Search Results (373)

Achieving precise fruit localization and fine branch segmentation simultaneously in unstructured orchard environments remains challenging due to variable lighting, occlusion, and complex backgrounds. This study proposed a joint detection–segmentation architecture based on an improved YOLOv11 network for collaborative perception of apples and tree branches. First, a dual-task dataset of spindle-type apple orchards was constructed with bounding-box annotations for fruits and pixel-level polygon masks for branches, encompassing diverse illumination and occlusion conditions. Second, Convolutional Block Attention Modules (CBAMs) are strategically embedded into the YOLOv11 backbone to enhance feature discrimination for slender branch structures while preserving high fruit detection accuracy. The enhanced model achieves precision of 0.981, recall of 0.986, and F1-score of 0.983 for apple detection, and precision of 0.803, recall of 0.715, mAP of 0.698, and IoU of 0.6066 for branch segmentation on the validation set. Comparative experiments against YOLOv8 and baseline YOLOv11 confirm improved segmentation continuity and finer branch delineation. The proposed integrated perception framework provides reliable visual guidance for collision-avoidance robotic harvesting and offers a practical reference for multi-task agricultural vision systems. Full article

This study compared the quality and bioactive composition of cloudy apple juices produced from four traditional and four conventional apple cultivars immediately after harvest and following cold storage of the fruit at 4 °C for three and six months. Apples were harvested at the ripening stage at the same criteria, stored as whole fruit, and processed into cloudy juice after harvest, three, and six months of storage. Physicochemical parameters and sugar composition were determined, while phenolic compounds were quantified by HPLC-PDA. Antioxidant activity, total phenolic, and flavonoid content were measured spectrophotometrically. All analyses were performed in technical triplicate. The results revealed notable differences between traditional and conventional cultivars. Juices produced from traditional apple cultivars exhibited significantly higher total polyphenol and flavonoid contents than those from conventional cultivars. Significant variations in catechin, myricetin, quercetin, and epigallocatechin levels were also observed among cultivars. The traditional apple cultivar ‘Mašanka’ showed higher concentrations of quercetin (0.09 ± 0.01 µg/mL), chlorogenic acid (486.58 ± 5.48 µg/mL), catechin (8.76 ± 0.54 µg/mL), epicatechin (20.22 ± 0.20 µg/mL), and phloridzin (13.48 ± 0.19 µg/mL) compared to the other cultivars. In contrast, conventional cultivars showed higher concentrations of myricetin and procyanidin B1. Moreover, the content of TA, sucrose, and glucose decreased, whereas pH, fructose, TSS (except for ‘Fuji’ and ‘Granny Smith’) increased. The TFC decreased in traditional apple cultivars, while it increased in conventional cultivars; however, the TFC in conventional cultivars remained lower than in traditional ones. Overall, these findings demonstrate that the cold storage of apples significantly affects juice composition and highlight the advantages of traditional apple cultivars for producing juices with enhanced phenolic content and antioxidant activity. Full article

Rising global food demand, increasing labor costs, and farm labor shortages have created significant challenges for specialty crop production, particularly in labor-intensive tasks such as fruit harvesting. Robotic harvesting offers a promising long-term solution, yet its adoption in orchard environments remains limited due to unstructured conditions, variable lighting, and difficulties in fruit recognition and manipulation. This study presents an improved robotic fruit harvesting system, Orchard roBot (OrBot), developed by the Robotics Vision Lab at Northwest Nazarene University, with the goal of advancing autonomous apple harvesting applications. The updated OrBot platform integrates a dual-camera vision system consisting of an eye-to-hand stereo camera with a wide field of view for fruit detection and an eye-in-hand RGB-D camera for precise manipulation. The control architecture was redesigned using Robot Operating System 2 (ROS2) and Python, enabling modular subsystem development and coordination. Fruit detection was performed using a YOLOv5 deep learning model, and visual servoing was employed to guide the robotic manipulator toward the target fruit. System performance was evaluated through laboratory experiments using artificial trees and field tests conducted in a commercial apple orchard in Idaho. OrBot achieved a 100% harvesting success rate in indoor tests and a 75–80% success rate in outdoor orchard conditions. Experimental results demonstrate that the dual-camera approach significantly enhances fruit search efficiency and harvesting efficiency. Identified limitations include sensitivity to lighting conditions, end effector performance with varying fruit sizes, and depth estimation errors. Overall, the results indicate a positive potential toward effective robotic fruit harvesting and highlight key areas for future improvement in vision, manipulation, and system robustness. Full article

