scispace - formally typeset
Search or ask a question
Author

Aneesh Chauhan

Bio: Aneesh Chauhan is an academic researcher from Wageningen University and Research Centre. The author has contributed to research in topics: Vocabulary & Robot learning. The author has an hindex of 10, co-authored 23 publications receiving 386 citations. Previous affiliations of Aneesh Chauhan include Spanish National Research Council & Technical University of Madrid.

Papers
More filters
Proceedings ArticleDOI
06 Jul 2014
TL;DR: A supervised learning approach for solving the tower detection and classification problem, where HOG (Histograms of Oriented Gradients) features are used to train two MLP (multi-layer perceptron) neural networks, shows that a learning-based approach is a promising technique for power line inspection.
Abstract: Inspection of power line infrastructures must be periodically conducted by electric companies in order to ensure reliable electric power distribution. Research efforts are focused on automating the power line inspection process by looking for strategies that satisfy the different requirements of the inspection: simultaneously detect transmission towers, check for defects, and analyze security distances. Following this direction, this paper proposes a supervised learning approach for solving the tower detection and classification problem, where HOG (Histograms of Oriented Gradients) features are used to train two MLP (multi-layer perceptron) neural networks. The first classifier is used for background-foreground segmentation, and the second multi-class MLP is used for classifying within 4 different types of electric towers. A thorough evaluation of the tower detection and classification approach has been carried out on image data from real inspections tasks with different types of towers and backgrounds. In the different evaluations, highly encouraging results were obtained. This shows that a learning-based approach is a promising technique for power line inspection.

90 citations

Proceedings ArticleDOI
27 May 2014
TL;DR: The proposed strategy, which is the combination of the tower detector and the tracker, is evaluated on videos from several real manned helicopter inspections and shows that the proposed strategy performs very well at detecting and tracking various types of electric towers in diverse environmental settings.
Abstract: This paper presents an approach towards autonomous aerial power line inspection. In particular, the presented work focuses on real-time autonomous detection, localization and tracking of electric towers. A strategy which combines classic computer vision and machine learning techniques, is proposed. A generalized detection and localization approach is presented, where a two-class multilayer perceptron (MLP) neural network was trained for Tower-Background classification. This MLP is applied over sliding windows for each camera frame until a tower is detected. The detection of a tower triggers the tracker. A hierarchical tracking methodology, especially designed for tracking towers in real-time, is presented. This methodology is based on the Hierarchical Multi-Parametric and Multi-Resolution Inverse Compositional Algorithm [1], and is proposed to be used for tracking and maintaining the tower in the field of view (FOV). The proposed strategy, which is the combination of the tower detector and the tracker, is evaluated on videos from several real manned helicopter inspections. Overall, the results show that the proposed strategy performs very well at detecting and tracking various types of electric towers in diverse environmental settings.

72 citations

Journal ArticleDOI
TL;DR: An object perception and perceptual learning system developed for a complex artificial cognitive agent working in a restaurant scenario that integrates detection, tracking, learning and recognition of tabletop objects and the Point Cloud Library is used in nearly all modules.

46 citations

Journal ArticleDOI
TL;DR: The results indicate that the robot’s representations are capable of incrementally evolving by correcting class descriptions, based on instructor feedback to classification results, which is comparable to those obtained by other authors.
Abstract: This paper addresses word learning for human–robot interaction. The focus is on making a robotic agent aware of its surroundings, by having it learn the names of the objects it can find. The human user, acting as instructor, can help the robotic agent ground the words used to refer to those objects. A lifelong learning system, based on one-class learning, was developed (OCLL). This system is incremental and evolves with the presentation of any new word, which acts as a class to the robot, relying on instructor feedback. A novel experimental evaluation methodology, that takes into account the open-ended nature of word learning, is proposed and applied. This methodology is based on the realization that a robot’s vocabulary will be limited by its discriminatory capacity which, in turn, depends on its sensors and perceptual capabilities. The results indicate that the robot’s representations are capable of incrementally evolving by correcting class descriptions, based on instructor feedback to classification results. In successive experiments, it was possible for the robot to learn between 6 and 12 names of real-world office objects. Although these results are comparable to those obtained by other authors, there is a need to scale-up. The limitations of the method are discussed and potential directions for improvement are pointed out.

