Accurate in-field non-destructive fruit sizing is a fundamental requirement for yield estimation and harvest logistics, yet it remains challenging due to complex fruit morphology and real-time processing constraints. This study investigates the allometric relationships of Fuji (cv. FujiKO) apples and their impact on computer vision performance. Morphological analysis reveals that fruits are generally wider than they are tall, with fruit height and minimum equatorial diameter acting as the primary sources of size underestimation. The results identify an intrinsic morphological uncertainty—or "geometric noise"—ranging from 8.8% to 10.5% of the reference equatorial diameter, below which further sizing precision is limited by fruit geometry rather than sensor/algorithm accuracy. To address this issue, a Mean Thresholded Absolute Error (MTAE) is proposed as an evaluation metric that accounts for fruit geometric variability when assessing remote fruit sizing computer vision algorithms.Evaluation of three algorithm versions demonstrates that, while image resizing and local entropy improve visual fruit fitting, they impose significant computational overhead. An optimized pipeline achieves a processing speed of 1.0 ± 0.7 ms per fruit detected on the CPU, enabling a near-real-time analysis of up to 63 fruits per frame. Despite intrinsic geometric variability, all algorithms achieve an MTAE below 0.5 mm, confirming that a high-throughput, statistically representative sampling approach is more advantageous than pursuing sub-millimeter precision on individual, randomly oriented fruits.

Yellow flesh kiwi fruit production normally follows protocols based on the green species, A. deliciosa, often resulting in low yields, attributable to small sized fruit, meaning A. chinensis seems more susceptible to water limitations. Understanding the species physiology and fruit vascular flows may help determine this crop’s evapotranspiration needs, to efficiently obtain satisfactory harvests. The presented work results from a 3-year trial (2019-2020-2021), where control irrigation vines were compared with deficit-irrigated and over-irrigated vines. Midday physiology, including plant water relations, leaf gas exchanges and fruit vascular flows were analysed, along with harvest parameters and dry matter content. Irrigation treatments influenced the vines’ responses only when soil water content was below certain levels, reflecting sensitivity of the crop to water changes in the soil. Although no significant differences were found in harvest parameters, dry matter content was higher for the less irrigated fruit. The less irrigated treatment performed less better, than the control and the over-irrigated, especially when water supply did not fulfil fruit transpiration. This occurred during the berry development phase (around 1 month after full bloom), a critical period during which the fruit has very high transpiration rates, which passively call photosynthates (phloem inflow) to provide energy for cell division. Fruit transpiration appears to influence phloem inflow during most of the season, even until 1 month before harvest, however the initial phases of fruit development and growth are pivotal for final yield. Vascular flows allowed to unveil a typical simplasmic behaviour in the early stages of berry development, meaning the microenvironment is intensely influencing fruit behaviour. Irrigation must respond to the needs of young fruit, taking into account soil water content and the phenological phase. The use of sensors, plant based and environmental, is an important technique for determining the necessary water volumes for yellow kiwi fruit.

Yellow-fleshed kiwifruit production guidelines normally follow protocols based on the green-fleshed species, A. deliciosa, often resulting in low yields, reduced dry matter and small fruit size, as A. chinensis seems more susceptible to water limitations. Clarifying this species fruit vascular flows may help improve yields, and fruit size, by satisfactorily meeting its irrigation needs. This 3-year trial (2019-2021) compared 100 % with deficitirrigated and over-irrigated vines. Plant water relations, leaf gas exchanges and fruit vascular flows were measured at midday during the season, and fruit were harvested and assessed for dry matter content. Irrigation influenced vine responses only when soil water content dropped below 25-30 %, reflecting the sensitivity of this crop to soil water changes. Although no differences were found for yield, dry matter content was higher for the least irrigated fruit, but overall vine performance in this treatment was limited throughout the study, in particular during the rapid growth phase (e.g., ca 1 month after full bloom), when very high fruit transpiration rates energize passive phloem unloading, to support fruit parenchyma cell division and expansion. Unlike A. deliciosa, A. chinensis appears to maintain fruit transpiration and associated phloem unloading later into the season. The symplasmic unloading in the early stages of berry development reveals a strong effect on fruit development by its microenvironment, thus irrigation must be carefully gauged to the needs of developing fruit, accounting for soil water content and the phenological phase. Plant and environmental sensors are quite important for determining the necessary water volumes for yellow-fleshed kiwifruit.

