US 11,893,798 B2
Method, system and computer readable medium of deriving crowd information
Arun Kumar Chandran, Singapore (SG); Wen Zhang, Shanghai (CN); and Yusuke Takahashi, Singapore (SG)
Assigned to NEC CORPORATION, Tokyo (JP)
Filed by NEC Corporation, Tokyo (JP)
Filed on Feb. 22, 2023, as Appl. No. 18/112,856.
Application 18/112,856 is a continuation of application No. 17/042,474, granted, now 11,615,626, previously published as PCT/JP2019/008725, filed on Feb. 27, 2019.
Claims priority of application No. 10201802668Q (SG), filed on Mar. 29, 2018.
Prior Publication US 2023/0196783 A1, Jun. 22, 2023
This patent is subject to a terminal disclaimer.
Int. Cl. G06V 20/52 (2022.01); G06V 10/50 (2022.01); G06F 18/2321 (2023.01); G06V 40/10 (2022.01); G06V 10/25 (2022.01)
CPC G06V 20/53 (2022.01) [G06F 18/2321 (2023.01); G06V 10/25 (2022.01); G06V 10/50 (2022.01); G06V 40/103 (2022.01)] 15 Claims
OG exemplary drawing
 
1. A method of deriving crowd information, the method comprising:
receiving a picture;
receiving information of a crowd density level regarding a first area of the picture;
receiving information of a crowd density level regarding a second area of the picture, the second area being different from the first area;
calculating first information of a crowd level regarding people in the first area based on the information of the crowd density level regarding the first area; and
calculating second information of a crowd level regarding people in the second area based on the information of the crowd density level regarding the second area,
wherein the calculating the first information and the second information comprises:
determining a similarity of the picture of the crowd and each of a plurality of models of crowd level spatial variations; and
estimating a crowd level of the crowd in the picture in response to a most similar one of the plurality of models of crowd level spatial variations,
wherein the most similar one of the plurality of models of crowd level spatial variations is determined in response to a probability density function of the similarity of the picture of the crowd and each of the plurality of models of crowd level spatial variations.