Image classification method and device for defending against attack, service decision-making method and device
A classification method, anti-attack technology, applied in the field of artificial intelligence, can solve problems such as user B loss
Pending Publication Date: 2020-10-16
ALIPAY (HANGZHOU) INFORMATION TECH CO LTD
View PDF5 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
For example, for the face recognition model, an attacker can implant a numerical vector of user B that cannot be recognized by the naked eye in user A's face image, so as to impersonate user B to complete identity verification, thereby causing user B's loss
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0051] In order to enable those skilled in the art to better understand the technical solutions in this specification, the technical solutions in the embodiments of this specification will be clearly and completely described below in conjunction with the drawings in the embodiments of this specification. Obviously, the described The embodiments are only some of the embodiments in this specification, not all of them. Based on the embodiments in this specification, all other embodiments obtained by persons of ordinary skill in the art without creative efforts shall fall within the protection scope of this specification.
[0052]As mentioned earlier, adversarial attacks have been one of the main factors threatening the security and stability of mechanical image classification in recent years. Taking the face recognition scenario as an example, an attacker can embed user B's numerical vector in user A's face image to impersonate user B to complete identity verification. For the 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
Login to View More Abstract
The embodiment of the specification provides an image classification method and device for defending against an attack, and a service decision-making method and device. The image classification methodcomprises the steps: obtaining to-be-classified target image information; inputting the target image information into an image restoration model to obtain original image information corresponding tothe target image information, the training data set of the image restoration model comprising training data taking anti-attack sample image information as input and taking the original image information corresponding to the anti-attack sample image information as output; and performing classification operation based on the original image information corresponding to the target image information toobtain a classification result of the target image information. The classification result of the target image information can execute a decision variable of a service decision.
Description
technical field [0001] This document relates to the field of artificial intelligence technology, and in particular to an image classification method for defending against attacks, a business decision method and a device. Background technique [0002] Adversarial attacks have been one of the main factors threatening the security and stability of mechanical image classification in recent years. Since the input form of the machine learning algorithm is a numerical vector (numeric vectors), the attacker will design a targeted numerical vector to make the image classification model make a misjudgment, which is called confrontation. attack. For example, for the face recognition model, an attacker can implant a numerical vector of user B that cannot be recognized by the naked eye in user A's face image to impersonate user B to complete identity verification, thereby causing user B's loss. [0003] Therefore, it is currently necessary to propose a technical solution capable of def...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/62G06F21/36
CPCG06F21/36G06V40/16G06V40/172G06F18/24G06F18/214
Inventor 曹佳炯李亮丁菁汀
Owner ALIPAY (HANGZHOU) INFORMATION TECH CO LTD