The Accuware Sentinel Test Page is a powerful tool that can be used to evaluate quickly and easily the results provided by the Sentinel algorithm directly on your video streams!
Using the interface it is possible to upload short videos coming from:
- cameras installed in the production environment.
- or cameras installed in a test area that matches the configuration that most likely will be used in the production environment.
1 – Requirements of the video input
The video streams that can be processed by the Sentinel Test Page MUST meet these requirements:
- Camera mounting: videos should be recorded from static cameras mounted above head height such that they point on the test area with an angle of about 45°.
- Resolution: the algorithm can process any resolution. To optimize the processing time we suggest to resize the video to the lowest resolution where people can be detected:
- Monitoring a close up view – even a resolution of 320 x 240 px may be sufficient.
- Monitoring a busy scene – at least a resolution of 1280 x 720 px / 720P / HD is required.
- People dimension: the Sentinel algorithm resizes the original video stream to 1366 x 768 pixels before searching for people. In order to be detected, people must have a minimum screen width of at least 30 pixels in the resized video stream. So, even if the resolution of the original video stream is super high, it could be be that people inside the resized video stream are so small that can’t detected (usually those in the background of the scene).
- Frame rate: at least 7 fps.
- Field of view (FOV): fields of view up to 90° are supported. Fish-eye lens are not supported!
- Color depth: color streams are recommended for best results.
- Scene lighting: must have good visible lighting conditions. Works fine in outdoors settings unless there are extreme lighting conditions like sun shining directly on the camera. Infrared streams are not supported.
- File format: MP4, AVI, MPEG
- Max file size: 2 GB.
- Max video length: 5 minutes.
Good example of video input
2 – How to test your video
These are the easy steps to follow in order to check the output produced by the Sentinel engine on your video:
- Open the Accuware Sentinel Test Page at this link.
- If you HAVE already an Accuware Site please enter the Username, Password, mark the check box Use an Accuware Dashboard account, enter the SiteID and click on Login.
- If you DON’T HAVE a Accuware Site AND you have a set of credentials for the Sentinel Test page enter the Username, Password and click on Login.
- If you DON’T HAVE a Accuware Site AND you DON’T HAVE a set of credentials for the Sentinel Test page click on Create an account, enter a valid email address and a password and click on Register. Click on the verification link inside the email received. Login to the Sentinel Test page.
- Click on Upload Videos, click on Browse and select a video that meets the requirements listed in the previous paragraph.
- Click on Upload.
Once the upload is completed, click on Videos. Depending on the size of your file, it might take a few minutes to get the results back. These are the possible status of a video that can be viewed by hovering the cursor on the hourglass:
- Copying (uploading)
- Waiting (queued for processing)
For each video file you can find:
- 2 buttons:
- one to download the processed video.
- one to download the CSV files with the data retrieved from the video.
- 1 link to access the page with list of tracklets detected.
3 – Download the processed video
For each individual shown inside the processed video it is possible to see:
- a number on top of each individual. This is the ID assigned to each individual (from now on tracklet) by the Accuware Sentinel algorithm.
- a box that represents the spatial occupancy of each individual.
4 – Download the CSV files
4.1 – Data structure
Data inside the CSV files are ordered by time-stamp and frame number.
1458596907992, 27663, 238, 1.755, 1.316, 416, 67, 64, 176
1458596907992, 27663, 236, 0.943, 2.493, 416, 67, 64, 176
1458596907992, 27663, 218, 1.402, 0.730, 416, 67, 64, 176
1458596907992, 27663, 196, 0.357, 0.233, 416, 67, 64, 176
1458596907992, 27663, 214, 2.404, 1.633, 416, 67, 64, 176
1458596907992, 27663, 244, 1.374, 2.965, 416, 67, 64, 176
1458596907992, 27663, 243, 0.327, -0.412, 416, 67, 64, 176
1458596907992, 27663, 226, 0.412, 2.806, 416, 67, 64, 176
Where the fields are:
- Column 1 – Timestamp – number of milliseconds elapsed since the beginning of the video.
- Column 2 – Frame number – every frame in a video stream is assigned a consecutive number. The frame number is shown inside the lower left corner of the processed video.
- Column 3 – Person ID – this is an auto-generated number that identifies each pedestrian on the video. More information about this can be found inside this page.
- Column 4 – Position X – this is the absolute position of the feet in screen coordinates (pixels).
- Column 5 – Position Y – this is the absolute position of the feet in screen coordinates (pixels).
- Columns 6 to 9 – represents the location in pixels of the tracking box (x, y, width, height).
- Column 10 – used internally by Accuware for debug purposes.
5 – List of tracklets
By clicking on the video link is it possible to access a new page that shows:
- the processed video.
- the list of tracklets detected.
For each tracklet you will see the first 10 images and the tracklet ID.
By clicking on the tracklet ID the processed video will jump to corresponding frame.