Papers by Adam Winstanley
Automatic structuring (feature coding and object recognition) of topographic data, such as that d... more Automatic structuring (feature coding and object recognition) of topographic data, such as that derived from air survey or raster scanning large-scale paper maps, requires the classification of objects such as buildings, roads, rivers, fields and railways. The recognition of objects in computer vision is largely based on the matching of descriptions of shapes. Fourier descriptors, moment invariants, boundary chain coding and scalar descriptors are methods that have been widely used and have been developed to describe shape irrespective of position, orientation and scale. The applicability of the above four methods to topographic shapes is described and their usefulness evaluated.
1st International Conference on …, Jan 1, 2000
The automatic classification of geometric topographical data into object types (and/or feature co... more The automatic classification of geometric topographical data into object types (and/or feature codes) can partially be done through recognition of shape [Keyes & Winstanley 2000]. Performance can be improved by extending the classification mechanism with contextual information. This improves the accuracy of automatic classification because we can frequently resolve ambiguous data by examining its context to provide evidence for category membership (thus informally, for example, we can say that a square on a map is more likely to depict a house if it is near a road.).
Citeseer
The Ordnance Survey has traditionally recorded the large-scale topography of Britain as Cartesian... more The Ordnance Survey has traditionally recorded the large-scale topography of Britain as Cartesian co-ordinatebased point, line and text label features within the tile-based Land-Line® Database. Under their Digital National Framework™ (DNF™) project, this data has been re-engineered into a topologically structured format known as OS MasterMap™ [Ordnance Survey]. This required the modelling of the area features enclosed by the line data as polygon objects. This new polygon-enriched data can be provided seamlessly for pre-defined areas and by theme. Each feature is assigned a unique Topographic Identifier (TOID™) number, allowing for the easy updating of a data holding, and the association of any topographic feature with external information. Each point object is classified with a particular feature code, such as post-box or bench-mark; likewise, a line feature could be labelled as a building outline or a public road edge. The feature-coding of polygons is the most difficult requirement of the DNF, as it requires the inferring of information that is not present in the Land-Line data. Properly classified area features greatly add to the intelligence of the resulting OS MasterMap data, allowing a myriad of valuable analyses to be carried out. The OS has accomplished high quality polygon classification semi-automatically, largely by examining the feature codes of the lines that bound each polygon. Using novel feature-coding techniques, the accuracy can be further improved.
Synchronised Encoding of GPS NMEA Messages onto High Definition Video Streams
International Machine Vision and Image Processing Conference (IMVIP 2007), 2007
Consumer awareness and retail penetration of high definition (HD) technologies has increased dram... more Consumer awareness and retail penetration of high definition (HD) technologies has increased dramatically over the last few years. This has led not only to increased interest and expectation of these technologies but also affordability of equipment. Global Positioning Systems (GPS) have also seen a very dramatic increase in consumer exposure and affordability with in car navigation systems, as an example, becoming more common place. The integration of HD Video (HDV) and GPS offers new research and development directions in many of the application areas that spatial video has played a role in to date. These include many mapping and navigation application areas such vehicular route corridor surveys, aerial and marine studies and location based services (LBS). In the past, work on integrating these systems has typically involved technologies that have lower quality spatial accuracy and much lower image resolutions, which has restricted spatial video to enhanced visualisation roles in geographical information systems (GIS). A number of research project have developed such system integration, while propriety commercial applications have also successfully deployed survey and analysis products in this area. Many application areas, especially in GIS, will benefit from using integrated HDV and GPS National Marine Electronics Association (NMEA) spatial data. The high resolution progressive scan nature of HDV will facilitate many research avenues that apply image processing techniques to GIS analysis and operations. A pilot project has begun by collecting two and half hours of route corridor surveys using a vehicle mounted system where GPS NMEA data and HDV progressive scan 1280 times 720 p format data sets have been simultaneously recorded. In integrating these data sets a number of difficulties have been encountered. These include the specific data integration methods that are possible and time synchronization problems. Existing GPS and Video systems integration include freq- - uency shift modulated audio encoded systems, however our approach is to investigate both MPEG-2 and MPEG-4 formats for Spatial data integration and the possibility of HDV in a MPEG-7 format which supports video and audio object metadata fields. Synchronising 1 Hz GPS spatial data signals with 24 to 60 Hz HDV frames is a problem affecting the positional accuracy of the location information with the HDV image location. Following on from this, the latency involved in signal propagation through an integrated capturing system will present some more technical problems.
