• Sure. Venkata. Padmavathi Devi
• D. Murugan

Abstract: This paper developed a robust automatic approach for retrieving the man-made objects (buildings) automatically from the given satellite images Retrieving features from remote sensing image is a significant duty prior to classifying that image. This article first applies preprocessing step on the satellite imagery. After preprocessed the image, then the partition phase is followed. At the stage of segmentation, the satellite image is divided into vegetation and non-vegetation regions. The efficient algorithms named Improved Dist Artificial Bee Colony (IDABC) have been designed for Segmentation purpose. The vegetation regions were left, and only the Non-vegetation area was remaining. For deriving the features, deep features are employed to illustrate each non-plant area. These deep features are retrieved by Transform Encoded Deep Local Binary Pattern (TEDLBP). After deep features are produced, the classifier which is trained earlier was used to classify non - plant areas into the buildings. The ELM method is employed for classification purpose. Finally, the change detection is done to the building image by distance measure. At last, the detection results are validated and analyzed through the evaluation metrics such as Detection Accuracy, and Error rate. From the evaluation, it is proved that the newly developed algorithms provide the best result in the building extraction process.

Building Extraction, Vegetation, Non Vegetation, Wavelet Shrinkage, IDABC, TEDLBP, ELM

"An Efficient Robust Approach for Change Detection of Buildings From High-Resolution Satellite Images using IDABC and TEDLBP", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 12, page no.264-276, December-2018, Available :http://www.jetir.org/papers/JETIR1812033.pdf

"An Efficient Robust Approach for Change Detection of Buildings From High-Resolution Satellite Images using IDABC and TEDLBP", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 12, page no. pp264-276, December-2018, Available at : http://www.jetir.org/papers/JETIR1812033.pdf