Search Results (20)

Residues in the grain tank of seed-production wheat harvesters often cause varietal admixture, challenging seed purity maintenance above 99%. To address this, an intelligent cleaning system was developed for automatic residue recognition and removal. The system utilizes an RGB-D camera and an embedded AI unit paired with an improved lightweight object detection model. This model, enhanced for feature extraction and compressed via LAMP, was successfully deployed on a Jetson Nano, achieving 92.5% detection accuracy and 13.37 FPS for real-time 3D localization of impurities. A D–H kinematic model was established for the 4-DOF cleaning manipulator. By integrating the PSO and FWA models, the motion trajectory was optimized for time-optimality, reducing movement time from 9 s to 5.96 s. Furthermore, a gas–solid coupled simulation verified the separation capability of the cyclone-type dust extraction unit, which prevents motor damage and centralizes residue collection. Field tests confirmed the system’s comprehensive functionality, achieving an average cleaning rate of 92.6%. The proposed system successfully enables autonomous residue cleanup, effectively minimizing the risk of variety mixing and significantly improving the harvest purity and operational reliability of seed-production wheat. It presents a novel technological path for efficient seed production under the paradigm of smart agriculture. Full article

Accurate identification of tea plant varieties during the harvest period is a critical prerequisite for developing intelligent multi-variety tea harvesting systems. Different tea varieties exhibit distinct chemical compositions and require specialized processing methods, making varietal purity a key factor in ensuring product quality. However, achieving reliable classification under real-world field conditions is challenging due to variable illumination, complex backgrounds, and subtle phenotypic differences among varieties. To address these challenges, this study constructed a diverse canopy image dataset and systematically evaluated 14 convolutional neural network models through transfer learning. The best-performing model was chosen as a baseline, and a comprehensive optimization of the training strategy was conducted. Experimental analysis demonstrated that the combination of Adamax optimizer, input size of 608 × 608, training and validation sets split ratio of 80:20, learning rate of 0.0001, batch size of 8, and 20 epochs produced the most stable and accurate results. The final optimized model achieved an accuracy of 99.32%, representing a 2.20% improvement over the baseline. This study demonstrates the feasibility of highly accurate tea variety identification from canopy imagery but also provides a transferable deep learning framework and optimized training pipeline for intelligent tea harvesting applications. Full article

Global food security is increasingly challenged by climate change and the availability of arable land. This situation calls for improved crop monitoring and management strategies. Rice is a staple food for nearly half of the world’s population and a significant source of calories. Accurately identifying rice varieties is crucial for maintaining varietal purity, planning agricultural activities, and enhancing genetic improvement strategies. This study evaluates the effectiveness of machine learning algorithms to identify the most effective approach to predicting rice varieties, using multitemporal Sentinel-2 images in the Marismas del Guadalquivir of Sevilla, Spain. Spectral reflectance data were collected from ten Sentinel-2 bands, which include visible, red-edge, near-infrared, and shortwave infrared regions, at two key phenological stages: tillering and reproduction. The models were trained on pixel-level data from the growing seasons of 2021 and 2024, and they were evaluated using a test set from 2022. Four classifiers were compared: random forest, XGBoost, K-nearest neighbors, and logistic regression. Performance was assessed based on accuracy, precision, recall, specificity and F1 score. Non-linear models outperformed linear ones. The highest performance was achieved with the Random Forest classifier during the reproduction phase, reaching an exceptional accuracy of 0.94 using all bands or only the most informative subset (red edge, NIR, and SWIR). This classifier also maintained excellent accuracy (0.93 and 0.92) during the initial tillering phase. This fact demonstrates that it is possible to perform reliable varietal mapping in the early stages of the growing season. Full article

