The angular orientation partition edge descriptor

Edges are one of the most important image visual features. They are highly related with shapes and can also be representative of the image textures. Edge orientations histograms are usually very reliable descriptors suitable for image analysis, search and retrieval. In this work edges detected with Canny algorithm are described by their angular orientations. The resulting descriptor is resilient to image rotation and image translation. It is also resilient to noise. An example of automatic image semantic annotation using this description method is reported using a database with 738 images. The K Nearest Neighbor is used as classifier and the Manhattan distance is used for image similarity computation. The annotation that results with this description method is compared with the provided with other well known descriptors. These examples show that a reliable high level automatic description based in the semantic content can be extracted.