In this project, we explain how an image can be recognized using Finite and Pushdown Automata.
Each step in the process of recognition is represented as a state in the automata. The image to be recognized can be of any dimension as the steps to interpret the image remain the same and only the algorithm needs to be tweaked for images of different dimensions.
We focus on how to derive the language and grammar from the finite automata and also check if it is regular. Further, we construct the pushdown automata from the transition function computed and also determine whether the automata constructed is deterministic or non-deterministic.
The approach we have implemented for determining and constructing the language, grammar and automata for the image is simpler and more understandable compared to the pre-existing studies. This enables it to be applied not only for image recognition but in various other fields.