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

Image recognition method and device, and model training method and device

An image recognition and image block technology, applied in the Internet field, can solve problems such as easy false alarm rate, unstable effect, and lower quality inspection accuracy rate, achieve high detection rate and accuracy rate, solve false alarm rate, reduce The effect of the false positive rate

Pending Publication Date: 2022-02-25
ALIBABA GRP HLDG LTD
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The traditional quality inspection of jelly relies on manual work, requiring quality inspection workers to pick up the jelly one by one, and repeatedly check whether there are impurities inside the jelly under the light
However, there are a series of problems in manual quality inspection: the effect is unstable and easily affected by artificial energy and emotions; quality inspection personnel need a long training period; quality inspection workers cannot work 24 hours a day, etc.
[0004] In the process of using machine vision to perform quality inspection on jelly, commonly used algorithms corresponding to computer vision are prone to false positives, thereby reducing the accuracy of quality inspection
[0005] In view of the above-mentioned problem that the algorithm applied in the process of quality inspection of jelly in the prior art is prone to false positives, thereby reducing the accuracy of quality inspection, no effective solution has been proposed yet

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 and device, and model training method and device
  • Image recognition method and device, and model training method and device
  • Image recognition method and device, and model training method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0041] According to an embodiment of the present invention, a method embodiment of an image recognition method is also provided. It should be noted that the steps shown in the flow chart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and , although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that shown or described herein.

[0042] The method embodiment provided in Embodiment 1 of the present application may be executed in a mobile terminal, a computer terminal, or a similar computing device. Take running on a computer terminal as an example, figure 1 It is a block diagram of the hardware structure of a computer terminal for an image recognition method in an embodiment of the present invention. Such as figure 1 As shown, the computer terminal 10 may include one or more (only one is shown in the figure) processors 102 (th...

Embodiment 2

[0091] According to another aspect of the embodiments of the present invention, a model training method is also provided, Figure 9 is a flow chart of the model training method according to Embodiment 2 of the present invention, such as Figure 9 As shown, the model training method provided by the embodiment of the present application includes:

[0092] Step S902, using the picture data with or without defects to learn the features of the defined detection category, and train the first detection network and the second detection network;

[0093] Step S904, using the trained first detection network and the second detection network to perform prediction, and output classification and detection scores;

[0094] Step S906, converting the detection score into the category and position of the defect frame.

[0095] Optionally, in step S902, the features of the defined detection category are learned by using the picture data with or without flaws, and the training of the first dete...

Embodiment 3

[0114] According to another aspect of the embodiments of the present invention, an image recognition method is also provided, Figure 10 is a schematic flow chart of an image recognition method according to Embodiment 3 of the present invention, as Figure 10 shown, including:

[0115] Step S1002, obtaining the uploaded object to be detected through the client;

[0116] In the above step S1002 of the present application, the object to be detected in the picture is acquired through the client or directly uploaded by the user through the web page. In the embodiment of the present application, the object to be detected may include jelly.

[0117] Step S1004, using the first detection network to identify the object to be detected for the first time to obtain a recognition result data set corresponding to the object to be detected, wherein the recognition result data set is used to indicate the detection category to which the object to be detected belongs;

[0118] In the above s...

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 an image recognition method and device, and a model training method and device. The image recognition method comprises the following steps: acquiring a to-be-detected object; through the first detection network, recognizing the to-be-detected object for the first time, and obtaining a recognition result data set corresponding to the to-be-detected object, wherein the recognition result data set is used for indicating the detection category to which the to-be-detected object belongs; acquiring an image block according to a target frame meeting a specified condition in the recognition result data set, and inputting the image block into a second detection network for secondary recognition to obtain a target object; and according to the recognition result data set and the target object, determining whether the to-be-detected object meets a detection requirement, and obtaining a recognition result. According to the invention, the technical problem that the accuracy of quality inspection is reduced due to the fact that the false alarm rate is easy to generate in the algorithm suitable for the quality inspection process of the jelly in the prior art is solved.

Description

technical field [0001] The present invention relates to the technical field of the Internet, in particular, to an image recognition method, a model training method and a device. Background technique [0002] During the industrial jelly production process, various harmful impurities, such as paint, metal shavings, material scale, hair, bugs, etc., will be mixed into the jelly and affect the quality of the jelly. In order to obtain qualified jelly, it is necessary to check the finished jelly to see if it contains harmful impurities, and if it does, it will be removed. In the production process of jelly, the quality inspection of jelly is a very critical link. [0003] The traditional quality inspection of jelly is all manual, requiring quality inspection workers to pick up the jelly one by one, and repeatedly check whether there are impurities inside the jelly under the light. However, there are a series of problems in manual quality inspection: the effect is unstable and ea...

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): G06T7/00
CPCG06T7/0004G06T2207/10004
Inventor 邹远鹏李海洋谢晓路金智勇刘伟
Owner ALIBABA GRP HLDG LTD
Features
  • Generate Ideas
  • 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