Image recognition method, training method, device, equipment, medium and program product

An image recognition and image technology, applied in the field of artificial intelligence or information security, can solve the problem that the image processing model cannot effectively defend against confrontation attacks, and achieve the effect of improving the defense against confrontation attacks

Pending Publication Date: 2022-06-21
INDUSTRIAL AND COMMERCIAL BANK OF CHINA
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, image processing models trained with such adversarial examples cannot effectively defend against adversarial attacks

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 recognition method, training method, device, equipment, medium and program product
  • Image recognition method, training method, device, equipment, medium and program product
  • Image recognition method, training method, device, equipment, medium and program product

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. It should be understood, however, that these descriptions are exemplary only, and are not intended to limit the scope of the present disclosure. In the following detailed description, for convenience of explanation, numerous specific details are set forth in order to provide a thorough understanding of the embodiments of the present disclosure. It will be apparent, however, that one or more embodiments may be practiced without these specific details. Also, in the following description, descriptions of well-known structures and techniques are omitted to avoid unnecessarily obscuring the concepts of the present disclosure.

[0033] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the present disclosure. The terms "comprising", "comprising" and the like as used herein indicate the presence o...

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 recognition method, and relates to the field of artificial intelligence or information security. The method comprises the steps of obtaining a to-be-recognized image; inputting the to-be-recognized image into a pre-trained image recognition model to obtain a recognition result of the first portrait; wherein the image recognition model is obtained through training according to the following modes: obtaining a training image, and the training image comprises a second portrait area; determining a face area from the second portrait area; adding image noise into the face region to obtain a disturbance region; and training the image recognition model based on the disturbance area. According to the embodiment of the invention, the adverse effect of the background outside the face region on the model can be reduced during training, and the anti-attack defense capability of the trained model on the to-be-recognized image is further improved in a region enhancement mode. The invention further provides a training method and device, equipment, a storage medium and a program product.

Description

technical field [0001] The present disclosure relates to the field of artificial intelligence or the field of information security, and more particularly, to an image recognition method, training method, apparatus, device, medium and program product. Background technique [0002] The adversarial attack method can make the image recognition model output a feature vector with a large deviation by adding disturbance to the image to be recognized, and finally give a wrong recognition result. In order to improve the robustness of the image recognition model, adversarial samples can be used for training in advance to achieve the ability to defend against adversarial attacks. [0003] In the related art, adversarial samples are obtained by adding global perturbations to training images, and then image processing models are trained. However, image processing models trained with this class of adversarial samples cannot effectively defend against adversarial attacks. SUMMARY OF THE...

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): G06V40/16G06V10/26G06V10/764G06V10/774
Inventor 许啸吕博良程元鸿张诚
Owner INDUSTRIAL AND COMMERCIAL BANK OF CHINA
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