A comparison method of trademark similarity based on interval extraction

A technology of similarity and trademark, applied in the fields of instruments, computing, electrical digital data processing, etc., can solve the problems of cumbersome trademark retrieval steps, low accuracy and efficiency, and achieve easy visual distinction, improve work efficiency, and reduce complexity. Effect

CN109670072BActive Publication Date: 2022-08-05广州企图腾科技有限公司
5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Publication Date
2022-08-05

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention discloses a trademark similarity comparison method based on interval extraction, which comprises the following steps: obtaining a user's trademark image to be searched; All of them perform equal scaling, matrix storage and binarization processing in turn to obtain the original matrix of the user image and the original matrix of all trademark library images, and compare and analyze the original matrix of the user image and all trademark library images in turn; if they are completely consistent, Then, the trademark information corresponding to the original matrix of the trademark library image is sent to the user. By comparing the original matrix of the trademark database image with the original matrix of the user image, the invention retains the overall characteristics of the trademark, is easy to visually distinguish, and has more accurate retrieval results, and simplifies the operation steps of trademark retrieval.
Need to check novelty before this filing date? Find Prior Art

Description

Technical field

[0001] The invention involves the field of trademark search, and specifically involves a method of comparison of trademark similarity based on interval withdrawal. Background technique

[0002] Trademarks are used to distinguish the labels of a commodity or service of one operator and other operators, which is an intangible asset of an enterprise. According to the "Trademark Law", trademark registered people have the right to use trademarks or permits third -party use in the form of rewards, all of which are protected by law. With the comprehensive implementation of my country's trademark strategy, the number of registered trademarks received by the Trademark Office each year is as high as millions of pieces, and the identification and ruling of similar trademarks is a major difficulty, which requires a lot of manpower and material resources.

[0003] In response to the above situation, trademark retrieval becomes a necessary condition for trademark identification...

Examples

Embodiment 1

[0033] Example 1: Please refer to figure 1 Show, a method of comparison of a trademark similarity based on interval extraction, including the following steps:

[0034] Step S100: Get the user's to retrieve the trademark image; the average scaling, matrix storage, and two -value processing of the trademark image to be retrieved in order are performed to obtain the original matrix of the user image;

[0035] Specific methods for scaling, matrix storage, and two -value processing are specifically included:

[0036] Substalling the image to the pixel ratio of the default fixed size, and the grid matrix stored on the MXN (the grid matrix of the mxn indicates that the line of the matrix is ​​m and the number of columns is n). Among them, a pixel point is stored in a grid Then perform two -value processing to obtain the original matrix of the image.

[0037] 二值化的方法具体包括:当网格矩阵像素点的灰度值小于或等于阀值时,该网格定义为0;当网格矩阵像素点的灰度值大于灰度阈值时,该网The definition is 1, where the grayscale value = (the gray value of t...

Embodiment 2

[0051] Example 2: Different from the Example One, please refer to figure 2 It is shown that the invention is based on the interval withdrawal of trademark similarity comparison methods.

[0052] If you do not retrieve the completely consistent trademark library image matrix of the original matrix of the user image, execute the following steps:

[0053] Step S500: The raw matrix of all trademark library images is reduced in turn to reduce the level in turn to obtain all levels of the trademark library of all trademark library images;

[0054] Step S600: Reproduct the increase in levels in order of the original matrix of the user image to obtain the reduction matrix of each level of the trademark to be retrieved;

[0055] The reduction method of incremental levels in turn includes the following steps: the identification value of the grid extraction of the original matrix of the trademark library or the original matrix of the user image, corresponding to obtain a level A image narrow...