Scene Text Recognition (STR) enables the Advanced Driver Assistance System (ADAS) to recognize text in natural context, such as object labels, instructions, and text-based traffic signs. STR helps self-driving cars make informed decisions such as which direction to take, how fast to go, and what to do next. Traffic signs are categorized into three categories: traffic lights based on symbols and texts, and additional traffic. Traffic signs recognition is a very important task in ADAS, although many researchers have had impressive success with symbol-based traffic signs, there are very few researchers working on the other types of signs due to the difficulties they encounter, chief among which is the lack of publicly available datasets. In addition to the many factors that make text-traffic signs difficult to recognize, including complex backgrounds, noise, lightning conditions, different fonts, and geometric distortions in the signs. In this paper, we will survey some modern and effective methods of scene text recognition and discuss some of the problems they face, taking a closer look at the problem of text recognition of traffic signs in the first place.