Search

Traffic Sign Recognition (TSR) is one of the many utilities made possible by embedded systems with internet connections. Through the usage of vehicular cameras, it's possible to capture and classify traffic signs in real time with Artificial Intelligence (AI), more specifically, Convolutional Neural Networks (CNNs) based techniques. This article discusses the implementation of such TSR systems, and the building process of datasets for AI training. Such datasets include a brand new class to be used in TSR, vegetation occlusion. The results show that this approach is useful in making traffic sign maintenance faster since this application turns vehicles into moving sensors in that context. Leaning on the proposed technique, identified irregularities in traffic signs can be reported to a responsible body so they will eventually be fixed, contributing to a safer traffic environment. This paper also discusses the usage and performance of different YOLO models according to our case studies.

Yield forecasting is of immeasurable value in modern viticulture to optimize harvest scheduling and quality management. Traditionally, this task is carried out through manual and destructive sampling of production components and their accurate assessment is expensive, time-consuming, and error-prone, resulting in erroneous projections. The number of inflorescences and flowers per vine is one of the main components and serves as an early predictor. The adoption of new non-invasive technologies can automate this task and drive viticulture yield forecasting to higher levels of accuracy. In this study, different Single Stage Instance Segmentation models from the state-of-the-art You Only Look Once (YOLO) family, such as YOLOv5 and YOLOv8, were benchmarked on a dataset of RGB images for grapevine inflorescence detection and segmentation, with the aim of validating and subsequently implementing the solution for counting the number of inflorescences and flowers. All models obtained promising results, with the YOLOv8s and the YOLOv5s models standing out with an F1-Score of 95.1% and 97.7% for the detection and segmentation tasks, respectively. Moreover, the low inference times obtained demonstrate the models' ability to be deployed in real-time applications, allowing for non-destructive predictions in uncontrolled environments. [ABSTRACT FROM AUTHOR]

Currently, Unmanned Aerial Vehicles (UAVs) are considered in the development of various applications in agriculture, which has led to the expansion of the agricultural UAV market. However, Nano Aerial Vehicles (NAVs) are still underutilised in agriculture. NAVs are characterised by a maximum wing length of 15 centimetres and a weight of fewer than 50 g. Due to their physical characteristics, NAVs have the advantage of being able to approach and perform tasks with more precision than conventional UAVs, making them suitable for precision agriculture. This work aims to contribute to an open-source solution known as Nano Aerial Bee (NAB) to enable further research and development on the use of NAVs in an agricultural context. The purpose of NAB is to mimic and assist bees in the context of pollination. We designed this open-source solution by taking into account the existing state-of-the-art solution and the requirements of pollination activities. This paper presents the relevant background and work carried out in this area by analysing papers on the topic of NAVs. The development of this prototype is rather complex given the interactions between the different hardware components and the need to achieve autonomous flight capable of pollination. We adequately describe and discuss these challenges in this work. Besides the open-source NAB solution, we train three different versions of YOLO (YOLOv5, YOLOv7, and YOLOR) on an original dataset (Flower Detection Dataset) containing 206 images of a group of eight flowers and a public dataset (TensorFlow Flower Dataset), which must be annotated (TensorFlow Flower Detection Dataset). The results of the models trained on the Flower Detection Dataset are shown to be satisfactory, with YOLOv7 and YOLOR achieving the best performance, with 98% precision, 99% recall, and 98% F1 score. The performance of these models is evaluated using the TensorFlow Flower Detection Dataset to test their robustness. The three YOLO models are also trained on the TensorFlow Flower Detection Dataset to better understand the results. In this case, YOLOR is shown to obtain the most promising results, with 84% precision, 80% recall, and 82% F1 score. The results obtained using the Flower Detection Dataset are used for NAB guidance for the detection of the relative position in an image, which defines the NAB execute command. [ABSTRACT FROM AUTHOR]

