Fast Prototyping of In-Pavement Airport Navigation Lamp Prism Classification

The paper presents an analysis of fast selection of neural network for the purpose of visual analysis of mechanical wear on prism lenses of in-pavement airport navigational lighting systems. This issue is particularly important in terms of aviation safety and navigational lighting control, regulated by EASA and ICAO. The article is the next stage of the development of the system for the vision control of lamps, in which the concept of using a different neural network with an increased data set prepared by the authors is presented. The Deep Network Designer tool included in the Matlab 2022b environment was used. The solution using the GoogLeNet neural network allows for the classification of lamps with an accuracy of 88.37%.