Détail de l'auteur
Auteur Li |
Documents disponibles écrits par cet auteur (1)
Ajouter le résultat dans votre panier Affiner la recherche Interroger des sources externes
Titre : Semi-automatic road extraction from satellite and aerial images Type de document : Thèse/HDR Auteurs : Li, Auteur Editeur : Zurich : Institut für Geodäsie und Photogrammetrie IGP - ETH Année de publication : 1997 Collection : IGP Mitteilungen, ISSN 0252-9335 num. 061 Importance : 164 p. Format : 21 x 30 cm ISBN/ISSN/EAN : 978-3-906513-96-6 Note générale : Bibliographie Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Photogrammétrie numérique
[Termes IGN] extraction semi-automatique
[Termes IGN] fonction spline
[Termes IGN] image satellite
[Termes IGN] photographie aérienne
[Termes IGN] réseau routier
[Termes IGN] traitement d'imageIndex. décimale : 33.30 Photogrammétrie numérique Résumé : (auteur) Of all parts in the process of GIS data generation from satellite and aerial images, the actual mapping phase is one of the most time consuming and expensive procedure. Research is therefore increasingly focusing on the development of efficient methods to automatically extract man-made objects like houses and roads from digital images. As fully automatic methods for mapping are still far out of reach, semi-automatic methods for feature extraction that interact with a human operator are considered to be a good compromise, combining the mensuration speed and accuracy of a computer algorithm with the interpretation skills of a human operator. This dissertation deals with semi-automatic linear feature extraction from digital images for GIS data capture, where the identification task is performed manually on a single image, while a special automatic digital module performs the high precision line tracking. More specifically, a human operator is used to identify the object from an on-screen display of a digital image, selects the particular class this object belongs to and provides some very few seed points coarsely distributed. This is done through activation of a mouse in a convenient interactive graphics-image user interface. Subsequently, with these seed points as an approximation of the position and shape, the linear feature will be extracted precisely and automatically by either a dynamic programming approach or LSB-Snakes. These techniques can be used in a monoplotting mode, which combines one image with its underlying DTM. The LSB-Snakes approach is also implemented in a multi-image mode, which uses multiple images simultaneously and provides for a robust and mathematically sound full 3-D approach. Firstly, we propose a semi-automatic road extraction scheme which combines the wavelet decomposition for road sharpening and a model driven linear feature extraction algorithm based on dynamic programming. With a wavelet transform interesting image structures can be enhanced and a multiresolution representation can be obtained by selection of a special wavelet. We have built a particular wavelet for road sharpening, which has been implemented as a fast pyramidal algorithm. In the model driven feature extraction scheme, a road is represented by a generic road model with six photometric and geometric properties. This model is formulated by some constraints and a merit function which embodies a notion of the "best road segment", and evaluated by a "time-delayed" dynamic programming algorithm. In order to reduce the computational complexity, a strategy of dynamic vertex insertion and deletion is developed. In such a way, even a long road segment can be handled efficiently. The mathematical foundation and issues relating to its practical implementation are discussed in detail. This approach has been applied very successfully to extract complete road structures from single SPOT scenes and small scale aerial images. Experimental results show that the algorithm is very robust in case of gaps and other distortions because of use of global photometric information and geometric constraints. Then, a general approach for linear feature extraction with active contour models is investigated. In general, the Snakes or active contour models feature extraction algorithm integrates both photometric and geometric constraints, with an initial estimate of the location of the feature, by an integral measure referred to as the total energy of Snakes. The local minimum in this energy defines the feature of interest. In this dissertation, active contour models are approximated by B-spline curves and formulated in terms of a combined least squares adjustment. The observation equations consist of the equations formulating the matching of a generic object model and image data, and those that express the geometric constraints and operator-given seed points. We call this novel concept of Snakes "LSB-Snakes" (Least Squares B-spline Snakes). LSB-Snakes considerably improve active contour models by using three new elements: (i) the possibility for internal quality control through computation of the covariance matrix of the estimated parameters, (ii) the exploitation of any a priori known geometric and photometric information to constrain the solution and (iii) the simultaneous use of any number of images. The least squares approach allows for precision and reliability assessment of the estimated 3-D feature via covariance matrix evaluation. This is in clear contrast to conventional methods of Snakes, which due to their particular theoretical background and formulation, do not provide any measures for the qualitative control of their results. Instead of a set of points on the feature, a B-spline representation of the linear feature is estimated. Through the connection of image and object space, assuming that the interior and exterior orientation of the sensors are known, any number of images can be simultaneously accommodated and the feature can be extracted in a 2-D as well as in a fully 3-D mode. Thus blunders in image data, like occlusions, can be controlled very well. At the same time, LSB-Snakes can be considered a new application and extension of the least squares template matching (LSM) techniques. Our LSB-Snakes concept is not restricted to road extraction. Other linear features, e.g. edges, can be modelled and extracted. In fact, anything which can be geometrically modelled by B-splines can be handled. This makes it a powerful general concept for semiautomated feature extraction, not only for the processing of aerial and space images, but also for a variety of close-range (machine vision) applications. The results obtained so far are very encouraging. Further studies will make use of more extensive data sets and will focus on the quality assessment and automated performance evaluation. Numéro de notice : 66887 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse étrangère DOI : 10.3929/ethz-a-001766570 En ligne : http://dx.doi.org/10.3929/ethz-a-001766570 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=61672 Exemplaires(2)
Code-barres Cote Support Localisation Section Disponibilité 66887-01 33.30 Livre Centre de documentation Photogrammétrie - Lasergrammétrie Disponible 66887-02 33.30 Livre Centre de documentation Photogrammétrie - Lasergrammétrie Disponible