Search Results (87)

Strawberry harvesting perception in greenhouse environments requires visual models that remain reliable under occlusion while staying compact enough for edge-side inference. To address this requirement, this study develops StrawPose-Lite, a lightweight pose network for strawberry picking point prediction based on YOLOv11n-pose. The network combines ADown and C3Ghost to reduce redundant computation while preserving informative structure, and it adopts a six-keypoint pose definition derived from strawberry phenotypic characteristics. In this representation, the pedicel–fruit junction is used as the final visual picking point, whereas the remaining peak, curvature, and bottom keypoints provide geometric support when the visible contour is incomplete. The keypoint branch is further enhanced by P2-guided multi-scale fusion and SimAM-based refinement to improve sensitivity to fine pedicel-related cues under strict lightweight constraints. On the public validation split, StrawPose-Lite contains 0.73 M parameters and requires 3.0 GFLOPs while achieving a pose mAP@0.5:0.95 of 79.2%. In the independent field deployment set, the TensorRT INT8 version achieved a pure network inference throughput of 277 FPS on a Jetson Orin NX 16G Super platform, with a measured total software latency of 5.01 ms under the embedded pipeline. These results indicate that StrawPose-Lite provides an effective balance between pose accuracy, model compactness, and edge-side inference speed for strawberry picking point perception on edge devices. Full article

To address challenges such as severe occlusion caused by the dense growth of blueberry fruits in natural environments, complex backgrounds, and the limited computational resources of agricultural edge devices, this study proposes BR-DETR-Prune, a lightweight object detection model oriented towards edge computing environments. Based on the RT-DETR architecture, the model introduces a PConv-based FasterNet as the backbone network, which effectively reduces memory access latency and floating-point operation costs. Furthermore, it utilizes a “Gather-and-Distribute” (GD) mechanism to reconstruct the feature fusion neck. Through the unified aggregation and multi-branch distribution of global information, it significantly enhances the model’s feature extraction capability for dense and overlapping targets. An AIFI-RepBN encoder is designed, integrating re-parameterization technology into the attention module to further reduce computational redundancy. For lightweight processing, a random channel pruning strategy based on the “Lottery Ticket Hypothesis” is adopted to perform structural compression and fine-tuning on the model, achieving a significant reduction in the number of parameters while inversely improving accuracy. The experimental results demonstrate that BR-DETR-Prune achieves an mAP@0.5 of 97.1% on a self-built blueberry dataset, with only 15.52 M parameters and a computational load reduced to 34.0 GFLOPs. Its comprehensive performance is superior to mainstream models such as YOLOv8, YOLO11, and the original RT-DETR. Particularly, deployment testing on the NVIDIA Jetson Orin Nano Super embedded edge computing platform reveals that the model achieves a real-time inference speed of 20.5 FPS under FP16 precision, exhibiting smooth detection frames and strong robustness against occlusion. This study provides an effective optimization solution for the deployment of high-precision Transformer architectures on low-computational-power devices, offering an efficient and reliable visual perception approach for automated blueberry harvesting and yield estimation. Full article

Within the domain of intelligent picking robotics, fruit recognition and positioning are essential. Challenging conditions such as varying light, occlusion, and limited edge-computing power compromise fruit maturity detection. To tackle these issues, this paper proposes a lightweight algorithm YOLO-ELS based on YOLOv8n. Specifically, we reconstruct the backbone by replacing the bottlenecks in the C2f structure with Edge-Information-Enhanced Modules (EIEM) to prioritize morphological cues and filter background redundancy. Furthermore, a Large Separable Kernel Attention (LSKA) mechanism is integrated into the SPPF layer to expand the effective receptive field for multi-scale targets. To mitigate occlusion-induced errors, a Spatially Enhanced Attention Module (SEAM) is incorporated into the decoupled detection head to enhance feature responses in obscured regions. Finally, the Inner-GIoU loss is adopted to refine bounding box regression and accelerate convergence. Experimental results demonstrate that compared to the YOLOv8n baseline, the proposed YOLO-ELS achieves a 14.8% reduction in GFLOPs and a 2.3% decrease in parameters, while attaining a precision, recall, and mAP@50% of 92.7%, 83.9%, and 92.0%, respectively. When compared with mainstream models such as DETR, Faster-RCNN, SSD, TOOD, YOLOv5s, and YOLO11n, the mAP@50% is improved by 7.0%, 4.7%, 11.4%, 8.6%, 3.1%, and 3.2%. Deployment tests on the NVIDIA Jetson Orin Nano Super edge platform yield an inference latency of 25.2 ms and a detection speed of 28.2 FPS, successfully meeting the real-time operational requirements of automated harvesting systems. These findings confirm that YOLO-ELS effectively balances high detection accuracy with lightweight architecture, providing a robust technical foundation for intelligent fruit picking in resource-constrained greenhouse environments. Full article

