Precision Agriculture最新文献

In-season crop growth and yield prediction at high spatial resolution are essential for informing decision-making for precise crop management, logistics and market planning in horticultural crop production. This research aimed to establish a plant-level cabbage yield prediction system by assimilating the leaf area index (LAI) estimated from UAV imagery and a segmentation model into a crop simulation model, the WOrld FOod STudies (WOFOST). The data assimilation approach was applied for one cultivar in five fields and for another cultivar in three fields to assess the yield prediction accuracy and robustness. The results showed that the root mean square error (RMSE) in the prediction of cabbage yield ranged from 1,314 to 2,532 kg ha^–1 (15.8–30.9% of the relative RMSE). Parameter optimisation via data assimilation revealed that the reduction factor in the gross assimilation rate was consistently attributed to a primary yield-limiting factor. This research further explored the effect of reducing the number of LAI observations on the data assimilation performance. The RMSE of yield was only 107 kg ha^–1 higher in the four LAI observations obtained from the early to mid-growing season than for the nine LAI observations over the entire growing season for cultivar ‘TCA 422’. These results highlighted the great possibility of assimilating UAV-derived LAI data into crop simulation models for plant-level cabbage yield prediction even with LAI observations only in the early and mid-growing seasons.

This study presents the development and the evaluation of a low-cost sensor-based system to optimize the management of surface irrigation at the field level. During a surface irrigation event, water flows according to the slope of the field and it is difficult and time-consuming to predict the optimal time when inflow should be stopped. In such systems, measurement tools are uncommon and those existing are far too complex and expensive to be used as decision support tools on small farms. This article presents the development of an Open Source system, based on low-cost technologies, Internet of Things and LoRaWAN network, that allows: (i) detection of water at the sensor location in the field, (ii) sending an alert by phone to the user and (iii) remote control of surface irrigation gates. The metrological characteristics of the system and its suitability were tested in real conditions during one irrigation season of hay fields in the Mediterranean region. The results highlighted the reliability of the low-cost sensor system for detecting water and transmitting information remotely, with a 100% success rate. Remote control of irrigation gates was successful in 89% of trials carried out in the field, and adjustments resulted in a 100% success rate. The savings in labour time for the farmer and in irrigation water volumes made possible by the use of this system, as well as the inevitable trade-offs between accessibility, reliability and robustness of new technologies for agriculture, are finally discussed.

Introduction

This study explores the integration of precision agriculture technologies (PATs) in rainfed cereal production within semi-arid regions.

Methods

utilizing the Veris 3100 sensor for apparent soil electrical conductivity (ECa) mapping, differentiated management zones (MZs) were established in experimental plots in Valsalada, NE Spain. Site-specific variable dose technology was applied for seed and fertilizer applications, tailoring inputs to distinct fertility levels within each MZ. Emphasizing nitrogen (N) management, the study evaluated the impact of variable-rate applications on crop growth, yield, nitrogen use efficiency (NUE), and economic returns. For the 2021/2022 and 2022/2023 seasons, seeding rates ranged from 350 to 450 grains/m², and basal fertilizer dosages varied between high and low levels. Additionally, the total nitrogen units were distributed differently between the two seasons, while maintaining a uniform topdressing fertilizer dose across all treatments.

Results

Results revealed a significant increase in yield in MZ 2 (higher fertility) compared to MZ 1 (lower fertility). NUE demonstrated notable improvement in MZ 2, emphasizing the effectiveness of variable-rate N applications. Economic returns, calculated as partial net income, showed a considerable advantage in MZ 2 over MZ 1, resulting in negative outcomes for low-fertility areas in several of the analyzed scenarios, and highlighting the financial benefits of tailored input management.

Conclusion

This research provides quantitative evidence supporting the viability and advantages of adopting PATs in rainfed cereal production. The study contributes valuable insights into optimizing input strategies, enhancing N management, and improving economic returns in semi-arid regions.

