Search

Within the scope of precision agriculture, many applications have been developed to support decision making and yield enhancement. Fruit detection has attracted considerable attention from researchers, and it can be used offline. In contrast, some applications, such as robot vision in orchards, require computer vision models to run on edge devices while performing inferences at high speed. In this area, most modern applications use an integrated graphics processing unit (GPU). In this work, we propose the use of a tensor processing unit (TPU) accelerator with a Raspberry Pi target device and the state-of-the-art, lightweight, and hardware-aware MobileDet detector model. Our contribution is the extension of the possibilities of using accelerators (the TPU) for edge devices in precision agriculture. The proposed method was evaluated using a novel dataset of peaches with three cultivars, which will be made available for further studies. The model achieved an average precision (AP) of 88.2% and a performance of 19.84 frames per second (FPS) at an image size of 640 × 480. The results obtained show that the TPU accelerator can be an excellent alternative for processing on the edge in precision agriculture. [ABSTRACT FROM AUTHOR]

Portable devices play an essential role where edge computing is necessary and mobility is required (e.g., robots in agriculture within remote-sensing applications). With the increasing applications of deep neural networks (DNNs) and accelerators for edge devices, several methods and applications have been proposed for simultaneous crop and weed detection. Although preliminary studies have investigated the performance of inference time for semantic segmentation of crops and weeds in edge devices, performance degradation has not been evaluated in detail when the required optimization is applied to the model for operation in such edge devices. This paper investigates the relationship between model tuning hyperparameters to improve inference time and its effect on segmentation performance. The study was conducted using semantic segmentation model DeeplabV3 with a MobileNet backbone. Different datasets (Cityscapes, PASCAL and ADE20K) were analyzed for a transfer learning strategy. The results show that, when using a model hyperparameter depth multiplier (DM) of 0.5 and the TensorRT framework, segmentation performance mean intersection over union (mIOU) decreased by 14.7% compared to that of a DM of 1.0 and no TensorRT. However, inference time accelerated dramatically by a factor of 14.8. At an image resolution of 1296 × 966 , segmentation performance of 64% mIOU and inference of 5.9 frames per second (FPS) was achieved in Jetson Nano's device. With an input image resolution of 513 × 513 , and hyperparameters output stride OS = 32 and DM = 0.5, an inference time of 0.04 s was achieved resulting in 25 FPS. The results presented in this paper provide a deeper insight into how the performance of the semantic segmentation model of crops and weeds degrades when optimization is applied to adapt the model to run on edge devices. Lastly, an application is described for the semantic segmentation of weeds embedded in the edge device (Jetson Nano) and integrated with the robotic orchard. The results show good spraying accuracy and feasibility of the method. [ABSTRACT FROM AUTHOR]

Precision irrigation and optimization of water use have become essential factors in agriculture because water is critical for crop growth. The proper management of an irrigation system should enable the farmer to use water efficiently to increase productivity, reduce production costs, and maximize the return on investment. Efficient water application techniques are essential prerequisites for sustainable agricultural development based on the conservation of water resources and preservation of the environment. In a previous work, an off-policy deep reinforcement learning model, Deep Q-Network, was implemented to optimize irrigation. The performance of the model was tested for tomato crop at a site in Portugal. In this paper, an on-policy model, Advantage Actor–Critic, is implemented to compare irrigation scheduling with Deep Q-Network for the same tomato crop. The results show that the on-policy model Advantage Actor–Critic reduced water consumption by 20% compared to Deep Q-Network with a slight change in the net reward. These models can be developed to be applied to other cultures with high production in Portugal, such as fruit, cereals, and wine, which also have large water requirements. [ABSTRACT FROM AUTHOR]

The concept of the Internet of Things (IoT) in agriculture is associated with the use of high-tech devices such as robots and sensors that are interconnected to assess or monitor conditions on a particular plot of land and then deploy the various factors of production such as seeds, fertilizer, water, etc., accordingly. Vine trunk detection can help create an accurate map of the vineyard that the agricultural robot can rely on to safely navigate and perform a variety of agricultural tasks such as harvesting, pruning, etc. In this work, the state-of-the-art single-shot multibox detector (SSD) with MobileDet Edge TPU and MobileNet Edge TPU models as the backbone was used to detect the tree trunks in the vineyard. Compared to the SSD with MobileNet-V1, MobileNet-V2, and MobileDet as backbone, the SSD with MobileNet Edge TPU was more accurate in inference on the Raspberrypi, with almost the same inference time on the TPU. The SSD with MobileDet Edge TPU achieved the second-best accurate model. Additionally, this work examines the effects of some features, including the size of the input model, the quantity of training data, and the diversity of the training dataset. Increasing the size of the input model and the training dataset increased the performance of the model. [ABSTRACT FROM AUTHOR]

Fruit detection is crucial for yield estimation and fruit picking system performance. Many state-of-the-art methods for fruit detection use convolutional neural networks (CNNs). This paper presents the results for peach detection by applying a faster R-CNN framework in images captured from an outdoor orchard. Although this method has been used in other studies to detect fruits, there is no research on peaches. Since the fruit colors, sizes, shapes, tree branches, fruit bunches, and distributions in trees are particular, the development of a fruit detection procedure is specific. The results show great potential in using this method to detect this type of fruit. A detection accuracy of 0.90 using the metric average precision (AP) was achieved for fruit detection. Precision agriculture applications, such as deep neural networks (DNNs), as proposed in this paper, can help to mitigate climate change, due to horticultural activities by accurate product prediction, leading to improved resource management (e.g., irrigation water, nutrients, herbicides, pesticides), and helping to reduce food loss and waste via improved agricultural activity scheduling. [ABSTRACT FROM AUTHOR]

In the field of agriculture, the water used for irrigation should be given special treatment, as it is responsible for a large proportion of total water consumption. Irrigation scheduling is critical to food production because it guarantees producers a consistent harvest and minimizes the risk of losses due to water shortages. Therefore, the creation of an automatic irrigation method using new technologies is essential. New methods such as deep learning algorithms have attracted a lot of attention in agriculture and are already being used successfully. In this work, a Deep Q-Network was trained for irrigation scheduling. The agent was trained to schedule irrigation for a tomato field in Portugal. Two Long Short Term Memory models were used as the agent environment. One predicts the total water in the soil profile on the next day. The other one was employed to estimate the yield based on the environmental condition during a season and then measure the net return. The agent uses this information to decide the following irrigation amount. An Artificial Neural Network, a Long Short Term Memory, and a Convolutional Neural Network were used to estimating the Q-table during training. Unlike the Long-Short Terms Memory model, the Artificial Neural Network and the Convolutional Neural Network could not estimate the Q-table, and the agent's reward decreased during training. The comparison of the performance of the model was done with fixed base irrigation and threshold based irrigation. The trained model increased productivity by 11% and decreased water consumption by 20–30% compared to the fixed method. • The creation of a system that can help farmers tell them how much and when to irrigate. • After training, the model can be used by farmers and does not need an expert to use the system. • The training of the model is end-to-end, and no manual adjustments are required. • It uses climate data as input which is easy to collect. • The trained model adjusts the irrigation based on climate change. [ABSTRACT FROM AUTHOR]