Calcium (Ca) is one of the most important nutrients involved in fruit quality and storability; therefore, its application in fruit trees is often used in pre- and post-harvest. The aims of this study were to manipulate soil Ca, K, and N availability, photosynthetic active radiation, and fruit transpiration rate to understand their implication on fruit Ca accumulation on green-flesh kiwifruit grown in calcareous soil. Our results show that Ca partitioning into the fruit is not affected by the applications of Ca, K, and N, as well as the increase of photosynthetic active radiation. However, the presence of reflective films reduced fruit firmness and increased soluble solid content at harvest and during cold storage, thus enhancing fruit quality. Fruit calcium accumulation is decreased by the reduction of fruit transpiration rate; however, it has the possibility to recover, even close to fruit harvest, when the fruit transpiration is restored. The presence of bags reduced fruit weight from 84 to 63 g even though bags were removed. Our data provide evidence of the inefficiency of calcium fertilization in kiwifruit in calcareous soils and demonstrate the extension of calcium transportation into the fruit, which seems to occur during the entire growing season.

In recent years, reduced summer precipitation frequencies related to climate change have raised the probability of water scarcity, even in the Po Valley of Italy, thus requiring an optimization of the irrigation management for walnut cultivation which has become very present in the area. The aim of the present study was to evaluate, during four consecutive seasons (2018–2021), the physiological (stem water potential Ψw, leaf photosynthesis A and stomatal conductance gs), yield (nut weight, shelled yield, kernel colour) and water use efficiency (WUE) responses of walnut trees to different irrigation levels (100

Fruit size is crucial for growers as it influences consumer willingness to buy and the price of the fruit. Fruit size and growth along the seasons are two parameters that can lead to more precise orchard management favoring production sustainability. In this study, a Python-based computer vision system (CVS) for sizing apples directly on the tree was developed to ease fruit sizing tasks. The system is made of a consumer-grade depth camera and was tested at two distances among 17 timings throughout the season, in a Fuji apple orchard. The CVS exploited a specifically trained YOLOv5 detection algorithm, a circle detection algorithm, and a trigonometric approach based on depth information to size the fruits. Comparisons with standard-trained YOLOv5 models and with spherical objects were carried out. The algorithm showed good fruit detection and circle detection performance, with a sizing rate of 92

Apricot (Prunus armeniaca L.) is among the most cultivated fruit worldwide, appreciated for its taste and appearance. Previous studies on the application of regulated deficit irrigation (RDI) strategies have demonstrated how this species can successfully respond to water stress, also increasing some qualitative traits. This work aims to investigate the physiological response of plants in absence of irrigation water during the last weeks before harvest. A total of 12 plants were considered in this study, 6 subjected to water stress and 6 control, divided in two blocks per treatment. In the "stress" treatment the irrigation was withheld on June 28 (108 days after full bloom, DAFB) and remained so until harvest, at 131 DAFB, while control trees were irrigated according to ETc. Stem and leaf water potentials as well as leaf gas exchange were measured at 108 and at 131 DAFB and fruit vascular flows data were registered every 15 min at 115 and 128 DAFB. Leaf water potential did not show differences between the two treatments, while stem water potential reached statistically significant lower values at harvest (131 DAFB). In the plants subjected to water stress, the photosynthetic rate was not affected, while transpiration and stomatal conductance presented statistically significant lower values at harvest. Control plants were able to transpire more, with higher xylem and phloem inflow. However, fruit quality traits were similar in the two treatments, with the only difference in the chlorophyll degradation level, meaning that stress conditions slightly delayed fruit ripening. These results lead to the conclusion that apricot can afford important water shortages without negative impacts on productivity, while increasing its water use efficiency.

The application of exclusion nets is gaining interest for apple modern production, since it can limit the use of pesticides. From the physiological point of view, information related to final fruit harvest is missing. This two-year study compared a classical anti-hail net (A, 20% shading) with an exclusion net (E, 40% shading), and their effects on Gala apple trees, under two irrigation treatments each. Physiology, production and quality parameters were tested. In both years, midday stem water potentials and leaf gas exchanges were unaffected. The higher shading properties of E created a more favourable microclimate for the trees, allowing them to improve marketable fruit weight, compared to the A net. Fruit quality was influenced by different shading and water treatments, visual red color especially in 2021; however, the other quality traits did not have similar trends over the two years. A remarkable commercial impact was gained, since higher shading provided by exclusion netting lowered water requirements. Fruit productivity was sustained or, even elevated, under water limitations when exclusion netting was used. These results are promising in the view of the increasing demand for sustainability in fruit production.

