Web Content Viewer


Crowd People Counter Version 1.0

Software Solutions

Item Code: 3913C008

To request more information about this product, click here!


For Crowd People Counter, cameras can be installed depending on the environment.1 Coupling the wide coverage range from 65° to 10° with high resolution network cameras, fewer cameras are needed which helps to save on equipment and installation costs.

Crowd People Counter’s video content analysis technology utilizes deep learning technology for highly accurate human detection. This includes dense crowds, people who are overlapping, and people who are very small due to distance.1

Crowd People Counter is compatible with Windows Operating System, including:

  • Windows 10 Pro/Windows 10 Enterprise/Windows 10 Home 64-bit
  • Windows Server 2012 Standard 64-bit
  • Windows Server 2012 R2 Standard 64-bit
  • Windows Server 2016 Standard 64-bit
The software is also compatible with Canon Network Cameras and Axis Network Cameras.

Crowd People Counter Version 1.0

Operating Environment Recommended PC Specifications CPU : Intel Core i7-6700 or higher (4 cores or more, HT technology compatible)
Memory : 16 GB or more
Storage: The amount of space needed for data output

When outputting JPEG files, one JPEG file is output for each analysis interval, the amount of data from the start till the end of analysis can be calculated by the following formula.
Amount of output data = JPEG image file size* x period between start and end of analysis (sec.) / analysis interval (sec.)

*A JPEG image file size can vary widely, depending on the resolution of the input video, background and color of the image.
Operating System Windows 10 Pro/Windows 10 Enterprise/Windows 10 Home 64-bit
Windows Server 2012 Standard 64-bit
Windows Server 2012 R2 Standard 64-bit
Windows Server 2016 Standard 64-bit
Software .NET Framework 4.5.2
Display Language Japanese/English/Italian/Spanish
*Linked to the language of OS. Displayed in English for other regions.
Requirements for Input Data [Live Video]
Compatible Cameras: Canon Network Cameras, Axis Network Cameras

<Canon Network Cameras>
ME20F-SHN, VB-H45, VB-M44, VB-H730F Mk II, VB-S30D Mk II, VB-S31D Mk II, VB-S800D Mk II, VB-S900F Mk II, VB-S805D Mk II, VB-S905F Mk II, VB-H761LVE-H, VB-H751LE-H, VB-M741LE-H, VB-S30VE, VB-S800VE, VB-S910F, VB-R13VE, VB-R13, VB-R12VE, VB-M50B, VB-H652LVE, VB-H651VE, VB-H651V, VB-H761LVE, VB-H760VE, VB-H751LE, VB-R11VE, VB-R11, VB-R10VE, VB-M641VE, VB-M641V, VB-M640VE, VB-M640V, VB-M741LE, VB-M740E, VB-H43, VB-H630VE, VB-H630D, VB-H730F, VB-M42, VB-M620VE, VB-M620D, VB-M720F, VB-S30D, VB-S31D, VB-S800D, VB-S900F, VB-S805D, VB-S905F (As of December 2018)

<Axis Network Cameras>
M1124, M1124-E, M1125, M1125-E, M1145, M1145-L, M2025-LE, M2026-LE Mk II, M3015, M3016, M3024-LVE, M3025-VE, M3026-VE, M3044-V, M3044-WV, M3045-V, M3045-WV, M3046-V, M3104-L, M3104-LVE, M3105-L, M3105-LVE, M3106-L Mk II, M3106-LVE Mk II, M5054, M5065, M5525-E, P1364, P1364-E, P1365 Mk II, P1365-E Mk II, P1367, P1367-E, P1368-E, P1435-LE, P1447-LE, P1448-LE, P3224-LV Mk II, P3224-V Mk II, P3225-LV Mk II, P3225-LVE Mk II, P3225-V Mk II, P3225-VE Mk II, P3227-LV, P3227-LVE, P3228-LV, P3228-LVE, P3367-V, P3367-VE, P3374-LV, P3374-V, P3375-LV, P3375-LVE, P3375-V, P3375-VE, P5414-E, P5415-E, P5624-E Mk II, P5635-E Mk II, Q1615 Mk II, Q1615-E Mk II, Q1645, Q1645-LE, Q1647, Q1647-LE, Q1659, Q1798-LE, Q3515-LV, Q3515-LVE, Q3517-LV, Q3517-LVE, Q3615-VE, Q6054 Mk II, Q6054 Mk III, Q6054-E Mk III, Q6055, Q6055-C, Q6055-E, Q6124-E, Q6125-LE, Q6128-E, Q6155-E, Q8414-LVS, Q8685-E, Q8685-LE (As of January 2020)

