Feature extraction and edge detection using a fractal-based metric

Fractal error is an image processing metric that can be used to locate man-made features in aerial images, as well as edges in industrial images. The metric can aid photo interpreters in locating targets in aerial reconnaissance images. Since the development of fractal error, it has been shown that the fractal error metric also works well for extracting features in synthetic aperture radar (SAR) images. A novel method is presented for automatic 2-D entropic segmentation using fractal error as a feature.; Previous work has also shown that the fractal error metric is useful for locating edge pixels in industrial images. An introduction to edge detection using fractal error is presented; the results of the fractal error edge detection algorithm are compared with the Canny edge operator for robustness and accuracy.; Fractal error has many useful applications; however, some applications require real-time image analysis. The main disadvantage of the fractal error algorithm is that it can take several seconds to compute on large images. Therefore, it is desirable to create an approximation of fractal error to provide real-time image analysis. A novel approximation of fractal error using a genetic algorithm is also presented.