Design Of Airport Runway Foreign Object Debris System

Foreign Object Debris(FOD)are objects left on the airport runway.With the development of the global aviation industry,the safety of airport runways has also attracted much attention.In order to ensure flight safety and reduce the occurrence of accidents,countries have put forward higher standards and requirements for foreign object debris on airport runways.At present,the detection of foreign objects on airport runways is mainly based on computer vision technology.However,the existing automatic detection equipment is limited by the way of installation and detection accuracy,and cannot effectively detect small foreign objects.At present,small foreign objects are still checked manually.In order to improve the detection accuracy of automatic detection equipment,this paper conducts research on the mobile airport runway foreign object detection system.The main work content of this paper is as follows:1.A mobile detection method suitable for foreign object debris on airport runways is proposed.By improving the acquisition method of the image required by the target detection algorithm and replacing the multi-frame image with multi-eye images,the problem of algorithm failure in the current double-frame method/three-frame method in low-speed detection(the extreme case is a static state)and high-speed detection can be solved.The existing "double shadow" problem improves the robustness of the mobile detection system,enabling the system to perform target detection at any speed.2.Aiming at the non-uniform field of view problem caused by the use of multi-eye cameras,through the analysis of the image registration algorithm and the existing feature point algorithm,it is determined that the SIFT algorithm is used to achieve accurate matching of image feature points,and the perspective transformation is used to achieve image alignment.Fast registration enables fast and precise adjustment of the multi-camera field of view.3.According to the proposed foreign object detection method on the airport runway,every link from the real-time image data acquisition to the final foreign object labeling is analyzed.Firstly,the image brightness adjustment algorithm and image real-time acquisition system are designed to ensure the real-time acquisition of four-eye images;secondly,the four-eye image registration process based on perspective transformation algorithm and SIFT algorithm is given;then,the image-based spatial domain and frequency domain to optimize the image to improve the robustness of the detection system;finally,to extract the foreign object area in the image after effects,a threshold separation method based on the exponential function characteristics is proposed,and the extraction of the foreign object foreground is realized by using the scaling characteristics of the exponential function.4.Design and implement the FOD detection software based on Python and Py Qt platform,and apply it to the FOD detection system.The test results show that the airport runway foreign object detection method proposed in this paper has a high accuracy rate in small foreign objects.