Published Paper Details:

Here a framework has been proposed that peruses the content experienced in regular scenes with the mean to give help to the visually impaired people. This paper portrays the framework plan and assesses a few character extraction techniques. Programmed content acknowledgment from normal pictures gets a developing consideration due to potential applications in picture recovery, mechanical autonomy and clever transport framework. Camera-based archive examination turns into a genuine plausibility with the expanding determination and accessibility of advanced cameras. Notwithstanding, on account of a visually impaired individual, finding the content is the primary imperative issue that must be tended to, in light of the fact that it can't be expected that the obtained picture contains just characters. At to start with, our framework tries to discover in the picture regions with little characters. At that point it zooms into the discovered regions to retake higher determination pictures vital for character acknowledgment. In the present paper, It has been proposed four character-extraction techniques in light of associated segments. The execution of the diverse techniques relies upon character measure. In the information, greater characters are more common and the best extraction technique turns out to be the succession: Sobel edge discovery, Otsu binarization, connected component extraction and rule-based connected component filtering.