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

Incremental small sample target detection method based on meta-learning

A target detection, small sample technology, applied in the field of incremental small sample target detection based on meta-learning, can solve the problems of training time overhead, basic category data storage and privacy, etc., to achieve less sample demand, data privacy protection, data privacy protection Effect

Active Publication Date: 2021-02-05
TONGJI UNIV
View PDF12 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] However, these methods require the target detector to use not only the samples of the new category but also the samples of the basic category when learning to recognize the new category, which causes the problem of training time overhead and the storage and privacy of the basic category data.

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
  • Incremental small sample target detection method based on meta-learning
  • Incremental small sample target detection method based on meta-learning
  • Incremental small sample target detection method based on meta-learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments. This embodiment is carried out on the premise of the technical solution of the present invention, and detailed implementation and specific operation process are given, but the protection scope of the present invention is not limited to the following embodiments.

[0046] Such as figure 1 As shown, the present invention proposes an incremental small-sample target detection method based on meta-learning, comprising the following steps:

[0047] 1) Feature extraction step: Given the initial image, after random cropping, flipping and other data enhancement methods and normalization operations, extract the abstract features of the image as the input of the feature extractor;

[0048] 2) Target positioning step: Target positioning includes three parallel working ends, which are heat map end, size end and compensation end. The three working ends all t...

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 relates to an incremental small sample target detection method based on meta-learning, and the method comprises the following steps: 1) constructing an incremental small sample target detection model which comprises a feature extractor, a target locator and a meta-learning device; 2) training an incremental small sample target detection model; 3) carrying out new target positioning and classification according to the trained incremental small sample target detection model. Compared with the prior art, the method has the advantages of few sample requirements, avoidance of forgetting, data privacy protection and the like.

Description

technical field [0001] The invention relates to the field of target detection, in particular to an incremental small-sample target detection method based on meta-learning. Background technique [0002] At present, object detection methods based on convolutional neural networks have made significant progress, and one of the very important reasons is that there is sufficient human-labeled data. However, one of the capabilities that humans have but that traditional object detectors do not have is the ability to learn quickly from a small amount of data. Therefore, a training method using meta-learning is needed to enable object detectors to learn from sufficient basic categories of data. A pattern is obtained so that the target detector can learn the ability to recognize new types of objects based on a small amount of new type of data. [0003] Effective small sample target detection methods at this stage can be divided into the following two categories: [0004] (1) Small sa...

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): G06K9/62G06N20/00
CPCG06N20/00G06V2201/07G06F18/241G06F18/214
Inventor 王瀚漓程孟
Owner TONGJI UNIV
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