1,721,011 research outputs found
BUP-ST20: Weakly Labelled Spatial Temporal Sweet Pepper Data
No full textAccurate monitoring of crop phenotypic traits is essential for efficient farm management and automation in agriculture. Multi-object tracking (MOT) and video instance segmentation (VIS) offer promising approaches to enhance agricultural robotic vision systems, yet a major limitation is the scarcity of high-quality spatial-temporal datasets. We introduce BUP-ST20, a novel weakly labelled spatial-temporal dataset for sweet pepper tracking and segmentation captured on a robotic platform. BUP-ST20 contains 16,240 images from 275 sequences, each with bounding boxes, instance segmentation masks, and temporal identities.The dataset has weakly labelled training and validation sets, while the evaluation set includes 3810 frames with hand-labelled ground truth annotations
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Variations on the Author
Full text link“Variations on the Author” discusses two of Eduardo Coutinho’s recent films (Um Dia na Vida, from 2010, and Últimas Conversas, posthumously released in 2015) and their contribution to the general question of documentary authorship. The director’s filmography is characterized by a consistent yet self-effacing form of authorial self-inscription: Coutinho often features as an interviewer that rather than express opinions propels discourses; an interviewer that is good at listening. This mode of self-inscription characterizes him as an author who is not expressive but who is nonetheless markedly present on the screen. In Um Dia na Vida, however, Coutinho is completely absent form the image, while Últimas Conversas, on the contrary, includes a confessional prologue that moves the director from the margins to the center of his films. This article examines the ways in which these works stand out in the filmography of a director who offers new insights into the notion of cinematic authorship
Appropriate Similarity Measures for Author Cocitation Analysis
Full text linkWe provide a number of new insights into the methodological discussion about author cocitation analysis. We first argue that the use of the Pearson correlation for measuring the similarity between authors’ cocitation profiles is not very satisfactory. We then discuss what kind of similarity measures may be used as an alternative to the Pearson correlation. We consider three similarity measures in particular. One is the well-known cosine. The other two similarity measures have not been used before in the bibliometric literature. Finally, we show by means of an example that our findings have a high practical relevance.information science;Pearson correlation;cosine;similarity measure;author cocitation analysis
Precision Weed Management Enabled by Robotic and Robotics Vision
Full text linkIn recent decades, traditional crop and weed management has heavily relied on herbicides and mechanical weeding. These methods have caused significant environmental and agricultural challenges. Over 2 million tons of herbicides are used annually globally, raising concerns about food safety, environmental harm, and human health risks. Weed resistance to herbicides is a growing problem, with over 500 cases reported worldwide. Meanwhile, consumer demand for organic, chemical-free food pushes farmers to reduce agrochemical use while maintaining high yields. This situation highlights the urgent need for innovative, sustainable farming solutions.
This thesis explores precision agriculture technologies, focusing on biodiversityaware robotic systems for plant-level weeding in arable farms. We tried to address the limitations of conventional weed management, by proposing advanced robotic solutions using machine vision, deep learning, and autonomous navigation for sustainable and targeted interventions in the real world. The core innovation is centered on developing a novel precision weeding and crop-monitoring robot platform called BonnBot-I. This platform is equipped with advanced sensors and computational tools to conduct autonomous operations in diverse arable farming environments.
One of the main topics in agricultural autonomy is performing reliable autonomous navigation in cluttered farming environments with poor global localization accessibility like GPS. Considering the fact that still a large portion of the arable farms are not seeded using GPS-guided systems, integration of local observations-based navigation methods could relieve environmentally posed challenges for robots to achieve reliable navigation and minimize crop damage. Hence, we introduce a vision-based navigation approach that guides the BonnBot-I through rows of crops with different canopy types and cultivars relying only on the real-time camera data.
