A Fast Retrieval and Analysis Method for Advertisement Sequence Images

A technology of sequence images and analysis methods, applied in the field of image matching and retrieval, can solve the problems of low computational complexity and decreased accuracy, and achieve the effects of good robustness, improved accuracy, and increased computing speed.

CN106066887BActive Publication Date: 2019-05-17BEIJING INSTITUTE OF TECHNOLOGYGY
4 Cites 1 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2019-05-17

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
Patent Text Reader

Abstract

The invention discloses an advertisement sequence image fast retrieval and analysis method comprising the following steps: using image complexity characteristics and a decision classification tree method to divide an advertisement image database into LOGO and scene images; extracting and storing HOG characteristics and SIFT characteristics of images in the advertisement database; extracting HOG characteristics and SIFT characteristics of a to-be-coupled image, calculating HOG characteristic vector Euclidean distances between the to-be-coupled image and all images of same type in the advertisement database, ranking the Euclidean distances from small to big, and screening the front S candidate images; calculating SIFT characteristic coupling point pairs between the to-be-coupled image and the S candidate images, and using the image with the most pairs as the image corresponding to the to-be-coupled image; calculating time length of each advertisement according to the to-be-coupled image of the advertisement sequence, thus obtaining advertisement playing information. The advertisement sequence image fast retrieval and analysis method can ensure coupling accuracy, can save calculating time, and can improve image retrieval efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The invention belongs to the technical field of image matching and retrieval, and in particular relates to a method for fast retrieval and analysis of advertisement sequence images. Background technique

[0002] At present, with the rapid development of social economy, advertising has become a very important means of commercial publicity and marketing. Companies and enterprises place a large number of advertisements on public media platforms such as TV and movies to obtain huge economic benefits. With the increase in the amount of advertisements and the variety of advertisements, in order to monitor the advertisements and obtain relevant data information, how to accurately and quickly retrieve advertisement fragments has become an urgent problem to be solved. At present, advertisement retrieval is mostly done manually, which wastes time and consumes manpower, especially for massive databases, the labor and time costs are too high. Therefore, a fast au...

Examples

Embodiment Construction

[0041]The present invention will be described in detail below with reference to the accompanying drawings and examples. The implementation of the method of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

[0042] figure 1 It is a flow chart of an advertisement sequence image retrieval and analysis method in a specific embodiment of the present invention, specifically comprising the following steps:

[0043] Step 1: The classification of images in the advertising database mainly includes extracting the edge information of the image, determining the central area (rectangle) of the image content, calculating the edge ratio between the central area and the non-central area, and using the method of decision classification tree to divide the image into LOGO and scene images. ...