Smart agricultural technology最新文献

This study investigates the impact of Chinese smallholders’ adoption of Information and Communication Technologies—the use of smartphones and computers connected to the internet—on their commercial orientation, land, and labor productivity. Commercial orientation is the share of farm output for sales in the market. We used a control function approach and a selectivity-corrected model. The study uses national survey data from rural sample households, the China Household Database, and the China Household Finance Survey and Research Center. Findings reveal that the adoption of information and communication technologies by Chinese farmers increased the commercial orientation of farming. Furthermore, adopting information and communication technologies increases land and labor productivity by about 21.3 % and 28.2 %, respectively. Farm households’ commercial orientation improved labor productivity by about 35.9 %. Heterogeneity indicates that the adoption of information and communication technologies has a more significant effect on improving productivity for young household heads and small farmers. Policymakers should establish information and communication technologies training programs, develop digital infrastructure, and promote smallholder commercial production to increase agricultural productivity.

The Qazaq Aqbas beef breed is the most important in Kazakhstan. The breed is very well adapted to the harsh conditions in Central Asia. Other more productive breeds need additional costs to ensure their survival and productivity. However, their production levels are lower than other beef breeds globally. It may be possible to improve this by selecting bulls that have greater feed efficiency. This case study reports analyses of feed intakes and weight gains by this breed on farms in Kazakhstan. Twenty-nine bulls were selected, and fed using the GrowSafe system that measures and records intakes and weights. The ranking by Residual feed intakes (RFI) identified those bulls that were most efficient regarding weight gains compared to their feed intakes. While there was a positive correlation between ADG and DMI (P = 0.011), there was no correlation between RFI and ADG. Relying simply on weight gains for breeding decisions is therefore not supported by this evidence. The daily feed intakes of the breed recorded (11.03 kg/d) were similar to those of non-native popular beef breeds, while the weight gains (0.95 kg/d) were smaller. Therefore, the selection for breeding of beef bulls could focus on feed efficiency and not only feed intakes or daily weight gains.

Vegetable breeding companies invest a considerable amount of their resources in phenotyping. The advancement of computer vision technology has made it possible to digitalize these processes, leading to improved efficiency and quality. However, phenotyping activities often take place in outdoor fields or greenhouses, where the environmental/illumination conditions are constantly changing. This lack of standardization presents a problem for automatically isolating the relevant elements in the images, which is an important first step for phenotyping. Classical image analysis methods have shown not to be robust enough for that in these changing conditions. However, in the last years, deep learning models have demonstrated to be able to identify and learn meaningful features that are more robust and representative of the underlying patterns, enabling them to handle diverse and changeable conditions effectively.

In this work, we propose a pepper instance segmentation solution based on deep learning after harvest under field conditions. We implement the method and validate it for three pepper varieties: Blocky Bell, Jalapeño and Lamuyo. We compare the performance of this new method for each variety with a previous solution based on classical image processing techniques, with the objective of measuring and demonstrating the superiority of deep learning-based instance segmentation over traditional methods as a first step for phenotyping.

The instance segmentation deep learning based models outperform the results obtained by classical image processing algorithms for the three pepper varieties: in Blocky Bell mAP is increased from 0.63 to 0.97, in Jalapeño from 0.39 to 0.52 and in Lamuyo from 0.67 to 0.97.

Accurate detection and localization of fruits in natural environments is a key step for fruit picking robots to achieve precise harvesting. However, existing banana detection and positioning methods have two main limitations in practical applications: a large number of model parameters that make deployment difficult, and a need for performance improvement. To tackle the above issues, a high-precision and lightweight banana bunch recognition and localization method was proposed and deployed on edge devices for application. First, a Slim-Banana model was proposed based on the improvement of YOLOv8l. In order to reduce the model calculation amount and maintain high performance, GSConv was introduced in the Slim-Banana model to replace the standard convolution, and combined with grouped convolution and spatial convolution. At the same time, the cross-stage local network (GSCSP) module was designed to reduce the computational complexity and the complexity of the network structure through a single-stage aggregation method. Then, the RealSense depth sensor is combined with TOF technology to perform image registration and 3D localization of the banana. Finally, the pipeline is deployed on the Nvidia Orin NX edge device and its performance and resource consumption in actual work are deeply analyzed. Experimental results show that the detection precision, recall, mAP and inference time of our method are 0.947, 0.948, 0.98 and 113.6 ms respectively, the network memory size required is 4449MiB, and the average localization errors in the X-axis, Y-axis and Z-axis directions are 13.47 mm, 12.87 mm and 13.87 mm respectively. To our knowledge, this is the first work that implements banana detection and localization on edge devices. Experimental results show that compared with existing methods, our method achieves better performance in complex orchard environments, achieving efficient and lightweight banana recognition and localization.

