Multivariate Alteration Detection Method Research Articles

Unsupervised Scene Change Detection via Latent Dirichlet Allocation and Multivariate Alteration Detection

Scene change detection is the process of identifying the differences between the multitemporal image scenes, which has significant potential in the application of urban development and land management at the semantic level. Traditional scene change detection methods are based on the supervised scene classification, and then directly compare the independent classification results without considering the temporal correlation between the unchanged regions. However, few studies have focused on detecting the semantic changes of multitemporal image scenes with unsupervised methods. In this paper, we propose a novel unsupervised scene change detection method by using latent Dirichlet allocation (LDA) and multivariate alteration detection (MAD). First, the scene is represented by the bag-of-visual-words model, and adopt the union dictionary to ensure the consistency of dictionary space. Then, LDA is used to achieve the middle-level feature dimension reduction, and generate the common topic space of the two multitemporal image scene datasets. And finally, the MAD method was applied to detect the semantic changes of corresponding multitemporal image scenes. Two experiments with high-resolution remote sensing image scene datasets demonstrated that our proposed approach can get a better performance in unsupervised scene change detection without prior knowledge.

Relative radiometric normalization of bitemporal very high-resolution satellite images for flood change detection

Change-detection analysis using bitemporal satellite imagery is a reliable method for providing and assessing information about flood-induced changes over a wide area in a timely and cost-effective manner. Accurate radiometric normalization between bitemporal imagery is a critical component in the application of change-detection techniques to flood mapping because the accuracy of the change detection is directly affected by the quality of radiometric normalization. A methodology based on multivariate alteration detection (MAD) is introduced as an approach that enables reliable radiometric normalization of bitemporal very high-resolution (VHR) images for detecting flood-induced changes. The method uses a weighting function to adaptively identify weights based on open water features, which are estimated by the normalized difference water index, in the computation of the covariance matrices of the MAD transform. To quantitatively evaluate and test the performance of the proposed method, a comparison is made between it and the iteratively reweighted (IR)-MAD method based on statistical tests and the accuracy of flood change detection. Change vector analysis- and MAD-based change-detection methods were used for the comparison of the proposed and IR-MAD methods. Experimental results on KOMPSAT-2 bitemporal VHR images prove that the proposed method produced better results than the IR-MAD method in the statistical tests and also increased the overall accuracy of flood change detection by 1.8% and 12.6% for the two study sites.

변화지역 탐지를 위한 시계열 KOMPSAT-2 다중분광 영상의 MAD 기반 상대복사 보정에 관한 연구

원격탐사 방법을 활용한 변화지역 탐지, 재난재해 지도 작성, 작황 모니터링 등 다중시기의 위성영상을 활용한 결과를 도출하기 위해서는 시계열 영상 정보를 서로 비교할 수 있는 공통의 스케일로 정규화 하는 것이 필요하다. 다중시기 영상에 대한 정규화 방법은 절대복사보정과 상대복사 보정으로 나눌 수 있으며, 본 연구에서는 상대복사 보정을 통한 시계열 위성영상처리 기법을 다루고자 한다. 2011년 3월 해일 피해가 발생했던 일본 센다이 지역을 연구대상지로 선정하였고, KOMPSAT-2 다중분광영상을 이용한 사고 전, 후의 피해지역 탐지에 있어 상대복사 보정의 실효성을 분석하였다. 다양한 상대복사 보정 기법 중에서 정준상관분석을 통해 PIFs(Pseudo Invariant Features) 지역을 자동으로 추출하는 MAD(Multivariate Alteration Detection) 기법을 적용하였다. 본 사례연구 분석결과 MAD 방식에 의한 자동 PIFs 지역의 추출은 비교적 높은 정확도 수준에서 이루어짐을 확인할 수 있었으며, 상대복사 보정된 시계열 위성영상을 사용함으로써 변화지역 자동탐지의 신뢰수준을 높일 수 있는 것으로 나타났다. It is necessary to normalize spectral image values derived from multi-temporal satellite data to a common scale in order to apply remote sensing methods for change detection, disaster mapping, crop monitoring and etc. There are two main approaches: absolute radiometric normalization and relative radiometric normalization. This study focuses on the multi-temporal satellite image processing by the use of relative radiometric normalization. Three scenes of KOMPSAT-2 imagery were processed using the Multivariate Alteration Detection(MAD) method, which has a particular advantage of selecting PIFs(Pseudo Invariant Features) automatically by canonical correlation analysis. The scenes were then applied to detect disaster areas over Sendai, Japan, which was hit by a tsunami on 11 March 2011. The case study showed that the automatic extraction of changed areas after the tsunami using relatively normalized satellite data via the MAD method was done within a high accuracy level. In addition, the relative normalization of multi-temporal satellite imagery produced better results to rapidly map disaster-affected areas with an increased confidence level.

