Today, in the 21st century, China's national economy continues to maintain stable growth, and the public generally pursues comfort and short-term travel, so the airplane has become the preferred means of transportation for the public. To realize the rapid recognition and classification of aircraft targets in aerial images, an aircraft target recognition method based on YOLOv5 framework was proposed. As a typical representative of fine-grained image classification, the accuracy of aircraft classification is seriously affected by its shape, color, and strong interference factors of arbitrary modification. In order to improve the accuracy of target recognition, overlapping cropping, rotation, scaling, adjusting saturation and brightness, and increasing noise were used to increase the number of sample features: in order to solve the problem of excessive change in the size of the aircraft target, samples of different scales were put into the YOLOv5 framework for training, and the corresponding detectors were obtained, and the prediction results of multiple detectors were synthesized, and the position and category of the target frame were determined by non-maximum suppression. A small sample set containing a variety of aircraft types was selected and annotated from the public remote sensing dataset, and the results were tested. Experimental results show that the deep learning algorithm performs well in the target recognition task of the airport flight area, and can realize the task of classification of active targets in the flight area.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Research on Recognition and Classification of Active Targets in Airport Flight Area Based on Improved Deep Learning Algorithm


    Contributors:


    Publication date :

    2024-03-29


    Size :

    804874 byte




    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English




    Aircraft image recognition in airport flight area based on deep transfer learning

    Yang, Lijun / Tao, Zheng | British Library Conference Proceedings | 2021


    Airport Flight Movement Area Safety Inspection System

    European Patent Office | 2018

    Free access

    Airport Flight Movement Area Safety Inspection System

    KIM DONG CHUL / KIM DOO YOON / LIM HYE RIN et al. | European Patent Office | 2018

    Free access