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

Image classification method, device and system

A classification method and image technology, applied in the field of image processing, can solve problems such as large differences in data sets, large distribution distances, and difficulty in accurately identifying classification models

Active Publication Date: 2019-01-01
NANJING KUANYUN TECH CO LTD +2
View PDF3 Cites 22 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Most of the traditional fine-grained image classification methods are only based on a single scene. The trained classification model can only obtain more accurate classification results when the distribution of the test data set and the training data set are consistent. For example, the trained classification model can only identify A target vehicle in the same scene, once the target vehicle is located in another scene, it is difficult for the classification model to accurately identify
Due to the large differences in data sets in different scenarios (that is, the distribution distance of different domain sets is large), it is difficult for traditional fine-grained image classification methods to accurately classify data sets in different scenarios, and the universality is poor.

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
  • Image classification method, device and system
  • Image classification method, device and system
  • Image classification method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0048] First, refer to figure 1 An example electronic device 100 for implementing the image classification method, device and system of the embodiments of the present invention will be described.

[0049] like figure 1 Shown is a schematic structural diagram of an electronic device. The electronic device 100 includes one or more processors 102, one or more storage devices 104, an input device 106, an output device 108, and an image acquisition device 110. These components pass through a bus system 112 and / or other forms of connection mechanisms (not shown). It should be noted that figure 1 The components and structure of the electronic device 100 shown are only exemplary, not limiting, and the electronic device may also have other components and structures as required.

[0050] The processor 102 can be implemented in at least one hardware form of a digital signal processor (DSP), a field programmable gate array (FPGA), and a programmable logic array (PLA), and the processor...

Embodiment 2

[0057] see figure 2 A flow chart of an image classification method is shown, the method can be executed by the electronic device provided in the foregoing embodiment, and the method specifically includes the following steps:

[0058] Step S202, acquiring the target image to be processed. The target image contains objects to be classified, for example, the target image contains animals such as birds or cats of specific species to be identified, or the target image contains vehicles of specific models to be identified.

[0059] Step S204, input the target image into the pre-trained classification main network; wherein, the classification main network is used to extract the key features of the target image, and perform fine-grained classification based on the key features; the key features are related to the target object to be classified in the target image related, and the key features are not related to the scene in the target image.

[0060] It can be understood that usual...

Embodiment 3

[0101] This embodiment proposes a specific application example based on the image classification method proposed in Embodiment 2:

[0102] First, in a specific application, this embodiment may use CaffeNet as the classification main network, that is, the network structure of the classification main network in this embodiment may refer to CaffeNet. When CaffeNet performs feature extraction on images to be classified, it can use different features to classify them. The structure of CaffeNet is similar to AlexNet (so this embodiment can also use AlexNet as the classification main network). ReLU can be used as the activation function of CNN in the network structure of AlexNet, which successfully solves the gradient dispersion problem of Sigmoid when the network is deep. Moreover, AlexNet also uses overlapping max pooling. Compared with the common use of average pooling in traditional CNNs, AlexNet uses maximum pooling to avoid the fuzzy effect of average pooling. In addition, A...

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 provides an image classification method, a device and a system, which relate to the technical field of image processing. The method comprise inputting a target image to a pre-trained classification master network; among them, the classification master network being used to extract the key features of the target image, and fine-grained classification being carried out based on the keyfeatures. The key features are related to the object to be classified in the target image, and the key features are independent of the scene in the target image. Classification result of target imageis obtained by classification master network. The invention can be adapted to use a plurality of scenes and accurately classify images under different scenes, and has certain general applicability.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image classification method, device and system. Background technique [0002] Fine-grained image classification is a very important research topic in computer vision. The main purpose of fine-grained image classification is to distinguish object subcategories under the same object category, such as identifying different types of birds or different models of cars. Usually, intra-class differences are much smaller than inter-class differences, so the difficulty of fine-grained image classification for identifying intra-class differences is much higher than traditional image classification for identifying inter-class differences, such as the difficulty of distinguishing different breeds of cats Higher than the difficulty of distinguishing cats from dogs. [0003] Most of the traditional fine-grained image classification methods are only based on a single scene....

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): G06K9/62G06N3/04
CPCG06N3/045G06F18/241
Inventor 魏秀参王易木
Owner NANJING KUANYUN TECH CO LTD
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