Search Results (15)

In the context of blueberry orchard management and automated harvesting, this study introduces an improved YOLOv8 model, ADE-YOLO, designed for precise blueberry ripeness detection, enhancing automated picking efficiency. Built on the YOLOv8n architecture, ADE-YOLO features a dimensionality-reducing convolution at the backbone’s end, reducing computational complexity while optimizing input features. This improvement enhances the effectiveness of the AIFI module, particularly in multi-scale feature fusion, boosting detection accuracy and robustness. Additionally, the neck integrates a dynamic sampling technique, replacing traditional upsampling methods, allowing for more precise feature integration during feature transfer from P5 to P4 and P4 to P3. To further enhance computational efficiency, CIOU is replaced with EIOU, simplifying the aspect ratio penalty term while maintaining high accuracy in bounding box overlap and centroid distance calculations. Experimental results demonstrate ADE-YOLO’s strong performance in blueberry ripeness detection, achieving a precision of 96.49%, recall of 95.38%, and mAP scores of 97.56% (mAP50) and 79.25% (mAP50-95). The model is lightweight, with just 2.95 M parameters and a 6.2 MB weight file, outpacing YOLOv8n in these areas. ADE-YOLO’s design and performance underscore its significant application potential in blueberry orchard management, providing valuable support for precision agriculture. Full article

The automated harvesting of blueberries using a picking robot places a greater demand on the 3D spatial perception performance, as the robot’s grasping mechanism needs to pick blueberry fruits accurately at specific positions and in particular poses. To achieve this goal, this paper presents a method for blueberry detection, 3D spatial localization, and pose estimation using visual perception, which can be deployed on an OAK depth camera. Firstly, a blueberry and calyx scar detection dataset is constructed to train the detection network and evaluate its performance. Secondly, the blueberry and calyx scar detection model based on a lightweight YOLOv11 (the eleventh version of You Only Look Once) network with an improved depth-wise separable convolution (DSC) module is designed, and a 3D coordinate system relative to the camera is established to calculate the 3D pose of the blueberry fruits. Finally, the above detection model is deployed using the OAK depth camera, leveraging its depth estimation module and three-axis gyroscope to obtain the 3D coordinates of the blueberry fruits. The experimental results demonstrate that the method proposed in this paper can accurately identify blueberry fruits at various maturity levels, achieving a detection accuracy of 95.8% mAP50-95, a maximum positioning error of 7.2 mm within 0.5 m, and an average 3D pose error of 19.2 degrees (around 10 degrees at the ideal picking angle) while maintaining a detection frame rate of 13.4 FPS (frames per second) on the OAK depth camera, providing effective picking guidance for the mechanical arm of picking robots. Full article

Identifying fruit maturity accurately and achieving damage-free harvesting are challenges in designing blueberry-picking robots. This paper presents an intelligent flexible picking system. First, we trained a deep learning-based YOLOv8n network to locate the position of the fruit and determine fruit ripeness. We used a neural network to establish the relationship between fruit hardness and shape parameters, achieving an adaptive gripping force for different fruits. To address the issue of dense clusters in some blueberry varieties, we designed a fruit separation subsystem using a combination of flow field analysis and pressure-sensitive experiments. The results show that the mean average precision can reach 84.62%, the precision is 94.49%, the recall is 83.85%, the F1 score is 88.85%, and the test time is 0.12 s, which can meet the requirements for blueberry fruit recognition accuracy and speed. The spacing between closely packed fruits can increase by 4 mm, and the damage-free picking rate exceeds 92%, achieving stable, damage-free harvesting. Full article

In China, most blueberry varieties are characterized by tightly clustered fruits, which pose challenges for achieving precise and non-destructive automated harvesting. This complexity limits the design of robots for this task. Therefore, this paper proposes adding a jetting step during harvesting to separate fruit clusters and increase the operational space for mechanical claws. First, a combined approach of flow field analysis and pressure-sensitive experiments was employed to establish design criteria for the number, diameter, and inclination angle parameters of two types of nozzles: flat tip and round tip. Furthermore, fruit was introduced, and a fluid–structure coupling method was employed to calculate the deformation of fruit stems. Simultaneously, a mechanical analysis was conducted to quantify the relationship between jet characteristics and separation gaps. Simulation and pressure-sensitive experiments show that as the number of holes increases and their diameter decreases, the nozzle’s convergence becomes stronger. The greater the inclination angle of the circular nozzle holes, the more the gas diverges. The analysis of the output characteristics of the working section indicates that the 8-hole 40° round nozzle is the optimal solution. At an air compressor working pressure of 0.5 MPa, force analysis and simulation results both show that it can increase the picking space for the mechanical claw by about 5–7 mm without damaging the blueberries in the jet area. The final field experiments show that the mean distance for Type I (mature fruit) is 5.41 mm, for Type II (red fruit) is 6.42 mm, and for Type III (green fruit) is 5.43 mm. The short and curved stems of the green fruit are less effective, but the minimum distance of 4.71 mm is greater than the claw wall thickness, meeting the design requirements. Full article