The introduction of Super-High-Density (SHD) olive orchards represents a crucial innovation in modern olive growing, enhancing sustainability. However, the long-term success of these planting systems depends strongly on cultivar selection, combining suitable vegetative and reproductive traits. This three-year field study investigated key floral biology parameters—flowering phenograms, gynosterility, and self-compatibility—of ten olive cultivars grown under irrigated conditions in southern Italy: ‘Arbequina’, ‘Arbosana’, ‘Cima di Bitonto’, ‘Coratina’, ‘Don Carlo’, ‘Frantoio’, ‘Favolosa’ (=‘Fs-17’), ‘I-77’, ‘Koroneiki’, and ‘Urano’ (=‘Tosca’). Flowering phenograms varied significantly across years and cultivars, showing temporal shifts related to chilling accumulation and yield of the previous year. Early blooming cultivars (‘Arbequina’, ‘Arbosana’, and ‘Coratina’) exhibited partial flowering overlap with mid-season ones, enhancing cross-pollination opportunities. Quantitative analysis of flowering overlap revealed that most cultivar combinations exceeded the 70% threshold required for effective pollination, although specific genotypes (‘Coratina’, ‘Fs-17’, and especially ‘I-77’) showed critical mismatches, while ‘Frantoio’ and ‘Arbequina’ emerged as the most reliable pollinizers. Gynosterility exhibited statistical differences among cultivars and canopy positions: ‘I-77’ showed the highest values (71.4%), while ‘Coratina’ and ‘Cima di Bitonto’ showed the lowest ones (7.3 and 8.4%, respectively). The median portions of the canopies generally displayed a greater number of sterile flowers (29.4%), reflecting the combined effect of genetic and environmental factors such as light exposure. In the inflorescence, the majority of gynosterile flowers were concentrated in the lower part, for all canopy portions (modal value). Self-compatibility tests were performed considering a fruit set of 1% as a threshold to discriminate. For open pollination, the fruit set was highly variable among cultivars, ranging from 0.5% in ‘I-77’ to 4.7% in ‘Arbosana’. Apart from ‘I77’, all varieties achieved a fruit set greater than 1%. Instead, for the self-pollination, only ‘Arbequina’, ‘Koroneiki’, ‘Frantoio’, and ‘Cima di Bitonto’ could be identified as pseudo-self-compatible, whereas ‘Coratina’, ‘Fs-17’, and the others were clearly self-incompatible and therefore unsuitable for monovarietal orchards in areas with limited availability of pollen. By integrating self-compatibility and gynosterility data, the cultivars were ranked according to reproductive aptitude, identifying ‘Cima di Bitonto’ and ‘Frantoio’ as the most fertile genotypes, whereas ‘Don Carlo’ and particularly ‘I-77’ showed severe genetic sterility constraints. The findings underline the critical role of floral biology in defining reproductive efficiency and varietal adaptability in SHD systems. This research provides valuable insights for optimizing cultivar selection, orchard design, and management practices, contributing to the development of sustainable, climate-resilient olive production models for Mediterranean environments. Full article

The pomegranate cultivar Barski slatki, the most widely cultivated on the Eastern Adriatic coast, was evaluated over one growing season across four growing areas to assess its pomological and chemical properties and antioxidant activity. Results showed that location significantly influenced fruit weight, volume, number of arils per fruit, and both total and individual aril weight, with the Kaštela (CRO) site producing the largest fruits and highest aril yields. Climatic factors, such as precipitation during bud differentiation, flowering, and early fruit development, were found to impact fruit set, aril number, and fruit size. Aril and juice yields, however, remained relatively stable across sites. Notable differences were observed in total soluble solids, titratable acidity, pH, total phenolic content, and anthocyanin profiles. Location with higher rainfall occurring during fruit growth favored enhanced phenolic accumulation. Although total anthocyanin content remained consistent among locations, significant variation occurred in aril coloration and composition of individual anthocyanins. In conclusion, microclimatic factors, particularly rainfall distribution, temperature, and altitude, play a decisive role in shaping the physical, chemical, and visual attributes of ‘Barski slatki’. Despite being cultivated under similar Mediterranean conditions, the observed differences across sites highlight the strong adaptability of this cultivar to diverse agroecological environments, while maintaining stable quality. Full article

