انتخاب مكان مناسب براي فرود اضطراري وسايل نقليه ي هوايي بدون سرنشين با استفاده از تكنيك هاي بينايي ماشين و پردازش تصوير

چكيده انگليسي :

Landing Site Selection for UAVs Forced Landing Using Machine Vision and image processing Parvin Ahmadi p ahmadi@ec iut ac ir Date of Submission 03 09 2011 Department of Electrical and Computer Engineering Isfahan University of Technology Isfahan 84156 83111 Iran Degree M Sc Language Persian Supervisors Dr Saeed Sadri Sadri@cc iut ac ir Dr Rasul Amirfattahi fattahi@cc iut ac ir Abstract While using unmanned systems in combat is not new what will be new in the foreseeable future is how such systems are used and integrated in the civilian space The potential use of Unmanned Aerial Vehicles in civil and commercial applications is becoming a fact and is receiving considerable attention by industry and the research community The majority of Unmanned Aerial Vehicles performing civilian tasks are restricted to flying only in segregated space that are typically not above populated areas which in turn are one of the areas most useful for civilian applications The reasoning behind the current restrictions is mainly due to the fact that current UAV technologies are not able to demonstrate an equivalent level of safety to manned aircraft particularly in the case of an engine failure which would require an emergency or forced landing Piloted aircraft in the same scenario have a human on board that is able to engage in the complex decision making process involved in the choice of a suitable landing location If UAVs are to ever fly routinely in civilian airspace then it is argued that the problem of finding a safe landing location for a forced landing is an important unresolved problem that must be addressed This thesis presents the results of an investigation into the feasibility of using machine vision and image processing techniques to locate candidate landing sites for an autonomous UAV forced landing To this end some areas are extracted from aerial images which are suitable for UAV landing and others are reported as unsuitable or unknown areas Aerial image segmentation is the main part of this thesis Defining the surface type of each segment is also necessary Recently Active Contours and their implementation by Level Set method have received a great deal of attention within the image community especially as a framework for dealing with image partitioning issues This method allows introducing different kinds of information as constraints texture features color features etc to image partitioning process A new model for active contours called region based active contour Chan Vese model has been proposed to detect objects in a given image which is more powerful than classical edge based active contours In this thesis for finding the suitable landing sites firstly aerial images are segmented into a number of homogeneous areas using Chan Vese model implemented by Narrow Band Level Set method with re initialization together with extracting color and texture features For this purpose different kinds of color and texture features have been tested and resulted in choosing Gabor filters in HSV color space for the best segmentation result After aerial images segmentation surface type of individual areas is classified to an accuracy of about 97 using k nearest neighbor KNN algorithm and the same color and texture features Finally a geometric test for an appropriately sized landing site is performed Kay Words aerial image segmentation features extraction Active Contours surface type classification Unmanned Arial Vehicles UAVs emergency or forced landing

احمدي، پروين

انتخاب مكان مناسب براي فرود اضطراري وسايل نقليه ي هوايي بدون سرنشين با استفاده از تكنيك هاي بينايي ماشين و پردازش تصوير

قطعه بندي تصاوير هوايي , استخراج ويژگي هاي كانتورهاي فعال , طبقه بندي نوع سطح