WO2007131385A1 - Instrument de calcul pour le décompte d'entrées et de sorties de personnes - Google Patents

Instrument de calcul pour le décompte d'entrées et de sorties de personnes Download PDF

Info

Publication number
WO2007131385A1
WO2007131385A1 PCT/CN2006/000953 CN2006000953W WO2007131385A1 WO 2007131385 A1 WO2007131385 A1 WO 2007131385A1 CN 2006000953 W CN2006000953 W CN 2006000953W WO 2007131385 A1 WO2007131385 A1 WO 2007131385A1
Authority
WO
WIPO (PCT)
Prior art keywords
target
image
module
background
ccd camera
Prior art date
Application number
PCT/CN2006/000953
Other languages
English (en)
Chinese (zh)
Inventor
Wei Yao
Original Assignee
Shanghai Yaowei Industry Co, Ltd.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Shanghai Yaowei Industry Co, Ltd. filed Critical Shanghai Yaowei Industry Co, Ltd.
Priority to PCT/CN2006/000953 priority Critical patent/WO2007131385A1/fr
Publication of WO2007131385A1 publication Critical patent/WO2007131385A1/fr

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects

Definitions

  • the present invention relates to a method for counting the number of people entering and leaving the entrance and exit counters and the number of entrance and exit counters. Background technique
  • the purpose of the invention of the wood is to provide an apparatus for automatically obtaining an image of a person entering and exiting an area by real-time, using a computer to automatically recognize the number of people entering and leaving the area through the number of people entering and exiting the area, and providing an instrument for counting the number of people entering and leaving the area.
  • the method of counting the number of entrances and exits is to provide an apparatus for automatically obtaining an image of a person entering and exiting an area by real-time, using a computer to automatically recognize the number of people entering and leaving the area through the number of people entering and exiting the area, and providing an instrument for counting the number of people entering and leaving the area.
  • the wood invention provides such an entrance and exit number counter, which is composed of a CCD camera and a PC.
  • the CCD camera is fixedly mounted on the top of the entrance and exit, the CCD camera is connected to the PC through a line, and the image is mounted on the PC.
  • the acquisition card and the image recognition algorithm software, the image recognition algorithm software package comprises five modules, namely a target background separation module, a background reconstruction module, a post-processing and recognition module of the target image, a target detection tracking module, and a data statistics output module.
  • the present invention provides such a method for counting the number of people entering and exiting the number of meters, and the steps of the human body image recognition counting are as follows:
  • the device starts, initializes, the program starts to initialize, the image collection card, the counter starts to initialize;
  • the target existence judgment to determine whether an object has passed, that is, whether there is a target exists? If it is determined that no object has passed, the target image is re-set;
  • the image is saved, if there is a target, the image is saved;
  • the target background reconstruction adding a monochrome background to the current image after removing the fixed scene
  • the adhesion target is separated, the spots near the target image are eliminated, the target range is further clarified, and the target edge is trimmed within a clear target range to conform to the judgment of the standard target image;
  • the entrance and exit number counter and the counting method thereof can be adapted to the number of people entering and leaving the behavior in various order situations, and the most accurate recognition result can be obtained under the most disorderly guest entry and exit behavior conditions. . DRAWINGS
  • 1 is a system block diagram of the entrance and exit number counter of the present invention.
  • FIG. 2 is a system block diagram of the entrance and exit number counter of the present invention.
  • FIG. 3 is a flow chart of a method for counting the number of entrances and exits by the entrance and exit number counter of the present invention. detailed description
  • the present invention will be further described in detail below by means of a specific embodiment of installing the entrance and exit number counter of the present invention on a long-distance passenger car door and applying the counting method thereof.
  • a single CCD camera is mounted on the ceiling near the door.
  • the CCD camera is connected to the PC through a line.
  • the PC is equipped with an image acquisition card and image recognition algorithm software.
  • the image recognition algorithm software package contains five modules, namely the target background separation module. , background reconstruction module, post-processing and recognition module of target image, target detection and tracking module, and data statistics output module.
  • CCD camera lens direction is perpendicular to the horizontal plane and faces The entrance or exit or the lens facing the entrance and exit is at a fixed angle to the horizontal. The principle of determining the angle is to ensure that the mutual occlusion between people is as small as possible.
  • the camera uses visible light imaging during the day, and at night it can add some kind of light-filling measures such as infrared light.
  • the CCD camera uses auto iris, manual focus, auto shutter, PAL system, the lens is spherical, the lens is 1/2 inch, the lens resolution is 38 lines / mm, the minimum working distance is 200 mm, the standard C / CS interface, The specific interface needs to match the selected camera.
  • Camera CCD size is 1/2 inch, CCD resolution is 360 lines, color information is color TV signal output, working system is supported by interlaced PAL system, standard 625 lines, 25 frames (50 fields) / sec, signal to noise ratio 50 decibels
  • the minimum working illumination is 0. lLux, the output interface is RS170, the synchronous mode is internal synchronization or external synchronization, and the power supply is DC 24V or 12V.
  • the image acquisition card on the PC supports PAL and NTSC formats, interlaced scanning, and supports image formats such as RGB 888, YUV, and pure grayscale.
  • the maximum size of the output image is 768 X 576, NTSC is 640 X 480, and the output image is The size is programmable, the synchronization mode is self-synchronization or program setting, and the crystal frequency is preset or programmable.
  • the target background separation module in the image recognition algorithm software adopts the depth fusion target background separation, that is, the current frame image is compared with the background image, the current frame image is compared with the previous frame image, and then the depth fusion is performed.
  • Background The reconstruction module uses multi-step median filtering background reconstruction or maximum probability background reconstruction.
  • the post-processing and recognition module of the target image uses three kinds of processing methods: connectivity analysis, false target elimination, and adhesion target separation, as well as comparison of the shape and configuration structure of the contour.
  • the target detection tracking module Based on the target background separation module processing, the target detection tracking module performs histogram correlation matching according to the nearest neighbor priority principle, and directly matches the corresponding pixel values to match the failed target, and simultaneously tracks the target motion direction.
  • the data statistics output module records and updates the statistical result of the number of entrances and exits according to the recognition result and outputs it.
  • the camera can monitor all the stepping areas near the passenger's access door through visible light imaging. Since the lens is perpendicular to the horizontal plane or faces the door and the horizontal plane at a fixed angle r, the passengers are less likely to block each other. It is beneficial to reduce the resulting passenger counting error rate and improve the correctness of the recognition algorithm as a whole.
  • the shape of the passenger's head and the shoulder or the upper body of the human body and its configuration are set as standard human target images, and the direction of movement of the feature is tracked to distinguish the passenger's getting on and off.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Theoretical Computer Science (AREA)
  • Image Processing (AREA)