Blueberries are one of the more economically rewarding fruits for fruit growers. Identifying blueberry fruit at different stages of maturity is economically important and can aid fruit growers in planning pesticide applications, estimating yields, and efficiently conducting harvesting operations, among other benefits. Visual methods for identifying the different ripening stages of fruits are increasingly receiving widespread attention. However, due to the complex natural environment and the serious shading caused by the growth characteristics of blueberries, the accuracy and efficiency of blueberry detection are reduced to varying degrees. To address the above problems, in the study presented herein, we constructed an improved YOLOv9c detection model to accurately detect and identify blueberry fruits at different ripening stages. The size of the network was reduced by introducing the SCConv convolution module, and the detection accuracy of the network in complex and occluded environments was improved by introducing the SE attention module and the MDPIoU loss function. Compared to the original model, the mAP0.5 and mAP0.5:0.95 of the improved YOLOv9c network improved by 0.7% and 0.8%, respectively. The model size was reduced by 3.42 MB, the number of model parameters was reduced by 1.847 M, and the detection time of a single image was reduced by 4.5 ms. The overall performance of the detection model was effectively improved to provide a valuable reference for accurate detection and localization techniques for agricultural picking robots. Full article

Fresh blueberries are delicate, hand-picked, packaged, and refrigerated fruits vulnerable to spoilage and contamination. Cold atmospheric plasma (CAP) is a promising antimicrobial technology; therefore, this study evaluated the CAP treatment effect on acid-tolerant Listeria innocua and Listeria monocytogenes and evaluated changes in the quality of the treated fruit. Samples were spot-inoculated with pH 5.5 and 6.0 acid-adapted Listeria species. Samples were treated with gliding arc CAP for 15, 30, 45, and 60 s and evaluated after 0, 1, 4, 7, and 11 days of storage at 4 °C and 90% humidity for the following quality parameters: total aerobic counts, yeast and molds, texture, color, soluble solids, pH, and titratable acidity. CAP treatments of 30 s and over demonstrated significant reductions in pathogens under both the resistant strain and pH conditions. Sixty-second CAP achieved a 0.54 Log CFU g⁻¹ reduction in L. monocytogenes (pH 5.5) and 0.28 Log CFU g⁻¹ for L. monocytogenes (pH 6.0). Yeast and mold counts on day 0 showed statistically significant reductions after 30, 45, and 60 s CAP with an average 2.34 Log CFU g⁻¹ reduction when compared to non-CAP treated samples. Quality parameters did not show major significant differences among CAP treatments during shelf life. CAP is an effective antimicrobial treatment that does not significantly affect fruit quality. Full article

Blueberries have high nutritional and economic value and are easy to cultivate, so they are common fruit crops in China. There is a high demand for blueberry in domestic and foreign markets, and various technologies have been used to extend the supply cycle of blueberry to about 7 months. However, blueberry grows in clusters, and a cluster of fruits generally contains fruits of different degrees of maturity, which leads to low efficiency in manually picking mature fruits, and at the same time wastes a lot of manpower and material resources. Therefore, in order to improve picking efficiency, it is necessary to adopt an automated harvesting mode. However, an accurate maturity detection model can provide a prerequisite for automated harvesting technology. Therefore, this paper proposes a blueberry ripeness detection model based on enhanced detail feature and content-aware reassembly. First of all, this paper designs an EDFM (Enhanced Detail Feature Module) that improves the ability of detail feature extraction so that the model focuses on important features such as blueberry color and texture, which improves the model’s ability to extract blueberry features. Second, by adding the RFB (Receptive Field Block) module to the model, the lack of the model in terms of receptive field can be improved, and the calculation amount of the model can be reduced at the same time. Then, by using the Space-to-depth operation to redesign the MP (MaxPool) module, a new MP-S (MaxPool–Space to depth) module is obtained, which can effectively learn more feature information. Finally, an efficient upsampling method, the CARAFE (Content-Aware Reassembly of Features) module, is used, which can aggregate contextual information within a larger receptive field to improve the detection performance of the model. In order to verify the effectiveness of the method proposed in this paper, experiments were carried out on the self-made dataset “Blueberry—Five Datasets” which consists of data on five different maturity levels of blueberry with a total of 10,000 images. Experimental results show that the mAP (mean average precision) of the proposed network reaches 80.7%, which is 3.2% higher than that of the original network, and has better performance than other existing target detection network models. The proposed model can meet the needs of automatic blueberry picking. Full article