Automatic structuring (feature coding and object recognition) of topographic data, such as that d... more Automatic structuring (feature coding and object recognition) of topographic data, such as that derived from air survey or raster scanning large-scale paper maps, requires the classification of objects such as buildings, roads, rivers, fields and railways. The recognition of objects in computer vision is largely based on the matching of descriptions of shapes. Fourier descriptors, moment invariants, boundary chain coding and scalar descriptors are methods that have been widely used and have been developed to describe shape irrespective of position, orientation and scale. The applicability of the above four methods to topographic shapes is described and their usefulness evaluated.
Automatic structuring (feature coding and object recognition) of topographic data, such as that d... more Automatic structuring (feature coding and object recognition) of topographic data, such as that derived from air survey or raster scanning large-scale paper maps, requires the classification of objects such as buildings, roads, rivers, fields and railways. The recognition of objects is largely based on the matching of descriptions of shapes. Fourier descriptors, moment invariants, boundary chain coding and scalar descriptors
Use of Graph Databases in Tourist Navigation Application
Lecture Notes in Computer Science, 2014
Lecture Notes in Computer Science, 2014
Big Data encompasses collection, management, processing and analysis of the huge amount of data t... more Big Data encompasses collection, management, processing and analysis of the huge amount of data that varies in types and changes with high frequency. Often data component of Big Data has a positional component as an important part of it in various forms, such as postal address, Internet Protocol (IP) address and geographical location. If the positional components in Big Data extensively used in storage, retrieval, analysis, processing, visualization and knowledge discovery (geospatial Big Data) the Big Data systems need certain type of techniques and algorithms for management, analytics and sharing.
Spatial Uncertainty Management in Pedestrian Navigation
Lecture Notes in Geoinformation and Cartography, 2014
International Journal of Navigation and Observation, 2014
Vehicle navigation systems usually simply function by calculating the shortest fastest route over... more Vehicle navigation systems usually simply function by calculating the shortest fastest route over a road network. In contrast, pedestrian navigation can have more diverse concerns. Pedestrians are not constrained to road/path networks; their route may involve going into buildings (where accurate satellite locational signals are not available) and they have different priorities, for example, preferring routes that are quieter or more sheltered from the weather. In addition, there are differences in how people are best directed: pedestrians noticing landmarks such as buildings, doors, and steps rather than junctions and sign posts. Landmarks exist both indoors and outdoors. A system has been developed that uses quick response (QR) codes affixed to registered landmarks allowing users to localise themselves with respect to their route and with navigational instructions given in terms of these landmarks. In addition, the system includes images of each landmark helping users to navigate visually in addition to through textual instructions and route maps. The system runs on a mobile device; the users use the device's camera to register each landmark's QR code and so update their position (particularly indoors) and progress through the route itinerary.
Removing the texture feature response to object boundaries
Computer Vision Theory and Applications, 2007
Delivering Real-Time Bus Tracking Information on Mobile Devices
Communications in Computer and Information Science, 2011
The recent technological advances in mobile communication, computing and geo-positioning technolo... more The recent technological advances in mobile communication, computing and geo-positioning technologies have made real-time transit vehicle information systems an interesting application area. In this paper we present a transit application system which displays the transit information on an OpenStreetMap (OSM) web interface and delivers this information on the Google Android mobile device. The content is in the form of predicted arrival/departure times for buses at user-selectable geographic locations within a transit region. This application uses the real-time information such as current location and timestamp of both bus and users to provide bus route information. The public interface provides a graphical view which used to display and updates the vehicle locations and to allow users to see routes, stops and moving buses. The mobile device provides the user with the expected arrival/departure time of the next bus at the bus stop based on the user’s current position.
Efficient Online Sharing of Geospatial Big Data Using NoSQL XML Databases
2013 Fourth International Conference on Computing for Geospatial Research and Application, 2013
ABSTRACT form only given: Today a huge amount of geospatial data is being created, collected and ... more ABSTRACT form only given: Today a huge amount of geospatial data is being created, collected and used more than ever before. The ever increasing observations and measurements of geo-sensor networks, satellite imageries, point clouds from laser scanning, geospatial data of Location Based Services (LBS) and location-based social networks has become a serious challenge for data management and analysis systems. Traditionally, Relational Database Management Systems (RDBMS) were used to manage and to some extent analyze the geospatial data. Nowadays these systems can be used in many scenarios but there are some situations when using these systems may not provide the required efficiency and effectiveness. More specifically when the geospatial data has high volume, high frequency of change (in both data content and data structure) and variety of structures, the conventional data storage systems cannot provide needed efficiency in online systems in terms of performance and scalability. In these situations, NoSQL solutions can provide the efficiency necessary for applications using geospatial data. This paper provides an overview of the characteristics of geospatial big data, possible solutions for managing and processing them. Then the paper provides an overview of the major types of NoSQL solutions, their advantages and disadvantages and the challenges they present in managing geospatial big data. Then the paper elaborates on serving geospatial data using standard geospatial web services with a NoSQL XML database as a backend.
