compounds or optimise texture. Beyond genetic improvements, multi-omics data also inform the optimisation of post-harvest handling and processing methods (e.g., germination and fermentation) to enhance desirable sensory profiles and mitigate undesirable ones. This holistic approach, spanning from the genetic blueprint to the final sensory experience, will accelerate the development of new legume cultivars and products with enhanced palatability, thereby fostering increased consumption and ultimately contributing to healthier diets and more resilient food systems worldwide. Full article

Mozambique soils are known for having an unbalanced agronomic and environmental composition that results in poor agricultural production yields. However, agriculture is the main economic activity of Mozambique, and soils must be characterised for their elemental deficiencies and/or excesses. This paper sampled nine farms from the Manica and Sussundenga districts (Manica province) in three campaigns in 2021/2022, 2022/2023, and 2023/2024 (before and after the rainy seasons). They were subjected to a physical–chemical analysis to assess their quality from the fertility and environmental contamination point of view. Attending to the physical–chemical properties analysed, and for all the soils and sampling campaigns, a low concentration below the limit of detection for B of <0.2 mg/Kg for the majority of soils and a low concentration of Al < 0.025 mg/Kg for all the soils were obtained. Also, higher concentrations for the majority of soils for the Ca between 270 and 1634 mg/Kg, for the Mg between 41 and 601 mg/Kg, for the K between 17 and 406 mg/Kg, for the Mn between 13.6 and 522 mg/Kg, for the Fe between 66.3 and 243 mg/Kg, and for the P between <20 and 132 mg/Kg were estimated. In terms of texture and for the sand, a high percentage between 6.1 and 79% was found. In terms of metal concentrations and for all the soils of the Sussundenga district and sampling campaigns, a concentration above the reference value concentration for the Cr (76–1400 mg/Kg) and a concentration below the reference value concentration for the Pb (5–19 mg/Kg), Ba (13–120 mg/Kg) and for the Zn (10–61 mg/Kg) were evaluated. A multivariate data analysis methodology was used based on cluster and discriminant analysis. The analysis of twenty-three physical–chemical variables of the soils suggested four clusters of soils characterised by deficiencies and excess elements that must be corrected to improve the yield and quality of agricultural production. Moreover, the multivariate analysis of the metal composition of soil samples from the second and third campaigns, before and after the rainy season, suggested five clusters with a pristine composition and different metal pollutant compositions and concentrations. The information obtained in this study allows for the scientific comprehension of agricultural soil quality, which is crucial for designing agronomic and environmental corrective measures to improve food quality and quantity in the Manica and Sussundenga districts and ensure environmental, social, and economic sustainability. Full article

The use of microwaves (MWs) has been proposed as an energy-efficient method for reducing checking. Along with understanding moisture distribution, it is essential to consider structural characteristics to explain how MWs reduce checking. The influence of MWs on these characteristics depends on the food matrix’s dielectric and viscoelastic properties, which vary significantly between fresh and pre-baked dough. This study investigates the effects of MW treatment applied before (MW-O) or after conventional oven baking (O-MW) on low-fat biscuits that are prone to checking. Color (CIELab), thickness, moisture content and distribution, checking rate, texture, sensory properties, energy consumption and baking time were analyzed. The findings suggest that MWs reduce checking rate by eliminating internal moisture differences, while also changing structural properties, as evidenced by increased thickness and hardness. MW-O eliminated checking (control samples showed 100%) but negatively affected color, texture (increased hardness and breaking work), and sensory quality. The O-MW checking rate (3.41%) was slightly higher than in MW-O, probably due to the resulting different structural properties (less thickness, less hardness and breaking work). O-MW biscuits were the most preferred by consumers (54.76% ranked them first), with color and texture close to the control samples. MW-O reduced total energy consumption by 16.39% and baking time by 25.00%. For producers, these improvements could compensate for the lower biscuit quality. O-MW did not affect energy consumption but reduced baking time by 14.38%. The productivity improvement, along with the reduction in checking and the satisfactory sensory quality, indicates that O-MW could be beneficial for the bakery sector. Full article

This paper proposes a novel automatic recognition method for distinguishing voids and severe loosening in road structures based on features of ground-penetrating radar (GPR) B-scan images. By analyzing differences in image texture, the intensity and clarity of top reflection interfaces, and the regularity of internal waveforms, a set of discriminative features is constructed. Based on these features, we develop the FKS-GPR dataset, a high-quality, manually annotated GPR dataset collected from real road environments, covering diverse and complex background conditions. Compared to datasets based on simulations, FKS-GPR offers higher practical relevance. An improved ACF-YOLO network is then designed for automatic detection, and the experimental results show that the proposed method achieves superior accuracy and robustness, validating its effectiveness and engineering applicability. Full article