Video Format: JPEG
Highest Resolution: 5472 x 3648
Other Settings: The camera’s image quality and angle settings, other than the resolution and frame rate, need to be set using the camera beforehand.
Continuous time: Not limited. Continues as long as the camera’s “Maximum Connection Time” setting allows.
Communication Protocol: IPv4 (IPv6 not supported), HTTP. SSL not supported
User Authentication: Basic Authentication/Digest Authentication
Connection via proxy: Not supported
View area can be specified (only Axis Camears supporting the function).
Functions Analysis interval Select from [Not Specified, 1/30, 1/25, 1/20, 1/15, 1/10, 1/5, 1/2, 1, 2, 5, 10, 15, 20, 25, 30, 60, 600, 900, 1800, 3600](sec.)
Detection area Set the detection areas for counting people.
  • Polygon (polygon consisting of 3 to 8 points)
  • Up to 4 detection areas
Output Analyzed still image file (JPEG), analysis results file (CSV), aggregated results file (CSV), log (CSV)
Analysis result file for creating an interval Select from [Do Not Create New File, 1min., 5min., 10min., 15min., 30min., 1hour, 2hours, 4hours, 8hours, 12hours, 24hours].
Aggregation interval of analysis result files Select from [10 min., 15 min., 30 min., 1 hour].
Preview of analysis results
Display counted result on the screen. Display aggregated result on the chart.
Settings information Import/Export
Log output
Specifications Number of Pixels Necessary for Detecting a Person 15 pixels for a person's width
  • Maximum detectable number of people at highest resolution: approx. 88,000
  • Maximum detectable number of people in full HD: approx. 9,000
Angle Looking Downward on People (Angle of Depression) 10° – 65°*
* Even within the angle of depression, depending on the degree of congestion of the crowd, the people overlapping may affect the count.
Difficult Scenes and Subjects Scenes that may decrease the detection rate of people appearing in the video People of largely different sizes are within the same level in the video
  • Ex. - Distance between the people in front and people in the back is large, and the angle of depression is shallow
  • When there is a large difference in the distance from the camera between left and right of the video
  • Fisheye image, distortion of the peripheral part of the video, etc.
An image in which the size of the faces are irregular in the vertical direction of the angle of view
  • Ex. Different floors (stairwell etc.)
  • People having big difference in height
Wearing a hat or helmet, special hairstyle or color, baldness Black-and-white image
  • Ex. When shooting in night mode
Hidden person
  • Ex. The head of the person is hidden by another person, an umbrella, a banner, eaves, etc. and unable to be seen
Situations where people generally cannot be seen clearly in the video
  • Same color clothing as the fence or wall, people around, etc. unable to determine the body and their surroundings
  • Insufficient illuminance
  • People barely visible due to smoke or fog etc.
  • Out of focus, blur
Scenes in which objects other than people may mistakenly be counted as people
Scenes or objects which can cause recognition error of people are as follows:
Ex. - Objects' shape/color/size similar to the human body and head

This is to make Misc appear on the page when needed.




  1. Count result varies depending on the installation environment and how people are captured
  2. Depending on the width of person and congestion in the video, count may not be successful.

† Prices and specifications subject to change without notice. Actual prices are determined by individual dealers and may vary.

 Return to Top