Research On The Measurement Of Navigation Light Signage

With the rapid development of Civil Aviation Industry of China in recent years,China has become the second largest civil aviation market in the world after the United States.With the continuous expansion of aviation scale,security pressure of civil aviation will inevitably become more and more serious.As a set of information indicating systems in the airport,the signage can help aircraft and vehicle drivers to drive safely and efficiently in the airport.Therefore,it is of great significance to check the qualifications of the tags for ensuring the safe and effective operation of the airport.Based on this demand,the key measurement technologies of signage information are studied in this paper.Combining with advanced visual measurement technology,this paper designs a measurement system which can automatically measure the key information of the signage.Aiming at the unstable performance of classical image algorithms for text detection and recognition at different scales,an improved ORB operator-based text detection algorithm is studied in this paper.This method makes full use of the ORB operator detection performance,and improves the robustness of the algorithm to different scale text through feature fusion,multi-scale detection and edge removal algorithm.In order to improve the generalization of the existing model and the performance of the model in the real scene,a network structure with domain invariance is designed for detection task of signage board text,which helps improve the accuracy and efficiency of the algorithm without increasing the computation.In the process of evaluating the color and illumination performance of signages,the key technologies involved in the research,such as image signal processing and camera internal parameter distortion correction,are summarized and analyzed.Finally,a set of better measurement schemes and evaluation criteria are put forward according to the current research status.The results show that the improved ORB operator-based text detection algorithm proposed in this paper can extract more effective feature points for different scales of text,and has higher positioning accuracy.The cross-domain tag text detection algorithm can effectively make the model pay more attention to the target itself,reduce the feature divergence caused by appearance differences,and improve the robustness and generalization ability of the model in different test environments.The method of measuring the label's license and chroma information effectively reduces the measurement error,and improves the accuracy and scientificity of the performance evaluation experiment of the navigational light label.