| 主权项 |
1. An image processing system comprising:
a first data input unit to which a captured image, which is obtained by capturing an image of a scene viewed from a vehicle, is input; a first feature point extraction unit that extracts image feature points from the captured image input to the first data input unit; an image-capturing situation information obtaining unit that obtains image-capturing situation information indicating a possibility that a specific subject is included in the captured image input to the first data input unit; a feature point importance degree determination unit that assigns one of a plurality of importance degrees to each of the image feature points extracted by the first feature point extraction unit, based on the image-capturing situation information, each of the plurality of importance degrees corresponding to a different importance level; an image feature point data generation unit that generates image feature point data for each captured image using the image feature points extracted by the first feature point extraction unit, based on the importance degrees, the generated image feature point data including important ones of the extracted image feature points and excluding unimportant ones of the extracted image feature points, the importance of each extracted image feature point being determined by the importance degree of the extracted image feature point; and a reference data database creation unit that generates reference data used when scenic image recognition is performed, by associating the image feature point data with image-capturing attribute information, and creates a reference data database that is a database of the reference data, wherein the image-capturing attribute information includes an image-capturing position at which the image is captured to obtain the captured image corresponding to the image feature point data, the image feature point data in the reference data database being searchable using the image-capturing position. |
