Image search program

Abstract

An image search program includes a feature amount data extraction unit for extracting feature amount data of a plurality of types (such as layout feature amount, texture feature amount, and shape feature amount) from key image data, adaptation judgement unit for judging presence / absence of adaptation depending on whether the similarity calculation is matched with the human similarity sense for the types of the feature amount data extracted by the feature amount data extraction unit, a similarity calculation unit for calculating similarity between the key image data and each image data to be searched, according to the result of judgement of the adaptation judgement unit, and a search unit for outputting the image data corresponding to the similarity as a search result.

Description

technical field [0001] The present invention relates to a technique for searching a desired image from an image database based on a keyword image. Background technique [0002] So far, many image retrieval devices have been developed that have the function of sequentially fetching and displaying image data similar to a keyword image given by a user as a retrieval request from an image database storing a large amount of image data, so that The user can efficiently find desired image data. [0003] In such an image retrieval device, a similar image retrieval technique is used, and since the similarity between keyword image data and image data to be retrieved is calculated numerically, a feature quantity representing a characteristic of image data is defined numerically. This feature amount is used in the similarity calculation of images. [0004] Here, among the above-mentioned feature quantities, there are arrangements based on the colors possessed by the captured image, te...

AI Technical Summary

Problems solved by technology

[0011] However, when an image data set containing image data of various contents, such as image data collected using the Internet that has spread in recent years, is used as a search object, many images that do not match the type of feature amount are included. data, there will be a problem of decreased retrieval accuracy

Images