Chestnut crop is regaining its fame worldwide with powerful investment perspectives. Unluckily the climate change effects are posing high threat to its cultivation with less available resources and increased production cost both in traditional and specialized orchards. Additionally, the chestnut physiological knowledge is still limited, especially as concern the burr development (i.e., the economical production target) and its relationship with the environmental parameters. The aim of the present study was to evaluate the seasonal, daily, and hourly burr growth pattern associated to environmental parameters for improving physiological knowledge on this species. The study was carried out in a traditional rainfed sweet chestnut orchard located in the Tuscan-Emilian Apennines (Monterenzio, Italy). The chestnut burr growth was measured, along the entire season, both with a digital calliper and through the use of plant-based sensors (fruit-gauges) that permitted to measure, in real-time, the burr growth pattern. Environmental data were recorded by a weather station placed in the middle of the orchard. Results evidenced a higher burr growth rate, in the last part of the season (from middle-end of August to full fall) while the daily growing pattern was characterized by increased oscillation, along the season, of night-swelling and daily-shrinkage. The night-swelling was found to be influenced by high nocturnal air relative humidity while the daily-shrinkage was influenced by the higher wind speed, solar radiation and vapour pressure deficit. Thus, the burr daily net growth can be associated, depending on the phenological stages, to environmental parameters. Precipitation but especially the atmosphere humidity, in September and October, were the main external drivers of burr daily net growth. These results could be promising for the adoption of sustainable (e.g., late season grass mowing, sprinkler irrigation) and smart practices for improving chestnut management in both traditional and specialized orchards.

Disease control of fruit trees is still primarily based on pesticide applications. In the apple orchard many chemical treatments are applied to control apple scab (Venturia inaequalis), fire blight (Erwinia amylovora) and pests such as the codling moth (Cydia pomonella) and the brown marmorated stink bug (Halyomorpha halys). This research, carried out at the experimental farm of the University of Bologna, aimed at the development and optimization of an automated, fixed point, above canopy distribution system to reduce pesticide drift to non-target plants. The fixed spray system was installed along rows of trees planted both under anti-hail, and rainproof nets, thus allowing to also investigate the effect of the cover. The fixed point system was compared with the conventional sprayer (positive control) and untreated trees were used as negative control. The cultivars tested were 'Gala-Buckeye' and 'Pink Lady (R) Rosy Glow*'. The fixed point system reduced disease incidence, but with lower efficacy than the conventional sprayer. Efficacy was improved under the rain cover, which reduced scab incidence up to 22% in 'Gala' especially in early spring, compared to the hail net. Alternative products to chemical pesticides, such as thyme and oregano essential oils, sodium and ammonium bicarbonate, were tested against V. inaequalis and E. amylovora. The biological control agent, Clonostachys rosea was tested against V. inaequalis. All these methods reduced V. inaequalis growth, while only essential oils were effective against E. amylovora. These results suggest that fixed point pesticide application represents a promising alternative to conventional pesticide sprayers, particularly under rain covers. Finally, alternative, green compounds may be used in this system to further reduce pesticide impact on the environment and fruit residues.

Accurate, continuous and reliable data gathering and recording about crop growth and state of health, by means of a network of autonomous sensor nodes that require minimal management by the farmer will be essential in future Precision Agriculture. In this paper, a low-cost multi-channel sensor-node architecture is proposed for the distributed monitoring of fruit growth throughout the entire ripening season. The prototype presented is equipped with five independent sensing elements that can be attached each to a sample fruit at the beginning of the season and are capable of estimating the fruit diameter from the first formation up to the harvest. The sensor-node is provided with a LoRa transceiver for wireless communication with the decision making central, is energetically autonomous thanks to a dedicated energy harvester and an accurate design of power consumption, and each measuring channel provides sub-mm 9.0-ENOB effective resolution with a full-scale range of 12 cm. The accurate calibration procedure of the sensor-node and its elements is described in the paper, which allows for the compensation of temperature dispersion, noise and non-linearities. The prototype was tested on field in real application, in the framework of the research activity for next-generation Precision Farming performed at the experimental farm of the Department of Agricultural and Food Science of the University of Bologna, Cadriano, Italy.

Superficial scald (SS) is one of the main physiological disorders affecting postharvest of pears. Its onset is linked to oxidative processes. Antioxidant compounds such as ascorbic acid and phenolics could play a key role in preventing SS. Growing environment and fruit quality also have an influence on SS symptoms occurrence. The aim of this project is to understand the relationship between antioxidant activity, phenolic content, and development of SS in ‘Abate Fétel’ pear. Moreover, the effect on SS of fruit maturity at harvest was assessed using multivariate statistical approach. Data were collected in thirty orchards in the Emilia-Romagna region (Italy) in three seasons (2018, 2019 and 2020), and the fruit were stored in a regular atmosphere for 120 days. Antioxidant capacity was determined by 2,2-diphenyl-1-picrylhydrazy (DPPH) method and total phenol content by Folin-Ciocalteau colorimetric protocol. The results showed that 340 mg of ascorbate/100 g of FW and 300 mg of gallic ac./100 g of FW at least provide good protection against SS. Multivariate analysis indicated that pulp firmness and index of absorbance difference ( lAD ) seem to keep low the SS occurence, when at harvest are higher than 6.3 kg and 1.9, respesctively. In conclusion, it would be possible to build a forecasting model to control SS that considers pre-harvest data and content of antioxidants in different orchards, to improve the postharvest management of ‘Abate Fétel’.