Advances in Earth observation technology using multispectral imagery from satellite Earth observation systems and sensors mounted on unmanned aerial vehicles (UAVs) are enabling more accurate crop monitoring. These images, once processed, facilitate the analysis of crop health by enabling the study of crop vigour, the calculation of biomass indices, and the continuous temporal monitoring using vegetation indices (VIs). These indicators allow for the identification of diseases, pests, or water stress, among others. This study compares images acquired with the Altum PT sensor (UAV) and Super Dove (satellite) to evaluate their ability to detect specific problems in super-intensive olive groves at two critical times: January, during pruning, and April, at the beginning of fruit development. Four different VIs were used, and multispectral maps were generated for each: the Normalized Difference Vegetation Index (NDVI), the Green Normalized Difference Vegetation Index (GNDVI), the Normalized Difference Red Edge Index (NDRE) and the Leaf Chlorophyll Index (LCI). Data for each plant (n = 11,104) were obtained for analysis across all dates and sensors. A combined methodology (Spearman’s correlation coefficient, Student’s t-test and decision trees) was used to validate the behaviour of the variables and propose predictive models. The results showed significant differences between the sensors, with a common trend in spatial patterns and a correlation range between 0.45 and 0.68. Integrating both technologies enables multiscale assessment, optimizing agronomic management and supporting more sustainable precision agriculture. Full article

Over the past two decades, hyperspectral imaging (HSI) systems have shown significant potential in agriculture, from disease detection to the assessment of plant and fruit nutritional status. However, most applications remain confined to laboratory analyses under controlled conditions, with only a limited fraction implemented in field environments. In this scenario, spectral reconstruction techniques may serve as a bridge between the high accuracy of HSI and the challenges of on-field or even real-time applications. This review outlines the current state of the art of on-field HSI in the agrifood sector, highlighting existing limitations and potential advantages. It then introduces the problem of spectral reconstruction and reviews current techniques used to address it. Laboratory and on-field studies will be taken into account. The final section offers our perspective on the limitations of HSI and the promising potential of spectral super-resolution to overcome current barriers and enable broader adoption of hyperspectral technology in precision agriculture. Full article

Pruning is labor-intensive and increases production costs, while mechanical pruning offers a promising alternative. However, research on its effectiveness remains limited. To address this gap, we evaluated the effects of mechanical pruning over two consecutive years (2023 and 2024) on tree growth, yield, and fruit quality of ‘Arisoo’ apple trees. The treatment included hand (manual) pruning (HP), mechanical pruning (MP), and combined mechanical and hand pruning (MP + HP) applied during winter pruning in a super-spindle-slender-shaped apple orchard. MP significantly reduced pruning time; however, the amount of plant biomass removed was lower in the MP treatment than in the HP and MP + HP treatments. Canopy volume was higher in the HP treatment than in MP and MP + HP treatments; however, the pruning treatments did not affect trunk cross-sectional area or tree yield. Leaf chlorophyll and nitrogen contents were slightly lower in the MP treatment than in the HP treatment in 2023 but were not affected in 2024. The MP treatment also noticeably reduced light penetration within the canopy and produced smaller fruits with lower soluble solids content and poorer coloration at harvest compared to the HP and MP + HP treatments. In contrast, the HP and MP + HP treatments showed similar effects on light penetration, yield, fruit size, and fruit quality; however, the MP + HP treatment significantly reduced the pruning time compared with the HP treatment. Overall, this study found that MP reduced light penetration and produced smaller and poorly colored fruits, whereas a follow-up combination of HP after MP improved pruning efficiency, light penetration, fruit size, and fruit quality. Full article

The adaptability of several Italian olive cultivars to high-density cultivation was evaluated from 2020 to 2024 in central Italy by assessing their agronomic behavior, with the aim of identifying which Italian olive cultivars can combine high productivity and suitability for intensive mechanization—through high- and very high-density planting systems—allowing biodiversity valorization. The cultivars were Borgiona, Don Carlo, FS17, Gentile di Anghiari, Gentile di Montone, Giulia, Leccio del Corno, Maurino, Moraiolo, Pendolino, Piantone di Falerone, and Piantone di Mogliano. The international cultivar Arbequina was used as a reference. The olive orchard was planted in 2015, at a tree spacing of 5 m × 2 m (1000 trees/ha). Arbequina was found to have limited vigor and high production efficiency, as reported in other works, therefore confirming its suitability for high-density and super-high-density cultivation. Some cultivars, such as Leccio del Corno, Maurino, FS17, Piantone di Mogliano, and Piantone di Falerone, had a production and yield efficiency that was not different from or even higher than Arbequina. Other cultivars found to be promising were Don Carlo and Gentile di Anghiari, which had a slightly lower productive performance than Arbequina. Overall, the results are encouraging and suggest that some of these cultivars may be suitable candidates for high- and super-high-density olive orchards. This suitability is further supported by their favorable fruit characteristics, which appear to facilitate efficient mechanical harvesting. However, additional data is necessary to enable a more comprehensive assessment of these cultivars, particularly their capacity to maintain canopy dimensions compatible with straddle harvester operation, while maintaining a stable vegetative–reproductive balance over time. Full article

