Population size counting method of double cameras

A technology of people counting and dual-camera, which is applied in the field of crowd counting with dual-camera, can solve the problem of inaccurate fusion of pixel features and texture features

Inactive Publication Date: 2015-09-23
XIAN UNIV OF TECH
View PDF4 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method for counting the number of people with dual cameras, whi

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
  • Population size counting method of double cameras
  • Population size counting method of double cameras
  • Population size counting method of double cameras

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0030] The present invention will be described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0031] The present invention is a method for counting the number of people with a dual camera, the flow chart of which is as follows figure 1 As shown, there are target camera A and auxiliary camera B in the monitoring scene. The target camera A video frame is used to count the number of people in the crowd, and the video frame data in auxiliary camera B is used to assist the data in A for training. The target crowd data set D A In order to count the pixel features and texture features of the crowd block in the video frame of the target camera A and the actual number of people in the target block; the auxiliary crowd data set D B In order to count the pixel features and texture features of the crowd block in the video frame of the auxiliary camera B and the actual number of people in the target block; the migration crowd data set D R is from...

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 a population size counting method of double cameras. The counting method comprises the steps as follows: setting a target camera A and an assistant camera B in a monitoring scene and respectively calculating characteristics of crowd block data sets in video frames of two cameras; using a transfer learning algorithm to extract sample data adjacent to the characteristic of a target population data set from an assistant population data set as a migration population data set; combining the migration population data set and the target population data set as a new population data set; fusing pixel feature and textural feature of the population data set according to a difference value between a predicated population number and real population in shaded population block in the new population data set; training a population size statistic model according to a fusion feature; using the trained population size statistic model to predict the population of newly input population image. The population size counting method of the invention solves the problem of a large number of population data scaling, unbalanced population data distribution and data missing via transfer learning, and gives full play of characteristic of different type features via feature fusion, thereby improving accuracy of the population size statistic model.

Description

technical field [0001] The invention belongs to the technical field of intelligent security monitoring, and in particular relates to a method for counting the number of people of a crowd with dual cameras. Background technique [0002] In the process of crowd counting, the task of traditional machine learning is to learn a model based on sufficient training data, and then use the model to predict the number of people. However, traditional machine learning needs to calibrate a large amount of training data, and the calibrated data may have unbalanced data distribution and missing data. Therefore, the present invention uses a transfer learning method to extract calibrated auxiliary crowd data associated with the target camera from other cameras to solve the above problems. [0003] In the process of people counting, pixel features can effectively describe the characteristics of low-density crowds, and texture features can effectively express the characteristics of high-densit...

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): G06K9/00G06K9/62
CPCG06V40/10G06F18/2411
Inventor 张二虎王海强
Owner XIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap