Signboard Detection and Text Recognition Using Artificial Neural Networks

A sub-field of Artificial Intelligence (AI) is machine learning (ML) which allows the computer to learn and adopt new rules. With the ML algorithm, the computer can identify samples of the observations, figures and other formats, which may describe the model or other structure. The machine observes various events from the world and presents things with/without explaining the pre-programmed rules. This paper presents the recognition of scene text from the outside environment focusing signboards. A framework for text detection and recognition of text from the natural environment is presented. Advance steps from paper-based text recognition; the text recognition from natural scenes are divided into several levels. Firstly, the image is captured from the outside environment with a smart device, followed by detection of the edges of a signboard. The next phase is the detection of text and the recognition of the text into two languages such as Urdu and English. Final phase uses Artificial Neural Network for the classification and recognition of the text extracted from the natural scenes or an outside environment. Experimental results have been generated on the developed image database created as the part of this research. The effort is a multilingual and produces output in Urdu and English. The system performs well as compared to traditional recognition systems and delivers an overall 85% accuracy in image results.