آشكارسازي اتوماتيك تغييرات ساختمان ها ناشي از زلزله با استفاده از تركيب داده هاي برداري و تصاوير ماهواره اي با مقياس بزرگ

Automatic change detection of buildings using high resolution satellite imagery and vector dataset

تعيين ميزان تخريب ساختمان ها در مدت زمان كوتاهي پس از وقوع زلزله،نقش مهمي در برنامه ريزي جهت اعزام گروه هاي امداد رساني در محل حادثه دارد.بدين منظور نياز به اطلاعات مكاني از قبل و بعد از وقوع زلزله مي باشد كه در اين تحقيق،از نقشه مربوط به قبل و تصوير ماهواره اي مربوط به بعد از زلزله استفاده مي شود و با استفاده از باند هاي طيفي و آناليز بافت،فضاي توصيف براي پيكسل هاي ساختماني تشكيل شده و با استفاده از الگوريتم ژنتيك فضاي توصيف بهينه ايجاد مي شود و پيكسل ها با دو روش طبقه بندي بيشترين شباهت و شبكه عصبي براي دو كلاس تخريب و عدم تخريب طبقه بندي مي شوند و دقت طبقه بندي حاصل از اين دو روش مقايسه مي شوند به طوري كه دقت كلي طبقه بندي به روش بيشترين شباهت 97 درصد و به روش شبكه عصبي 99 درصد براي داده هاي چك بدست مي آيند.درنهايت پلي گون هاي ساختماني بر اساس تعداد پيكسل هاي تخريب شده هر پلي گون به سه كلاس تخريب بالاي 70%،تخريب بين 30%تا70% و تخريب زير 30% تقسيم مي شوند.

Damage building assessment after an earthquake is an essential analysis for saving people and their assets. For doing this, it is needed to have before and after earthquake spatial data. The aim of this study to assess the building damage distribution in the urban area of Bam, Iran, using postearthquake QuickBird high-resolution optical satellite images and before earthquake vector map to produce a damage map.This paper investigates the effect of spectral and texture features for detecting damage buildings from after earthquake QuickBird image and select optimum features using of genetic algorithm then, the pixels in the each building area are classified with ML and ANN classification method as ANN classifiers resulted better classification accuracy with 99% overall accuracy rather than ML 97% overall accuracy on check data. Finally all buildings are classified in three classes: Upper 70% damage, between 30% to 70% damage and lower 30% damage.
 Abstracts of Papers in English simple DLT, are in use. However, the great difficulties with these formulations are their dependence on large number of well distributed GCPs. The most precise generic model which is used as a substitute for rigorous model is rational function model. It has been proven that RFMs in terrain independent mode, which is calculated by fitting to rigorous model, have the accuracy of rigorous model. Nevertheless, a very interesting parametric approach, which uses simple 2D to 3D affine transformation, has been experimentally proven to be very promising. This approach has been already evaluated by different researchers worldwide and reasonably accurate results in both flat and hilly terrains have been reported using only few numbers of GCPs. The theoretical basis that justifies the achieved accuracy is the fact that with the high resolution satellite images the very small camera field of view and the high altitude makes the incoming signals almost parallel. This renders the perspective geometry along the scan lines to approach the parallel geometry and effectively a homogenous geometry in the scan line and the direction of the satellite motion is produced. This particular geometry provides a simple linear relationship between the image space and the object space and makes a simple eight parameters affine transformation optimum for geo referencing applications. Simplicity of the formulation (i.e. only eight affine parameters for the entire scene and linear form of the equations), few numbers of required GCPs and the achieved accuracy makes this approach very attractive from the mapping point of view. Affine transformation is followed, along terrain independent RFMs, by many investigators who claimed that it has the potential to achieve high accuracy same as RFMs. This paper shows and discusses the fitting accuracy of affine model, applying for Cartosat-1 images over Roodehen city in Iran, to rational terrain independent model and concentrates on its limitations then works on different correction models, topography and curvature correction, to improve planimetric as well as altimetric accuracies.