Shihong Du

Fuzzy description of topological relations I: a unified fuzzy 9-intersection model

First, the impacts of uncertainty of position and attribute on topological relations and the disadvantages of qualitative methods in processing the uncertainty of topological relations are concluded. Second, based on the above point, the fuzzy membership functions for dividing topology space of spatial object and for describing uncertainty of topological relations are proposed. Finally, the fuzzy interior, exterior and boundary are defined according to those fuzzy membership functions, and then a fuzzy 9-intersection model that can describe the uncertainty is constructed. Since fuzzy 9-intersection model is based on fuzzy set, not two-value logic, the fuzzy 9-intersection model can describe the impacts of position and attribute of spatial data on topological relations, and the uncertainty of topological relations between fuzzy objects, relations between crisp objects and fuzzy objects, and relations between crisp objects in a united model.

Reasoning about topological relations between regions with broad boundaries

Uncertain regions can be represented as having broad boundaries (BBRs) and their topological relations can be modeled by the extended 9-intersection. In order to satisfy the need for querying, managing, and processing BBRs, this study presents a 4-tuple representation of topological relations between BBRs, and a method in which the relations between simple regions with broad boundaries (SBBRs) are used to infer new topological information. The 4-tuple representation can distinguish the same topological relations as identified by the extended 9-intersection. Since the 4-tuple uses combinations of the basic topological relations between crisp regions to describe the relations between uncertain regions, the reasoning of topological relations between SBBRs can be obtained by combining the results of those between crisp regions. The reasoning mechanism can be used in several applications, such as to evaluate the consistency of topological relations between uncertain regions in multi-resolution spatial databases and to assess the consistency of a complete or incomplete symbolic description of a spatial scene.

Scene classification using multi-scale deeply described visual words

ABSTRACT This article presents a deep learning-based Multi-scale Bag-of-Visual Words (MBVW) representation for scene classification of high-resolution aerial imagery. Specifically, the convolutional neural network (CNN) is introduced to learn and characterize the complex local spatial patterns at different scales. Then, the learnt deep features are exploited in a novel way to generate visual words. Moreover, the MBVW representation is constructed using the statistics of the visual word co-occurrences at different scales, which are derived from a training data set. We apply our technique to the challenging aerial scene data set: the University of California (UC) Merced data set consisting of 21 different aerial scene categories with sub-metre resolution. The experimental results show that the statistics of deeply described visual words can characterize the scene well and improve classification accuracy. It demonstrates that the proposed method is highly effective in the scene classification of high-resolution remote-sensing imagery.

Semantic Classification of Heterogeneous Urban Scenes Using Intrascene Feature Similarity and Interscene Semantic Dependency

Semantic classification of urban scenes aims to classify scenes composed of many different types of objects into predefined semantic classes. To learn the association between urban scenes and semantic classes, five tasks are needed: 1) segmenting the image into scenes; 2) establishing semantic classes of scenes; 3) extracting and transforming features; 4) measuring the intrascenes feature similarity; and 5) labeling each scene by a semantic classification method. Despite many efforts on these tasks, most existing works consider only visual features with inconsistent similarity measurement, while ignore semantic features inside scenes and the interactions between scenes, leading to poor classification results for high heterogeneous scenes. To solve these problems, this study combines intrascene feature similarity and interscene semantic dependency to form a two-step classification approach. For the first step, visual and semantic features are first optimized to be invariant to affine transformation, and then are employed in K-Nearest Neighbor to initially classify scenes. For the second step, multinomial distribution is presented to model both the spatial and semantic dependency between scenes, and then used to improve the initial classification results. The implementations conducted in two study areas indicate that the proposed approach produces better results for heterogeneous scenes than visual interpretation, as it can discover and model the hidden information between scenes which is often ignored by existing methods. In addition, compared with the initial classification, the optimized step improves accuracies by 3.6% and 5% in the two study areas, respectively.

Evaluating structural and topological consistency of complex regions with broad boundaries in multi-resolution spatial databases

Multi-resolution or multi-scale spatial databases store and manage multiple representations of spatial objects in the same area, so consistency among multiple representations of the same objects should be evaluated and maintained. Although many approaches have been proposed to check inconsistencies in multi-resolution databases, there is still a lack of effective approaches working for complex objects, especially for regions with broad boundaries which is a general model for representing various types of uncertainties. This paper presents approaches for evaluating structural and topological consistency among multiple representations of complex regions with broad boundaries (CBBRs) based on map generalization operators: merging, dropping, and hybrid of these two. For evaluation of structural consistency, all possible multiple representations of a CBBR are generated automatically and organized into a structured neighborhood graph, and then correspondences and equivalences among the multiple representations are defined to determine whether two representations at different levels of detail are structurally consistent. For evaluation of topological consistency, the topological relations between all pairs of regions in two CBBRs are considered, and their variation with change of spatial scale is analyzed. Since the approaches in this paper are built on a hiearchical representation of CBBRs with arbitrarily complex structure, they will also work well for evaluating consistency among multiple representations of complex objects.