In this study, three novel machine learning algorithms of additive regression-random forest (AR-RF), Iterative Classifier Optimizer (ICO-AR-RF), and multi-scheme (MS-RF) were explored for carbon dioxide (CO₂) flux rate prediction from three agricultural fields. To build the dataset, 401 samples were collected from two fields in Prince Edward Island (PEI) and 122 samples from the New Brunswick (NB), Canada. In addition, soil moisture (SM), temperature (ST), and electrical conductivity (EC), alongside eight climatic variables including wind speed (WS), solar radiation (SR), relative humidity (RH), precipitation (PCP), air temperature (AT), dew point (DP), vapour pressure difference (VPD) and reference evapotranspiration (ET_o) were also collected. Greedy stepwise (GS) approach was implemented for feature selection. Finally, different qualitative (scatter plot, line graph, Taylor diagram, box plot, and Rug plot), and quantitative (uncertainty analysis, root mean square error (RMSE), percent of BIAS (PBIAS), Nash Sutcliff efficiency (NSE) and RMSE-observations standard deviation ratio (RSR)) techniques were used for model evaluation and comparison. Results of feature selection approaches revealed that DP, AT, SM, and ST are the four most effective variables at CO₂ prediction in PEI, while AT, RH, DP, and ST are the most effective in the NB study area. For optimum input scenario, the GS algorithm was applied, and results showed that a combination of DP, AT, ST, SM, and ET_o was the best for the PEI study area, while for NB, all input variables should be involved. Our analysis, for prediction of CO₂ fluxes, confirmed that the ICO-AR-RF model performed the best at both PEI (RMSE=0.70, NSE=0.76, PBIAS=-5.11, RSR=0.48) and NB (RMSE=0.74, NSE=0.75, PBIAS=3.23, RSR=0.50), followed by MS-RF and AR-RF. Uncertainty analysis showed that CO₂ prediction is more sensitive to input scenario selection than models in both study areas. Results revealed that climatic variables are more effective in CO₂ prediction than soil characteristics and the developed hybrid model ICO-AR-RF can be a promising tool for decision-makers and beneficial for stakeholders.

In recent years, consumers have shown interest in non-destructive methods to assess the fruit's internal quality during ripening. The objective of this study is to construct an E-nose system using low-cost MQ sensors and evaluate fruit quality, specifically soluble sugar content (SSC) and hardness of mango during ripening. The correlation test was performed to compare sensor readings with SSC and hardness, and multiple linear regression (MLR) was used to establish linear equations for mango quality indices based on sensor variation. Over the storage period, the hardness of mango was decreased from the value of 15.4 kgf/cm² to 12.25 kgf/cm². Similarly, the SSC for mangoes increased from 19.7 %Brix to a final value of 24.66 %Brix. The sensor values also showed positive correlation with SSC and negative correlation with hardness of mango, respectively. Using the MLR analysis, the hardness and SSC of mango during the ripening stage, the correlation coefficient (r) of 0.847, standard error of 1.49 kgf/cm² and 0.815, standard error of 1.696 %Brix for hardness and SSC prediction, respectively. These results indicate that MQ-based E-nose is the rapid and non-destructive method for predicting mango qualities during ripening stage.

Tomato yield is significantly affected by diseases, which are a continuous challenge for its production and pose threats to its global supply chain. Automatic and early detection of these diseases could help growers to swiftly adopt mitigation strategies to limit the disease spread, leading to improved production. Deep learning-based CNN approaches have been widely applied to detect tomato diseases. However, deep learning models are highly computationally demanding, resulting in a computational bottleneck for practical adaptation for agricultural applications such as disease detection and monitoring. Over the last few years, developments of open-source Edge systems have provided opportunities for low-cost and low-power consumption practical solutions for deep learning applications for agriculture. Therefore, the primary goal of this study was to evaluate the performance of standalone Edge-AI solutions for tomato leaf disease detection. To achieve this goal, firstly, this study employed lightweight deep learning networks to detect and differentiate tomato leaf diseases (bacterial spot, early blight, healthy, late blight, leaf mold, septoria leaf spot, two spotted spider mites, target spot, and yellow leaf curl virus). Then, these deep learning networks were deployed on low-cost and low-power consumption Edge devices to investigate their performance capabilities as standalone Edge-AI solutions for the early detection of tomato leaf diseases. Lightweight CNN based GoogleNet and MobileNetV2 deep learning networks achieved accuracies of up to 98.25 % compared to accuracies of 98.13 %, 98.13 %, 94.62 %, and 90.67 % of EfficientNetB0, ResNet-18, SqueezeNet, and NasNetMobile, respectively, in detecting tomato diseases. NVIDIA Jetson ORIN AGX and Nano significantly outperformed Raspberry Pi and Raspberry Pi with AI accelerator (Google Coral) in image classification, achieving classification times of 3.5 ms and 5.2 ms respectively, using SqueezeNet, compared to 15.3 ms and 10.5 ms on the Raspberry Pi devices. In addition, Raspberry Pi with Google Coral achieved the best cost/FPS performance of 0.14 compared to other Edge devices NVIDIA Jetson AGX Orin and NVIDIA Jetson Nano Orin with cost/FPS of 0.7 and 0.26, respectively. These results showed the potential of standalone Edge-AI solutions using low-cost and low-power consuming software and hardware resources for early tomato disease detections.

