Road network extraction from high-resolution multi spectral satellite images

Karaman, Ersin
In this thesis, an automatic road extraction algorithm for multi-spectral images is developed. The developed model extracts elongated structures from images by using edge detection, segmentation and clustering techniques. The study also extracts non-road regions like vegetative fields, bare soils and water bodies to obtain more accurate road map. The model is constructed in a modular approach that aims to extract roads with different characteristics. Each module output is combined to create a road score map. The developed algorithm is tested on 8-band WorldView-2 satellite images. It is observed that, the proposed road extraction algorithm yields 47 % precision and 70 % recall. The approach is also tested on the lower spectral resolution images with four-band, RGB and gray level. It is observed that the additional four bands provide an improvement of 12 % for precision and 3 % for recall. Road type analysis is also in the scope of this study. Roads are classified into asphalt, concrete and unpaved using Gaussian Mixture Models. Other linear objects such as railroads and water canals may also be extracted by this process. An algorithm that classifies drive roads and railroads for very high resolution images is also investigated. It is based on the Fourier descriptors that identify the presence of railroad sleepers. Water canals are also extracted in multi-spectral images by using spectral ratios that employ the near infrared bands. Structural properties are used to distinguish water canals from other water bodies in the image.


Alignment of uncalibrated images for multi-view classification
Arık, Sercan Ömer; Vural, Elif; Frossard, Pascal (2011-12-29)
Efficient solutions for the classification of multi-view images can be built on graph-based algorithms when little information is known about the scene or cameras. Such methods typically require a pairwise similarity measure between images, where a common choice is the Euclidean distance. However, the accuracy of the Euclidean distance as a similarity measure is restricted to cases where images are captured from nearby viewpoints. In settings with large transformations and viewpoint changes, alignment of im...
An integrated approach to computational vision - The edge strength function and the nested symmetries
Tarı, Zehra Sibel (1999-01-01)
A new development in local symmetry extraction and its connections to segmentation functionals and the fronts propagating with curvature-dependent speed are examined. The basic tool is a new distance function that attains its maximum value at the shape boundary and decays rapidly away from there. It is shown that the Hessian of the distance function captures perceptual information that can be extracted easily, efficiently, and robustly in the form of nested local shape symmetries at multiple scales. The mos...
A Shadow based trainable method for building detection in satellite images
Dikmen, Mehmet; Halıcı, Uğur; Department of Geodetic and Geographical Information Technologies (2014)
The purpose of this thesis is to develop a supervised building detection and extraction algorithm with a shadow based learning method for high-resolution satellite images. First, shadow segments are identified on an over-segmented image, and then neighboring shadow segments are merged by assuming that they are cast by a single building. Next, these shadow regions are used to detect the candidate regions where buildings most likely occur. Together with this information, distance to shadows towards illuminati...
Kose, Kivanc; Yılmaz, Erdal; ÇETİN, AHMET ENİS (2009-07-17)
In this paper a new Digital Elevation Map (DEM) image compression algorithm is proposed. DEM image can be threated as a grayscale image, whose pixel values are the elevation values of the map points. The grayscale DEM image is compressed using an adaptive wavelet based image compression algorithm. The method, which is an extension of the progressive mesh compression takes advantage of the multiresolution property of the wavelets while coding the map images. This makes it possible to decode different resolut...
Kilickaya, Mert; Erdem, Erkut; Erdem, Aykut; İKİZLER CİNBİŞ, NAZLI; Çakıcı, Ruket (2014-04-25)
Automatic image captioning, the process cif producing a description for an image, is a very challenging problem which has only recently received interest from the computer vision and natural language processing communities. In this study, we present a novel data-driven image captioning strategy, which, for a given image, finds the most visually similar image in a large dataset of image-caption pairs and transfers its caption as the description of the input image. Our novelty lies in employing a recently' pr...
Citation Formats
E. Karaman, “Road network extraction from high-resolution multi spectral satellite images,” Ph.D. - Doctoral Program, Middle East Technical University, 2012.