45 citations

01 Jan 2007
TL;DR: In this article, a lifelong learning system based on one-class learning was developed to make a robotic agent aware of its surroundings, by having it learn the names of the objects it can find.
Abstract: This paper addresses word learning for human–robot interaction. The focus is on making a robotic agent aware of its surroundings, by having it learn the names of the objects it can find. The human user, acting as instructor, can help the robotic agent ground the words used to refer to those objects. A lifelong learning system, based on one-class learning, was developed (OCLL). This system is incremen tal and evolves with the presentation of any new word, which acts as a class to the robot, relying on instructor feedback. A novel experimental evaluation methodology, that takes into account the open-ended nature of word learning, is proposed and applied. This methodology is based on the realization that a robot’s vocabulary will be limited by its discriminatory capacity which, in turn, depends on its sensors and perceptual capabilities. The results indicate that the robot’s representations are capable of incrementally evolving by correcting class descriptions, based on instructor feedback to classification results. In succes sive experiments, it was possible for the robot to learn between 6 and 12 names of real-world office objects. Although these results are comparable to those obtained by other authors, there is a need to scale-up. The limitations of the method are discussed and potential directions for improvement are pointed out.

37 citations


Cited by
More filters
Book
26 Aug 2021
TL;DR: The use of unmanned aerial vehicles (UAVs) is growing rapidly across many civil application domains, including real-time monitoring, providing wireless coverage, remote sensing, search and rescue, delivery of goods, security and surveillance, precision agriculture, and civil infrastructure inspection.
Abstract: The use of unmanned aerial vehicles (UAVs) is growing rapidly across many civil application domains, including real-time monitoring, providing wireless coverage, remote sensing, search and rescue, delivery of goods, security and surveillance, precision agriculture, and civil infrastructure inspection. Smart UAVs are the next big revolution in the UAV technology promising to provide new opportunities in different applications, especially in civil infrastructure in terms of reduced risks and lower cost. Civil infrastructure is expected to dominate more than $45 Billion market value of UAV usage. In this paper, we present UAV civil applications and their challenges. We also discuss the current research trends and provide future insights for potential UAV uses. Furthermore, we present the key challenges for UAV civil applications, including charging challenges, collision avoidance and swarming challenges, and networking and security-related challenges. Based on our review of the recent literature, we discuss open research challenges and draw high-level insights on how these challenges might be approached.

901 citations

Journal ArticleDOI
TL;DR: The review shows that most previous studies have concentrated on the mapping and analysis of network components, and more attention should be given to an integrated use of various data sources to benefit from the various techniques in an optimal way.
Abstract: To secure uninterrupted distribution of electricity, effective monitoring and maintenance of power lines are needed This literature review article aims to give a wide overview of the possibilities provided by modern remote sensing sensors in power line corridor surveys and to discuss the potential and limitations of different approaches Monitoring of both power line components and vegetation around them is included Remotely sensed data sources discussed in the review include synthetic aperture radar (SAR) images, optical satellite and aerial images, thermal images, airborne laser scanner (ALS) data, land-based mobile mapping data, and unmanned aerial vehicle (UAV) data The review shows that most previous studies have concentrated on the mapping and analysis of network components In particular, automated extraction of power line conductors has achieved much attention, and promising results have been reported For example, accuracy levels above 90% have been presented for the extraction of conductors from ALS data or aerial images However, in many studies datasets have been small and numerical quality analyses have been omitted Mapping of vegetation near power lines has been a less common research topic than mapping of the components, but several studies have also been carried out in this field, especially using optical aerial and satellite images Based on the review we conclude that in future research more attention should be given to an integrated use of various data sources to benefit from the various techniques in an optimal way Knowledge in related fields, such as vegetation monitoring from ALS, SAR and optical image data should be better exploited to develop useful monitoring approaches Special attention should be given to rapidly developing remote sensing techniques such as UAVs and laser scanning from airborne and land-based platforms To demonstrate and verify the capabilities of automated monitoring approaches, large tests in various environments and practical monitoring conditions are needed These should include careful quality analyses and comparisons between different data sources, methods and individual algorithms

