Test scenario for road sign recognition systems with special attention on traffic sign anomalies

The accurate identification and recognition of horizontal and vertical traffic signs can cause problems for autonomous vehicles. That is why have checked anomalies of traffic signs in the topic. Because for the safely navigate an autonomous vehicle, it is necessary to have a very accurate environment recognition. However, there are many problems with the sensors recognitions ability. The current article presents a test using a previously created anomaly classification table. The table divides the signals into five anomalies groups that are real and can cause problems with the recognition systems. This article with the tests highlights the limit values and circumstances that make it challenging to identify the traffic sign. The environment, weather, and visibility conditions, as well as different traffic situations, are typical problems. The test was done on the ZalaZONE, SMART city track in Hungary. The study can further develop a critical test environment for traffic signs to help develop future autonomous vehicle systems. It can use for environment recognition systems tests and validation procedures.