A central aim of this thesis is to establish a robust framework for developing robots capable of conducting precise, plant-specific weed and crop management in arable farms that feature a variety of cultivars and weed densities. Hence an accurate crop and weed monitoring system is needed to shape weeding strategies based on the presence of plant instances. To fulfill this requirement, BonnBot-I incorporates cutting-edge instance-based semantic segmentation and trackingvia-segmentation methods. Our approach enables the identification and tracking of individual plants in real time, categorizing them by species, size, growth stage, and precise location under actual field conditions. These advanced systems allow us to implement eco-friendly weeding strategies tailored to specific plants in real agricultural settings. This innovation enables plant-level prioritization and the execution of targeted interventions based on each plants unique needs using BonnBot-I’s novel weeding tool. BonnBot-Iis equipped with a specialized weeding tool, including independently controllable linear axes and spray nozzles, facilitating these selective interventions. This design enables BonnBot-I to perform highly precise applications, significantly reducing the need for agrochemicals and minimizing the environmental impact associated with traditional broadcast methods.
In conclusion, this thesis demonstrates how robotics and artificial intelligence (AI) can profoundly reshape the future of crop management through innovative biodiversity-aware and plant-specific weeding practices. By integrating advanced machine vision, deep learning, and autonomous navigation, BonnBot-I provides a unique approach to sustainable agriculture that respects biodiversity and prioritizes environmental health. Unlike traditional weeding methods that rely on uniform herbicide application or mechanical removal, which often harm surrounding crops and ecosystems, BonnBot-I offers precision interventions tailored to individual plants.In den letzten Jahrzehnten hat sich das traditionelle Unkraut- und Pflanzenmanagement stark auf Herbizide und mechanische Unkrautbekämpfung
verlassen. Diese Methoden haben erhebliche ökologische und landwirtschaftliche Herausforderungen mit sich gebracht. Weltweit werden jährlich über 2 Millionen Tonnen Herbizide eingesetzt, was Bedenken hinsichtlich der Lebensmittelsicherheit, Umweltschäden und Gesundheitsrisiken für den Menschen aufwirft. Die Resistenz von Unkräutern gegen Herbizide ist ein wachsendes Problem, mit über 500 gemeldeten Fällen weltweit. Gleichzeitig drängen Verbraucher auf organische, chemiefreie Lebensmittel, was Landwirte dazu zwingt, den Einsatz von Agrochemikalien zu reduzieren und gleichzeitig hohe Erträge aufrechtzuerhalten. Diese Situation verdeutlicht die dringende Notwendigkeit innovativer, nachhaltiger Lösungen für die Landwirtschaft.
Diese Dissertation untersucht Technologien der Präzisionslandwirtschaft, mit einem besonderen Fokus auf biodiversitätsbewusste robotische Systeme für pflanzenindividuelle Unkrautbekämpfung auf Ackerflächen. Wir haben versucht, die Einschränkungen konventioneller Unkrautmanagementmethoden zu überwinden, indem wir fortschrittliche robotische Lösungen unter Einsatz von maschinellem Sehen, Deep Learning und autonomer Navigation für nachhaltige und gezielte Interventionen in realen Anwendungen vorschlagen. Die zentrale Innovation konzentriert sich auf die Entwicklung einer neuartigen Plattform für Präzisionsunkrautbekämpfung und Pflanzenüberwachung namens BonnBot-I. Diese Plattform ist mit fortschrittlichen Sensoren und Rechenwerkzeugen ausgestattet, um autonome Operationen in verschiedenen Ackerbauumgebungen durchzuführen.
Ein Hauptthema der landwirtschaftlichen Autonomie ist die zuverlässige autonome Navigation in unübersichtlichen landwirtschaftlichen Umgebungen mit eingeschränkter globaler Lokalisierung, wie GPS. Angesichts der Tatsache, dass immer noch ein groSSer Teil der Ackerflächen nicht mit GPS-gestützten Systemen eingesät wird, könnte die Integration von Navigationsmethoden, die auf lokalen Beobachtungen basieren, dazu beitragen, Umweltprobleme zu lösen, die zuverlässige Navigation ermöglichen und Pflanzenschäden minimieren. Daher stellen wir einen visionbasierten Navigationsansatz vor, der den BonnBot-Idurch Pflanzenreihen mit unterschiedlichen Kronentypen und Kultivaren allein auf Basis von Echtzeitkameradaten führt.
