Wojciech Drzewiecki

Comparison of selected textural features as global content-based descriptors of VHR satellite image

The paper presents the results of the study of usefulness of selected textural features for describing the content of image chips (1024 by 1024 pixels) cut from WorldView-2 panchromatic images. Several texture analysis techniques were applied to calculate 290 different global image descriptors. These features together with 4 histogram-based characteristics formed the set of classification features. Decision trees were used to classify the images into broad land cover categories. Performance of each group of textural characteristics was assessed and compared to the results of classification based on all calculated characteristics.

Fractal and multifractal characteristics of very high resolution satellite images

In our work we analyse fractal and multifractal characteristics for description and extraction of information from very high spatial resolution satellite images. In particular, we propose the degree of multifractality as a parameter for extraction of four land cover types and investigate its usefulness in comparison with the fractal dimension. Results show that degree of multifractality designated for individual fragments of images differs depending on the present land cover type. The highest multifractality level is observed for urban area, the lowest for water, which can be treated as a monofractal. In general multifractal parameter allows for automatic assignment of land cover types to specific classes. Some deviations take place in case of discrimination between agricultural areas and forests. This problem is also considered by using additional measures during the process of multifractal parameter computation. Conducted analysis shows that multifractal formalism creates additional possibilities for the description and automatic classification of satellite images.

Change Detection Using Global and Local Multifractal Description

In this letter, we apply the multifractal formalism to land cover change detection on very high spatial resolution data. Specifically, multifractal spectra are determined and, with modifications, are used as an initial general indicator of change on the subsets of IKONOS and Pleiades images. Next, we calculate Hölder exponents for each pixel in the images and use them to generate a change mask. Our analysis shows that Hölder exponents enable a detailed evaluation of changes in land cover. A comparison with change detection based solely on panchromatic images shows that the multifractal description method has significant advantages as it reduces the number of false positives. In addition, we show that our change detection results are comparable with other multiscale techniques.

Two-stage subpixel impervious surface coverage estimation: comparing classification and regression trees and artificial neural networks

The paper presents accuracy comparison of subpixel classification based on medium resolution Landsat images, performed using machine learning algorithms built on decision and regression trees method (C.5.0/Cubist and Random Forest) and artificial neural networks. The aim of the study was to obtain the pattern of percentage impervious surface coverage, valid for the period of 2009-2010. Imperviousness index map generation was a two-stage procedure. The first step was classification, which divided the study area into categories: i) completely permeable (imperviousness index less than 1%) and ii) fully or partially impervious areas. For pixels classified as impervious, the percentage of impervious surface coverage in pixel area was estimated. The root mean square errors (RMS) of determination of the percentage of the impervious surfaces within a single pixel were 11.0% for C.5.0/Cubist method, 11.3% for Random Forest method and 12.6% using artificial neural networks. The introduction of the initial hard classification into completely permeable areas (with imperviousness index <1%) and impervious areas, allowed to improve the accuracy of imperviousness index estimation on poorly urbanized areas covering large areas of the Dobczyce Reservoir catchment. The effect is also visible on final imperviousness index maps.

A comparative study of Landsat TM and RapidEye imagery for two-stage impervious surface coverage estimation

The paper presents accuracy comparison of sub-pixel classification based on medium resolution Landsat data and high resolution RapidEye satellite images, performed using machine learning algorithms built on decision and regression trees method (C.5.0 and Cubist). The research was conducted in southern Poland for the catchment of the Dobczyce Reservoir. The aim of the study was to obtain image of percentage impervious surface coverage and assess which data sets can be more applicable for the purpose of impervious surface coverage estimation. Imperviousness index map generation was a two-stage procedure. The first step was classification, which divided the study area into two categories: a) completely permeable (imperviousness index less than 1%) and b) fully or partially impervious areas. For pixels classified as impervious, the percentage of impervious surface coverage within a single pixel was estimated. Decision and regression trees models construction was done based on training data set derived from Landsat TM pixels as well as for fragments of RapidEye images corresponding to the same Landsat TM training pixels. In order to obtain imperviousness index maps with the minimum possible error we did the estimation of models accuracy based on the results of cross-validation. The approaches guaranteeing the lowest means errors in terms of training set using C5.0 and Cubist algorithm for Landsat and RapidEye images were selected. Accuracy of the final imperviousness index maps was checked based on validation data sets. The root mean square error of determination of the percentage of the impervious surfaces within a single Landsat pixel was 9.9% for C.5.0/Cubist method. However, the root mean square error specified for RapidEye test data was 7.2%. The study has shown that better results of two-stage imperiousness index map estimation using RapidEye satellite images can be obtained.

A study of selected textural features usefulness for impervious surface coverage estimation using Landsat images

The aim of our research was to evaluate the applicability of textural measures for sub-pixel impervious surfaces estimation using Landsat TM images based on machine learning algorithms. We put the particular focus on determining usefulness of five textural features groups in respect to pixel- and sub-pixel level. However, the two-stage approach to impervious surfaces coverage estimation was also tested. We compared the accuracy of impervious surfaces estimation using spectral bands only with results of imperviousness index estimation based on extended classification features sets (spectral band values supplemented with measures derived from various textural characteristics groups). Impervious surfaces coverage estimation was done using decision and regression trees based on C5.0 and Cubist algorithms. At the stage of classification the research area was divided into two categories: i) completely permeable (imperviousness index less than 1%) and ii) fully or partially impervious areas. At the stage of sub-pixel classification evaluation of percentage impervious surfaces coverage within single pixel was done. Based on the results of cross-validation, we selected the approaches guaranteeing the lowest means errors in terms of training set. Accuracy of the imperviousness index estimation was checked based on validation data set. The average error of hard classification using spectral features only was 6.5% and about 4.4% for spectral features combining with absolute gradient-based characteristics. The root mean square error (RMSE) of determination of the percentage impervious surfaces coverage within a single pixel was equal to 9.46% for the best tested classification features sets. The two-stage procedure was utilized for the primary approach involving spectral bands as the classification features set and for the approach guaranteeing the best accuracy for classification and regression stage. The results have shown that inclusion of textural measures into classification features can improve the estimation of imperviousness based on Landsat imagery. However, it seems that in our study this is mainly due higher accuracy of hard classification used for masking out the completely permeable pixels.

Applicability of multifractal features as global characteristics of WorldView-2 panchromatic satellite images

Abstract In this work we analyse fractal and multifractal characteristics for description and extraction of information from VHR satellite images. We propose the degree of multifractality as a global descriptor of satellite image content and investigate its usefulness for classification of WorldView-2 image chips into main land cover types. The research confirmed that it is possible to use the textural features as efficient global descriptors of WorldView-2 panchromatic image content. Results show that the degree of multifractality is related to land cover type prevailing in the imaged area. It was also proved that multifractal parameters should be considered as valuable textural features in the context of land cover classification.