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

An Active Sampling Method for Target Detection Based on Timing Variance Threshold

A target detection and active sampling technology, which is applied in the direction of instruments, computing, character and pattern recognition, etc., can solve the problems of high cost and loss of key information, and achieve the effect of reducing negative impact and labeling cost

Active Publication Date: 2022-06-28
NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The labeling cost of the former is too high, while the sampling of the latter is random, and key information may be lost

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
  • An Active Sampling Method for Target Detection Based on Timing Variance Threshold
  • An Active Sampling Method for Target Detection Based on Timing Variance Threshold
  • An Active Sampling Method for Target Detection Based on Timing Variance Threshold

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0041] like figure 1 Shown is a flow diagram of the mechanism of the present invention. It is assumed that initially there is a data set L composed of a small number of labeled time series data images, and a data set U composed of a large number of unlabeled time series data images. First, initialize the target detection model with a small set of labeled time series data images L, set the number of planned query samples n, and the variance threshold δ, and initialize the number of selected samples q to 0. Then, the target detection model is initialized using the labeled time series data image set L, and the prediction results for the image frames in the unlabeled time series data image set U are output. Next, according to the prediction result, the model uncertainty size of each iteration is calculated for each image frame in the unlabeled time series data image set U, and arranged in descending order. Take the sample with the largest model uncertainty. If the time series va...

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 active sampling method for target detection based on time series variance threshold. Including: 1. Collect a large amount of unlabeled time series data and a small amount of labeled data; 2. Set the number of query samples n and the variance threshold δ; 3. Initialize the model; 4. The target detection model outputs the prediction results for unlabeled frames; 5. 1. Calculate the model uncertainty of each iteration for the unlabeled frame according to the prediction result; 6. Take the sample with the largest model uncertainty, if the timing variance is greater than the threshold and the adjacent frame is not selected, mark the sample to the query; 7. Update the labeled image set, unlabeled image set and prediction model; 8. Return to step 4 or have queried enough samples and output the target detection model f. Aiming at the target detection task of time series data in the automatic driving scene, the present invention sets a special active learning index to reduce the labeling cost.

Description

technical field [0001] The invention belongs to the technical field of automatic labeling of digital images, and in particular relates to an active sampling method for target detection based on a time series variance threshold. Background technique [0002] In the actual industrial application process, data has always been valued as the core resource of the Industrial Internet. Massive data is generated after the interconnection of people, machines, and things, which promotes the development of the industry. However, these data usually contain a large amount of redundant data, and data extraction and cleaning have become the top priority. One of the outstanding areas is the time series data of automatic driving. The automatic driving data is usually intercepted from multiple driving clips recorded by the camera. Each clip contains consecutive pictures in the driving scene, that is, time series frames. For this type of data, using active learning algorithm for selective samp...

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 Patents(China)
IPC IPC(8): G06V20/56G06V10/778
Inventor 黄圣君罗世发
Owner NANJING UNIV OF AERONAUTICS & ASTRONAUTICS
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com