At present, remote sensing technology is the best weapon to get information from the earth surface, and it is very useful in geo- information updating and related applications. Extracting road from remote sensing images is one of the biggest demand of rapid city development, therefore, it becomes a hot issue. Roads in high-resolution images are more complex, patterns of roads vary a lot, which becomes obstacles for road extraction. In this paper, a machine learning based strategy is presented. The strategy overall uses the geometry features, radiation features, topology features and texture features. In high resolution remote sensing images, the images cover a great scale of landscape, thus, the speed of extracting roads is slow. So, roads' ROIs are firstly detected by using Houghline detection and buffering method to narrow down the detecting area. As roads in high resolution images are normally in ribbon shape, mean-shift and watershed segmentation methods are used to extract road segments. Then, Real Adaboost supervised machine learning algorithm is used to pick out segments that contain roads' pattern. At last, geometric shape analysis and morphology methods are used to prune and restore the whole roads' area and to detect the centerline of roads.

In this paper, a unique approach for road extraction utilizing pixel spectral information for classification and image segmentation-derived object features was developed. In this approach, road extraction was performed in two steps. In the first step, support vector machine (SVM) was employed merely to classify the image into two groups of categories: a road group and a non-road group. For this classification, support vector machine (SVM) achieved higher accuracy than Gaussian maximum likelihood (GML). In the second step, the road group image was segmented into geometrically homogeneous objects using a region growing technique based on a similarity criterion, with higher weighting on shape factors over spectral criteria. A simple thresholding on the shape index and density features derived from these objects was performed to extract road features, which were further processed by thinning and vectorization to obtain rood centerlines. The experiment showed the proposed approach worked well with images comprised by both rural and urban area features.

ABSTRACT An approach for automatic road extraction from digital aerial imagery is presented. The extraction is based on a semantic model for roads. The images are divided into different so-called "global contexts": rural, forest, and urban. Different parts of the road model and different strategies are used in the different global contexts. In rural areas, a multi-scale approach is employed to find initial hypotheses for roadsides, which are then grouped into road segments using local context knowledge. In urban areas, road markings and DEM information are used to extract road segments. In addition to the local grouping, road segments are selected and linked into a global road network. An external evaluation shows the high quality of the results that are obtainable automatically with the proposed approach. 1 INTRODUCTION In the past, the automation of road extraction from digital imagery has received considerable attention. Research on this issue is often motivated by the increasing importanc...

Automatic estimation of salient object regions across images, without any prior assumption or knowledge of the contents of the corresponding scenes, enhances many computer vision and computer graphics applications. We introduce a regional contrast based salient object detection algorithm, which simultaneously evaluates global contrast differences and spatial weighted coherence scores. The proposed algorithm is simple, efficient, naturally multi-scale, and produces full-resolution, high-quality saliency maps. These saliency maps are further used to initialize a novel iterative version of GrabCut, namely SaliencyCut, for high quality unsupervised salient object segmentation. We extensively evaluated our algorithm using traditional salient object detection datasets, as well as a more challenging Internet image dataset. Our experimental results demonstrate that our algorithm consistently outperforms 15 existing salient object detection and segmentation methods, yielding higher precision and better recall rates. We also show that our algorithm can be used to efficiently extract salient object masks from Internet images, enabling effective sketch-based image retrieval (SBIR) via simple shape comparisons. Despite such noisy internet images, where the saliency regions are ambiguous, our saliency guided image retrieval achieves a superior retrieval rate compared with state-of-the-art SBIR methods, and additionally provides important target object region information.

Abstract Figure-ground segmentation from bounding box input, provided either automatically or manually, has been extremely popular in the last decade and influenced various applications. A lot of research has focused on high-quality segmentation, using complex formulations which often lead to slow techniques, and often hamper practical usage. In this paper we demonstrate a very fast segmentation technique which still achieves very high quality results. We propose to replace the time consuming iterative refinement of global colour models in traditional GrabCut formulation by a densely connected crf . To motivate this decision, we show that a dense crf implicitly models unnormalized global colour models for foreground and background. Such relationship provides insightful analysis to bridge between dense crf and GrabCut functional. We extensively evaluate our algorithm using two famous benchmarks. Our experimental results demonstrated that the proposed algorithm achieves an order of magnitude (10脳) speed-up with respect to the closest competitor, and at the same time achieves a considerably higher accuracy.