Effects of enhanced image quality in infrastructure monitoring through micro aerial vehicle stabilization

Traditional monitoring of large infrastructure such as towers of churches or for cooling, chimneys or any other type of tall buildings can require extreme effort and hence become very risky and costly since it mainly requires people to be moved around. An interesting alternative in that regard is the use of rotary wing micro aerial vehicles (MAV) equipped with sensors such as digital cameras to capture series of images and stitch them on to a 3D model. However the images recorded always have "noise" which is caused by the vehicle movement during the inspection process. Experiments have been carried out using a multi-rotor copter with an off the shelf camera for building inspection (aerial photography) for years [1]. To validate the effect of MAV flight stabilization a vector thrust principle recently developed at Saarland University (patent application in progress [18]) has been implemented onto a MAV for flight stabilization. The paper describes the procedure of MAV-ba sed infrastructure monitoring as well as the image stitching process in general before explaining the vector thrust design principle and the enhancements achieved in terms of image resolution and processing.