Mining parameter information for building extraction and change detection with very high-resolution imagery and GIS data

Although much efforts have been made to develop automatic methods for building extraction from very high-resolution (VHR) imagery during the past 30 years; the methods with high performance are still unavailable due to the three issues: uncertainty of segmentation scales, selection of effective features, and sample selection. In this study, by introducing GIS data, a parameter mining approach is proposed to (1) mine parameter information for building extraction, and (2) detect changes of buildings between VHR imagery and GIS data. For the first target, the learning mechanism is proposed for identifying optimal segmentation scales, feature subsets, and samples. For the second target, the discovered information (i.e., optimal segmentation scales, feature subsets, and selected samples) is applied to classify the VHR imagery with a multilevel random forest (RF) classifier. The proposed approach is validated on two datasets: Dataset 1 and Dataset 2. The knowledge of building extraction is first learned from Dataset 1 and then used to classify both datasets, and change detection is conducted on Dataset 1. Results of change detection in Dataset 1 indicate that the false alarm ratio and omission error of increased buildings are 20.1% and 8.4%, while the false alarm ratio and omission error of destroyed buildings are 19.1% and 11.3%, respectively. Results of building extraction in Dataset 2 revealed scores of 81.50% and 81.09% at pixel- and object-based evaluation levels. Accordingly, our proposed method is successful in building extraction and change detection.

Global and local indicators of spatial association between points and polygons: a study of land use change.

The existing indicators related to spatial association, especially the K function, can measure only the same dimension of vector data, such as points, lines and polygons, respectively. We develop four new indicators that can analyze and model spatial association for the mixture of different dimensions of vector data, such as lines and points, points and polygons, lines and polygons. The four indicators can measure the spatial association between points and polygons from both global and local perspectives. We also apply the presented methods to investigate the association of temples and villages on land-use change at multiple distance scales in the Guoluo Tibetan Autonomous Prefecture in Qinghai Province, PR China. Global indicators show that temples are positively associated with land-use change at large spatial distances (e.g., >6000 m), while the association between villages and land-use change is insignificant at all distance scales. Thus temples, as religious and cultural centers, have a stronger association with land-use change than the places where people live. However, local indicators show that these associations vary significantly in different sub-areas of the study region. Furthermore, the association of temples with land-use change is also dependent on the specific type of land-use change. The case study demonstrates that the presented indicators are powerful tools for analyzing the spatial association between points and polygons.

Object-Based Convolutional Neural Network for High-Resolution Imagery Classification

Timely and accurate classification and interpretation of high-resolution images are very important for urban planning and disaster rescue. However, as spatial resolution gets finer, it is increasingly difficult to recognize complex patterns in high-resolution remote sensing images. Deep learning offers an efficient strategy to fill the gap between complex image patterns and their semantic labels. However, due to the hierarchical abstract nature of deep learning methods, it is difficult to capture the precise outline of different objects at the pixel level. To further reduce this problem, we propose an object-based deep learning method to accurately classify the high-resolution imagery without intensive human involvement. In this study, high-resolution images were used to accurately classify three different urban scenes: Beijing (China), Pavia (Italy), and Vaihingen (Germany). The proposed method is built on a combination of a deep feature learning strategy and an object-based classification for the interpretation of high-resolution images. Specifically, high-level feature representations extracted through the convolutional neural networks framework have been systematically investigated over five different layer configurations. Furthermore, to improve the classification accuracy, an object-based classification method also has been integrated with the deep learning strategy for more efficient image classification. Experimental results indicate that with the combination of deep learning and object-based classification, it is possible to discriminate different building types in Beijing Scene, such as commercial buildings and residential buildings with classification accuracies above 90%.

Road extraction from ETM panchromatic image based on Dual-Edge Following

Research on road extraction from digital imagery is motivated by the need for data acquisition and update for geographic information systems (GIS). Roads usually have parallelism of road sides, and on the images, especially the edge map, there are dual edges for each road. In this paper, we propose an approach for automatically extracting road from ETM panchromatic image with a resolution of 15 meters based on Dual-Edge Following. Our approach uses the edge detector with embedded confidence (EDEC, Peter Meer, 2001) to detect road edge, then traces road to generate road candidates by Dual-Edge Following, next exploits the perceptual organization based on probability to link the road segments. Dual-Ddge Following use edge information of both road sides to search for road segments which can improve the precision of road segments. The experiment with ETM panchromatic image in Xinjiang, China shows the validity of the approach.

A Direction-Guided Ant Colony Optimization Method for Extraction of Urban Road Information From Very-High-Resolution Images

Typical object-based classification methods only take image object properties as criteria to classify roads, leaving the associated edge information unused. These methods often lead to fragmented road areas and inconsistent road widths and smoothness. Meanwhile, very-high-resolution (VHR) images contain a large amount of edge information and different types of geographic objects, thus, it is challenging to extract roads by typical edge-based extraction or grouping methods. In this study, a globally optimized method is developed to integrate both object and edge features to extract urban road information from VHR images. This novel method extends ant colony optimization (ACO) through deploying and moving ants (artificial agents) along roads with the guidance of comprehensive object and edge information. As ants spread pheromone along their paths, roads are recognized based on aggregated pheromone levels. A set of experiments on VHR images showed that our method significantly outperforms object-based classification methods with not only improved road extraction quality but also enhanced stability when applied to large and complex images.