•In orchards, acquiring the maximum data possible is essential for growers.•Propose a deep learning model to help growers measure and acquire crop data.•Construct a large apple tree detection, ...segmentation, and measurement dataset.•The proposed model achieved a high count accuracy of 94.1% and a high segmentation accuracy of 97.1%.•The average measurement accuracy for the proposed model is over 92%.
Manual measurement and visual inspection is a common practice for acquiring crop data in orchards and is a labor-intensive, time-consuming, and costly task. Accurate and rapid acquisition of crop data is vital for monitoring the dynamics of tree growth and optimizing farm management. In this work, we present a technique for orchard data acquisition and analysis that uses remote imagery acquired from unmanned aerial vehicles (UAVs) combined with deep learning convolutional neural networks to automatically detect and segment individual trees and measure the crown width, perimeter, and crown projection area of apple trees. By using an UAV platform, 50 high-resolution images of apple trees were collected from an orchard during dormancy (bare branches), and then each apple tree was detected by using a Faster R-CNN object detector. Based on these results, each tree was segmented by using a U-Net deep learning network. After convex tree boundaries were extracted from the semantic segmentation results by using an efficient pruning strategy, the crown parameters were automatically calculated, and the accuracy was compared with that obtained by manual delineation. The results show that the proposed remote sensing technique can be used to detect and count apple trees with precision and recall of 91.1% and 94.1%, respectively, segment their branches with an overall accuracy of 97.1%, and estimate crown parameter with an overall accuracy exceeding 92%. We conclude that this method not only saves labor by avoiding field measurements but also allows growers to dynamically monitor the growth of orchard trees.
The current study focused on the branches Cherry Gala variety of apple in Azrouregion.The objective is to characterize the length and the insertion distance of different structures of apple trees ...based on nondestructive and exhaustive measurements of the branches. In total 2982 branches were identified and measured in the two stages (stage 1, stage 2) on different 6 levels present at each stage known as A, B, C, D, E, and F. The length of the branches in each stage and level was classified into 1 to 5 homogeneous classes. The length of vegetative branch variation was not significant between stages 1 and 2 but the insertion distance was significant for level D and not for levels A, B, and C. For five different length structures bearing fruit, the variation was significantly different for B, C, and D levels between stages 1 and 2 (length of fruiting spurs/ Dard/ Bourse; insertion distance of Dard and Bourse). The analysis showed that the length of the branches decreases while passing from one first stage to the second. The average branch length for stage 1 is 4,84 cm against 4,05 cm for stage 2. In each stage, the length decreases progressively while passing from one level to a higher level, except for level E of stage 1 and levels E and F of stage 2.
•The Rainwater-Collecting and Deep-Infiltration (RWCI) technology was introduced.•The RWCI technology increased soil water content in 40–80cm of apple orchards.•The RWCI technology reduced total root ...dry weight in deep soils.
Water is a key factor for the sustainable development of rain-fed apple orchards on the Loess Plateau. This study was conducted to investigate the spatial distribution of soil moisture and roots in the soil profiles of an apple orchard employing a Rainwater Collection and Infiltration (RWCI) system on the Loess Plateau of China. The results showed that: 1) There was a low soil moisture content (SMC) zone between depths of 40 and 80cm and the RWCI treatment significantly increased SMC in these depths; 2) The RCWI system increased SMC in the 20–140cm depth interval and depths greater than 140cm when the RCWI pit depths were 60cm (RWCI60) and 80cm (RWCI80), respectively; 3) The total dry root density under the control (CK), RWCI60 and RWCI80 treatments was 372.12, 594.76 and 491.82g/m3, respectively. The mass percentage root distributions in the 0–100, 100–200 and 200–300cm depths were: CK (69.88%, 13.74% and 16.38%); RWCI60 (70.35%, 24.08% and 5.58%); and RWCI80 (46.54%, 15.04% and 38.42%), respectively. We found that root mass was concentrated close to the wetted soil depths affected by the RWCI treatment. The RWCI system apparently reduced drought stress through increased water and nutrients uptake because of increased subsoil SMC in the RWCI treatments. This study of the RWCI systems and the effects of the RCWI pit depth are important with respect to increasing the efficient use of rainwater, reducing drought stress and demonstrating the effectiveness of RWCI water saving technology in the rain-fed orchards in the Loess Plateau, China.
