A Contour Tracing Algorithm that Avoids Duplicate Tracing Common Boundaries between Regions

The popularity of digital devices has greatly increased the chance of manipulating images on computers. In general, image data can be largely divided into raster format and function format. A raster image is composed by a group of pixels arranged in a square tessellation, with each having a certain value or label denoting the color information. This format is suitable for representing images on modern CRT and LCD displays. However, when raster images are subjected to affine transformations like enlargement, reduction, rotation, and translation, the image quality deteriorates visually in the form of jaggy-ness occurring along the region boundaries (Figure 1). Compared to this, the function format is suitable for representing images that exhibit clear region boundaries. Moreover, the function format has the added advantage that the visual quality of images can be maintained even after being subjected to affine transformations. In our research, a series of techniques on automatic function approximation of raster images have been developed, two of which were published in this journal . In our techniques, contours are traced and their shapes are approximated by a combination of functions including straight line, arc, and quadratic curve. Because our previous works had targeted bi-level or two-labeled images, conventional contour tracing algorithms that focus on a single region at a time during tracing were suffiA Contour Tracing Algorithm that Avoids Duplicate Tracing Common Boundaries between Regions