Watershed Segmentation Using a Multiscale Ramp Edge Merging Strategy
International Machine Vision and Image Processing Conference (IMVIP 2007), 2007
For the watershed segmentation algorithm to be successful it must be implemented on a realistic g... more For the watershed segmentation algorithm to be successful it must be implemented on a realistic gradient image. In most watershed implementations, gradients are extracted using an operator optimal for ideal step edges. However, image edges are never ideal steps and more ...
A Pedestrian Navigation System Using Navigation Sentence Based on XML
2009 International Conference on Information Management, Innovation Management and Industrial Engineering, 2009
ABSTRACT Pedestrian navigation system has been widely used in many fields. In order to offer navi... more ABSTRACT Pedestrian navigation system has been widely used in many fields. In order to offer navigation information intelligible for pedestrian, we provide navigation sentences to users, which can be easily displayed on cellular phones with small display area. In addition, the navigation sentences can be converted into voice, which is useful for users in some special situations or with visually impairment. Navigation sentences can separately help pedestrians to go to the destination, or can be combined with electronic map to provide more intuitive navigation information. Compared with guidance maps, navigation sentences are easy to transmit and manage. The navigation sentences are described by XML (extensible markup language) and can be created on the server automatically based on the users' position and destination. This paper will discuss the structure of the pedestrian navigation system using navigation sentences. The structure and generating process of navigation sentences are also described.
An Extended System for Labeling Graphical Documents Using Statistical Language Models
Lecture Notes in Computer Science, 2006
This paper describes a proposed extended system for the recognition and labeling of graphical obj... more This paper describes a proposed extended system for the recognition and labeling of graphical objects within architectural and engineering documents that integrates Statistical Language Models (SLMs) with shape classifiers. Traditionally used for Natural Language ...
Feature Selection Based on Run Covering
Lecture Notes in Computer Science, 2006
Abstract. This paper proposes a new feature selection algorithm. First, the data at every attribu... more Abstract. This paper proposes a new feature selection algorithm. First, the data at every attribute are sorted. The continuously distributed data with the same class labels are grouped into runs. The runs whose length is greater than a given threshold are selected as valid runs, ...
Pattern Recognition, 2008
The goal of feature selection is to find the optimal subset consisting of m features chosen from ... more The goal of feature selection is to find the optimal subset consisting of m features chosen from the total n features. One critical problem for many feature selection methods is that an exhaustive search strategy has to be applied to seek the best subset among all the possible n m feature subsets, which usually results in a considerably high computational complexity. The alternative suboptimal feature selection methods provide more practical solutions in terms of computational complexity but they cannot promise that the finally selected feature subset is globally optimal. We propose a new feature selection algorithm based on a distance discriminant (FSDD), which not only solves the problem of the high computational costs but also overcomes the drawbacks of the suboptimal methods. The proposed method is able to find the optimal feature subset without exhaustive search or Branch and Bound algorithm. The most difficult problem for optimal feature selection, the search problem, is converted into a feature ranking problem following rigorous theoretical proof such that the computational complexity can be greatly reduced. The proposed method is invariant to the linear transformation of data when a diagonal transformation matrix is applied. FSDD was compared with ReliefF and mrmrMID based on mutual information on 8 data sets. The experiment results show that FSDD outperforms the other two methods and is highly efficient. ᭧
Spatial Characteristics of Walking Areas for Pedestrian Navigation
2009 Third International Conference on Multimedia and Ubiquitous Engineering, 2009
Abstract We consider how to handle open walking areas in pedestrian navigation applications. Wal... more Abstract We consider how to handle open walking areas in pedestrian navigation applications. Walking areas are a key type of feature which is a major difference between networks of pedestrian navigation and that of road vehicle navigation. Up to recently, little work has ...
Structuring vector maps using Computer vision technologies
Uploads
Papers by Adam Winstanley