Ein zentrales Ziel dieser Dissertation ist es, ein robustes Framework für die Entwicklung von Robotern zu etablieren, die in der Lage sind, präzise, pflanzenin dividuelle Unkraut- und Pflanzenpflege auf Ackerflächen mit einer Vielzahl von Kultivaren und Unkrautdichten durchzuführen. Dafür ist ein genaues Überwachungssystem für Pflanzen und Unkraut erforderlich, um Unkrautbekämpfungsstrategien auf Grundlage der vorhandenen Pflanzeninstanzen zu gestalten. Um dieses Ziel zu erreichen, integriert BonnBot-Ifortschrittliche instanzbasierte semantische Segmentierungs- und Tracking-via-Segmentierungs-Methoden. Unser Ansatz ermöglicht es, einzelne Pflanzen in Echtzeit zu identifizieren und zu verfolgen, sie nach Art, GröSSe, Wachstumsstadium und genauer Position unter realen Feldbedingungen zu kategorisieren. Diese fortschrittlichen Systeme erlauben es, umweltfreundliche Unkrautbekämpfungsstrategien umzusetzen, die auf spezifi sche Pflanzen in realen landwirtschaftlichen Umgebungen zugeschnitten sind. Diese Innovation ermöglicht die Priorisierung auf Pflanzenebene und die Durchführung gezielter Interventionen basierend auf den individuellen Bedürfnissen jeder Pflanze mithilfe des neuartigen Unkrautbekämpfungswerkzeugs von BonnBot-I.
BonnBot-I ist mit einem spezialisierten Unkrautbekämpfungswerkzeug ausgestattet, das unabhängig steuerbare Linearschienen und Sprühdüsen umfasst, um diese selektiven Interventionen zu erleichtern. Dieses Design ermöglicht es BonnBot-I, hochpräzise Anwendungen durchzuführen, den Bedarf an Agrochemikalien erheblich zu reduzieren und die mit herkömmlichen Methoden verbundenen Umweltauswirkungen zu minimieren. AbschlieSSend zeigt diese Dissertation, wie Robotik und künstliche Intelligenz (KI) die Zukunft des Pflanzenmanagements durch innovative biodiversitätsbewusste und pflanzenindividuelle Unkrautbekämpfungspraktiken grundlegend verändern können. Durch die Integration fortschrittlicher maschineller Bildverarbeitung, Deep Learning und autonomer Navigation bietet BonnBot-Ieinen einzigartigen Ansatz für nachhaltige Landwirtschaft, der die Biodiversität respektiert und die Umweltgesundheit priorisiert. Im Gegensatz zu herkömmlichen Unkrautbekämpfungsmethoden, die auf eine einheitliche Anwendung von Herbiziden oder mechanische Entfernung setzen und oft umliegende Pflanzen und Ökosysteme schädigen, bietet BonnBot-Ipräzise Eingriffe, die individuell auf einzelne Pflanzen abgestimmt sind
Efficient Semantic Scene Understanding for Mobile Robots
Full text linkOver the last few years, robots have been slowly making their way into our everyday lives. From robotic vacuum cleaners picking up after us already working in our homes, to the fleets of robo-taxis and self-driving vehicles lurking on the horizon, all of these robots are designed to operate in conjunction with, and in an environment designed for us, humans. This means that unlike traditional robots working in industrial settings where the world is designed around them, mobile robots need to acquire an accurate understanding of the surroundings in order to operate safely, and reliably. We call this type of knowledge about the surroundings of the robot semantic scene understanding. This understanding serves as the first layer of interpretation of the robot's raw sensor data and provides other tasks with useful and complete information about the status of the surroundings. These tasks include the avoidance of obstacles, the localization of the robot in the world, the mapping of an unknown environment for later use, the planning of trajectories, and the manipulation of objects in the scene, among others.
In this thesis, we focus on semantic scene understanding for mobile robots. As their mobility usually requires these robots to be powered by batteries, the key characteristics they require from perception algorithms are to be computationally, as well as energy efficient. Efficient means that the approach can exploit all the information available to it to run fast enough for the robot's online operation, both in power- as well as compute-constrained embedded computers. We approach this goal through three different avenues. First, in all of the algorithms presented in this thesis, we exploit background knowledge about the task we are trying to solve to make our algorithms fast to execute and at the same time, more accurate. Second, we instruct the approaches to exploit peculiarities of the particular sensor used in each application in order to make the processing more efficient. Finally, we present a software infrastructure that serves as an example of how to implement said scene understanding approaches on real robots, exploiting commercially available hardware accelerators for the task, and allowing for scalability. Because of this, every method presented in this thesis is capable of running faster than the frame rate of the sensor, both when using cameras or laser sensors.