Blueberries are widely planted because of their rich nutritional value. Due to the problems of dense adhesion and serious occlusion of blueberries during the growth process, the development of automatic blueberry picking has been seriously hindered. Therefore, using deep learning technology to achieve rapid and accurate positioning of blueberries in the case of dense adhesion and serious occlusion is one of the key technologies to achieve the automatic picking of blueberries. To improve the positioning accuracy, this paper designs a blueberry recognition model based on the improved YOLOv5. Firstly, the blueberry dataset is constructed. On this basis, we design a new attention module, NCBAM, to improve the ability of the backbone network to extract blueberry features. Secondly, the small target detection layer is added to improve the multi-scale recognition ability of blueberries. Finally, the C3Ghost module is introduced into the backbone network, which reduces the number of model parameters while ensuring the accuracy, thereby reducing the complexity of the model to a certain extent. In order to verify the effectiveness of the model, this paper conducts experiments on the self-made blueberry dataset, and the mAP is 83.2%, which is 2.4% higher than the original network. It proves that the proposed method is beneficial to improve the blueberry recognition accuracy of the model. Full article

Blueberries, which are recognized by their colored fruits and exquisite flavor and taste, are a great source of bioactive substances with potential functional properties. For the purpose of this study, the blueberry cultivars ‘Duke’, ‘Chandler’ and ‘Bluecrop’ were picked at four different times. The aim of the study was to compare the cultivars and determine the best time for picking fruits for table consumption and to produce berries that can be used as functional foods with elevated levels of bioactive compounds. According to principal component analysis (PCA), the most influential traits for distinguishing different times of harvest in the ‘Duke’ cultivar were sorbitol, glucose, sucrose, and turanose; for the cultivar ‘Chandler’, they were caffeic acid, aesculetin, and quercetin; for the ‘Bluecrop’, they were fructose, maltose, radical scavenging activity, and quercetin. Blueberry fruits aimed for table consumption were those harvested in the first two pickings of the cultivar ‘Duke’, in the first and third of the ‘Bluecrop’, and in the third picking time of the cultivar ‘Chandler’, due to the highest fruit size and very high level of sugar (mostly glucose and fructose). ‘Duke’ berries from the second and third harvest (high level of total phenolic content, radical scavenging activity, total anthocyanins, aesculin, quercetin, and isorhamnetin), ‘Chandler’ from the first and third (the highest p-hydroxybenzoic acid, aesculetin, caffeic acid, phloridzin, kaempferol, kaempferol 3-O-glucoside, quercetin 3-O-rhamnoside, rutin, and quercetin) and ‘Bluecrop’ from the third harvest (highest level of total phenolics, radical scavenging activity, quercetin, rutin, quercetin 3-O-glucoside, kaempferol, quercetin 3-O-rhamnoside, kaempferol 3-O-glucoside, and isorhamnetin) had the highest levels of health-promoting compounds. Full article

