Method for searching image based on image and video content in webpage

Abstract

The invention relates to a method for searching an image based on an image and video content in a webpage, which comprises the following steps of: 1) receiving an image research request, 2) pre-treating a researched image, 3) normalizing the image research, 4) searching a target image, 5) setting a normalized true target and turning back to step two if user requires to adjust, or skipping to step six, 6) evaluating a user community and 7) listing a final researched targeted images. The method of the invention has better property, is more intelligent and is more practical.

Description

technical field [0001] The invention relates to an information search method in the Internet, in particular to an image search method for information such as images and videos in the Internet. Background technique [0002] As the Internet has increasingly become a commonly used and important source of information in daily life, many e-commerce has grown and matured day by day based on search engines based on Internet information. [0003] The usual pattern is: people find a certain type of things they are interested in from massive information, and then use general or vertical application search engines to find them, and then go to e-commerce websites to search, that is, continue to use e-commerce websites to provide The site's search engine to locate the specific things you want to buy. The whole process is a process of consumers' thinking and behavior from rough to fine, in which search engines play a very key role. In this process from information to business, user beha...

AI Technical Summary

Problems solved by technology

[0006] It can be seen that most of the so-called "image search" image search engines are essentially text keyword searches, usually collecting images from the Internet, and then extracting some text information of the webpage where the image is located to create a tag index for the image, which is simpler Even just extract the title of the web page where the image is located as the image index, when the user submits a query to the search engine, only consider the similarity between the query and the pre-extracted information and return the result

[0007] In the field of real image search, there are currently two levels that are not considered

[0008] First: At the technical level of image recognition, due to the limitation of technological development, the current solutions only consider the color distribution and texture distribution of the image, and do not consider the block characteristics, shape characteristics, The role of more dimensional rich information such as contrast features in image recognition, and use these elements to enrich and build image feature libraries

[0009] Second: At the business application level of image search engines, the current image submission and result ranking methods often adopt a relatively simple method like text search output results, and the precision rate of image search is not ideal, and only focuses on In terms of accuracy, it does not consider that users may only be interested in the part of the image they see, nor does it consider the copyright issue of the image, and it does not consider what the user wants to obtain through the image search service from the perspective of user needs.

[0010] Moreover, the existing image search engines hardly consider the following two situations:

[0011] 1) The copyright issue of the image, because image search will inevitably process the image in the background, but finding the searched image through the network robot and directly storing the image as a template is not feasible under commercial copyright regulations;

Images