Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Fine-grained image classification method based on multilayer focusing attention network

A focusing network, multi-layer focusing technology, applied in the field of computer vision, can solve the problems of similarity between classes, limited number of parts, limited classification accuracy, etc., to achieve the effect of improving robustness and reducing computational complexity

Active Publication Date: 2021-04-20
KUNMING UNIV OF SCI & TECH
View PDF17 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the number of parts of the object in this method is limited (2 or 4), which will limit the accuracy of the classification
Previous work usually uses positioning or segmentation to solve intra-class differences, but inter-class similarities still affect feature learning;

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
  • Fine-grained image classification method based on multilayer focusing attention network
  • Fine-grained image classification method based on multilayer focusing attention network
  • Fine-grained image classification method based on multilayer focusing attention network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] Embodiment 1: as Figure 1-3 Shown, a kind of fine-grained image classification method based on multi-layer focused attention network, the concrete steps of described method are as follows:

[0032] Step1, the public data set CUB-200-2011 contains a total of 11788 images from 200 bird species, including 5994 training and verification images, and 5794 test images. Input the training image into the first-layer focus network, which is a single-layer focus convolution network combined with the convolution block attention feature module, which generates a feature and attention product matrix, and outputs the positioning area at the same time;

[0033] Step2, cropping and occlusion operation: the positioning area is obtained after the operation of Step1, and the cropping operation cuts the original image according to the positioning area to obtain a cropped image; the occlusion operation blocks the corresponding position of the original image according to the positioning area...

Embodiment 2

[0053] Example 2, such as Figure 1-3 As shown, a fine-grained image classification method based on a multi-layer focused attention network, this embodiment is the same as Embodiment 1, the difference is that in this embodiment, the public data set CUB-200-2011 (200 categories 11788 Bird images), FGVC-Aircraft (10,000 aircraft images of 100 categories) and Stanford Cars (16,185 car images of 196 categories) were evaluated respectively, and the Top-1 accuracy of 89.7%, 93.6%, and 95.1% were respectively obtained. The results obtained on the three fine-grained public datasets are compared with the current mainstream fine-grained image classification methods (VGG-19, ResNet-101, etc.) as shown in Table 2. The experimental results show that the classification accuracy of this method is higher than the current mainstream method.

[0054] Table 2 Comparison with the accuracy of the current method

[0055]

[0056]

[0057] The present invention proposes a fine-grained image ...

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 relates to a fine-grained image classification method based on a multilayer focusing attention network, and belongs to the technical field of computer vision. The method comprises the following steps: firstly, accurately and effectively focusing on an identification local area and generating a positioning area through a first-layer focusing network; cutting and shielding the original image according to the positioning area and then inputting into a focusing network of the next layer to be trained and classified, wherein a single-layer focusing network focuses an effective positioning area through a convolution attention feature module and a positioning area selection mechanism on the basis of an InceptionV3 network; then extracting features of each local part by using bilinear attention maximization pooling; and finally performing classification prediction. Experimental results show that the classification accuracy of the method is higher than that of a current mainstream method.

Description

technical field [0001] The invention relates to a fine-grained image classification method based on a multi-layer focused attention network, which belongs to the technical field of computer vision. Background technique [0002] With the continuous development of deep learning and convolutional network technology, deep learning networks have been widely used in the field of computer vision, such as image retrieval, scene analysis, target tracking, etc. In the field of fine-grained image recognition, deep convolutional networks have also been widely studied and applied. Because in fine-grained image recognition, intra-class differences are easily affected by factors such as pose, viewing angle and position. The second category has similarities. In the end, manually marking the position is unstable and labor-intensive. Therefore, the fine-grained recognition task is more challenging. Zhang et al. proposed a strongly supervised fine-grained image classification model (Part-b...

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 Applications(China)
IPC IPC(8): G06K9/20G06K9/62G06N3/04G06T7/11
Inventor 乔伟晨黄青松王波单文琦刘利军黄冕
Owner KUNMING UNIV OF SCI & TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com