Image super-resolution (SR) is essential for enhancing image quality in critical applications, such as medical imaging and satellite remote sensing. However, existing methods were often limited in their ability to effectively process and integrate multi-scales information from fine textures to global structures. To address these limitations, this paper proposes DSCNN, a dynamic snake convolution neural network for enhanced image super-resolution. DSCNN optimizes feature extraction and network architecture to enhance both performance and efficiency: To improve feature extraction, the core innovation is a feature extraction and enhancement module with dynamic snake convolution that dynamically adjusts the convolution kernel’s shape and position to better fit the image’s geometric structures, significantly improving feature extraction. To optimize the network’s structure, DSCNN employs an enhanced residual network framework. This framework utilizes parallel convolutional layers and a global feature fusion mechanism to further strengthen feature extraction capability and gradient flow efficiency. Additionally, the network incorporates a SwishReLU-based activation function and a multi-scale convolutional concatenation structure. This multi-scale design effectively captures both local details and global image structure, enhancing SR reconstruction. In summary, the proposed DSCNN outperforms existing methods in both objective metrics and visual perception (e.g., our method achieved optimal PSNR and SSIM results on the Set5 $\times 4$ dataset). Full article

As consumer preferences tend toward safer, chemical residue-free, and nutritionally rich fruits, preharvest bagging has gained attention as a sustainable method for improving fruit quality and protecting produce from environmental and biological stressors and pesticide residues. This study assessed the impact of preharvest bagging using paper bags with inner aluminum coating on the physicochemical traits, storability, and browning susceptibility after cutting or bruising of ‘Afrata Volou’ quince (Cydonia oblonga Mill.) fruit grown in central Greece. Fruits were either bagged or left unbagged approximately 60 days before harvest, and evaluations were conducted at harvest and after three months of cold storage, plus two days of shelf-life. Fruit bagging reduced the quince’s flesh temperature on the tree crown. Bagging had minor effects on fruit and nutritional quality, except for more yellow skin and higher titratable acidity (TA). Also, at harvest, bagging did not significantly affect fruit flesh browning after cutting or bruising. After three months of storage, unbagged and bagged quince fruit developed more yellow skin color, without significant alterations in most quality characteristics and nutritional value, but increased total tannin content (TTC). After three months of storage, the quince flesh color determined immediately after cutting or bruising was brighter and more yellowish compared to that at harvest, due to continuation of fruit ripening, but it darkened faster with time after cutting or skin removal. Therefore, fruit bagging appears to be a sustainable practice for improving the aesthetic and some chemical quality characteristics of quince, particularly after storage, without negative impacts on other characteristics such as texture and phenolic content. Full article

Low glycemic index (GI) cookie bars were prepared with soft wheat flour substituted with 10–50% soybean flour and 10–50% resistant starch. The effects of increased levels of soybean flour and resistant starch on the quality of low glycemic index cookie bars were investigated (i.e., moisture, cookie spread, texture (breaking force), surface color, and in vitro starch digestibility). It was found that increasing soybean flour substitution increased the breaking force, moisture, protein content, and yellowish color of the low GI cookie bars but decreased the cookie bar spread and the lightness of the cookie bars (p < 0.05). The addition of soybean flour and resistant starch by up to 50% did not significantly change the in vitro starch digestibility of the cookie bars. The overall acceptability of the cookie bars was lower when the soybean flour blend went beyond 10%. When soft wheat flour in the cookie bar formulation was replaced at the following levels (10%, 30%, and 50%) by resistant starch, the cookie spread and lightness of the cookie bars increased but the breaking force was decreased along with the yellowish color (p < 0.05). When resistant starch was combined with soft wheat flour at levels of up to 50%, this significantly increased the content of total dietary fiber and spread ratio of cookie bars. Sensorial analysis showed that resistant starch presence had an acceptable impact on overall acceptability of the low GI cookie bars. Resistant starch represents a viable dietary fiber source when substituted for 50% of soft wheat flour in formulations. While this substitution may result in increased spread ratio and decreased crispness in cookie bars, the addition of 10% soybean flour can mitigate these textural changes. Full article