The precise identification, classification, sorting, and rapid and accurate quality assessment of soybean seeds are extremely important in terms of the continuity of agricultural production, varietal purity, seed processing, protein extraction, and food safety. Currently, commonly used methods for the identification and quality assessment of soybean seeds include morphological analysis, chemical analysis, protein electrophoresis, liquid chromatography, spectral analysis, and image analysis. The use of image analysis and artificial intelligence is the aim of the presented research, in which a method for the automatic classification of soybean varieties, the assessment of the degree of damage, and the identification of geometric features of soybean seeds based on numerical models obtained using a 3D scanner has been proposed. Unlike traditional two-dimensional images, which only represent height and width, 3D imaging adds a third dimension, allowing for a more realistic representation of the shape of the seeds. The research was conducted on soybean seeds with a moisture content of 13%, and the seeds were stored in a room with a temperature of 20–23 °C and air humidity of 60%. Individual soybean seeds were scanned to create 3D models, allowing for the measurement of their geometric parameters, assessment of texture, evaluation of damage, and identification of characteristic varietal features. The developed 3D-CNN network model comprised an architecture consisting of an input layer, three hidden layers, and one output layer with a single neuron. The aim of the conducted research is to design a new, three-dimensional 3D-CNN architecture, the main task of which is the classification of soybean seeds. For the purposes of network analysis and testing, 22 input criteria were defined, with a hierarchy of their importance. The training, testing, and validation database of the SB3D-NET network consisted of 3D models obtained as a result of scanning individual soybean seeds, 100 for each variety. The accuracy of the training process of the proposed SB3D-NET model for the qualitative classification of 3D models of soybean seeds, based on the adopted criteria, was 95.54%, and the accuracy of its validation was 90.74%. The relative loss value during the training process of the SB3D-NET model was 18.53%, and during its validation process, it was 37.76%. The proposed SB3D-NET neural network model for all twenty-two criteria achieves values of global error (GE) of prediction and classification of seeds at the level of 0.0992. Full article

The gene flow rate in rice (Oryza sativa L.) is a critical factor for establishing safe isolation distances between genetically modified (GM) and non-GM varieties and for ensuring varietal purity in rice breeding programs. This study refines existing gene flow models by disentangling two key components of rice pollen dynamics: quantitative pollen competition and genetic competitiveness. We define B as the proportion of GM pollen within mixed pollen, representing quantitative pollen competitiveness. The outcrossing parameter C_b reflects the likelihood of successful fertilization and seed development by foreign pollen, while the hybrid compatibility parameter C_p captures the relative fertilization success of GM versus non-GM pollen within the same pollen pool. Together, C_b and C_p characterize the genetic competitiveness of rice pollen. Our findings reveal a nonlinear relationship between B and the observed GM pollen rate G, which may exhibit either upward or downward curvature. A nonlinear model provides a significantly better fit to this relationship than a linear model, improving R² by 4.1–21.4% and reducing RMSE by 9.9–47.8%. The parameters C_b and C_p play central roles in determining gene flow; higher values correspond to stronger GM pollen competitiveness, resulting in higher gene flow rates and greater dispersal distances. Specifically, C_b sets the range of the B–G curve, while C_p determines its curvature. Full article

Wine DNA fingerprinting (WDF), retrieved from the amplification of a wider panel of Simple Sequence Repeat (SSR) marker mappings in the Vitis vinifera L. genome, was used to assess the monovarietal nature of Brunello di Montalcino wine. The reliability of the varietal assessment was carried out by estimating the PI values associated with resolutive unrooted dendrograms depicting the correct varietal nature of different wines. As few as five SSR DNA markers associated with a PI value of one over a million or less, PI ≤ 10⁻⁶, can identify the purity of Sangiovese against Merlot, Pinot Noir, Cabernet Sauvignon, Primitivo (Zinfandel), and genetic variants of the Sangiovese as plant references. WDF was used on other monovarietal wines obtained from Cabernet Sauvignon, Merlot, Chardonnay, and Pinot Noir to test the feasibility of the method. In blended wines, the test was able to trace the main varietal component in a three-variety blend, keeping the varietal fingerprint detectable when the main variety was at least 75% (v/v). The data confirm how local genetic variants of Sangiovese can be tracked in commercial wines, becoming, at wine makers’ demand, part of an evidence synthesis of geographical origin. Full article