Additive manufacturing, also known as 3D printing, is an amazing innovation with a wide range of uses in intelligent agriculture and food processing. Along with adjustable farming equipment and autonomous agricultural instruments like drones and robots, it offers real-time data on plant health, nutrient levels, and soil state. 3D printing has reinvented food processing by enabling personalized nutrition solutions, particularly in the field of medicinal nutrition. It also makes it possible to alter the textures and structures of food, creating novel sensory experiences and better-quality goods. 3D printing contributes to sustainable food production by reducing food waste (10–30 %) and using alternative protein sources. According to the study, AI and 3D-assisted IoT sensors can help increase yield by 10 % to 15 % while significantly reducing crop deterioration. They can also help reduce water usage by 20 % to 25 %, labor requirements by 20 % to 30 %, and overall power consumption by 20 %. However, high costs, complex technical and design knowledge, and limitations on production speed and scale are obstacles to broader use. It's also necessary to handle safety and regulatory concerns. 3D printing has a promising future in various fields thanks to advancements in bioprinting, multifunctional materials, blockchain, and artificial intelligence integration. These advancements could boost 3D printing's potential and result in higher output, more sustainable practices, and higher-quality products.

Energy savings in Wireless Sensor Networks for fish farms is necessary and beneficial. We present a novel energy savings method that minimizes the interpolation errors of sensors' measurements. Sensors follow a working cycle in which they are active when measuring data, and inactive or suspended when data is interpolated from the above measurements and consumed energy is reduced. To our knowledge, we are the first to implement interpolation for energy savings. We improved that model to describe the non-linear property of the consumed energy in the batteries, adding a new variable that explains their real behavior. Several experiments with a prototype of a Wireless Sensor Network with pH, water temperature, and ambient temperature sensors are implemented to validate our method. We made a series of measurements to determine the actual energy savings at each sensor and compared these savings with those predicted by each theory model. The results show that the model is more accurate, presenting less than 5 % prediction errors which does not affect fish growth. Furthermore, our paper introduces an energy-saving method for extending WSN lifetime by modeling the non-linear power consumption of sensors' batteries. We propose a new mathematical optimization formulation using an efficient interpolation mechanism that operates in real-time. A real-scale WSN prototype installed over water validates and refines our method. Finally, we showed that the number of interpolated values is of a broader range for aquatic sensors than for outdoor sensors such as ambient temperature. That is, energy savings for fish farming is acceptable.

Globally, livestock farmers are challenged with reducing greenhouse gas emissions to mitigate climate change. A potential option for pasture-based dairy farmers involves including methane-inhibiting compounds in the diet. A novel approach to deliver these compounds with the required frequency and precision is via smart-feeders, an existing smart farming technology used to feed supplements automatically to animals in-paddock. For this innovation to be successful, however, it must integrate with farm systems and provide farmers with a positive value proposition. The aim of this study was to examine the farm system and technology factors influencing potential uptake of in-paddock smart technologies for delivering methane inhibitors in pasture-grazed systems. We utilized an adoption prediction tool (ADOPT) to model the adoption outcomes of smart-feeders as methane inhibitor delivery mechanisms on dairy farms, with input from industry experts and farmers via focus groups. The results indicated low adoption of smart-feeders in a pasture-based system context. This was further explored with a sensitivity analysis of seven critical ADOPT factors which were identified as influential through the farmer focus groups. We modelled the impact of the seven critical ADOPT factors for two smart-feeder concepts to evaluate their relative adoption potential. The adoption modelling showed that while factors such as technology cost and function were important, adoption would also be highly influenced by future regulation settings, innovation uncertainty, and the alignment with farmer values and worldviews about their farm system. This research highlighted that in-paddock delivery technology, and processes for its use on-farm, represents an early-stage innovation and therefore is vital that farmers and other stakeholders are involved in further development to ensure adoption factors are addressed.