The super high-density (SHD) production system has recently been introduced to the Chilean European plum (Prunus domestica L.) industry, but the potential of applying regulated deficit irrigation (RDI) in this system remains unexplored. As irrigation water availability in Chile has been strongly jeopardized by climate change, there is an urgent need to validate water-conserving practices in modern production systems. A field study was conducted in a commercial SHD European plum orchard (cv. French grafted on Rootpac-20 rootstock) for two consecutive seasons in Peralillo, O’Higgins Region, Chile. The objective of this study was to assess the impact of a late water deficit (LD) on water productivity, fruit quality, plant water relations, and soil microbiota. The results showed that implementing LD enhanced water productivity by 40% without compromising fresh and dry fruit quality. Moderate to severe water stress induced no changes in physiological parameters such as stomatal conductance and photochemical efficiency. Additionally, the LD treatment significantly reduced soil moisture but increased the abundance of certain groups of beneficial soil microbiota and fine roots. These results highlight the potential of LD as a viable water-conserving practice in modern SHD European plum orchards, particularly in regions facing water scarcity due to climate change. Full article

Abiotic stresses like salinity are proven to be crucial limiting factors in the seed germination of many plant species and the later establishment of cultivation regarding plant growth, yield and fruit quality. Therefore, there is a pressing need to find practices and materials to enhance abiotic stress tolerance from early stages such as germination so that plants can overcome these stresses as soon as possible. A total of six treatments of seaweed extracts [1, 2 and 3% of Algit Super (Ascophyllum nodosum) and Alga 300 (Sargassum spp.)] and three controls were tested, with 20 seeds per replication soaked in each extract concentration for 15′; four replications were carried out per treatment and seeds were placed on Petri dishes in the dark. Speed and percentage of germination, vigor index I and II, dry weight and average lengths of roots and shoots were evaluated under 75 mM NaCl stress. All treatments positively affected all parameters evaluated, whether significant or not. Results indicate that soaking tomato seeds in seaweed extracts of various concentrations led to a significantly increased speed and percentage of germination, vigor index I and II, dry weight and average lengths of roots and shoots. The best combination of concentration and seaweed species is concluded to be 2% Sargassum spp. for all parameters evaluated. Full article

The circular economy with compost fertilization is included in the sustainable orchard paradigm, creating a holistic production ecosystem. Modern orchards are mostly intensive and super-intensive, requiring different rootstocks. This research presents the response to compost fertilization of two specific pear rootstocks (quince ‘CTS 212’ and ‘Farold^® 40’) and own-rooted trees, analyzing six resistant cultivars in a circular production system. The dynamic of nitrogen and carbon concentration in leaves, soil respiration coefficient, the evolution of the fruit maturity stage in the field, and some biometric parameters such as trunk cross-section area, the annual vegetative growth, and fruiting shoots annual number were analyzed. The results highlighted that compost fertilization led to increased leaf nitrogen concentration over the first two years while carbon concentration remained relatively stable. Rootstock and compost fertilization influenced the fruit maturity dynamic, but a single pattern was not identified. Quince, as pear rootstock, expressed a higher sensitivity to compost application; the biometric parameters, such as trunk cross-section area, and almost all cultivars’ annual vegetative growth were higher than the controls’. Positive output can lead to future model upscaling in farms and households. Full article

