Automatic Identification and Mapping of Cone-Shaped Volcanoes Based on the Morphological Characteristics of Contour Lines

Cone-shaped volcanoes have important research significance and application value due to their typical cone shape unique structural features. The existing methods for recognizing are mainly morphological feature matching machine learning. In general, the former has low recognition accuracy, while latter requires a large number of training samples. contour lines cone-shaped distributed in concentric circles. Furthermore, from center outwards, elevation increases first then decreases. Based on characteristics Hough transform algorithm, main algorithm includes (1) preliminary filtering lines, (2) circular based random transform, (3) grouping trees, (4) concentric-circle (5) automatically mapping volcanoes. Case studies demonstrate effectiveness this method detecting Western Galapagos shield Mariana Trench submarine volcano group. proposed missed false alarm rates, which is basically consistent with manual results. This can effectively recognize landscapes powerful means support deep-space deep-sea exploration.