The Research Of Building Earthquake Damage Change Detection Based On Object-Oriented Technology With Remote Sensing Image

An accurate and quick detection and seismic classification of the building earthquake damage is significant for disaster emergency and rescue.Along with the continuous development of very high resolution(VHR)remote sensing for Earth observation,the method of earthquake information extraction has step into a new stage and provided an economical and fast way to earthquake emergency and damage estimation method.This paper aimed at quickly,precisely and efficiently detecting building damage information.According to the analysis of the problems existing in present research,a technical process for the VHR remote sensing image of the building earthquake damage information object-oriented change detection was proposed for its extraction and quantitative classification through different degrees of improvement and innovation for the key technologies.The main research achievements and innovations are summarized as follows:(1)In view of the disadvantage of existing methods including tedious parameter tuning of image segmentation scales,and the insufficient exploration of features,a multi-feature joint image segmentation approach was proposed.By improving the fast scanning and merging(FSAM)algorithm to construct the initialized over segmentation elements,the heterogeneity of each region was measured by combined using the spectrum,shape and the texture features.The optimal segmentation parameter was automatic iteratively calculated according to the supervised training for the feature index of selected targets segmentation samples using the Fuzzy-based Segmentation Parameter optimizer,which get a more accurate and objective segmentation results according with a higher efficiency.The experiment was proved and verified that the proposed strategy ensured that small size target elements were not merged by mistake and meanwhile the description integrity of the large size ones were reserved.(2)In view of the traditional object-oriented change detection technology using a single classifier to extract change information of the complex scene without considering different classifiers having different performance on the classification of heterogeneous objects,a multiple classifiers integration change detection system was proposed which including a new sample selection method and a multiple classifier system.A more uniform and objective training samples were gained by using a Double Threshold CVA(DT-CVA)method instead of a single threshold to select samples.The performance of multiple classifier system was improved by extracting,selecting and integrating the multi-source feature,and the extreme learning machine,multinomial logistic regression and K-nearest neighbor model were integrated by using the stochastic subspace identification technology.The classification results were fused based on the layer by layer method.The final building earthquake damage information result was output by accuracy evaluation through using classification evaluation criteria and other indicators.The change detection accuracy of the Yushu small area is 88.45% with the Kappa coefficient was 0.8411 and the whole Jiegu Town is 87.2%.It was proved and verified that the proposed method can make up for the classification deficiency based on the single data source,and realized the complementary advantages among the classifiers,which can improve the classification accuracy.(3)In view of the active and passive remote sensing data collection situation and the comprehensive analysis of the Imaging,gray level,texture and statistics of the building earthquake damage information,of which a quantitative classification technological process was proposed.The intensity and texture correlation based on single SAR and the change detection results of proposed method were entered the quantitative classification model.The quantitative earthquake damage classification accuracy of the methods were 70.5%,76.7% and 86.4%.Results indicated that the proposed method can be effectively used to evaluate the damage of earthquake disaster,and meanwhile provided a qualitative and quantitative basis for earthquake emergency decision-making.