Background/Objectives: The biostimulant SuperFifty, produced from the brown algae Ascophyllum nodosum, can improve crop quality and yield and mitigate stress tolerance in model and crop plants such as Arabidopsis thaliana, pepper, and tomato. However, the effect of SuperFifty on raspberries and strawberries has not been well studied, especially in terms of nutritional properties and yield. The aim of this study was to investigate the effect of SuperFifty on the quality and quantity of raspberry and strawberry fruits, with a focus on metabolic composition and essential elements, which together determine the nutritional properties and total yield of these two crops. Methods: Metabolome analysis was performed by liquid chromatography–mass spectrometry analysis (LC-MS), and essential elements analysis was performed by inductively coupled plasma-mass spectrometry (ICP-MS). Results: Here, we demonstrate that SuperFifty increases the fruit size of both raspberries and strawberries and enhances the yield in these two berry crops by 42.1% (raspberry) and 33.9% (strawberry) while preserving the nutritional properties of the fruits. Metabolome analysis of 100 metabolites revealed that antioxidants, essential amino acids, organic acids, sugars, and vitamins, such as glutathione, alanine, asparagine, histidine, threonine, serine, tryptophan, sucrose, citric acid, pantothenic acid (vitamin B5), as well as other primary metabolites, remain the same in the SuperFifty-primed fruits. Secondary metabolites, such as caffeic acid, p-coumaric acid, kaempferol, and quercetin, also maintained their levels in the SuperFifty-primed fruits. Analysis of essential elements demonstrated that elements important for human health, such as Zn, Mn, Fe, B, Cu, K, and Ca, maintain the same levels in the raspberry and strawberry fruits obtained from the biostimulant-primed plants. Magnesium, an important element known as a co-factor in many enzymatic reactions related to both plant physiology and human health, increased in both raspberry and strawberry fruits primed with SuperFifty. Finally, we discuss the potential financial and health benefits of the SuperFifty-induced priming for both growers and consumers. Conclusions: We demonstrate that SuperFifty significantly enhances the yield of both raspberries and strawberries, improves the marketable grade of the fruits (larger and heavier fruits), and enhances the nutritional properties by elevating Mg content in the fruits. Altogether, this biostimulant-induced molecular priming offers an environmentally friendly, efficient, and sustainable way to enhance the yield and quality of berry crops, with clear benefits to both berry producers and customers. Full article

In many European countries the consumption of tropical fruit is constantly growing, and people are increasingly turning to diets rich in fruit and vegetables. In this context, mango is considered a super-food for its nutritional medium-high energy value. Produced mainly in developing countries, tropical fruits animate an interesting international market. Production in Mediterranean countries is also growing and is increasingly requested in European markets. The aim of this work is to investigate the factors that drive the inclination to purchase fresh mango and mango food and drinks in Italy and Brazil in order to observe consumer preferences in the two countries. The personal experiences, motivations and choices of consumers regarding fresh mango and mango-based products were taken into consideration. Through an online survey, a semi-structured questionnaire was administered in Italy and Brazil which led to a total sample of 453 participants. The data were statistically analyzed, and a PLS-SEM model was used to empirically examine the factors influencing the consumption of fresh mango and mango food and drinks. The research hypotheses are all supported. For a comparison between the two countries, a multigroup analysis (PLS-MGA) was performed. In Italy, consumers are attentive to the quality and safety of the fruit; they choose the point of sale where they buy fresh mango or mango foods because they trust the seller to guarantee the fruit’s origin and transformation. In Brazil, new consumer trends are emerging especially in gastronomy; since they are local foods, they are considered safe, sustainable and healthy by consumers. The study addresses a little-explored topic and aims to enrich the debate on consumer orientations, preferences and reasons for buying mango and mango products. Full article

Object detection models are commonly used in yield estimation processes in intelligent walnut production. The accuracy of these models in capturing walnut features largely depends on the quality of the input images. Without changing the existing image acquisition devices, this study proposes a super-resolution reconstruction module for drone-acquired walnut images, named Walnut-SR, to enhance the detailed features of walnut fruits in images, thereby improving the detection accuracy of the object detection model. In Walnut-SR, a deep feature extraction backbone network called MDAARB (multilevel depth adaptive attention residual block) is designed to capture multiscale information through multilevel channel connections. Additionally, Walnut-SR incorporates an RRDB (residual-in-residual dense block) branch, enabling the module to focus on important feature information and reconstruct images with rich details. Finally, the CBAM (convolutional block attention module) attention mechanism is integrated into the shallow feature extraction residual branch to mitigate noise in shallow features. In 2× and 4× reconstruction experiments, objective evaluation results show that the PSNR and SSIM for 2× and 4× reconstruction reached 24.66 dB and 0.8031, and 19.26 dB and 0.4991, respectively. Subjective evaluation results indicate that Walnut-SR can reconstruct images with richer detail information and clearer texture features. Comparative experimental results of the integrated Walnut-SR module show significant improvements in mAP50 and mAP50:95 for object detection models compared to detection results using the original low-resolution images. Full article