Abstract

Procédé et instrument pour le décompte d'entrées et de sorties de personnes, comprenant: caméra CCD et ordinateur personnel. La caméra est fixée sur le sommet d'un passage et reliée à l'ordinateur via des lignes, et l'ordinateur fonctionne avec un logiciel à carte de collecte d'image et algorithme de reconnaissance d'image qui comprend les cinq modules suivants: séparation cible-fond, reconstruction de fond, post-traitement et reconnaissance d'image cible, suivi-détection de cible et décompte-sortie de données. Le procédé de décompte utilise la carte de collecte d'image pour la collecte des images, reconnaissant de façon circulaire et en temps réel chaque image collectée, puis assurant la séparation cible-fond et effectuant une analyse de connectivité et une élimination de fausse cible sur les images cibles, après la séparation des images cibles liées, et enfin comparant ces images cibles avec les images cibles standard des corps humains pour en évaluer le comportement, ce qui permet en dernier ressort de décompter les personnes entrant et sortant et de fournir ce décompte. On peut adapter l'invention au décompte des entrées et sorties de personnes selon différents ordres.
PCT/CN2006/000953 2006-05-12 2006-05-12 Instrument de calcul pour le décompte d'entrées et de sorties de personnes WO2007131385A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2006/000953 WO2007131385A1 (fr) 2006-05-12 2006-05-12 Instrument de calcul pour le décompte d'entrées et de sorties de personnes

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2006/000953 WO2007131385A1 (fr) 2006-05-12 2006-05-12 Instrument de calcul pour le décompte d'entrées et de sorties de personnes

