Contour pattern recognition through auditory labels of freeman chain codes for people with visual impairments

Tactile devices are commonly used for pattern recognition learning by people with visual impairments. However, it is often difficult for such persons to recognize the detailed and mathematical shape of objects when using such devices. The primary difficulty is gaining a sufficiently accurate understanding of the shape of an object to permit its correct reconstruction based on tactile device feedback. Furthermore, people with visual impairments have less opportunity to discern visual images and must therefore apply their own sense of understanding to the shape of an object as gained from the tactile experiences. Thus, they are often not sure if it reflects the actual shape. To solve this problem, we propose a pattern recognition technique utilizing auditory labels based on a contour chain code. Using a Freeman chain, object shapes are represented as a sequence of steps, each in one of eight directions; each step is designated by an integer from 0 to 7. Our proposed method takes advantage of Freeman chain codes in order to present the detailed shape of objects. This report describes our proposed method and presents experimental results that verify its effectiveness in support of mathematical education efforts for visually impaired persons.