Abstract:Object based image analysis (OBIA) is an approach increasingly used in classifying high spatial resolution remote sensing images. Object based image classifiers first segment an image into objects (or image segments), and then classify these objects based on their attributes and spatial relations. Numerous algorithms exist for the first step of the OBIA process, i.e. image segmentation. However. less research has been conducted on the object classification part of OBIA, in particular the spatial relations between objects that are commonly used to construct rules for classifying image objects and refining classification results. In this paper, we establish a context where objects are areal (not points or lines) and non-overlapping (we call this "single-valued" space), and propose a framework of binary spatial relations between segmented objects to aid in object classification. In this framework, scale-dependent "line-like objects" and "point-like objects" are identified from areal objects based on their shapes. Generally, disjoint and meet are the only two possible topological relations between two non-overlapping areal objects. However, a number of quasitopological relations can be defined when the shapes of the objects involved are considered. Some of these relations are fuzzy and thus quantitatively defined. In addition, we define the concepts of line-like objects (e.g. roads) and point-like objects (e.g. wells), and develop the relations between two line-like objects or two point-like objects. For completeness, cardinal direction relations and distance relations are also introduced in the proposed context. Finally, we implement the framework to extract roads and moving vehicles from an aerial photo. The promising results suggest that our methods can be a valuable tool in defining rules for object based image analysis. Published by Elsevier B.V. on behalf of International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS).

A fuzzy set representation approach for object-based

3D Object Recognition by Fast Spherical Correlation Between Combined View EGIs and PFT.

Rough-fuzzy set approach for color and texture based image segmentation

Feature Representation for 3D Object Retrieval Based on Unconstrained Multi-View

AN image retrieval method based on multiple hyperspheres OC-SVM hashing

Object-Based Image Retrieval Using Spatial Context

An Object-Level Feature Representation Model for the Multi-target Retrieval of Remote Sensing Images.

Object retrieval using configurations of salient regions

An image representation method based on the similarity of feature points

A Framework of Region-Based Spatial Relations for Non-Overlapping Features and Its Application in Object Based Image Analysis

Image Retrieval Based on Fuzzy Semantic Relevance Matrix

Fast Object Retrieval Using Direct Spatial Matching

Scalable Object Retrieval with Compact Image Representation from Generic Object Regions

Complex spatial region representation and similar matching for multi-object image retrieval

Fuzzy Image Interpretation Based on Genetic Searching

Fast and Robust Spatial Matching for Object Retrieval

Building Fuzzy Areal Geographical Objects from Point Sets

Object Retrieval Based On Spatially Frequent Items With Informative Patches

A Novel Coarse-to-Fine Scheme for Remote Sensing Image Registration Based on SIFT and Phase Correlation

A parameterized evaluation model of similarity on multi-target retrieval for remote sensing image

A Stepwise Similarity Approximation of Spatial Constraints for Image Retrieval