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

Object analysis model learning device and method based on data enhancement

A technology for analyzing models and learning devices, applied in image analysis, machine learning, image enhancement, etc., can solve problems such as setting keywords, achieve high performance, ensure high quality, and be easy to learn

Pending Publication Date: 2022-07-29
URBANBASE INC
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, since the number of products uploaded to the top 10 online shopping malls in Korea exceeds 400 million, it is difficult to set keywords for each product, so online shopping malls need to set product keywords only by using product image files The functional solution

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
  • Object analysis model learning device and method based on data enhancement
  • Object analysis model learning device and method based on data enhancement
  • Object analysis model learning device and method based on data enhancement

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0057] The advantages, features and implementation methods of the present invention can be clarified with reference to the accompanying drawings and the following embodiments specifically described in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but can be implemented in various ways. This embodiment only makes the disclosure of the present invention more complete, and fully conveys the invention to those skilled in the art to which the present invention belongs. Scope, the scope of the invention is defined by the scope of the claims.

[0058] In describing the embodiments of the present invention, detailed descriptions of well-known functions or configurations are omitted unless they are actually required. In addition, the following terms are terms defined in consideration of the functions in the embodiments of the present invention, and may vary depending on the user, the operator's intention, 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 object analysis model learning device based on data enhancement according to the embodiment of the invention comprises a processor, the processor may acquire a first spatial image including an object image to generate a second spatial image in which pixel information included in the first spatial image is changed, and designate a bounding box in a region including the first object image in the first spatial image. The method comprises the following steps: marking an operation for representing a first category of a first object image in a bounding box, inputting a first space image into a model designed based on an image classification algorithm, and primarily learning a weighted value of a model for deriving a correlation between the first object image and the first category in the bounding box; an operation of generating a model that represents an object image included in the spatial image according to the correlation and discriminates a category, and inputting a second spatial image to the model that has been learned for the first time; the annotation model represents a bounding box of a second object image in the second space image and operation of a second category discriminated by the model, and a model for performing secondary learning on a weighted value of the model is generated according to the second space image.

Description

technical field [0001] The present invention relates to a device and method for learning an analysis model of things based on data enhancement. Background technique [0002] According to the Korea Internet Agency (KISA), the estimated size of the Korean online shopping market in 2019 was about 133 trillion won, an increase of about 20% compared to 111 trillion won in 2018. With the substantial increase in the growth rate of the online shopping market, the number of stores and products registered on the online shopping platform has increased rapidly, and the proportion of consumers purchasing products through online shopping instead of offline has increased significantly. [0003] In addition, offline shopping is a way for consumers to choose shopping malls and use their eyes to confirm the goods in the mall to buy their favorite products, while online shopping is a way for consumers to search for and purchase products through the keywords of the desired products. With the c...

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): G06N20/00G06N5/02G06T19/00G06T7/187G06T7/11G06T5/00G06T19/20
CPCG06N3/08G06V10/25G06V10/774G06N3/045G06F18/214G06V10/82G06V10/235G06N20/00G06N5/022G06T19/006G06T7/187G06T7/11G06T19/20G06T2207/20081G06T2210/12G06T5/70
Inventor 金秀珉白允雅
Owner URBANBASE INC
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