The objective of this 2-year study (2017, 2019) was to evaluate the influence of photoselective nets on apple fruit growth, focusing on the initial fruit growth stages of "Pink Lady". Trees were subjected to four photoselective nets (Blue, Red, White and Yellow) and a standard black one (serving as Control), resulting in 5 light environments (LE), all shading at 20%. From 20 to 90 DAFB, 32 fruit and extension shoots, for each LE, were measured for a total of 11 times during the season. For each LE, fruit gauges were also installed to monitor fruit daily growth parameters, from 50 to 90 DAFB. At harvest, all fruit from each light environment were weighed and quality parameters were measured on 40 fruit per treatment. For each year, correlations were made to test the influence of LEs on final fruit weight and quality, and the impact that extension shoot growth had on fruit growth. In both years, white and control nets led to the production of fruits with higher weight compared to the other treatments (blue, red and yellow nets). In 2019, LEs did not appear to influence the relationship between shoot and fruit growth, probably due to the higher crop load and the massive pruning carried out the previous year. The higher vegetative outburst in this year might have unbalanced resources towards the shoots, than towards the fruit. Results show how weather and orchard management heavily influenced the trees responses. Yet, the consistent findings of final fruit weight, for both years, indicate that different wavelengths influence fruit and shoot behavior, even at early phenological stages. Hence, growers searching for bigger fruit should refer to white and black nets. Further studies approaching this technology can help improve apple production management and knowledge of the use of photoselective nets.

Fruit growth is a complex mechanism resulting from biochemical and biophysical events leading water and dry matter to accumulate in the fruit tissues. Understanding how fruits choose their growth strategies can help growers optimizing their resource management for a more sustainable production and a higher fruit quality. This paper compares the growth strategies adopted by different fruit crops, at different times during the season and relates their fruit surface conductance to key physiological parameters for fruit growth such as phloem and xylem inflows as well transpiration losses. Our results show how fruits capacity to transpire (determined by their surface conductance) is a key driver in determining the growth strategy adopted by a species and explains the inter-species variability existing among different crops. Indeed, fruits change their surface conductance depending on the species and the phenological stage. This has an impact on the fruit's ability to lose water due to transpiration, affecting fruit pressure potential and increasing the force with which the fruit is able to attract xylem and phloem flows, with a considerable impact on fruit growth rate.

In 2017, two multi-location apple rootstock trials were established at 16 sites in 12 European countries. The evaluations are performed by members of the EUFRIN (European Fruit Research Institute Network) Apple & Pear Variety & Rootstock Testing Working Group. Two separate trials were arranged, grouping rootstocks into dwarf and semi-dwarf rootstocks according to the expected vigour; 'Galaval' was used as scion cultivar. The trial of dwarf rootstocks includes 'G.11' and 'G.41' (US), 'EM_02', 'EM_03', 'EM_04', 'EM_05' and 'EM_06' (UK), '62-396-B10 (R)' (Russia), 'P 67' (Poland), 'PFR4' and 'PFR5' (New Zealand) and 'Cepiland-Pajam (R) 2' as control. The trial of semi-dwarf rootstocks includes 'G.202' and 'G.935' (US), 'PFR1' and 'PFR3' (New Zealand), 'EM_01' (UK) and 'G.11' as a control for both trials. Part of the rootstocks (from dwarf and semidwarf rootstock trials) was planted in replanting conditions to test their tolerance to apple replant disease. All test trees came from the same nursery, and a common standardised evaluation protocol was used. Based on preliminary results averaged across sites, dwarf rootstocks can be ranked in terms of vigour in the following order: 'EM_04' < 'EM_03', 'EM_05' < '62-396-B10 (R)', 'P 67', 'EM_02', 'G.11' < 'G.41', 'Cepiland-Pajam (R) 2' < 'EM_06', 'PFR4' < 'PFR5'. On average, semi-dwarf rootstocks can be ranked in terms of vigour in the following order: 'G11' < 'G.935', 'G.202' < 'PFR3', 'EM_01' < 'PFR1'. The highest cumulative yield in the young orchard was registered for trees on 'PFR5', 'PFR4', 'G.11', 'G.41', 'Cepiland-Pajam (R) 2' and 'EM_02', while the lowest production was found for trees on 'EM_04'. In the group of semi-dwarf rootstocks, the highest yield was on 'PFR3', 'G.935' and 'PFR1'. Rootstocks also had a significant effect on fruit weight and fruit quality parameters. Results from the young orchards revealed interactions between sites and rootstock, potentially leading to site-specific rootstock choice based on the combination of rootstock, soil conditions and climate.