Mahamadou Idrissa

Texture classification using Gabor filters

Abstract An unsupervised texture classification scheme is proposed in this paper. The texture features are based on the image local spectrum which is obtained by a bank of Gabor filters. The fuzzy clustering algorithm is used for unsupervised classification. In many applications, this algorithm depends on assumptions made about the number of subgroups present in the data. Therefore we discuss ideas behind cluster validity measures and propose a method for choosing the optimal number of clusters.

Detecting urbanization changes using SPOT5

An automatic system to estimate the urbanization changes on the Belgian territory, using SPOT5 images and the National Geographic Institute vectorial database is proposed. The images and the vectorial data are first co-registered. Then, the vectorial database is projected and dilated to produce a mask representing the old status of the database. On the other hand, a fusion of two classification processes on the images enables to extract the built-up area and the communication network, providing a mask representing the actual state of the urbanization in the zone. The comparison between the two masks gives a coarse information of the changes.

Digital terrain models derived from digital surface model uniform regions in urban areas

Digital terrain models (DTMs) are of significant interest for applications such as environment planning, flood risk assessment or building detection. A digital surface model (DSM) can be obtained efficiently in both time and cost from light detection and ranging (lidar) acquisition or from digital photogrammetry with aerial or satellite stereoscopic imagery. A DTM can be derived from a DSM if the distinction between ground and non-ground pixels can be automated. We propose in this article a new automatic DSM-to-DTM transform targeting urban areas. Our approach segments the DSM twice: first to get large uniform regions normally corresponding to the road network and attached town squares, and second to obtain smoother areas. Smoother DSM areas overlapping the large regions are selected to populate the DTM, which is then completed by a hierarchical interpolation procedure. As a refinement step, unused smoother regions lying under this DTM are added to create, after interpolation, the final DTM. This approach is positioned relative to the literature about segmentation-based lidar ground filtering. The procedure was developed for DSM rasters. Since the DTM extraction is intended to be applied to large images, special attention was devoted to optimize image processing tasks relative to memory usage and execution time. The proposed development was integrated in a building detection procedure and validated qualitatively in the context of a benchmark on urban object detection of the International Society for Photogrammetry and Remote Sensing (ISPRS). It was also applied to Brussels data for which lidar DTMs are available. The DTM comparison supports the correctness of our solution although difficulties may be encountered in off-terrain regions surrounded by higher regions and some interiors of city blocks. A final test on a rural and peri-urban scene opens positive perspectives for scenes more general than urban areas.

A multiresolution-MRF approach for stereo dense disparity estimation

We present a method that reduces the computational cost of the MRF-based stereo algorithm and increases the quality of the final disparity map. In a first step, using window-based method we compute successive disparity maps at different resolutions by varying the correlation window size, in order to estimate for each pixel the set of most probable disparity values. Thus, by replacing the initial disparity range — which may exceed hundreds of pixels for some applications — by the small set of valid disparities, we increase the probability of choosing the right value for each pixel and thus speed up the MRF optimization process.

Generic epipolar resampling method for perspective frame camera and linear push-broom sensor

Epipolar rectification aims at resampling stereoscopic images so that conjugate points are located along the same horizontal x-axis. For most stereoscopic digital surface model production algorithms, this rectification is a necessary preprocessing task because it allows reducing the 2D matching to a 1D matching problem. Most epipolar rectification algorithms exist in literature and are mainly for a given type of sensor (frame camera or push-broom sensor). In this paper, we present a generic method for epipolar rectification. This means that the method is applicable for both frame and push-broom cameras. The only difference lies in the projection function used to generate epipolar curves.

Building Change Detection by Histogram Classification

This paper presents a supervised classification method applied to building change detection in VHR aerial images. Multi-spectral stereo pairs of 0.3m resolution have been processed to derive elevation, vegetation index and colour features. These features help filling a 5-dimensional histogram whose bins finally hold the ratio of built-up and non built-up pixels, according to the vector database to be updated. This ratio is used as building confidence at each pixel to issue a building confidence map from which to perform building verification and detection. The implementation based on histogram is very simple to code, very fast in execution and compares in this application to a state-of-the-art supervised classifier. It has been tested for the Belgian National Mapping Agency (IGN) to identify areas with high probability of change in building layers.

Contribution of nDSM derived from VHR stereo imagery to urban land-cover mapping in Sub-Saharan Africa

Mapping the urban land cover from VHR optical imagery remains a challenging task, more particularly in cities that present complex landscapes and patterns. In this study, we assessed the contribution of height data derived from WorldView-3 stereo imagery for mapping the land cover of Sub-Saharan African cities. Our case study is located in Ouagadougou, Burkina Faso. An OBIA approach was implemented using an open-source semi-automated processing chain. The use of the nDSM as input to the segmentation and/or to the classification in addition to the four VNIR WorldView-3 optical bands was evaluated. The quantitative and qualitative analysis of the results indicate an improvement for a number of classes, among which the class ‘Buildings’ that is of particular interest in many applications. Visually, this improvement is more noticeable in planned settlements and industrial areas than in unplanned settlements.

Urban features classification using 3D hyperspectral data

The surface classification of heterogeneous urban areas can be refined using the integration of spectral and 3D information. However, pixel-classification based fusion requires semi-pixel geo-registration accuracy. In this paper the 3D information is obtained from the hyperspectral data set itself. This study presents an adaptation of optimized MRF based stereo matching for the creation of 3D scenes using hyperspectral data. The obtained 3D information is integrated into a SVM classifier procedure. The results obtained in this study show the potential in the creation of 3D scenes using hyperspectral data and the benefit of combining this data with spectral information for better classification of the urban materials.

Fusion of multispectral and stereo information for unsupervised target detection in VHR airborne data

Very high resolution multispectral imaging reached a high level of reliability and accuracy for target detection and classification. However, in an urban scene, the complexity is raised, making the detection and the identification of small objects difficult. One way to overcome this difficulty is to combine spectral information with 3D data. A set of (very high resolution) airborne multispectral image sequences was acquired over the urban area of Zeebrugge, Belgium. The data consist of three bands in the visible (VIS) region, one band in the near infrared (NIR) range and two bands in the mid-wave infrared (MWIR) region. Images are obtained images at a frame rate of 1/2 frame per second for the VIS and NIR image and 2 frames per second for the MWIR bands. The sensors have a decimetric spatial resolution. The combination of frame rate with flight altitude and speed results in a large overlap between successive images. The current paper proposes a scheme to combine 3D information from along-track stereo, exploiting the overlap between images on one hand and spectral information on the other hand for unsupervised detection of targets. For the extraction of 3D information, the disparity map between different image pairs is determined automatically using an MRF-based method. For the unsupervised target detection, an anomaly detection algorithm is applied. Different methods for inserting the obtained 3D information into the target detection scheme are discussed.

Building Change Detection from Uniform Regions

This paper deals with building change detection by supervised classification of image regions into ’built’ and ’non-built’ areas. Regions are the connected components of low gradient values in a multi-spectral aerial image. Classes are learnt from spectral (colour, vegetation index) and elevation cues relatively to building polygons and non building areas as defined in the existing database. Possible candidate building regions are then filtered by geometrical features. Inconsistencies in the database with the recent image are automatically detected. Tests in cooperation with the Belgian National Geographical Institute on an area with sufficient buildings and landscape variety have shown that the system allows for the effective verification of unchanged buildings, and detection of destructions and new candidate buildings.