Apple detection is a core perception task for harvesting robots operating in complex orchard environments. Targets are frequently affected by branch–foliage occlusion, alternating front/side/back lighting, and strong local illumination fluctuations, which blur object boundaries against background textures and substantially increase detection difficulty. To improve target perception under these conditions, we propose an improved detector, YOLOv11-CBMES. First, based on YOLOv11, we replace the original neck with a weighted BiFPN to enhance cross-scale feature fusion under occlusion. Second, we introduce a Contrast-Driven Feature Aggregation (CDFA) module at the P5 stage, using Haar wavelet decomposition to decouple low-frequency illumination components from high-frequency structural components. Third, we reconstruct spatial feature learning and the upsampling pathway using CSP-based multi-scale blocks and efficient upsampling blocks, and embed a zero-parameter Shift-Context strategy to strengthen local neighbourhood interaction. Finally, we formulate apple detection as a three-class occlusion classification task (No Occlusion, Soft Occlusion, and Hard Occlusion) to support occlusion-aware target recognition. On the apple occlusion dataset, YOLOv11-CBMES achieves ${\mathrm{m}\mathrm{A}\mathrm{P}}_{\mathrm{N}\mathrm{O}}$ = 83.50%, ${\mathrm{m}\mathrm{A}\mathrm{P}}_{\mathrm{S}\mathrm{O}}$ = 67.36%, and ${\mathrm{m}\mathrm{A}\mathrm{P}}_{\mathrm{H}\mathrm{O}}$ = 51.90% at IoU = 0.5. Compared with YOLOv11n under the same training protocol, the gains are +2.16 pp (NO), +3.68 pp (SO), and +5.31 pp (HO), with the largest improvement observed in Hard Occlusion (HO). The results indicate that introducing frequency-domain structural processing into the detection framework improves apple occlusion classification and object detection performance, and provides a theoretical basis for designing perception modules for end-effector operations in apple harvesting robots. Full article

Apple picking is an inherently labor-intensive, time-consuming, and costly task, and robotic harvesting represents a potential alternative to address this challenge. This study presents the development and field evaluation of an integrated robotic system for apple harvesting, which combines machine vision, a dual four-degree-of-freedom (DoF) manipulator, and a mobile platform. The harvesting mechanism employed a streamlined 4-DoF manipulator driven by closed-loop stepper motors, incorporating a differential gear mechanism to execute yaw and pitch motions. Trajectory planning utilized linear interpolation with a harmonic acceleration/deceleration profile to ensure smooth end-effector movement. Fruit detection and localization within the canopy were performed by a stereo vision system running a lightweight deep neural network, achieving a mean hand-eye calibration accuracy of 4.7 ± 2.7 mm. Three negative-pressure driven soft end-effector designs—a suction soft end-effector (SSE), a grasping soft end-effector (GSE), and a suction-grasping soft end-effector (SGSE)—were assessed for their harvesting performance. Field trials conducted in a commercial spindle orchard demonstrated that the GSE achieved the highest performance, with a harvesting success rate of 80.80% among reachable fruits, a full-process success rate (from detection to collection) of 61.59%, an overall fruit damage rate of 10.89%, and an average single-fruit cycle time of 5.27 s. In contrast, the SSE and SGSE showed lower success rates (49.21% and 64.71%, respectively). This work provides a practical robotic harvesting solution. It validates the feasibility of a zoned, multi-manipulator harvesting strategy and delivers comparative data to guide the development of more efficient and robust harvesting robots. Full article

