Method, apparatus, device, and storage medium for predicting the number of people of dense crowd

Patent metadata

Field	Value
Publication number	US-11302104-B2
Application number	US-201916458504-A
Country	US
Kind code	B2
Filing date	Jul 1, 2019
Priority date	Jul 2, 2018
Publication date	Apr 12, 2022
Grant date	Apr 12, 2022

Abstract

A method, apparatus, device, and storage medium for predicting the number of people of a dense crowd, including: converting a first image, in which the number of people is to be determined, into a corresponding first thermodynamic chart according to a thermodynamic chart conversion model; and determining the number of people in the first image according to the first thermodynamic chart, wherein the thermodynamic chart conversion model is obtained by training according to a pre-marked second image and a thermodynamic chart corresponding to each second image, thereby achieving prediction of the number of people of a dense crowd, improving the accuracy in predicting the number of people of the dense crowd while improving management efficiency.

First claim

What is claimed is:

1. A method for predicting the number of people of a dense crowd, comprising:
converting a first image, in which the number of people is to be determined, into a corresponding first thermodynamic chart according to a thermodynamic chart conversion model; and
determining the number of people in the first image according to the first thermodynamic chart,
wherein the thermodynamic chart conversion model is obtained by training according to a pre-marked second image and a thermodynamic chart corresponding to each second image, and the thermodynamic chart is a graphical representation describing a crowd distribution in real time using blocks of different colors and indicates crowd density; and
wherein each second image comprises a first area that is marked as ignored, and the thermodynamic chart corresponding to the second image does not comprise a portion of the thermodynamic chart corresponding to the first area.

2. The method according to claim 1 , wherein before the converting of the first image in which the number of people is to be determined into a corresponding first thermodynamic chart according to a thermodynamic chart conversion model, the method further comprises:
obtaining the thermodynamic chart conversion model by training according to the pre-marked second image and the thermodynamic chart corresponding to each second image.

3. The method according to claim 2 , wherein the obtaining the thermodynamic chart conversion model by training according to the pre-marked second image and the thermodynamic chart corresponding to each second image comprises:
obtaining a thermodynamic chart conversion model according to the pre-marked second image and the thermodynamic chart corresponding to each second image;
deforming the second image, and then training the thermodynamic chart conversion model to obtain a final thermodynamic chart conversion model.

4. The method according to claim 3 , wherein the deforming the second image comprises at least one of:
scaling each second image by a preset ratio;
enlarging each second image according to a preset ratio;
translating each second image according to a preset distance; and
rotating each second image by a preset angle.

5. An apparatus for predicting the number of people of a dense crowd, comprising: a processor, and a computer readable medium for storing program codes, which, when executed by the processor, cause the processor to:
convert a first image, in which the number of people is to be determined, into a corresponding first thermodynamic chart according to a thermodynamic chart conversion model; and
determine the number of people in the first image according to the first thermodynamic chart;
wherein the thermodynamic chart conversion model is obtained by training according to a pre-marked second image and a thermodynamic chart corresponding to each second image, and the thermodynamic chart is a graphical representation describing a crowd distribution in real time using blocks of different colors and indicates crowd density; and
wherein each second image comprises a first area that is marked as ignored, and the thermodynamic chart corresponding to the second image does not comprise a portion of the thermodynamic chart corresponding to the first area.

6. The apparatus according to claim 5 , wherein the program codes further cause the processor to:
obtain the thermodynamic chart conversion model by training according to the pre-marked second image and the thermodynamic chart corresponding to each second image.

7. The apparatus according to claim 6 , wherein when the program codes cause the processor to obtain the thermodynamic chart conversion model by training according to the pre-marked second image and the thermodynamic chart corresponding to each second image, the program codes cause the processor to:
obtain a thermodynamic chart conversion model according to the pre-marked second image and the thermodynamic chart corresponding to each second image; deform the second image, and then train the thermodynamic chart conversion model to obtain a final thermodynamic chart conversion model.

8. The apparatus according to claim 7 , wherein when the program codes cause the processor to obtain the thermodynamic chart conversion model by training according to the pre-marked second image and the thermodynamic chart corresponding to each second image, the program codes further cause the processor to:
scale each second image by a preset ratio;
enlarge each second image according to a preset ratio;
translate each second image according to a preset distance; or
rotate each second image by a preset angle.

9. An electronic device, comprising: a transmitter, a receiver, a processor, a memory, and a computer program, the computer program being stored in the memory, wherein the processor executes the computer program to implement the method for predicting the number of people of a dense crowd according to claim 1 .

10. The method according to claim 1 , wherein the prediction is made using a computer readable storage medium, storing a computer program therein.

Assignees

• Baidu online network technology beijing co ltd

Inventors

• Zhang Chengyue
• Liu Zeyu
• Chi Zhizhen
• Kang Le
• Chen Mingyu
• Bao Yingze
• Wang Jian
• Li Xubin
• Wen Shilei
• Ding Errui
• Liu Xiao
• Sun Hao

Classifications

• G06F18/2148

  characterised by the process organisation or structure, e.g. boosting cascade · CPC title

• G06V20/53 Primary

  Recognition of crowd images, e.g. recognition of crowd congestion · CPC title

• G06M11/00

  Counting of objects distributed at random, e.g. on a surface · CPC title

• G06V10/754

  involving a deformation of the sample pattern or of the reference pattern; Elastic matching · CPC title

• G06K9/6257

  Physics · mapped topic