350 citations

Journal ArticleDOI
TL;DR: A novel deep convolutional neural network (CNN) cascading architecture for performing localization and detecting defects in insulators is proposed, which uses a CNN based on a region proposal network to transform defect inspection into a two-level object detection problem.
Abstract: As the failure of power line insulators leads to the failure of power transmission systems, an insulator inspection system based on an aerial platform is widely used. Insulator defect detection is performed against complex backgrounds in aerial images, presenting an interesting but challenging problem. Traditional methods, based on handcrafted features or shallow-learning techniques, can only localize insulators and detect faults under specific detection conditions, such as when sufficient prior knowledge is available, with low background interference, at certain object scales, or under specific illumination conditions. This paper discusses the automatic detection of insulator defects using aerial images, accurately localizing insulator defects appearing in input images captured from real inspection environments. We propose a novel deep convolutional neural network (CNN) cascading architecture for performing localization and detecting defects in insulators. The cascading network uses a CNN based on a region proposal network to transform defect inspection into a two-level object detection problem. To address the scarcity of defect images in a real inspection environment, a data augmentation method is also proposed that includes four operations: 1) affine transformation; 2) insulator segmentation and background fusion; 3) Gaussian blur; and 4) brightness transformation. Defect detection precision and recall of the proposed method are 0.91 and 0.96 using a standard insulator dataset, and insulator defects under various conditions can be successfully detected. Experimental results demonstrate that this method meets the robustness and accuracy requirements for insulator defect detection.

324 citations

Journal ArticleDOI
TL;DR: Bloom as discussed by the authors argues that children learn words via cognitive abilities that already exist for other pur- poses, such as the ability to infer others' intentions and acquire concepts, and an appreciation of syntactic In structure.
Abstract: How Children Learn the Meanings of Words by Paul Bloom. Cambridge, Massachusetts: MIT Press, 2000, xii+300 pp. Reviewed by Masahiko Minami San Francisco State University How do children learn the meanings of words? In his new book, Paul Bloom examines a variety of issues associated with children's word learning, a process intricately connected with other aspects of language acquisition. Bloom claims that children learn words via cognitive abilities that already exist for other pur- poses, such as the ability to infer others' intentions, the ability to acquire concepts, and an appreciation of syntactic In structure. Bloom's book provides a series of el- egant and convincing arguments concerning how children learn words. briefly Chapter First Words, Bloom lays out the plan for the book and describes issues surrounding the overall topic. In Chapter 2 the author explores fast mapping, in which children make a quick guess about a word's denotation on the basis of limited experience. Chapter 3, Theory of Mind, deals with a wide range of topics, including the listener's ability to determine the references made by his or her interlocutor's choice of words; here also, Bloom investigates children's appreciation of the mental states of others, through which children acquire lexical items (and syntax as well) by means of associative acquire are nouns, learning. Because the majority of words that children initially Bloom gives special treatment to nouns and pronouns: Common nouns are discussed in Chapter 4, and pronouns and proper names are dealt with in Chapter 5. In Chapter 6, Concepts and Categories, Bloom extends his analysis to the conceptual foundations of word learning. In Chapter Naming Representations, he discusses a case study important to any theory of concepts and naming visual representations. From here, Bloom moves to other parts of speech: In Chapter 8, Learning Words through Linguistic Context, he offers an account of how chil- dren learn verbs and adjectives, as the development of syntactic abilities cannot be dissociated from the development of lexical abilities. Chapter 9 deals with learn the how we numbers and Chapter 10 with how the words we learn affect our mental life. In Chapter 11, Final Words, Bloom provides a brief summary and some general remarks. Throughout the book, the author weaves in ideas pro- posed by such linguists, psychologists, and philosophers as B. F. Skinner, Noam Chomsky, and Jean Piaget, who, through different lenses, have closely observed words for and analyzed how human beings develop and around them. issue long relevant to how they conceptualize the world As with most language acquisition texts, Bloom makes early reference to an human development: the nature/nurture debate. These alter- ISSN 1050-4273 Vol. 12 Issues in Applied Linguistics © 2001, Regents of the University of California No.

308 citations