In order to estimate the potential hazards and adopt strategies for preventing the accidents, this paper is about the research of auto-recognition pedestrians. Using the captured images of the vehicle recorder, based on the HOG feature processing technology and SVM classifier, the automatic detection and recognition of pedestrians are realized. In the past, the cost of pedestrian identification system is too high and its universality is poor. In this research, optimization methods of reasonable sample allocation and difficult training are used to make the training model more general and effective.


    Zugriff

    Zugriff prüfen

    Verfügbarkeit in meiner Bibliothek prüfen

    Bestellung bei Subito €


    Exportieren, teilen und zitieren



    Titel :

    Auto-recognition Pedestrians Research Based on HOG Feature and SVM Classifier for Vehicle Images


    Beteiligte:
    Yunsheng, Li (Autor:in) / Jie, Cao (Autor:in) / Xuewen, Chen (Autor:in) / Feng, Zhao (Autor:in) / Jingling, Li (Autor:in)


    Erscheinungsdatum :

    28.09.2020


    Format / Umfang :

    880959 byte




    Medientyp :

    Aufsatz (Konferenz)


    Format :

    Elektronische Ressource


    Sprache :

    Englisch




    Visual vehicle recognition system for pedestrians

    Europäisches Patentamt | 2022

    Freier Zugriff

    Visual vehicle recognition system for pedestrians

    CHOI TAE JUNE / HWANG YU JUNG / YU JI HEE et al. | Europäisches Patentamt | 2022

    Freier Zugriff

    Motion-based recognition of pedestrians

    Heisele, B. / Woehler, C. | Tema Archiv | 1998