The continued emergence of weedy rice (Oryza sativa L.) in Taiwan poses serious challenges to seed purity and commercial rice cultivation, particularly under transplanting systems. These off-type individuals, often marked by a red pericarp, reduce varietal integrity and complicate seed propagation. This study evaluated the morphological variation among 117 Taiwan weedy rice (TWR) accessions and 55 control cultivars, which include 24 temperate japonica cultivars (TEJ), 24 indica cultivars, and seven U.S. weedy rice (UWR) types. Principal component analysis (PCA) showed that TWR shares vegetative traits with modern cultivars but exhibits grain morphology resembling indica landraces—indicating weak artificial selection pressure on grain traits during nursery propagation. TWR was also found to possess a suite of adaptive weedy traits, including semi-dwarfism, delayed heading, high shattering, and superior seed storability, facilitating its persistence in field conditions. These findings provide critical insights for integrated weed management and cultivar purity strategies, emphasizing the importance of certified seed use, stringent field hygiene, and disruption of weedy rice reproductive cycles. Full article

Sorting corn seeds before sowing is crucial to ensure the varietal purity of the seeds and the yield of the crop. However, most of the existing methods for sorting corn seeds cannot detect both varieties and defects simultaneously. Detecting seeds in motion is more difficult than at rest, and many models pursue high accuracy at the expense of model inference time. To address these issues, this study proposed a real-time detection model, YOLO-SBWL, that simultaneously identifies corn seed varieties and surface defects by using images taken at different conveyor speeds. False detection of damaged seeds was addressed by inserting a simple and parameter-free attention mechanism (SimAM) into the original “you only look once” (YOLO)v7 network. At the neck of the network, the path-aggregation feature pyramid network was replaced with the weighted bi-directional feature pyramid network (BiFPN) to increase the accuracy of classifying undamaged corn seeds. The Wise-IoU loss function supplanted the CIoU loss function to mitigate the adverse impacts caused by low-quality samples. Finally, the improved model was pruned using layer-adaptive magnitude-based pruning (LAMP) to effectively compress the model. The YOLO-SBWL model demonstrated a mean average precision of 97.21%, which was 2.59% higher than the original network. The GFLOPs were reduced by 67.16%, and the model size decreased by 67.21%. The average accuracy of the model for corn seeds during the conveyor belt movement remained above 96.17%, and the inference times were within 11 ms. This study provided technical support for the swift and precise identification of corn seeds during transport. Full article

Varietal purity is a critical quality indicator for seeds, yet various production processes can lead to the mixing of seeds from different varieties. Consequently, seed variety classification is an essential step in seed production. Existing classification algorithms often suffer from limitations such as reliance on single information sources, constrained feature extraction capabilities, time consumption, low accuracy, and the potential to cause irreversible damage to seeds. To address these challenges, this paper proposes a fast and non-destructive classification method for corn seeds, named DualTransAttNet, based on multi-source image information and hybrid feature extraction. High-resolution hyperspectral images of various corn varieties were collected, and a sliding sampling approach was employed to capture feature information across all spectral bands, resulting in the construction of a hyperspectral dataset for corn seed classification. Hyperspectral and RGB image data were then integrated to complement one another’s information and mitigate the insufficient feature diversity caused by single-source data. The proposed method leverages the strengths of convolutional neural networks (CNNs) and transformers to extract both local and global features, effectively capturing spectral and image characteristics. The experimental results demonstrate that the DualTransAttNet model can achieve a compact size of only 1.758 MB and an inference time of 0.019 ms. Compared to typical machine learning and deep learning models, the proposed model exhibits superior performance with an overall accuracy, F1-score, and Kappa coefficient of 90.01%, 88.9%, and 88.4%, respectively. The model’s rapid inference capability and low parameter count make it an excellent technical solution for agricultural automation and intelligent systems, thereby enhancing the efficiency and profitability of agricultural production. Full article

