Tomáš Zemčík, Lukáš Kratochvíla, Šimon Bilík, Ondřej Boštík, Pavel Zemčík & Karel Horák International Journal of Image Processing (IJIP), Volume (15) : Issue (1) : 2021 1 Performance Evaluation of CNN Based Pedestrian and Cyclist Detectors On Degraded Images Tomáš Zemčík zemcikt@feec.vutbr.cz Faculty of Electrical Engineering and Communications Dept. of Automation and Instrumentation Brno University of Technology, Brno, Czech Republic Lukáš Kratochvíla kratochvila@feec.vutbr.cz Faculty of Electrical Engineering and Communications Dept. of Automation and Instrumentation Brno University of Technology, Brno, Czech Republic Šimon Bilík bilik@feec.vutbr.cz Faculty of Electrical Engineering and Communications Dept. of Automation and Instrumentation Brno University of Technology, Brno, Czech Republic Ondřej Boštík bostik@feec.vutbr.cz Faculty of Electrical Engineering and Communications Dept. of Automation and Instrumentation Brno University of Technology, Brno, Czech Republic Pavel Zemčík zemcik@fit.vutbr.cz Faculty of Information Technology Dept. of Computer Graphics and Multimedia Brno University of Technology, Brno, Czech Republic Karel Horák horak@feec.vutbr.cz Faculty of Electrical Engineering and Communications Dept. of Automation and Instrumentation Brno University of Technology, Brno, Czech Republic Abstract This paper evaluates the effects of input image degradation on performance of image object detectors. The purpose of the evaluation is to determine usability of the detectors trained on original images in adverse conditions. SSD and Faster R-CNN based pedestrian and cyclist detector performance with images degraded with motion blur, out-of-focus blur, and JPEG compression artefacts, most commonly occurring in mobile or static traffic systems. An experiment was designed to assess the effect of degradations on detection precision and cross class confusion. The paper describes the two datasets created for this evaluation, evaluation of a number of detectors on increasingly more degraded images, comparison of their performance, and assessment of their tolerance to different types of image degradation as well as a discussion of the results. Keywords: Object Detection, Image Degradation, Pedestrian Detection, Cyclist Detection, SSD, Faster R-CNN. 1. INTRODUCTION Image object detection is one of the keys to solving technical problems in machine vision-based traffic application. CNN object detection architectures currently seem to be the most promising ones towards robust and reliable image object detection in very difficult traffic environments. With