The world wine sector is a multi-billion dollar industry with a wide range of economic activities. Therefore, it becomes crucial to monitor the grapevine because it allows a more accurate estimation of the yield and ensures a high-quality end product. The most common way of monitoring the grapevine is through the leaves (preventive way) since the leaves first manifest biophysical lesions. However, this does not exclude the possibility of biophysical lesions manifesting in the grape berries. Thus, this work presents three pre-trained YOLO models (YOLOv5x6, YOLOv7-E6E, and YOLOR-CSP-X) to detect and classify grape bunches as healthy or damaged by the number of berries with biophysical lesions. Two datasets were created and made publicly available with original images and manual annotations to identify the complexity between detection (bunches) and classification (healthy or damaged) tasks. The datasets use the same 10,010 images with different classes. The Grapevine Bunch Detection Dataset uses the Bunch class, and The Grapevine Bunch Condition Detection Dataset uses the OptimalBunch and DamagedBunch classes. Regarding the three models trained for grape bunches detection, they obtained promising results, highlighting YOLOv7 with 77% of mAP and 94% of the F1-score. In the case of the task of detection and identification of the state of grape bunches, the three models obtained similar results, with YOLOv5 achieving the best ones with an mAP of 72% and an F1-score of 92%. [ABSTRACT FROM AUTHOR]

The efficiency of agricultural practices depends on the timing of their execution. Environmental conditions, such as rainfall, and crop-related traits, such as plant phenology, determine the success of practices such as irrigation. Moreover, plant phenology, the seasonal timing of biological events (e.g., cotyledon emergence), is strongly influenced by genetic, environmental, and management conditions. Therefore, assessing the timing the of crops' phenological events and their spatiotemporal variability can improve decision making, allowing the thorough planning and timely execution of agricultural operations. Conventional techniques for crop phenology monitoring, such as field observations, can be prone to error, labour-intensive, and inefficient, particularly for crops with rapid growth and not very defined phenophases, such as vegetable crops. Thus, developing an accurate phenology monitoring system for vegetable crops is an important step towards sustainable practices. This paper evaluates the ability of computer vision (CV) techniques coupled with deep learning (DL) (CV_DL) as tools for the dynamic phenological classification of multiple vegetable crops at the subfield level, i.e., within the plot. Three DL models from the Single Shot Multibox Detector (SSD) architecture (SSD Inception v2, SSD MobileNet v2, and SSD ResNet 50) and one from You Only Look Once (YOLO) architecture (YOLO v4) were benchmarked through a custom dataset containing images of eight vegetable crops between emergence and harvest. The proposed benchmark includes the individual pairing of each model with the images of each crop. On average, YOLO v4 performed better than the SSD models, reaching an F1-Score of 85.5%, a mean average precision of 79.9%, and a balanced accuracy of 87.0%. In addition, YOLO v4 was tested with all available data approaching a real mixed cropping system. Hence, the same model can classify multiple vegetable crops across the growing season, allowing the accurate mapping of phenological dynamics. This study is the first to evaluate the potential of CV_DL for vegetable crops' phenological research, a pivotal step towards automating decision support systems for precision horticulture. [ABSTRACT FROM AUTHOR]

Object identification, such as tree trunk detection, is fundamental for forest robotics. Intelligent vision systems are of paramount importance in order to improve robotic perception, thus enhancing the autonomy of forest robots. To that purpose, this paper presents three contributions: an open dataset of 5325 annotated forest images; a tree trunk detection Edge AI benchmark between 13 deep learning models evaluated on four edge-devices (CPU, TPU, GPU and VPU); and a tree trunk mapping experiment using an OAK-D as a sensing device. The results showed that YOLOR was the most reliable trunk detector, achieving a maximum F1 score around 90% while maintaining high scores for different confidence levels; in terms of inference time, YOLOv4 Tiny was the fastest model, attaining 1.93 ms on the GPU. YOLOv7 Tiny presented the best trade-off between detection accuracy and speed, with average inference times under 4 ms on the GPU considering different input resolutions and at the same time achieving an F1 score similar to YOLOR. This work will enable the development of advanced artificial vision systems for robotics in forestry monitoring operations. [ABSTRACT FROM AUTHOR]