Cultivated land quality is a key factor in ensuring sustainable agricultural development. Exploring differences in cultivated land quality under distinct cropping systems is essential for developing targeted improvement strategies. This study takes place in Shenyang City—located in the typical black soil region of Northeast China—as a case area to construct a cultivated land quality evaluation system comprising 13 indicators, including organic matter, effective soil layer thickness, and texture configuration. A minimum data set (MDS) was separately extracted for paddy and upland fields using principal component analysis (PCA) to conduct a comprehensive evaluation of cultivated land quality. Additionally, an obstacle degree model was employed to identify the limiting factors and quantify their impact. The results indicated the following. (1) Both MDSs consisted of seven indicators, among which five were common: ≥10 °C accumulated temperature, available phosphorus, arable layer thickness, irrigation capacity, and organic matter. Parent material and effective soil layer thickness were unique to paddy fields, while landform type and soil texture were unique to upland fields. (2) The cultivated land quality index (CQI) values at the sampling point level showed no significant difference between paddy (0.603) and upland (0.608) fields. However, their spatial distributions diverged significantly; paddy fields were dominated by high-grade land (Grades I and II) clustered in southern areas, whereas uplands were primarily of medium quality (Grades III and IV), with broader spatial coverage. (3) Major constraint factors for paddy fields were effective soil layer thickness (21.07%) and arable layer thickness (22.29%). For upland fields, the dominant constraints were arable layer thickness (27.57%), organic matter (25.40%), and ≥10 °C accumulated temperature (23.28%). Available phosphorus and ≥10 °C accumulated temperature were identified as shared constraint factors affecting quality classification in both systems. In summary, cultivated land quality under different cropping systems is influenced by distinct limiting factors. The construction of cropping-system-specific MDSs effectively improves the efficiency and accuracy of cultivated land quality assessment, offering theoretical and methodological support for land resource management in the black soil regions of China. Full article

Cereal-based plant beverages have gained attention as functional ingredients in bakery formulations, offering both nutritional and technological benefits. Replacing water with these beverages may improve the nutritional value of bread by increasing its fiber and unsaturated fatty acid content, while also introducing functional components that affect dough rheology and bread texture. This study examined the effects of substituting water with oat (BO), millet (BM), and spelt (BS) beverages in wheat bread formulations at 25%, 50%, 75%, and 100% levels. Thirteen bread variants were prepared: one control and four substitution levels for each of the three cereal-based beverages, using the straight dough method, with hydration adjusted according to farinograph results. Farinograph tests showed increased water absorption (up to 64.5% in BO100 vs. 56.9% in control) and improved dough stability (10.6 min in BS100). Specific bread volume increased, with BS75 reaching 3.52 cm³/g compared to 3.09 cm³/g in control. Moisture content remained stable during storage, and crumb hardness after 72 h was lowest in BO100 (9.5 N) and BS75 (11.5 N), indicating delayed staling. All bread variants received favorable sensory ratings, with average scores above 3.75 on a 5-point scale. The highest bread yield (149.8%) and lowest baking loss (10.9%) were noted for BS100. Although BO breads had slightly higher fat and energy content, their nutritional profile remained favorable due to unsaturated fatty acids. Overall, oat and spelt beverages demonstrated the greatest potential as functional water substitutes, improving dough handling, shelf-life, and sensory quality while maintaining consumer appeal. Full article

In this study, whole grain sorghum flour was used to partially substitute the gluten-free flour blend in cookie formulation at 20% (C20) and 40% (C40) replacement levels. The goal was to explore its potential to improve the nutritional value and sensory appeal of cookies relative to conventional and commercially available gluten-free alternatives. Nutritional analysis revealed that cookies with added sorghum flour showed increased levels of protein, ash, and polyphenolic compounds, while maintaining favorable macronutrient profiles. Notably, several bioactive compounds, such as gallic acid, caffeic acid, and apigenin, were detected exclusively in sorghum-containing samples, suggesting enhanced functional properties. Despite these compositional changes, textural measurements showed no significant differences in hardness or fracturability compared with the control. Sensory profiling using the Rate-All-That-Apply (RATA) method demonstrated that both samples (C20 and C40) achieved balanced results in terms of aroma as well as texture and were generally well accepted by the panel. The results indicate that moderate inclusion of sorghum flour (20% and 40%) can improve the sensory and nutritional profiles of gluten-free cookies without compromising product acceptability. Sorghum thus offers a promising pathway for the development of high-quality, health-oriented, gluten-free bakery products. Full article

This study investigated the effects of the addition of Rosa rugosa fruit pomace and drying methods on the properties of pasta, such as culinary properties, color, texture, microstructure, phenolics, antioxidant capacity, and sensory properties. In laboratory conditions, the pasta was produced using low-extraction wheat flour with the addition of pomace at 0, 2, 4, 6, and 8% (g/100 g flour) and dried using either convective or microwave–vacuum drying. The incorporation of pomace into the pasta caused a notable reduction in lightness and increased redness and yellowness, as well as a decrease in pasta hardness and sensory acceptability. The RFP addition also increased the polyphenol content and antioxidant potential. The microwave–vacuum drying resulted in pasta with shorter cooking times, lower cooking loss, and higher total phenolic content and antioxidant activity compared to convective drying. Although the drying method did not markedly affect sensory attributes, ultrastructural analysis revealed that samples subjected to convective drying had a more compact structure, while microwave–vacuum dried pasta exhibited larger pores and smaller starch granules. Total porosity was higher in microwave–vacuum dried pasta. Taking into account both the level of pomace enrichment and the drying technique, the most optimal outcomes were achieved when microwave–vacuum drying was applied and the pomace addition did not exceed 4%. Full article