Melatonin regulates the seasonal reproduction in photoperiodic sensitive animals. Its function in plants reproduction has not been extensively studied. In the current study, the effects of melatonin ...on the apple tree flowering have been systematically investigated. For consecutive 2‐year monitoring, it was found that the flowering was always associated with the drop of melatonin level in apple tree. Melatonin application before flowering postponed apple tree flowering with a dose‐dependent manner. The increased melatonin levels at a suitable range also resulted in more flowering. The data indicated that similar to the animals, the melatonin also serves as the signal of the environmental light to regulate the plant reproduction. It was mainly the blue and far‐red light to regulate the gene expression of melatonin synthetic enzymes and melatonin production in plants. The seasonal alterations of the blue and far‐red lights coordinated well with the changes of the melatonin levels and led to decreased melatonin level before flowering. The mechanism studies showed that melatonin per se inhibits all the four flowering pathways in apple. The results not only provide the basic knowledge for melatonin research, but also uncover melatonin as a chemical message of light signal to mediate plant reproduction. This information can be potentially used to control flowering period and prolong the harvest time, helpfully to open a new avenue for increasing crop yield by melatonin application.
Soil water status and fine root distribution is the basis of implementing water management in semiarid rain-fed orchards. Exploitation of rainwater is an effective avenue for alleviating water ...scarcity in semiarid regions since ground water is generally unavailable there. Through the method of space-for-time substitution, we investigated the soil moisture and root distribution along a range of stand ages (6, 9, 12, 18 and 21 years) in rain-fed apple (Malus pumila Mill) orchards and the effects of rainwater collection and infiltration systems (RWCI) on root-zone soil water and fine root distributions in matures apple orchards (21 years) in the semiarid Loess Plateau of China. The results showed that the mean soil moisture content (SMC) in the shallow layer (<2 m) decreased with apple tree age (6 to 18 years); the deep SMC (>2 m) was higher than shallow soil layers (<2 m) in most cases and the SMC increased with depth in stands of all ages. Fine roots (<2 mm diameter) showed an obvious trend of extending deeper with apple tree age - nearly 8 m in a 12-yr-old apple plantation and >8 m in plantations >12 years old. Dry root weight density (RWD) decreased sharply with depth, but densities at each depth were greater in older stands. The RWCI system significantly increased SMCs from the surface down to the maximum rainfall infiltration depth (MRID) (2 m depth) (P < 0.05), especially in the 0.2–1 m soil layer. Further, we found that apple tree water requirements could be sustainably met when RWCI system and a low-volume of irrigation water was applied. The distribution of root system was greatly affected by the RWCI system, which led to higher root densities close to the wetted area in the shallow soil layers (2 m soil depth) under RWCI system, down to a depth of 3 m in the soil. Overall, the application of RWCI system could be an effective water management strategy for providing sustainable water resources for semiarid orchards.
•The soil moisture and root distribution in stand ages' orchards were investigated.•The space-for-time substitution design was used to carry out our investigation.•Fine roots showed an obvious trend of extending deeper with apple trees' stand age.•The effects of RWCI on soil water and root distributions were investigated.
•Established a new dataset and proposed an innovative data augmentation algorithm.•Integrating diverse networks effectively segments multi-scale and irregular targets.•Highlighting the synergistic ...effect of different networks on the segmentation performance.•Superior performance with 82.58 % mP, 74.58 % mR, 77.94 % mDice, 64.91 % mIoU, and 79.75 % mAP.
The bud stage is a crucial period in the growth and development of apple trees. Accurate detections of the physiological changes in the organs (branches, buds, leaves, and the connecting parts between buds and branches) are essential for the scientific management of orchards. In the field of intelligent orchard management, image segmentation is a fundamental method for obtaining the phenotypes of fruit tree organs, making it especially critical. To address this, we have created a dataset for apple tree organ segmentation during the bud stage and have incorporated several advanced convolutional network modules (ConvNeXt V2, Multi-scale Extended Attention Module (MSDA), Dynamic Snake Convolution (DSConv)) to enhance YOLOv8 and improve the accuracy of organ segmentation in complex natural environments. In the backbone network, we have integrated ConvNeXt V2 and MSDA modules to increase the extraction of contextual information and improve the network’s ability to recognize multi-scale and multi-shaped targets. Additionally, we have embedded DSConv in the network’s head and utilized deformable convolution to enable adaptive sampling of the feature map, capturing a wider range of context information and improving the local feature processing capability for objects of varying sizes, ultimately leading to improved segmentation accuracy. Our models significantly outperform existing models, achieving 82.58 % mean Precision (mP), 74.58 % mean Recall (mR), 77.94 % mean Dice(mDice), 64.91 % mean IoU(mIoU), and 79.75 % mean Average Precision (mAP). Ablation studies confirm the contributions of each module to intelligent orchard management and suggest potential benefits for precise agricultural decision-making and operations.