Mobile robotics in forests is currently a hugely important topic due to the recurring appearance of forest wildfires. Thus, in-site management of forest inventory and biomass is required. To tackle this issue, this work presents a study on detection at the ground level of forest tree trunks in visible and thermal images using deep learning-based object detection methods. For this purpose, a forestry dataset composed of 2895 images was built and made publicly available. Using this dataset, five models were trained and benchmarked to detect the tree trunks. The selected models were SSD MobileNetV2, SSD Inception-v2, SSD ResNet50, SSDLite MobileDet and YOLOv4 Tiny. Promising results were obtained; for instance, YOLOv4 Tiny was the best model that achieved the highest AP (90%) and F1 score (89%). The inference time was also evaluated, for these models, on CPU and GPU. The results showed that YOLOv4 Tiny was the fastest detector running on GPU (8 ms). This work will enhance the development of vision perception systems for smarter forestry robots. [ABSTRACT FROM AUTHOR]

Visual perception enables robots to perceive the environment. Visual data is processed using computer vision algorithms that are usually time-expensive and require powerful devices to process the visual data in real-time, which is unfeasible for open-field robots with limited energy. This work benchmarks the performance of different heterogeneous platforms for object detection in real-time. This research benchmarks three architectures: embedded GPU—Graphical Processing Units (such as NVIDIA Jetson Nano 2 GB and 4 GB, and NVIDIA Jetson TX2), TPU—Tensor Processing Unit (such as Coral Dev Board TPU), and DPU—Deep Learning Processor Unit (such as in AMD/Xilinx ZCU104 Development Board, and AMD/Xilinx Kria KV260 Starter Kit). The authors used the RetinaNet ResNet-50 fine-tuned using the natural VineSet dataset. After the trained model was converted and compiled for target-specific hardware formats to improve the execution efficiency. The platforms were assessed in terms of performance of the evaluation metrics and efficiency (time of inference). Graphical Processing Units (GPUs) were the slowest devices, running at 3 FPS to 5 FPS, and Field Programmable Gate Arrays (FPGAs) were the fastest devices, running at 14 FPS to 25 FPS. The efficiency of the Tensor Processing Unit (TPU) is irrelevant and similar to NVIDIA Jetson TX2. TPU and GPU are the most power-efficient, consuming about 5 W. The performance differences, in the evaluation metrics, across devices are irrelevant and have an F1 of about 70 % and mean Average Precision (mAP) of about 60 %. [Display omitted] • RetinaNet ResNet50 and SSD ResNet50 can be successfully executed in embedded GPUs, TPUs, and FPGAs. • FPGAs are the fastest devices for executing deep learning models and ANNs. • TPUs are the most power-efficient devices for embedded applications. • GPUs offer a good balance between implementation and deployment of ANNs. [ABSTRACT FROM AUTHOR]

The harvesting operation is a recurring task in the production of any crop, thus making it an excellent candidate for automation. In protected horticulture, one of the crops with high added value is tomatoes. However, its robotic harvesting is still far from maturity. That said, the development of an accurate fruit detection system is a crucial step towards achieving fully automated robotic harvesting. Deep Learning (DL) and detection frameworks like Single Shot MultiBox Detector (SSD) or You Only Look Once (YOLO) are more robust and accurate alternatives with better response to highly complex scenarios. The use of DL can be easily used to detect tomatoes, but when their classification is intended, the task becomes harsh, demanding a huge amount of data. Therefore, this paper proposes the use of DL models (SSD MobileNet v2 and YOLOv4) to efficiently detect the tomatoes and compare those systems with a proposed histogram-based HSV colour space model to classify each tomato and determine its ripening stage, through two image datasets acquired. Regarding detection, both models obtained promising results, with the YOLOv4 model standing out with an F1-Score of 85.81%. For classification task the YOLOv4 was again the best model with an Macro F1-Score of 74.16%. The HSV colour space model outperformed the SSD MobileNet v2 model, obtaining results similar to the YOLOv4 model, with a Balanced Accuracy of 68.10%. [ABSTRACT FROM AUTHOR]

