A kind of trademark retrieval method and device based on improved fast algorithm

A trademark and algorithm technology, applied in the field of image processing, can solve the problem of low retrieval accuracy, achieve the effect of improving speed and accuracy, improving precision, and improving efficiency

Active Publication Date: 2022-07-26
NANJING UNIV OF POSTS & TELECOMM
View PDF2 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But the retrieval accuracy based on Euclidean distance is not high

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A kind of trademark retrieval method and device based on improved fast algorithm
  • A kind of trademark retrieval method and device based on improved fast algorithm
  • A kind of trademark retrieval method and device based on improved fast algorithm

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0035] The technical solutions of the present invention will be described in further detail below with reference to the accompanying drawings.

[0036] The present invention provides a trademark retrieval method based on the improved Fast algorithm, such as figure 1 shown, including the following steps:

[0037] Step 1: Preprocess the trademark image to be retrieved to solve the problem of over-segmentation and under-segmentation in the trademark image.

[0038] The trademark image data of this study includes the official images of each trademark brand, as well as trademark images shot from different angles in natural scenes. The images are divided into different regions by Graph Cut to generate a mask of the image, which is then sent to Grab Cut for segmentation. , get the final trademark image, and effectively solve the problem of over-segmentation and under-segmentation in image preprocessing.

[0039] Step 2: Extract feature points of trademark images based on the improv...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention discloses a trademark retrieval method and device based on an improved Fast algorithm. The trademark image to be retrieved is preprocessed; K-means clustering and trademark image entropy are used to determine the adaptive threshold in the Fast detection algorithm. ; Use the Hessian matrix to eliminate edge response and eliminate the interference caused by the edge information of the trademark image; Based on the non-maximum value suppression algorithm, eliminate the non-local maximum point and eliminate the noise interference of the trademark image; Establish the trademark image feature bag model, and pass Clustering generates descriptors for trademark feature points; SVM classifier training is performed on the descriptors of various trademark image feature points, and training files are obtained to calculate the probability of each category to which the trademark belongs. The invention utilizes the feature point clustering extracted by the improved Fast algorithm to generate feature descriptors, performs K-means clustering and SVM classification on the descriptors, obtains the probability of various trademarks to which the trademark belongs, and then realizes the purpose of trademark retrieval.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a trademark retrieval method and device based on an improved Fast algorithm. Background technique [0002] Trademarks are a product of contemporary social and economic development. With the rapid economic development, the number of trademark registrations and inquiries is also increasing. A trademark is not only the logo of a corporate product, but also contains many social and cultural backgrounds. The management and protection of enterprise trademarks is extremely important, so how to accurately and effectively retrieve trademark information or determine whether the trademark is infringed has become one of the key directions of people's attention and research. And due to the development of computers and the popularization of machine learning in the commercial field, the automatic trademark retrieval function has been widely used. [0003] The trademark detection sy...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06V10/26G06V10/762G06V10/764G06K9/62G06F16/583
CPCG06F16/583G06V10/267G06F18/23213G06F18/2411
Inventor 徐荣青陈思宇申景金常春耕岳英杰
Owner NANJING UNIV OF POSTS & TELECOMM
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap