Neural networks trained on RGB and monochromatic images are tested on images augmented by polarimetry for recognition of road-based objects. The goal of this work is to understand the scene conditions for which object detection and recognition can be improved by linear Stokes measurements. Shadows, windows, low albedo, and other object features which reduce RGB image contrast also decrease neural network detection performance. This work demonstrates specific cases for which linear Stokes images increase image contrast and therefore increase object detection by a neural network. Linear Stokes videos for five difference scenes are collected at three times of day and two driving directions. Although limited in scope, this work demonstrates some enhancement to object detection by adding polarimetry to neural networks trained on RGB images.


    Access

    Check access

    Check availability in my library

    Order at Subito €


    Export, share and cite



    Title :

    Road scene object detection using pre-trained RGB neural networks on linear Stokes images


    Contributors:

    Conference:

    Polarization: Measurement, Analysis, and Remote Sensing XIV ; 2020 ; Online Only,California,United States


    Published in:

    Proc. SPIE ; 11412


    Publication date :

    2020-05-26





    Type of media :

    Conference paper


    Type of material :

    Electronic Resource


    Language :

    English



    OBJECT DETECTION USING MULTIPLE NEURAL NETWORKS TRAINED FOR DIFFERENT IMAGE FIELDS

    IANCU SABIN DANIEL / WANG BEINAN / GLOSSNER JOHN | European Patent Office | 2022

    Free access


    Real-time road scene classification using infrared images

    Forslund, David / Cronvall, Per / Roll, Jacob | Tema Archive | 2010