Retrieval of forest attributes in complex successional forests of Central Indonesia: Modeling and estimation of bitemporal data

Quantification of forest parameters in different successional stages is required because of its importance as a source of global emissions and ecosystem changes. This study focuses on a successional tropical forest under logging practices in East Kalimantan province, Indonesia. We modeled the forest attributes using both a parametric multiple linear regression analysis and neural networks approach, with Landsat ETM data acquired in 2000 (ETM00). We compiled sample plot data using forest inventory data collected from 1997 to 1998. A total of 226 plots were used to train the models and 112 plots were used for the validation. The remote sensing data (spectral values, vegetation indices, texture, etc.) coupled with digital elevation model (DEM) were experimented with and selectively used to model basal area, stem volume and above ground biomass (AGB). We investigated the possibility to estimate the forest attributes from bitemporal ETM data by calibrating radiometric properties of the ETM image from 2003 (ETM03) using the multivariate alteration detection method. The Pearson correlations showed that the mean texture index is strongly correlated with the forest attributes. We show that neural networks resulted in a higher coefficient of determination ( r 2) and lower RMSE than multiple regressions for predicting the forest attributes. The estimated forest properties increased with the forest succession advancement (i.e. from the open forest to advanced secondary forest classes). The modeled basal area, stem volume and AGB varied from 10.7–15.1 m 2 ha −1, 123.2–181.9 m 3 ha −1, and 132.7–185.3 Mg ha −1, respectively. The RMSE r values of model fitting ranged from 11.2% to 13.3%, and the test dataset estimated slightly higher RMSE r which varied from 12% to 14.1%. The ETM03 forest attributes revealed favorable estimates, showing considerably higher estimates than the ETM00. The estimation of forest properties using neural networks makes Landsat data a valuable source of information for forest management, mainly with the recent free access to its historical dataset.

The Regularized Iteratively Reweighted MAD Method for Change Detection in Multi- and Hyperspectral Data

This paper describes new extensions to the previously published multivariate alteration detection (MAD) method for change detection in bi-temporal, multi- and hypervariate data such as remote sensing imagery. Much like boosting methods often applied in data mining work, the iteratively reweighted (IR) MAD method in a series of iterations places increasing focus on "difficult" observations, here observations whose change status over time is uncertain. The MAD method is based on the established technique of canonical correlation analysis: for the multivariate data acquired at two points in time and covering the same geographical region, we calculate the canonical variates and subtract them from each other. These orthogonal differences contain maximum information on joint change in all variables (spectral bands). The change detected in this fashion is invariant to separate linear (affine) transformations in the originally measured variables at the two points in time, such as 1) changes in gain and offset in the measuring device used to acquire the data, 2) data normalization or calibration schemes that are linear (affine) in the gray values of the original variables, or 3) orthogonal or other affine transformations, such as principal component (PC) or maximum autocorrelation factor (MAF) transformations. The IR-MAD method first calculates ordinary canonical and original MAD variates. In the following iterations we apply different weights to the observations, large weights being assigned to observations that show little change, i.e., for which the sum of squared, standardized MAD variates is small, and small weights being assigned to observations for which the sum is large. Like the original MAD method, the iterative extension is invariant to linear (affine) transformations of the original variables. To stabilize solutions to the (IR-)MAD problem, some form of regularization may be needed. This is especially useful for work on hyperspectral data. This paper describes ordinary two-set canonical correlation analysis, the MAD transformation, the iterative extension, and three regularization schemes. A simple case with real Landsat Thematic Mapper (TM) data at one point in time and (partly) constructed data at the other point in time that demonstrates the superiority of the iterative scheme over the original MAD method is shown. Also, examples with SPOT High Resolution Visible data from an agricultural region in Kenya, and hyperspectral airborne HyMap data from a small rural area in southeastern Germany are given. The latter case demonstrates the need for regularization.