Supercharge Your Innovation With Domain-Expert AI Agents!

Cross-domain small sample image classification model method focusing on fine-grained recognition

A classification model, small sample technology, applied in neural learning methods, character and pattern recognition, biological neural network models, etc., can solve problems such as fine-grained image recognition

Active Publication Date: 2021-05-07
BEIJING TECHNOLOGY AND BUSINESS UNIVERSITY
View PDF4 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0026] The present invention mainly aims at feature encoders, and proposes to improve the model structure of image features and metric function classification and recognition, and mainly solves the problem of fine-grained image recognition in cross-domain small-sample image classification

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
  • Cross-domain small sample image classification model method focusing on fine-grained recognition
  • Cross-domain small sample image classification model method focusing on fine-grained recognition
  • Cross-domain small sample image classification model method focusing on fine-grained recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0064] Below in conjunction with accompanying drawing, further describe the present invention through embodiment, but do not limit the scope of the present invention in any way.

[0065] The model structure is as figure 1 shown. The method of the present invention includes: 1) building a cross-domain small-sample image classification model focusing on fine-grained recognition: the model extracts image features through a front-end dedicated feature encoder, and implements classification and recognition using image features in the back-end bilinear metric function; 2) Pre-training Focused Feature Encoder (MFFE): Pre-training image classification and recognition on the mini-ImageNet data set, and then transferring the pre-trained MFFE model and parameters to the FFGR model of the present invention and combining with BMF, As the front end of FFGR, it is used to extract the feature information of the image; 3) Focus on fine-grained recognition (FFGR) model classification and recog...

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 cross-domain small sample image classification model method focusing on fine-grained recognition, a cross-domain small sample classification model FFGR focusing on fine-grained recognition is constructed, and the FFGR model adopts a two-step recognition method and comprises an image feature extraction module MFFE and an image feature classification recognition module BMF; and image features are extracted through a front-end focused feature encoder, and then image classification and recognition are carried out by using the image features through a rear-end bilinear metric function. By adopting the method provided by the invention, the small sample image feature information can be extracted more quickly and efficiently, the overall optimization of the model is quicker and more accurate, and the classification accuracy is high.

Description

technical field [0001] The invention relates to image processing and cross-domain small-sample image classification and recognition technology, in particular to a cross-domain small-sample image classification model method focusing on fine-grained recognition, which belongs to the technical field of computer vision and image processing. Background technique [0002] In recent years, due to the emergence of powerful computing devices (for example, distributed platforms and image processing units), and the birth of large data sets (for example, ImageNet image data set), deep learning models have achieved great success in computer vision classification tasks. Huge success. However, these supervised learning methods require a large number of labeled samples and sufficient iterations to train the deep learning model in order to make the model optimal. In practice, the cost of manually labeling a large amount of data is too high, and there are not so many data in some sample cate...

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
IPC IPC(8): G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/048G06F18/214G06F18/243G06F18/253Y02T10/40
Inventor 于重重萨良兵谢涛赵霞
Owner BEIJING TECHNOLOGY AND BUSINESS UNIVERSITY
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com