This study reports some preliminary results on mechanical blueberry harvesting for the fresh market of cv. Cargo^® in the Piedmont region (northwest Italy). The investigated area is one of the most productive areas of Italy, which specializes in fresh blueberry production. The automatization of harvesting operations could represent a competitive advantage for the area’s blueberry supply chain but could limit the quality of fresh-picked berries. A prototype machine and a commercial harvester (Easy Harvester^®) were compared with manual picking, considering the harvesting efficiency, labor productivity, harvesting cost and farm rentability. In this context, the labor cost for manual harvesting exceeds EUR 2.00 per kg of saleable product. The prototype allowed a 39% cost reduction, and the Easy Harvester^® reduced it by about half. Nevertheless, these positive performances do not consider the reduction in the net sale price of EUR 0.40 due to the selection costs in the warehouse. In this study, we highlight that the transition to mechanical harvesting requires the transformation of several farming and packhouse operations, such as new crop varieties, field configurations and cultivation techniques. However, a possible technical improvement of the Easy Harvester^® could represent an opportunity for Italian farms in the planning of berry production and marketing, involving all of the supply chain actors. Further research on the use of mechanization in the sector must continue and be supported. Full article

Mechanical harvesters with small box and semi-automated bin handling systems are increasingly being used for harvesting wild blueberries in Eastern Canada, and Northeastern, USA. However, their field capacity and performance have not been quantified and compared. Important measures of field capacity and efficiency for a traditional mechanical harvester were compared with a novel semi-automatic bin handling harvester. Data were obtained from on-farm field trials conducted at four sites in Nova Scotia, Canada in 2017 and 2018. Both harvesters had double head configurations, along with other similar engineering configurations: (i) 0.66 m picking reels; (ii) 16 picker bars per head and 65 teeth per bar; (iii) 1.72 m picking width; (iv) 21 rpm head speed; and (v) 0.31 ms⁻¹ ground speed. Each harvester was operated for 120 min and data such as berry harvesting time and box handling time were recorded, with six replications during each year. Statistical methods were used to compare the harvest efficiency of the two mechanical harvesters. Harvest time efficiency was significantly higher for the semi-automatic bin handling technology than for the small box handling technology both in 2017 (p < 0.001), and 2018 (p < 0.001). Weed coverage did not have a significant effect of harvest time in either 2017 (p = 0.694) or 2018 (p = 0.765), though it did significantly affect yield in both 2017 (p = 0.011) and 2018 (p = 0.045). The findings provide useful insights for decision-makers contemplating the choice of harvesting technology to sustain profits from wild blueberry production. Full article

Mechanical harvesting of wild blueberries remains the most cost-effective means for harvesting the crop. Harvesting of wild blueberries is heavily reliant on operator skill and full automation of the harvester will rely on precise and accurate determination of the picking reel’s height. This study looked at developing a control system which would provide feedback on harvester picking reel height on up to five harvester heads. Additionally, the control system looked at implementing three quality of life improvements for operators, operating multiple heads until the point when full automation is achieved. These three functions were a tandem movement function, a baseline function, and a set-to-one function. Each of these functions were evaluated for their precision and accuracy and returned absolute mean discrepancies of 3.10, 2.20, and 2.50 mm respectively. Both electric and hydraulic actuators were evaluated for their effectiveness in this system however, the electric actuator was simply too slow to be deemed viable for the commercial harvesters. To achieve the full 203.2 mm stroke required by the harvester head, the electric actuator required 13.96 s while the hydraulic actuator required only 2.30 s under the same load. Full article

Fresh market southern highbush blueberries are typically hand-harvested which requires an extensive labor force over a relative short period of time. With rising production costs and labor availability issues, interest in mechanical harvesting options is increasing. In 2017, an over-the-row (OTR) harvester was modified to reduce purchase cost while making hand labor more efficient. The picking heads were removed and dual worker stations were added on each side of the unit. Handheld olive shakers were suspended at each station. Experimental catch plates were installed on one side of the OTR harvester and soft, inclined surfaces over the rigid conveyors on both sides. ‘Meadowlark’ and ‘Farthing’ blueberries were harvested with this system and compared to those manually harvested by a commercial harvest crew. Samples from each harvest method were then commercially cooled and mechanically harvested fruit were commercially packed to determine packout data. Fruit firmness, bruise severity and composition were determined after one day at room temperature (22 °C) and after seven and fourteen days of storage at 1 °C. Average packout was very high for mechanically harvested fruit, 87% for ‘Meadowlark’ and 91% for ‘Farthing’. Initial firmness of both cultivars was lower for mechanically harvested fruit (208 g/mm) than hand-harvested fruit (243 g/mm). Fruit from the three treatments softened during storage, and although ‘Meadowlark’ remained firmer than ‘Farthing’ during storage, there were no differences due to catch surfaces. Hand-harvested fruit had no severe bruising (>20% of cut surface area) at harvest, increasing to 2% after seven days, while mechanically harvested fruit from both fruit collection surfaces had 3% initial severe bruising that increased to 22% during storage. ‘Farthing’ had slightly higher soluble solids content and significantly higher total titratable acidity compared to ‘Meadowlark’. Additional modifications must be made to the next-generation OTR harvester to further reduce blueberry harvest and handling impacts. Full article