The agricultural sector plays a fundamental role in our society, where it is increasingly important to automate processes, which can generate beneficial impacts in the productivity and quality of products. Perception and computer vision approaches can be fundamental in the implementation of robotics in agriculture. In particular, deep learning can be used for image classification or object detection, endowing machines with the capability to perform operations in the agriculture context. In this work, deep learning was used for the detection of grape bunches in vineyards considering different growth stages: the early stage just after the bloom and the medium stage where the grape bunches present an intermediate development. Two state-of-the-art single-shot multibox models were trained, quantized, and deployed in a low-cost and low-power hardware device, a Tensor Processing Unit. The training input was a novel and publicly available dataset proposed in this work. This dataset contains 1929 images and respective annotations of grape bunches at two different growth stages, captured by different cameras in several illumination conditions. The models were benchmarked and characterized considering the variation of two different parameters: the confidence score and the intersection over union threshold. The results showed that the deployed models could detect grape bunches in images with a medium average precision up to 66.96%. Since this approach uses low resources, a low-cost and low-power hardware device that requires simplified models with 8 bit quantization, the obtained performance was satisfactory. Experiments also demonstrated that the models performed better in identifying grape bunches at the medium growth stage, in comparison with grape bunches present in the vineyard after the bloom, since the second class represents smaller grape bunches, with a color and texture more similar to the surrounding foliage, which complicates their detection. [ABSTRACT FROM AUTHOR]

The development of robotic solutions for agriculture requires advanced perception capabilities that can work reliably in any crop stage. For example, to automatise the tomato harvesting process in greenhouses, the visual perception system needs to detect the tomato in any life cycle stage (flower to the ripe tomato). The state-of-the-art for visual tomato detection focuses mainly on ripe tomato, which has a distinctive colour from the background. This paper contributes with an annotated visual dataset of green and reddish tomatoes. This kind of dataset is uncommon and not available for research purposes. This will enable further developments in edge artificial intelligence for in situ and in real-time visual tomato detection required for the development of harvesting robots. Considering this dataset, five deep learning models were selected, trained and benchmarked to detect green and reddish tomatoes grown in greenhouses. Considering our robotic platform specifications, only the Single-Shot MultiBox Detector (SSD) and YOLO architectures were considered. The results proved that the system can detect green and reddish tomatoes, even those occluded by leaves. SSD MobileNet v2 had the best performance when compared against SSD Inception v2, SSD ResNet 50, SSD ResNet 101 and YOLOv4 Tiny, reaching an F1-score of 66.15 %, an mAP of 51.46 % and an inference time of 16.44 m s with the NVIDIA Turing Architecture platform, an NVIDIA Tesla T4, with 12 GB. YOLOv4 Tiny also had impressive results, mainly concerning inferring times of about 5 m s. [ABSTRACT FROM AUTHOR]