A chromosome segment substituted line (CSSL) represents an ideal resource for studying quantitative traits like thermotolerance. To develop wheat inter-varietal CSSLs with E6015-3S (a heat-sensitive genotype) being the recipient parent, genome-wide unique DNA markers are urgently needed for marker-assisted selection. In this study, 11,016 primer pairs targeting 5036 indel sites were successfully designed for E6015-3S, with an average density of 0.36 indels per Mbp. These primer pairs are believed to be unique and polymorphic in the wheat genome; as gathered from the evidence, (i) 76.18 to 99.34% of the 11,016 primer pairs yielded a single hit during sequence alignment with 18 sequenced genomes, (ii) 83.59 to 90.98% of 1042 synthesized primer pairs amplified a single band in 16 wheat accessions, and (iii) 59.69 to 99.81% of the tested 1042 primer pairs were polymorphic between E6015-3S and 15 individual wheat accessions. These primer pairs are also anticipated with excellent resolvability on agarose or polyacrylamide gels, since most of them have indel sizes from 15 to 46 bp, amplicon sizes from 141 to 250 bp, and polymorphism ratios from 6.0 to 25.0%. Collectively, these primer pairs are ideal DNA markers for inter-varietal CSSL development and more broad applications, like germplasm classification, seed purity testing, genetic linkage mapping, and marker-assisted breeding in wheat, owing to their uniqueness, polymorphism, and easy-to-use characteristics. Full article

Breeding is an effective method for the varietal development of pineapple. However, due to open pollination, it is necessary to conduct authentic identification of the hybrid offspring. In this study, we identified the authenticity of offspring and analyzed the genetic diversity within the offspring F₁ hybrids resulting from crosses between ‘Josapine’ and ‘MD2’ by single nucleotide polymorphism (SNP) markers. From the resequencing data, 26 homozygous loci that differentiate between the parents have been identified. Then, genotyping was performed on both the parents and 36 offspring to select SNP markers that are suitable for authentic identification. The genotyping results revealed that 2 sets of SNP primers, namely SNP4010 and SNP22550, successfully identified 395 authentic hybrids out of 451 hybrid offspring. We randomly selected two true hybrids and four pseudohybrids for sequencing validation, and the results have shown that two true hybrids had double peaks with A/G, while pseudohybrids had single peaks with base A or G. Further study showed that the identification based on SNP molecular markers remained consistent with the morphological identification results in the field, with a true hybridization rate of 87.58%. K-means clustering and UPGMA tree analysis revealed that the hybrid offspring could be categorized into two groups. Among them, 68.5% of offspring aggregated with MD2, while 31.95% were grouped with Josapine. The successful application of SNP marker to identify pineapple F₁ hybrid populations provides a theoretical foundation and practical reference for the future development of rapid SNP marker-based methods for pineapple hybrid authenticity and purity testing. Full article

This study emphasizes the importance of seed quality in the context of yield formation. Based on the research data, this paper emphasizes the role of proper diagnosis of seed-borne pathogens in ensuring high and stable grain yields. Particular attention is paid to the study of the effect of the treatment of mother plants with fullerenol-based nanopreparations on the qualitative characteristics of spring barley seeds. The results showed that such treatment contributes to the increase in varietal purity, weight of 1000 grains as well as to the increase of nutrient and moisture reserves in seeds. Phytopathological analysis confirmed the presence of various diseases such as Alternaria, helminthosporiosis, fusarium, mold and mildew on the seeds. However, some samples showed a high resistance to pathogens, presumably due to the use of carbon nanopreparations. These results open new perspectives for the development of strategies to improve barley yield and disease resistance through seed optimization. Full article