Severe foliage occlusion and dynamically changing lighting conditions in complex orchard environments pose significant challenges for visual perception systems in automated apple harvesting, including low detection accuracy, poor robustness, and insufficient real-time performance. To address these issues, this study proposes an improved lightweight detection network based on YOLOv11, named YOLO-WBL, along with a precise yield estimation algorithm based on 3D point clouds, termed CLV. The YOLO-WBL network is optimized in three aspects: (1) A C3K2_WT module integrating wavelet transform is introduced into the backbone network to enhance multi-scale feature extraction capability; (2) A weighted bidirectional feature pyramid network (BiFPN) is adopted in the neck network to improve the efficiency of multi-scale feature fusion; (3) A lightweight shared convolution separated batch normalization detection head (Detect-SCGN) is designed to significantly reduce the parameter count while maintaining accuracy. Based on this detection model, the CLV algorithm deeply integrates depth camera point cloud information through 3D coordinate mapping, irregular point cloud reconstruction, and convex hull volume calculation to achieve accurate estimation of individual fruit volume and total yield. Experimental results demonstrate that: (1) The YOLO-WBL model achieves a precision of 93.8%, recall of 79.3%, and mean average precision (mAP@0.5) of 87.2% on the apple test set; (2) The model size is only 3.72 MB, a reduction of 28.87% compared to the baseline model; (3) When deployed on an NVIDIA Jetson Xavier NX edge device, its inference speed reaches 8.7 FPS, meeting real-time requirements; (4) In scenarios with an occlusion rate below 40%, the mean absolute percentage error (MAPE) of yield estimation can be controlled within 8%. Experimental validation was conducted using apple images selected from the dataset under varying lighting intensities and fruit occlusion conditions. The results demonstrate that the CLV algorithm significantly outperforms traditional average-weight-based estimation methods. This study provides an efficient, accurate, and deployable visual solution for intelligent apple harvesting and yield estimation in complex orchard environments, offering practical reference value for advancing smart orchard production. Full article

The brown marmorated stink bug, Halyomorpha halys (Stål), is an invasive pest that increasingly threatens apple production in Europe by causing fruit damage, yield losses, and quality deterioration under commercial orchard conditions. This study investigated seasonal population dynamics, spatial patterns of fruit damage, yield effects, and post-harvest fruit responses of two apple cultivars (‘Cripps Pink’ and ‘Fuji’) in a commercial orchard over two consecutive seasons (2024–2025). Adult and nymphal activity was monitored using pheromone traps, while fruit damage was assessed at harvest across orchard positions and canopy layers. Potential yield losses were estimated based on damage incidence, and selected physicochemical properties of healthy and affected fruits were analysed. Clear cultivar-dependent differences were observed. ‘Fuji’ exhibited typical external feeding damage, with low but consistent damage levels and limited yield losses in both seasons. In contrast, ‘Cripps Pink’ showed substantially higher damage rates and potential yield losses, particularly in 2025; however, classical external feeding damage was not observed. Instead, fruits exposed to H. halys pressure expressed atypical responses, primarily as increased individual fruit mass and size, and atypical skin color patterns, including pronounced striping and uneven pigmentation. Damage in ‘Cripps Pink’ was strongly structured within the orchard, with higher incidence in the upper and middle canopy layers and in areas adjacent to the forest edge, whereas damage in ‘Fuji’ remained low and spatially uniform. Overall, the results demonstrate that the impact of H. halys depends not only on pest pressure but also on cultivar traits and within-orchard spatial heterogeneity. These findings support the development of cultivar-specific and spatially targeted integrated pest management (IPM) strategies that better reflect the uneven distribution and expression of stink bug injury in commercial apple orchards. Full article

