标题：Impact of Image Corruptions on the Reliability of Traffic Sign Recognition Using Machine Learning Technique
作者：Ao, Boyu; Ren, Jiachang; Guo, Chen
作者机构：[Ao, Boyu] Wuhan Univ, 299 Bayi Rd, Wuhan, Hubei, Peoples R China.; [Ren, Jiachang] Shandong Univ Qingdao, 72 Binhai Rd, Qingdao, Shandong, Peoples 更多
会议名称：3rd International Conference on Compute and Data Analysis (ICCDA)
会议日期：MAR 14-17, 2019
来源：PROCEEDINGS OF THE 2019 THE 3RD INTERNATIONAL CONFERENCE ON COMPUTE AND DATA ANALYSIS (ICCDA 2019)
关键词：Machine learning; traffic sign recognition; image corruption; neural; networks
摘要：As autonomous driving gets closer to be widely applied, it is important to guarantee that traffic signs are recognized correctly. Due to change of light conditions and blockage by some other objects, traffic signs can sometimes be partially corrupted. In this paper, we evaluated how machine learning would respond to different types of image corruption by various degrees. Removing a higher percentage of pixels will gradually harm the recognition accuracy, and removal by blocks causes the biggest harm, which is consistent with human observation. Changing to various colors or a single color doesn't seem to cause significant differences. This study, by building a model for traffic sign recognition and evaluating its robustness to various types of image corruptions, provides insights into corruptions of datasets for machine learning in general, and provide potential concern for applying the well-trained model to a new test set with corruptions, which could be a concern for applying autonomous driving in certain areas.