A hybrid texture-based and region-based multi-scale image segmentation algorithm
The objective of this research was the design and development of a region-based multi-scale segmentation algorithm with the integration of complex texture features, in order to provide a low level processing tool for object-oriented image analysis. The im
- PDF / 2,325,524 Bytes
- 16 Pages / 439.37 x 666.142 pts Page_size
- 19 Downloads / 200 Views
A. Tzotsos, C. Iosifidis, D. Argialas Laboratory of Remote Sensing, Department of Surveying, School of Rural and Surveying Engineering, National Technical University of Athens, Greece
KEYWORDS: Object-Based Image Analysis, MSEG, Grey Level Cooccurrence Matrix ABSTRACT: The objective of this research was the design and development of a region-based multi-scale segmentation algorithm with the integration of complex texture features, in order to provide a low level processing tool for object-oriented image analysis. The implemented algorithm is called Texture-based MSEG and can be described as a region merging procedure. The first object representation is the single pixel of the image. Through iterative pair-wise object fusions, which are made at several iterations, called passes, the final segmentation is achieved. The criterion for object merging is a homogeneity cost measure, defined as object heterogeneity, and computed based on spectral and shape features for each possible object merge. An integration of texture features to the region merging segmentation procedure was implemented through an Advanced Texture Heuristics module. Towards this texture-enhanced segmentation method, complex statistical measures of texture had to be computed based on objects, however, and not on rectangular image regions. The approach was to compute grey level co-occurrence matrices for each image object and then to compute object-based statistical features. The Advanced Texture Heuristics module, integrated new heuristics in the decision for object merging, involving similarity measures of adjacent image objects, based on the computed texture features. The algorithm was implemented in C++ and was tested on remotely sensed images of different sensors, resolutions and complexity levels. The results were satisfactory since the produced primi-
222
A. Tzotsos, C. Iosifidis, D. Argialas
tive objects, were comparable to those of other segmentation algorithms. A comparison between the simple algorithm and the texture-based algorithm results showed that in addition to spectral and shape features, texture features did provide good segmentation results.
1 Introduction 1.1 Recent developments in Remote Sensing Recently, remote sensing has achieved great progress both in sensor resolution and image analysis algorithms. Due to very high resolution imagery, such as IKONOS and Quick Bird, traditional classification methods, have become less effective given the magnitude of heterogeneity appearing in the spectral feature space of such imagery. The spectral heterogeneity of imaging data has increased rapidly, and the traditional methods tend to produce “salt and pepper” classification results. Such problems occur also to medium resolution satellite data, such as Landsat TM, SPOT etc. Another disadvantage of traditional classification methods is that they do not use information related to shape, site and spatial relation (context) of the objects of the scene. Context information is a key element to photointerpretation, and a key feature used by all
Data Loading...
