Applications of Fractal and Wavelet to Feature Extraction

Within this paper a new feature extraction techniques is presented which uses wavelet analysis and fractal theory for image recognition. The proposed method reduces the dimensionality of a twodimensional pattern by way of a central projection approach, and thereafter, performs Daubechies' wavelet transformation on the derived one-dimensional pattern to generate a set of wavelet transformation sub-patterns, namely, curves that are non-selfintersecting. Further from the resulting non-selfintersecting curves, the divider dimensions are computed with modi ed box-counting approach. These divider dimensions constitute a new feature vector for the original two-dimensional pattern, de ned over the curve's fractal dimensions.