•‘Red Jonaprince’ apple trees supplied with Ca-nitrate were sprayed with Mo•Pre- and bloom sprays of Mo improved seed set•Pre- and bloom sprays of Mo enhanced fruit Ca status and retarded their ...ripening•The postbloom spray of Mo increased nitrate reductase activity and chlorophyll in leaves•Leaves of Mo-sprayed trees after bloom had improved N, Mg, and Fe status
The aim of the study was to examine the responses of mature ‘Red Jonaprince’ apple (Malus domestica Borkh.) trees/M.26 to molybdenum (Mo) sprays. The experiment was conducted in Central Poland in 2017-2018 at high-density orchard growing on strongly acidified coarse-textured soil with low Mo availability. Trees were sprayed with Mo in the form of sodium molybdate before bloom (at pink bud), at bloom (when 5-20% of flowers were open) or after bloom (14 days after petal fall). In each spray, the Mo rate was 286 g ha-1. Trees that were not treated with Mo served as the control. Except for manganese (Mn), the concentrations of the essential macro- and micronutrients in the summer leaves of the control trees were within optimum/sufficient ranges. The summer leaf Mn concentrations were high but not toxic. Prebloom and bloom Mo sprays increased the activity of nitrate reductase (NR) and SPAD chlorophylometer values for leaves at 21 days after full bloom (DAFB), Mo concentrations in flowers and seeds, the number of seeds in an apple, fruit Ca concentration (on average by 23% for both Mo spray treatments), and the Streif index values of the fruit at harvest. The postbloom Mo spray increased the summer leaf concentrations of nitrogen, magnesium, iron and Mo, and the NR activity and SPAD values of leaves at 21 DAFB and at harvest. No Mo spray affected tree vigour, fruit set and yield, mean fruit weight, skin russeting and blushing, and the soluble solids concentration and flesh firmness of apples at harvest. It is concluded that when the apple flower Mo concentration, at least in triploid varieties, is as low as 1.5 mg kg-1 DW, a prebloom or bloom Mo spray is recommended to improve Ca-related apple quality.
This study aimed to determine the effect of irrigation amount ( W ), nitrogen ( N ), potassium ( K ), and zinc ( Zn ) on the net photosynthetic rate ( Pn ) of closely planted apple trees on dwarf ...rootstocks in arid areas of Xinjiang. Taking the “Royal Gala” apple as the experimental material, a mathematical model for Pn was established using the principle of four-factor five-level quadratic regression with a general rotation combination design. The results show that: (1) The regression equations reached significant levels (F = 37.06 > F0.01(11.11) = 4.54). (2) The effect of W , N , K , Zn on Pn is significant with relative importance W > N > Zn > K . (3) The results of single factor analysis showed that with an increase in W, N, K, and Zn, Pn exhibits an n-shaped parabolic response. (4) The positive coupling between W and N is significant, and the positive coupling between W and Zn is also significant. (5) Analysis of the interaction between sets of three factors revealed that W , N, and Zn could be combined to best effect, with the maximum value reaching 12.77 μ mol· m −2· s −1 . Compared with W×K×Zn and W×N×K , the combination of W×N×Zn reduces W by 9.2% and 6.3%, respectively, which indicates its suitability for use in the dry and water deficient planting environment in Xinjiang. (6) Within the 95% confidence level, when W is 258–294.75 mm, N is 33.44–39.51 kg/hm2 , K is 53.82–69.39 kg/hm2, and Zn is 6.46–7.84 kg/hm2 , the net photosynthetic rate reaches 11 μ mol· m −2· s −1 .
