A coarse-to-fine approach for industrial meter detection and its application
Object recognition is a rapidly developing area of research. In our work we concentrate on detection and recognition of planar objects, like images and logos. a planar object is located in an image by finding a homography between the images. We describe the overall methodology for finding this homography using local feature description and matching. We also describe the most prominent algorithms used for locating and describing local features. We present multiple criteria for the homography to reduce the number of false positives and measure their impact on hand-made ground truth datasets.