VOCAL VISION For VISUALLY IMPAIRED Shrilekha Bangar PreetamNarkhede Rajashree Paranjape Abstract This project is a vision substitute system designed to assist blind people for autonomous navigation. Its writing concept is based on 'image to sound' conversion. The vision sensor captures the image in front of blind user. This image is then fed to SIA71AB for processing. Processing unit processes the captured image and enhances the significant vision data. This processed image is then compared with the database kept in microcontroller. The processed information is then presented as a structured form of acoustic signal and it is conveyed to the blind user using a set of earphones. Colour information from the interested object is evaluated to determine the colour of the object. The colour output is informed to the blind user through headphones. Keyword: object identification, visually impaired, colour detection, edge detection INTRODUCTION World Health Organization (WHO) reported that there are currently about 45 million blind people worldwide and every war this number is increases by 1 to 2 million. Blind people's navigation is restricted because they do not receive enough information about the objects or obstacles in their environment. Blind and visual unpaired face many problems during their daily activities with navigation and finding objects. When processing an image to determine the color of an object within, it is necessary to extract the object from the image, thus removing the background If the background remains m the image when it is processed it will distort the analysis of the image and produce inaccurate results. To extract the object from its background we first have to determine where the object is or more specifically, which region of the image belongs to the object. Image of the scene in front of blind user is captured using video camera and it is transformed into sound pattern. The intensity of the pixel of the image is transformed into loudness. In real world scene, background fills more area than the objects. It is also noted that most of the background is of light intensity and the sound produced on it will be of high amplitude compared to the objects in the scene. As a result, the sound produced from the unprocessed image may contain more information of the background rather than object. This may be the reason for the blind user's difficulties to distinguish the object and the background of the environment This proposed design incorporates the interfacing of the vision sensor with ARM processor to capture the image of the object which is present in front of the visually impaired person. This processing results out in the identification of the object as well as to recognize its color in the captured image The extra feature which is included in this design is to measure the distance of the object from the person which facilitates them to know the exact distance of the object. LITERATURE SURVEY: Blind and visual impaired face many problems during their daily activities with navigation and finding objects Many technologies supports blind, visual impaired and people with other kinds physical challenges. For example, scanners for the blind, facsimile devices for the deaf and remote controllers for the physical challenged Some of the early developments relating to independent navigation of visually impaired have been well covered in historical publications by P Heckbert (1982), C. L. Novak, and S. A. Shafer (1987),P.B.L. Meijer. (1992) and, more recently, in IEEE by R Nagarajan,(2001). However, it is quite recently that individual navigation for visually impaired have appeared that are capable of delivering believable sensory stimuli at a reasonable cost using different technologies of a practical size.