The development of robotic solutions in unstructured environments brings several challenges, mainly in developing safe and reliable navigation solutions. Agricultural environments are particularly unstructured and, therefore, challenging to the implementation of robotics. An example of this is the mountain vineyards, built-in steep slope hills, which are characterized by satellite signal blockage, terrain irregularities, harsh ground inclinations, and others. All of these factors impose the implementation of precise and reliable navigation algorithms, so that robots can operate safely. This work proposes the detection of semantic natural landmarks that are to be used in Simultaneous Localization and Mapping algorithms. Thus, Deep Learning models were trained and deployed to detect vine trunks. As significant contributions, we made available a novel vine trunk dataset, called VineSet, which was constituted by more than 9000 images and respective annotations for each trunk. VineSet was used to train state-of-the-art Single Shot Multibox Detector models. Additionally, we deployed these models in an Edge-AI fashion and achieve high frame rate execution. Finally, an assisted annotation tool was proposed to make the process of dataset building easier and improve models incrementally. The experiments show that our trained models can detect trunks with an Average Precision up to 84.16% and our assisted annotation tool facilitates the annotation process, even in other areas of agriculture, such as orchards and forests. Additional experiments were performed, where the impact of the amount of training data and the comparison between using Transfer Learning and training from scratch were evaluated. In these cases, some theoretical assumptions were verified. [ABSTRACT FROM AUTHOR]

Robotics will significantly impact large sectors of the economy with relatively low productivity, such as Agri-Food production. Deploying agricultural robots on the farm is still a challenging task. When it comes to localising the robot, there is a need for a preliminary map, which is obtained from a first robot visit to the farm. Mapping is a semi-autonomous task that requires a human operator to drive the robot throughout the environment using a control pad. Visual and geometric features are used by Simultaneous Localisation and Mapping (SLAM) Algorithms to model and recognise places, and track the robot's motion. In agricultural fields, this represents a time-consuming operation. This work proposes a novel solution—called AgRoBPP-bridge—to autonomously extract Occupancy Grid and Topological maps from satellites images. These preliminary maps are used by the robot in its first visit, reducing the need of human intervention and making the path planning algorithms more efficient. AgRoBPP-bridge consists of two stages: vineyards row detection and topological map extraction. For vineyards row detection, we explored two approaches, one that is based on conventional machine learning technique, by considering Support Vector Machine with Local Binary Pattern-based features, and another one found in deep learning techniques (ResNET and DenseNET). From the vineyards row detection, we extracted an occupation grid map and, by considering advanced image processing techniques and Voronoi diagrams concept, we obtained a topological map. Our results demonstrated an overall accuracy higher than 85% for detecting vineyards and free paths for robot navigation. The Support Vector Machine (SVM)-based approach demonstrated the best performance in terms of precision and computational resources consumption. AgRoBPP-bridge shows to be a relevant contribution to simplify the deployment of robots in agriculture. [ABSTRACT FROM AUTHOR]

• Low-cost and low-power experimental device benchmark for Deep Learning. • Vine trunk detection using Convolutional Neural Networks on embedded devices. • Online Deep Learning inference in an agricultural robot. • Generation of landmark detection for Localization and Mapping in steep slope vineyards. Research and development in mobile robotics are continuously growing. The ability of a human-made machine to navigate safely in a given environment is a challenging task. In agricultural environments, robot navigation can achieve high levels of complexity due to the harsh conditions that they present. Thus, the presence of a reliable map where the robot can localize itself is crucial, and feature extraction becomes a vital step of the navigation process. In this work, the feature extraction issue in the vineyard context is solved using Deep Learning to detect high-level features – the vine trunks. An experimental performance benchmark between two devices is performed: NVIDIA's Jetson Nano and Google's USB Accelerator. Several models were retrained and deployed on both devices, using a Transfer Learning approach. Specifically, MobileNets, Inception, and lite version of You Only Look Once are used to detect vine trunks in real-time. The models were retrained in a built in–house dataset, that is publicly available. The training dataset contains approximately 1600 annotated vine trunks in 336 different images. Results show that NVIDIA's Jetson Nano provides compatibility with a wider variety of Deep Learning architectures, while Google's USB Accelerator is limited to a unique family of architectures to perform object detection. On the other hand, the Google device showed an overall Average precision higher than Jetson Nano, with a better runtime performance. The best result obtained in this work was an average precision of 52.98% with a runtime performance of 23.14 ms per image, for MobileNet-V2. Recent experiments showed that the detectors are suitable for the use in the Localization and Mapping context. [ABSTRACT FROM AUTHOR]