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 $1296\times 966$, 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 $513\times 513$, 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.

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.

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.

Currently, there is a growing demand for cleaner and sustainable technologies due to environmental issues. In this sense, there is a necessity to manage the assessment of production processes and the rationalization of energy consumption. In this study, an Energy Life-Cycle Assessment (ELCA) was carried out through energy efficiency indicators, directed to the characterization and renewability of the peach production system life-cycle in the Portuguese region of Beira Interior. The study intends to investigate the non-renewable energy inputs from fossil fuels, as well as the emissions resulting from machinery. In addition, warehouse energy inputs are analyzed, mainly cooling systems of refrigerated chambers where fruits are preserved. This analysis aims to find opportunities for technological, environmental and best practices improvements. Test scenarios were analyzed and revealing soil groundcover maintenance is the operation with the largest impact in the energy consumption of the production process (3176 MJ·ha⁻¹). In the post-harvest processes, the energy consumption largest impact is given by the warehouse’s operations (35,700 MJ·ha⁻¹), followed by transportation (6180 MJ·ha⁻¹). Concerning the emissions resulting from the fuels consumption, the largest impact is due to the plantation machinery and the transportation from warehouse to retailers.

Permanent crops rely heavily on pesticides. Their short-term behaviour has already been characterised under laboratory conditions. However, their degradation in the field is still poorly understood. Furthermore, studies that relate pesticide residues in soil with their application history are still scarce, despite their relevance for inferring the historical and diffuse dimensions of pollution. This work explores the spatial and temporal dynamics of pesticide residues in soil under the light of their application records and the accuracy of the predicted environmental concentrations (PEC). To achieve these purposes, 270 soil samples were collected at 5 sampling times over the 2022 growing season, in 18 peach orchards from east-central Portugal. Moreover, the data extracted from the application records was used to obtain predicted levels which were compared with the measured results. A total of 37 residues were analysed. Overall, 32 of these were detected back in, at least, one sample. Glyphosate and AMPA were the most frequently found and had the highest median concentrations. The comparison between predicted and measured environmental concentrations (MEC) suggests that pesticide decay was generally slower than expected. Glyphosate, AMPA, fluopyram and tebuconazole showed a high potential for resulting in background concentrations in soil. Our results highlight the need for regular and comprehensive monitoring, being recommended long-term assessments of the persistence and fate of pesticides under realistic conditions. Furthermore, our results also stress the necessity of more studies on reasons for PEC-MEC mismatch, a factor of concern considering the regulatory relevance of PEC values.