Publications (1)

Publication Number Publication Date
WO2007131385A1 true WO2007131385A1 (fr) 2007-11-22

Family

ID=38693513

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2006/000953 WO2007131385A1 (fr) 2006-05-12 2006-05-12 Instrument de calcul pour le décompte d'entrées et de sorties de personnes

Country Status (1)

Country Link
WO (1) WO2007131385A1 (fr)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107590431A (zh) * 2017-07-27 2018-01-16 广东数相智能科技有限公司 一种基于图像识别的数量统计方法及装置
CN111738125A (zh) * 2020-06-16 2020-10-02 中国银行股份有限公司 客户人数确定方法及装置
CN113542670A (zh) * 2021-02-24 2021-10-22 上海汉时信息科技有限公司 检测方法、检测装置和检测系统

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1687955A (zh) * 2005-04-19 2005-10-26 上海遥薇实业有限公司 出入口人数计数仪

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN1687955A (zh) * 2005-04-19 2005-10-26 上海遥薇实业有限公司 出入口人数计数仪

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107590431A (zh) * 2017-07-27 2018-01-16 广东数相智能科技有限公司 一种基于图像识别的数量统计方法及装置
CN107590431B (zh) * 2017-07-27 2020-12-01 广东数相智能科技有限公司 一种基于图像识别的数量统计方法及装置
CN111738125A (zh) * 2020-06-16 2020-10-02 中国银行股份有限公司 客户人数确定方法及装置
CN111738125B (zh) * 2020-06-16 2023-10-27 中国银行股份有限公司 客户人数确定方法及装置
CN113542670A (zh) * 2021-02-24 2021-10-22 上海汉时信息科技有限公司 检测方法、检测装置和检测系统
CN113542670B (zh) * 2021-02-24 2023-04-18 上海汉时信息科技有限公司 检测方法、检测装置和检测系统

Similar Documents

Publication Publication Date Title
US11790504B2 (en) Monitoring method and apparatus
CN107563347B (zh) 一种基于tof相机的客流计数方法和装置
US8712149B2 (en) Apparatus and method for foreground detection
WO2017028587A1 (fr) Procédé et appareil de surveillance de véhicule, processeur, et dispositif d'acquisition d'images
EP1768067B1 (fr) Dispositif électro-optique pour compter des personnes ou autres, basé sur la vision stéréoscopique, et procédé correspondant
CN113759431B (zh) 安检数据关联方法和装置、以及x射线安检系统
US9672434B2 (en) Video-based system and method for parking occupancy detection
US20050084179A1 (en) Method and apparatus for performing iris recognition from an image
KR100691348B1 (ko) 팬/틸트 제어기반의 스테레오 카메라를 이용한 이동표적추적방법 및 그 시스템
KR101821144B1 (ko) 깊이정보 기반의 안면인식 출입통제시스템
CN110633648B (zh) 一种自然行走状态下的人脸识别方法和系统
KR101625538B1 (ko) 도시방범이 가능한 다차선 자동차 번호판 인식시스템
CN109829997A (zh) 人员考勤方法及系统
JP2003150942A (ja) 目位置追跡方法
CN104112309A (zh) 一种采用视频监控自动记录公交车人流量的装置及方法
CN1687955A (zh) 出入口人数计数仪
JP2022189835A (ja) 撮像装置
WO2007131385A1 (fr) Instrument de calcul pour le décompte d'entrées et de sorties de personnes
JP2005135339A (ja) 通行人検出方法および装置ならびに通行人計数装置
KR102423934B1 (ko) 안면인식 및 유사한 옷 색상의 다수 객체 추적기술을 통한 스마트 휴먼검색 통합 솔루션
CN101625506B (zh) 数字取像装置的脸部自动对焦方法
CN110930432A (zh) 一种视频分析方法、装置及系统
CN115410261A (zh) 面部识别异构数据关联解析系统
CN110321782A (zh) 一种检测人体特征信号的系统
KR20200010691A (ko) 이종 카메라를 활용한 인물인식 및 교통통제시스템

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 06741843

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 06741843

Country of ref document: EP

Kind code of ref document: A1