Olive groves, often characterized by complex topography and highly variable soils, present challenges for delineating irrigation management zones (MZs). This study addresses this issue by examining the relevance of apparent electrical conductivity (ECa), elevation (Z), topographic wetness index (TWI) and time-series of Sentinel-2 NDVI imagery for delimiting MZs for variable rate irrigation (VRI) in a 40-ha olive grove in southern Spain. Principal Component Analysis (PCA) was employed to disentangle olive and grass cover NDVI patterns. PC1 represented the olive tree development patten and showed little relationship with soil properties, while PC2 was associated with the grass cover growth pattern and considered a proxy for water storage-related soil properties that are relevant for irrigation scheduling. An alternative analysis using NDVI percentiles yielded similar results but favored PCA for distinguishing between grass cover and olive tree development patterns. Correlation between NDVI and ECa varied seasonally (r > 0.60), driven by the grass cover dynamics. To assess also possible non-linear relationships, regression trees were used to estimate NDVI percentiles, emphasizing the importance of ECa, ECa_ratio, Z, and slope in predicting different NDVI percentiles. Fuzzy k-means zoning using ECa + Z resulted in four classes that best classified variables that are relevant for irrigation scheduling due to their relationship with soil water storage (e.g. clay content, P_0.95 and PC2). Zonings based on ECa, ECa + Z + TWI and ECa + Z + TWI + NDVI yielded two zones that classified P_0.95 and PC2 well, but not clay content. Therefore, the zoning based on ECa + Z was chosen as optimal in the context of this VRI applications. Our analysis showed how NDVI series can be used in combination with ECa and elevation to evaluate the effectiveness of different zoning approaches for developing VRI prescriptions in olive groves.

In the context of limited resources and a growing demand for food due to an increase in the worldwide population, irrigation plays a vital role, and the efficient use of water is a major objective. In pressurized irrigation systems, water management is linked to high energy requirements, which is especially relevant in sprinkler irrigation. Therefore, decision support models are important for optimizing the design and management of irrigation systems. In this study, a holistic model for solid set irrigation systems (SORA 2024) was developed. This new model integrates hydraulic models at the subunit and plot levels to evaluate the distribution of pressure (EPANET, Rossman in The EPANET programmer’s toolkit for analysis of water distribution systems, Tempe, Arizona, 1999), the discharge and water distribution for each emitter (SIRIAS, Carrion et al. in , Irrig Sci 20(2):73–84, 2001) and the distribution of water applied by all the emitters of the subunit (SORA, Carrión et al. in Irrig Sci 20(2): 73–84, 2001). The integrated model also includes crop simulation (AQUACROP, Steduto et al. in Agron J 101(3), 426–437, 2009). to assess the effect of water distribution on crop production. The objective of this holistic model is to assist in decision-making processes for designing, sizing, upgrading, and managing solid set irrigation systems at the sprinkler level. The new integrated model (SORA 2024) was applied to a 2.84 ha commercial plot with 2 irrigation sectors that grow onion crops (Allium cepa L.). It was used to analyse each irrigation event from a real irrigation season, considering the conditions (pressure, irrigation time/periods, environmental conditions, and so on). The analysis is based on the sprinkler–nozzle combination, working pressure and wind direction and intensity during each irrigation event. The model also accounts for the cumulative effect/impact of all irrigation events on the plot. The model was validated through field trials using the “crop as a sensor” approach (Sarig et al. in , Agron 11(3):2021). To demonstrate the effectiveness of the model, the choice of nozzles in each sprinkler of the subunit was optimized. This is a quick and cost-effective way for farmers to improve their irrigation systems. By using this method, farmers can achieve better uniformity of water application and a slight increase in crop yield while maintaining the same irrigation schedule and amount of water used. Furthermore, the model enables farmers to work at the emitter level while integrating the results for the entire plot. This allows for precise irrigation of variable dosages by using different sprinkler–nozzle combinations in the same subunit. Farmers can do this based on the prior zoning of the plot, which is determined by its productive potential. This justifies the use of different irrigation dosages in each zone.

Introduction

Satellite-sourced data have become a valuable resource for precision agriculture because they provide crucial insights into various parameters that are essential for effective crop management. An array of practical agricultural tools provides comprehensive data for assessing crop biomass, soil conditions, and plant stress symptoms, predicting yields, and performing other functions. Satellite data, when combined with in situ data from different sources, can significantly enhance biomass and yield estimations.

Material and Methods

The ability of the “PROcesses of radiation, Mass and Energy Transfer” (PROMET) model to predict crop biomass and grain yield and to optimize nitrogen fertilization during the vegetation period was investigated. Field trials were conducted to assess the accuracy and limitations of biomass and yield predictions.

Results and Conclusion

The predicted yields were sufficiently accurate on a whole-field basis, and site-specific values showed strong correlations. In additional field trials with different fertilization strategies, the highest yield and nitrogen efficiency were observed for the PROMET-based strategy. Additional experiments with different crops and greater durations are needed to draw a more reliable conclusion.