All parts of this thesis have been published in proceedings of international conferences or as journal articles, undergoing a thorough peer-reviewing process. Furthermore, the work presented in this thesis resulted in the publication of a large-scale dataset and benchmark for the community to develop, share, and compare their semantic scene understanding approaches, as well as four open-source libraries for this task, using multiple sensor modalities
Robotic Vision for Precision Intervention in Horticulture
No full textStriving towards optimal sustainable agriculture systems to address the world’s growing demand for food, precision agriculture has emerged as a key strategy. In recent years robotic systems have gained remarkable capabilities to automate various agricultural tasks. Frequently, agricultural robots make use of vision-based sensors such as color (RGB) cameras coupled with advanced deep learning models to provide a fine-grained understanding of the environment. However, these robots are deployed in challenging conditions where current techniques fall short in terms of required performance to estimate suitable high-precision fine-grained plant-level information. Yet, opportunities exist to greatly enhance the quality of these vision-based approaches by employing robotic vision techniques that exploit not just the RGB camera information but also the estimated scene structure (depth) as well as coarse robot localization data.
This thesis explores the use of robotic vision to automate agricultural surveillance, particularly focusing on horticulture glasshouse systems. To achieve this we develop a robotic platform called PATHoBot and demonstrate how this is an enabling system for tasks such as crop monitoring, 3D panoptic fruit mapping, 4D registration, fruit volume, quantity and quality estimation, autonomous harvesting, and large-scale phenotyping in commercial glasshouses. We then demonstrate how rich robotic information, specifically relative motion plus scene geometry (e.g. depth), can be fused with state-of-the-art vision deep learning approaches to make them robust to real-world challenges, yielding highly accurate crop detection, tracking, and segmentation results.
PATHoBot is a crop monitoring robot designed for commercial glasshouses, equipped with a global multi-modal camera array for on-the-fly surveillance of vertical crops and a robotic arm for proximity monitoring and intervention tasks. We first show its utility by generating 3D crop maps and improving a tracking-via-segmentation fruit counting system by exploiting multi-modal spatial-temporal data it captures. We also propose methods to improve crop monitoring systems by explicitly incorporating spatial-temporal information. Combining scene geometry and how the robot moved, we estimate how extracted vision features had moved spatially over time, allowing us to directly incorporate these features into a DNN segmentation model. These approaches achieved improved performance and robustness to real-world conditions across horticulture and arable farming domains.
Finally, we introduce a 3D semantic scene understanding model capable of identifying individual fruits and tracking them through strong occlusions addressing key challenges in agricultural monitoring while achieving impressive performance. This system allows us to take object detections and jointly resolve geometry, robot pose, object instances, and even object identities (tracking objects) in a single approach. The output of PAg-NeRF provides a spatial-temporal consistent understanding of a field provided by deep learnt models. Our contributions show how robot spatial-temporal information and multimodal data can be exploited to improve the performance of DNN crop monitoring systems and expand their capabilities, in particular for horticulture domains. This has a direct impact on improving the crop decision-making process and automated intervention tasks, ultimately leading to advancement in sustainable food production practices. The approaches discussed in this thesis have associated peer-reviewed publications listed below. Furthermore, Our paper on explicitly incorporating spatial-temporal information into recurrent models received the best AgriRobotics paper award at IEEE’s Intelligent Robots and Systems conference 2022 (IROS 2022). Finally, the datasets and implementation of our novel monitoring methods have been publicly released to enable further research
Dispelling the Myths Behind First-author Citation Counts
Full text linkWe conducted a full-scale evaluative citation analysis study of scholars in the XML research field to explore just how different from each other author rankings resulting from different citation counting methods actually are, and to demonstrate the capability of emerging data and tools on the Web in supporting more realistic citation counting methods. Our results contest some common arguments for the continued
use of first-author citation counts in the evaluation of scholars, such as high correlations between author rankings by first-author citation counts and other citation
counting methods, and high costs of using more realistic citation counting methods that are not well-supported by the ISI databases. It is argued that increasingly available digital full text research papers make it possible for citation analysis studies to go beyond what the ISI databases have directly supported and to employ more
sophisticated methods
- …