Grape pomace is one of the main by-products in the wine industry and contains some high-added-value compounds, such as polysaccharides. Considering the wide application possibilities of polysaccharides in wine and in the food industry, the revalorization of grape pomace to extract polysaccharides presents itself as an opportunity for by-product management. Therefore, the aim of this study was to characterize polysaccharide extracts obtained from pomace by-products of different white grape varieties. The type and content of polysaccharides, proteins and phenols were analyzed. Statistically significant differences were found between the varietal extracts in the types and concentrations of polysaccharides. The extracts obtained from the Verdejo and Puesta en Cruz varieties showed the highest polysaccharide purity and contents, but the type of polysaccharides was different in each case. The Verdejo provided extracts richer in non-pectic polysaccharides, while the Puesta en Cruz provided extracts richer in pectic polysaccharides. The protein and polyphenol contents were low in all extracts, below 2.5% and 3.7%, respectively. These results open up a new possibility for the revalorization of grape pomace by-products to obtain polysaccharide-rich extracts, although it would be interesting to improve both the yield and the purity of the extracts obtained by studying other extraction techniques or processes. Full article

Faba beans (Vicia faba L.) constitute a partially outcrossing species requiring an isolation distance to maintain genetic purity when more than one variety is grown in field conditions. This information is crucial for seed growers and faba bean breeders. A study was conducted at the University of Sydney’s Plant Breeding Institute, Narrabri, over two years to examine the extent of natural outcrossing using a creamy white flower characteristic as a morphological marker, which is controlled by a single recessive gene. The white-flowered genotype (IX225c) was grown in paired rows of 150 m length in four directions from a central 480 m² plot of the normal flowered genotype PBA Warda. A beehive was placed in the central plot at the flowering time and natural pollination was allowed. At maturity, seed samples were taken from the white-flowered genotype at designated intervals along each axis and 100 seeds from each sample were grown in the glasshouse/birdcage to the 4–5 leaf stage and the proportion of plants displaying a stipule spot pigmentation (normal flower color and spotted stipule are linked) was used to determine the percentage of outcrossing. Maximum outcrossing of 2.28% occurred where both genotypes were grown side by side (0 m) and the degree of outcrossing decreased as the distance along each axis from the central plot increased. At a 6 m distance, the outcrossing was less than 1%; however, on occasion, it increased to 1% beyond a distance of 100 m, indicating the volatile and unpredictable nature of bee flights. Distance had a major effect on outcrossing but the direction and its interaction had no effect. The results suggest that to limit outcrossing to below 0.5%, a distance of more than 150 m between plots of different faba beans cultivars would be required. It also indicated that Australian faba bean genotypes are mostly self-fertile and a relatively narrow isolation distance will ensure self-fertilization in seed production and breeding programs. Full article

The non-invasive analysis of seeds is of great interest to experimental biologists and breeders. To reach a high varietal identity and purity of seed material, it is often necessary to access features of individual seeds via the screening of mutant populations. While near-infrared spectroscopy (NIRS) and time-domain nuclear-magnetic-resonance spectroscopy (TD-NMR) are well-known in seed research and industry for bulk seed measurements, their application for individual seeds is challenging. Here we demonstrate how to overcome this limitation using a practical approach to cereal grains using oat (Avena sp.) as a model. For this, we generated a representative collection of oat seeds from the ex situ German federal gene bank, which includes wide variation in grain size, shape, and coloration. Next, we established a short experimental pipeline to exemplify how to improve the procedure for individual seed measurements. In its current state, the method is ready to use for the high-accuracy estimation of nitrogen (protein) content (R² = 0.877), water content (R² = 0.715), and seed weight (R² = 0.897) of individual oat grains. This work introduces the combination of NIRS and TD-NMR as an efficient, precise, and, most importantly, non-destructive analytic platform for a high throughput analysis of individual intact seeds. Full article