Road Network Detection Research Articles

Road Network Detection Using Probabilistic and Graph Theoretical Methods

Road network detection from very high resolution satellite and aerial images has diverse and important usage areas such as map generation and updating. Although an expert can label road pixels in a given image, this operation is prone to errors and quite time consuming. Therefore, an automated system is needed to detect the road network in a given satellite or aerial image in a robust manner. In this paper, we propose such a novel system. Our system has three main modules: probabilistic road center detection, road shape extraction, and graph-theory-based road network formation. These modules may be used sequentially or interchangeably depending on the application at hand. To show the strengths and weaknesses of our system, we tested it on several very high resolution satellite (Geoeye, Ikonos, and QuickBird) and aerial image sets. We compared our system with the ones existing in the literature. We also tested the sensitivity of our system to different parameter values. Obtained results indicate that our system can be used in detecting the road network on such images in a reliable and fast manner.

Delineation and geometric modeling of road networks

In this work we present a novel vision-based system for automatic detection and extraction of complex road networks from various sensor resources such as aerial photographs, satellite images, and LiDAR. Uniquely, the proposed system is an integrated solution that merges the power of perceptual grouping theory (Gabor filtering, tensor voting) and optimized segmentation techniques (global optimization using graph-cuts) into a unified framework to address the challenging problems of geospatial feature detection and classification. Firstly, the local precision of the Gabor filters is combined with the global context of the tensor voting to produce accurate classification of the geospatial features. In addition, the tensorial representation used for the encoding of the data eliminates the need for any thresholds, therefore removing any data dependencies. Secondly, a novel orientation-based segmentation is presented which incorporates the classification of the perceptual grouping, and results in segmentations with better defined boundaries and continuous linear segments. Finally, a set of gaussian-based filters are applied to automatically extract centerline information (magnitude, width and orientation). This information is then used for creating road segments and transforming them to their polygonal representations.

Automatic detection of main road network in dense urban area using microwave SAR images

A hybrid method of edge detection of radar images and classification of fused infrared and microwave SAR images is developed for automatic detection of the main road network in a dense urban area. Fusion of the infrared and microwave SAR images effectively reduces the noise in the edge detection and also effectively removes possible confusion with non-road objects, i.e. linearly featured rivers in the classification. Possible road candidates are further processed using the morphological thinning algorithm. Road linking and extension are carried out hierarchically according to the characteristics of the main roads as well as their connections. Finally, the main road network is successfully established from the SAR image. The approach is effective and accurate in detecting road segments and linking without complicated mathematic models, e.g. the Snake or Genetic Algorithms. As an example, using the ERS-2 SAR image data, automatic detection of the main road network in Shanghai Pudong area is presented.

Urban Road Network Extraction from Spaceborne SAR Image

A two-step method is developed for the extraction of road network from spaceborne SAR image: road candidates detection and connection. In the road candidates detection, classification of fused infrared and microwave SAR images effectively reduces the noise in the edge detection, and also well removes possible confusion of no-road objects, i.e. linearly-featured rivers in the edge image. Possible road candidates are further processed using the morphological thinning algorithm. Road candidates connection is carried out hierarchically according to road models we established. Finally, the main road network is established from the SAR image successfully. As an example, using the ERS-2 SAR image data, automatic detection of main road network in Shanghai Pudong area is presented.

Road network detection from SPOT imagery for updating geographical information systems in the rural–urban fringe

Abstract Visual interpretation of high-resolution satellite data has been useful for mapping linear features, such as roads and updating land-use changes. However, it would be beneficial to map new road networks digitally from satellite data to update digital databases using semi-automated techniques. In this paper, an algorithm called Gradient Direction Profile Analysis (GDPA) is used to extract road networks digitally from SPOT High Resolution Visible (HRV) panchromatic data. The roads generated are compared with a visual interpretation of the SPOT HRV multispectral and panchromatic data. The technique is most effective in areas where road development is relatively recent. This is due to the spectral consistency of new road networks. As new road networks are those of most interest to the land manager, this is a useful technique for updating digital road network files within a geographical information system of urban areas.

Heuristics for intermediate level road finding algorithms

This paper deals with detection of road networks in aerial imagery. In natural images, roadlike segments extracted by low level operators are highly fragmented. The main focus of this work is the development of an intermediate processing stage that addresses the task of partitioning and connecting the roadlike fragments. Heuristics based upon generally observed properties of roadlike features are used to cluster and integrate segments that appear perceptually continuous. The segments are represented as lists and processed symbolically. The proposed methodology is exemplified using three data sets: a suite of four synthetic segment sets which were used to test the sensitivity of the algorithm to thresholds and noise; a circularly polarized millimeter wave (TABILS 5) radar pseudo image; and a thematic mapper (LANDSAT) image.