Purpose

The use of Unmanned aerial vehicle (UAV) data for predicting crop above-ground biomass (AGB) is becoming a more feasible alternative to destructive methods. However, canopy height, vegetation index (VI), and other traditional features can become saturated during the mid to late stages of crop growth, significantly impacting the accuracy of AGB prediction.

Methods

In 2022 and 2023, UAV multispectral, RGB, and light detection and ranging point cloud data of wheat populations were collected at seven growth stages across two experimental fields. The point cloud depth features were extracted using the improved PointNet++ network, and AGB was predicted by fusion with VI, color index (CI), and texture index (TI) raster image features.

Results

The findings indicate that when the point cloud depth features were fused, the R² values predicted from VI, CI, TI, and canopy height model images increased by 0.05, 0.08, 0.06, and 0.07, respectively. For the combination of VI, CI, and TI, R² increased from 0.86 to a maximum of 0.9, while the root-mean-square error (RMSE) and mean absolute error were 1.80 t ha⁻¹ and 1.36 t ha⁻¹, respectively. Additionally, our findings revealed that the hybrid fusion exhibits the highest accuracy, it demonstrates robust adaptability in predicting AGB across various years, growth stages, crop varieties, nitrogen fertilizer applications, and densities.

Conclusion

This study effectively addresses the saturation in spectral and chemical information, provides valuable insights for high-precision phenotyping and advanced crop field management, and serves as a reference for studying other crops and phenotypic parameters.

The success of Variable Rate Application (VRA) techniques is closely linked to the algorithm used to calculate the different fertilizer rates. In this study, we proposed an algorithm based on the integration between some estimated agronomic inputs and crop radiometric data acquired by using a multispectral sensor. Generally, VRA algorithms are evaluated by comparing the yields, but they can often be affected by factors acting in the final phase of the crop cycle and not dependent on the fertilization treatments. Therefore, we decided to compare our algorithm (ALG) versus the traditional application of fertilizer (TRD) by evaluating the crop growth 1.5 months after the fertilization time. The algorithm was tested on a sorghum crop under organic farming, managed with or without manure. The saving of N obtained with ALG was equal to 14 and 5 kg ha^− 1 (-14 and − 10% for the non-manure and fertilized treatments, respectively). The NDVI values acquired after fertilization showed a remarkable reduction of relative standard deviation for ALG system (from 22 to 9% and from 34 to 14% for manured and not manured, respectively), which was not found for TRD system (from 16 to 17% and from 29 to 18% for manured and not manured, respectively). The above ground biomass produced was statistically equivalent for the two systems in the manured plots and significant higher for ALG in not-manured plots (+ 0.74 t ha^− 1 of dm, equal to + 23%). Finally, the indices calculated to evaluate the Nitrogen Use Efficiency (NUE) were consistently better in the ALG theses.

Purpose

Fruit detection and counting represent one of the most important steps toward yield estimation and a well-known practice for farmers, on which they base the management of the harvesting, storage, and distribution phases of agricultural products. In the era of precision agriculture, yield estimation, which was previously performed only by human operators, is currently being re-designed through the employment of Artificial Intelligence and Computer Vision techniques. Despite the impressive results that AI has demonstrated in fruit detection systems, they rely on large image datasets, whose availability is still limited if compared to the great number of crop typologies. For this reason, great interest has recently been devoted to weakly supervised algorithms, which can reduce the dataset annotation effort required by using simple image-level labels.

Method

Based on these considerations, this work proposes a new method relying on a sample-efficient weakly supervised approach. The proposed system, named MangoDetNet, is trained through a two-stage curriculum learning approach, first involving an image reconstruction task, and secondly an image binary classification task for heatmap generation. In particular, during the first stage, the network is trained in an unsupervised manner for the image reconstruction task, in order to promote the learning of robust feature extractors that are customized for the fruit scenarios. The second stage of training, instead, is performed to achieve image binary classification, employing presence/absence binary labels. This phase further refines the feature extractor from the previous stage and favors the computation of more refined and precise activation maps.

Conclusion

As demonstrated through the experimental campaign, performed on a mango orchard image dataset, MangoDetNet is able to outperform the state-of-the-art weakly supervised approaches, providing an F1 score equal to 0.861, which is on par with those of fully supervised methods, and an F1 score equal to 0.856 when halving the number of labeled samples needed for training.