This paper presents the design of a high-efficiency apple harvesting robot based on a hybrid pneumatic-electric drive system, capable of operating around the clock. The robotic system comprises a mobile platform with two degrees of freedom (DOF) and a five-DOF PRRRP manipulator for fruit picking. To meet the harvesting requirements, a spoon-shaped end-effector with pneumatic control was developed, enabling precise manipulator control and flexible grasping. The robot’s vision system integrates machine vision and deep neural network approaches. Additionally, an industrial computer and AC servo drivers were employed to control the manipulator and end-effector. An integrated nighttime illumination system allowed for all-weather operation. Initial experiments were conducted in a controlled laboratory. Subsequently, comprehensive identification and harvesting tests were performed in both laboratory and field environments to validate system robustness. Experimental results validated the effectiveness of the proposed system, demonstrating an apple harvesting success rate of 81% and an average harvesting time of 7.81 s per apple. The system achieved a fruit damage rate of less than 5% during field experiments, demonstrating its potential for gentle handling. The primary innovation of this work lies in its hybrid drive architecture and adaptive vision strategy, which together offer a cost-effective and robust solution for all-weather automated harvesting, addressing key limitations of high cost and environmental sensitivity in existing robotic harvesters. Full article

Flesh mealiness, a textural disorder in apples, reduces storage quality and consumer acceptance. The ‘Delicious’ and ‘Fuji’, prominent apple cultivars in China, exhibit contrasting susceptibility to mealiness, though the underlying mechanisms remain unclear. This study compared cytological, physiological and cell wall metabolic changes between mealy ‘Oregon Spur II Delicious’ and non-mealy ‘Miyazaki Spur Fuji’ during ambient storage. Toluidine blue staining and scanning electron microscopy revealed that ‘Delicious’ exhibited larger intercellular spaces and cell separation in contrast to ‘Fuji’. This observation aligns with the earlier onset of mealiness in ‘Delicious’: its mealiness degree increased from 3.06% at harvest to 19.62% after 28 d of storage (a 6.4-fold rise), whereas that of ‘Fuji’ only increased from 2.13% to 3.90% (1.8-fold). This pronounced increase in ‘Delicious’ was accompanied by a significant increase in air space volume and a reduction in expressible juice. Furthermore, the occurrence of mealiness in ‘Delicious’ involved a sharp increase in respiration rate and ethylene production, alongside rapid declines in firmness and starch content. Notably, there was a substantial accumulation of water-soluble pectin (WSP) and chelator-soluble pectin (CSP) in ‘Delicious’, whereas the content of Na₂CO₃-soluble pectin (NSP) remained consistently lower. Monosaccharide composition analysis confirmed significantly reduced arabinose and galactose levels across pectin fractions (WSP, CSP, and NSP) in ‘Delicious’. Correspondingly, immunofluorescence labeling showed a pronounced degradation of arabinan and galactan within the side chains of rhamnogalacturonan-I (RG-I). In addition, the activities of pectin methylesterase, α-L-Arabinofuranosidase, and β-D-Galactosidase remained significantly elevated in ‘Delicious’. Collectively, these findings demonstrate that cultivar differences in flesh mealiness are attributable to divergent physiological senescence and cell wall disassembly processes. Full article

New Zealand’s kiwifruit and apple industries generate substantial quantities of organic residues during thinning and harvest, much of which is composted or disposed of in landfills due to logistical constraints. This study evaluates the potential of these residues as feedstock for biomethane production via anaerobic digestion (AD), followed by hydrogen generation through steam methane reforming (SMR). Two feedstock mixtures were examined: a 50:50 kiwifruit–apple blend and a 40:40:20 kiwifruit–apple–potato mixture, designed to mitigate acidification. Cow manure served as a cost-effective inoculum. Physicochemical analysis confirmed high moisture and volatile solids content, indicating strong biodegradability, although low nitrogen content suggests the need for co-digestion in full scale systems. Biomethane potential (BMP) tests yielded up to 45 mL CH₄/gVS at an ISR of 4, corresponding to 46.5% carbon conversion. Scaling to an annual waste volume of 476 t suggests a potential biomethane yield of approximately 18,000 m³. SMR simulations demonstrated technical feasibility, with methane conversion increasing from 46% under baseline conditions to >85% under optimized steam to carbon ratios and residence times. Hydrogen yields of ~7600 m³/year were estimated. This study provides a practical foundation for valorizing fruit waste into renewable biomethane and hydrogen, supporting New Zealand’s circular economy and decarbonization goals. Full article