Improved blueberry mechanical harvesting (MH) equipment that maintains fresh market quality are needed due to rising costs and decreasing availability of laborers for harvesting by hand. In 2017, a modified over-the-row (OTR) blueberry harvester with experimental catch surfaces and plates designed to reduce fruit bruising was evaluated. The catch surfaces were made of neoprene (soft catch surface; SCS) or canvas (hard catch surface; HCS) and compared to hand-picked fruit (control). Early- and early/mid-season ‘Duke’ and ‘Draper’, respectively, were evaluated in Oregon, while late-season ‘Elliott’ and ‘Aurora’ were evaluated in Washington. Harvested berries were run through commercial packing lines with fresh pack out recorded and bruise incidence or fresh fruit quality evaluated during various lengths of cold storage. The fresh pack out for ‘Duke’ and ‘Draper’ were 83.5% and 73.2%, respectively, and no difference was noted between SCS and HCS. ‘Duke’ fruit firmness was highest among MH berries with SCS, but firmness decreased in storage after one week. Firmness was highest among hand harvested ‘Draper’ followed by MH with SCS. For ‘Elliott’ and ‘Aurora’, fruit firmness was the same across harvesting methods. ‘Draper’ exhibited more bruising than ‘Duke’, but bruise ratings and the incidence of bruising at ≤10% and ≤20% were similar between hand and MH ‘Draper’ with SCS after 24 h of harvest. ‘Aurora’ berries had similar bruise ratings after 24 h between hand harvesting and MH with SCS, while ‘Elliott’ showed more bruise damage by MH with both SCS and HCS than hand harvested fruit. Although our studies showed slightly lower fresh market blueberry pack outs, loss of firmness, and increased bruise damage in fruit harvested by the experimental MH system compared to hand harvested fruit, higher quality was achieved using SCS compared to HCS. We demonstrated that improved fresh market quality in northern highbush blueberry is achievable by using modified OTR harvesters with SCS and fruit removal by either hand-held pneumatic shakers or rotary drum shakers. Full article

Work-related musculoskeletal disorders (MSDs) accounted for 32% of days-away-from-work cases in private industry in 2016. Several factors have been associated with MSDs, such as repetitive motion, excessive force, awkward and/or sustained postures, and prolonged sitting and standing, all of which are required in farm workers’ labor. While numerous epidemiological studies on the prevention of MSDs in agriculture have been conducted, an ergonomics evaluation of blueberry harvesting has not yet been systematically performed. The purpose of this study was to investigate the risk factors of MSDs for several types of blueberry harvesting (hand harvesting, semi-mechanical harvesting with hand-held shakers, and over-the-row machines) in terms of workers’ postural loads and self-reported discomfort using ergonomics intervention techniques. Five field studies in the western region of the United States between 2017 and 2018 were conducted using the Borg CR10 scale, electromyography (EMG), Rapid Upper Limb Assessment (RULA), the Cumulative Trauma Disorders (CTD) index, and the NIOSH (National Institute for Occupational Safety and Health) lifting equation. In evaluating the workloads of picking and moving blueberries by hand, semi-mechanical harvesting with hand-held shakers, and completely mechanized harvesting, only EMG and the NIOSH lifting equation were used, as labor for this system is limited to loading empty lugs and unloading full lugs. Based on the results, we conclude that working on the fully mechanized harvester would be the best approach to minimizing worker loading and fatigue. This is because the total component ratio of postures in hand harvesting with a RULA score equal to or greater than 5 was 69%, indicating that more than half of the postures were high risk for shoulder pain. For the semi-mechanical harvesting, the biggest problem with the shakers is the vibration, which can cause fatigue and various risks to workers, especially in the upper limbs. However, it would be challenging for small- and medium-sized blueberry farms to purchase automated harvesters due to their high cost. Thus, collaborative efforts among health and safety professionals, engineers, social scientists, and ergonomists are needed to provide effective ergonomic interventions. Full article