Accurate apple detection and precise three-dimensional (3D) localisation are essential for autonomous robotic harvesting in orchard environments, where occlusion, illumination variation, depth noise, and the similar colour appearance of fruits and surrounding leaves present significant challenges. This paper proposes a dual-detector vision framework combined with depth-aware back-projection to achieve robust apple detection and metric 3D localisation in real time. The method integrates the complementary strengths of YOLOv8 and Mask R-CNN through confidence-weighted fusion of bounding boxes and pixel-wise union of segmentation masks, producing stabilised two-dimensional (2D) apple representations under visually ambiguous conditions. The fusion results are converted into dense 3D representations through depth-guided projection within the camera coordinate system representing the visible fruit surface. A depth-consistency weighting strategy assigns higher influence to depth-reliable pixels during centroid computation, thereby suppressing noisy or occluded depth measurements and improving the stability of 3D fruit centre estimation, while local intensity normalisation standardises neighbourhood-level pixel intensities to reduce the impact of shadows, highlights, and uneven lighting, enabling more consistent segmentation and detection across varying illumination conditions. Experimental results demonstrate an accuracy of 98.9%, an mAP of 94.2%, an F1-score of 93.3%, and a recall of 92.8%, while achieving real-time performance at 86.42 FPS, confirming the suitability of the proposed method for robotic harvesting in challenging orchard environments. Full article

Phytopathogenic fungi pose a critical threat to global food security through substantial pre- and post-harvest crop losses, intensified by climate change and fungicide resistance. To address this, we synthesized low-concentration chitosan–inorganic oxide nanocomposites (CS-SiO₂, CS-ZnO, CS-CuO) via ionic gelation, a green and scalable method. Comprehensive characterization (DLS, UV-Vis, FTIR, XRD, SEM) confirmed nanocomposite formation, CS-SiO₂ exhibited uniform particle sizes (200–250 nm), while CS-CuO showed slightly larger particles, all with excellent dispersity. Zeta potential analysis confirmed strong colloidal stability, with pure chitosan nanoparticles (CSNPs) displaying a surface charge of +12.9 mV, while all nanocomposites retained positive charges, enhancing adhesion to negatively charged fungal membranes. In vitro antifungal assays against Alternaria alternata, Botrytis cinerea, Colletotrichum graminicola, and Fusarium graminearum demonstrated hierarchical efficacy: CS-CuO > CS-ZnO > CS-SiO₂, with CS-CuO achieving >80% growth inhibition against B. cinerea and A. alternata. SEM revealed severe hyphal damage and spore collapse in CS-CuO-treated fungi, attributed to synergistic reactive oxygen species (ROS) generation and chitosan-mediated membrane disruption. In vivo trials on B. cinerea-infected apples showed CS-CuO reduced lesion area by 81% and elevated host defense markers, including a 1.5-fold increase in total phenolic content and higher DPPH radical scavenging activity compared to controls. These nanocomposites, particularly CS-CuO, offer a sustainable, dual-action solution direct antifungal activity and enhanced host resilience while minimizing environmental impact. By integrating scalable synthesis, eco-compatibility, and efficacy, this work advances chitosan–inorganic oxide nanocomposites as viable alternatives to conventional fungicides, with immediate potential for agricultural and postharvest applications. Full article

Real-time object detection on resource-constrained edge devices remains a critical challenge in precision agriculture and autonomous systems, particularly when integrating advanced multi-modal sensors (RGB-D, thermal, hyperspectral). This paper introduces FEGW-YOLO, a lightweight detection framework explicitly designed to bridge the efficiency-accuracy gap for fine-grained visual perception on edge hardware while maintaining compatibility with multiple sensor modalities. The core innovation is a Feature Complexity Descriptor (FCD) metric that enables adaptive, layer-wise compression based on the information-bearing capacity of network features. This compression-guided approach is coupled with (1) Feature Engineering-driven Ghost Convolution (FEG-Conv) for parameter reduction, (2) Efficient Multi-Scale Attention (EMA) for compensating compression-induced information loss, and (3) Wise-IoU loss for improved localization in dense, occluded scenes. The framework follows a principled “Compress, Compensate, and Refine” philosophy that treats compression and compensation as co-designed objectives rather than isolated knobs. Extensive experiments on a custom strawberry dataset (11,752 annotated instances) and cross-crop validation on apples, tomatoes, and grapes demonstrate that FEGW-YOLO achieves 95.1% mAP@0.5 while reducing model parameters by 54.7% and computational cost (GFLOPs) by 53.5% compared to a strong YOLO-Agri baseline. Real-time inference on NVIDIA Jetson Xavier achieves 38 FPS at 12.3 W, enabling 40+ hours of continuous operation on typical agricultural robotic platforms. Multi-modal fusion experiments with RGB-D sensors demonstrate that the lightweight architecture leaves sufficient computational headroom for parallel processing of depth and visual data, a capability essential for practical advanced sensing systems. Field deployment in commercial strawberry greenhouses validates an 87.3% harvesting success rate with a 2.1% fruit damage rate, demonstrating feasibility for autonomous systems. The proposed framework advances the state-of-the-art in efficient agricultural sensing by introducing a principled metric-guided compression strategy, comprehensive multi-modal sensor integration, and empirical validation across diverse crop types and real-world deployment scenarios. This work bridges the gap between laboratory research and practical edge deployment of advanced sensing systems, with direct relevance to autonomous harvesting, precision monitoring, and other resource-constrained agricultural applications. Full article

In this work, thermal imaging is employed to study the opto-thermal response of apples (Malus domestica Borkh.), assessing their post-harvest evolution through the estimation of thermal diffusivity. A non-destructive experimental procedure based on mid-wave infrared (MWIR) thermal camera (3–5 µm) and localized heating with a visible laser is developed, enabling spatially and temporally resolved surface temperature measurements. Temperature fields are recorded at different time points and radial distances from the heated spot. A theoretical model based on Fourier thermal diffusion equation is formulated to describe the spatio-temporal evolution of surface temperature. After validation on a reference sample, the method is applied to Golden and Red Delicious apples over a 28-day storage period at room temperature. Red Delicious apple exhibits higher mean diffusivity values without significant temporal changes, whereas a progressive increase in diffusivity is observed for Golden Delicious apples. These results show that thermal diffusivity is sensitive to post-harvest physiological changes in apple tissue and may be associated with intrinsic properties such as tissue density and water content. By relating laser-induced temperature fields to the estimation of thermal diffusivity, this approach enables the non-destructive, quantitative assessment of thermal diffusivity, showing potential for fruit maturity and quality assessment, which are of high importance in agri-food monitoring applications. Full article

Efficient apple orchard water management under climate variability requires understanding how fruit load and water supply regulate branch-scale water use to optimize irrigation, yield, and fruit quality. During the summer of 2014, sap flow (SF) and maximum daily shrinkage (MDS) were measured in one branch from six apple trees (Malus domestica Borkh. Cv. ‘Jazz™’) using the Compensation Heat Pulse method and diameter variation sensors in an orchard near Havelock North, New Zealand. One west-oriented branch per tree, with diameters of 1.5 to 2.3 cm, was monitored alongside midday stem (ψs) and leaf (ψl) water potentials, leaf gas exchanges, leaf area index (LAI), and fruit dry matter per branch at the end of the growing season. Half of the trees were subjected to irrigation withdrawal after day of year (DOY) 31 (non-irrigated treatment), resulting in a significantly lower midday stem water potential (ψs) by DOY 56 (−1.03 MPa). Pre-harvest, SF and MDS were tightly correlated (r² = 0.69), but this correlation decreased post-harvest (r² = 0.16) due to reduced fluctuations in both SF and branch variations (BV). SF was normalized per unit of leaf area, categorizing branches into high and low LAI: fruit dry matter ratio. SF values were approximately 2.2 times higher for FI pre-harvest and remained 2-fold higher post-harvest, associated with lower ψl and higher midday leaf transpiration for FI. MDS was identified as a better indicator of mild water deficit compared to SF, with both measurements responding effectively to midday vapor pressure deficit and reference evapotranspiration values. Overall, MDS proved to be a more sensitive indicator of mild water deficit than SF, while fruit load exerted a persistent influence on branch water use, highlighting the value of branch-scale measurements for improving irrigation management in apple orchards. Full article