Detection settings: Automated setup
Aim
To quickly instruct EthoVision XT how to recognize the subject that moves in the arena.
tip For common tests like the Open field and the Morris water maze, choose Automated Setup. If Automated Setup does not work well enough, see Detection settings: Advanced setup.
Prerequisites
▪The subject must be released in the arena. If you work with pre-recorded video, play the video to a point when the subject is well visible.
▪The contrast of the subject with the background is crucial for success with video tracking software. Choose Help > Video Tutorial and watch the video chapter Set up your Test Environment for tips on how to improve the contrast between the subject and the background.
Procedure
1.In the Detection Settings pane, click the Automated Setup button.
2.Choose the type of animal you work with, then click Next.
3.Drag the mouse to draw a rectangle around the subject.
▪If you have multiple subjects, draw a rectangle around each of them. Make sure that the subjects are not in contact with one another.
▪If you have multiple arenas, draw a rectangle around each subject in every arena.
▪For rodents, draw the rectangle around the animal’s body. Try to leave the tail out of the rectangle.
▪For fish, always include the caudal fin in the rectangle.
4.Move the Finetune slider until detection quality is satisfactory.
Moving the slider to the left detects more from the subject, for example it results in including the legs and tail of a rodent, but also more noise. Moving the slider to the right results in detecting less noise, but also less of the subject.
Below you see a detailed image of the same rat as in the picture above, when detected correctly (the red dot marks the detected center point).
5.Check in the Detection Performance pane under Subject not found if detection is good. As a guideline, the percentage Subject not found should be lower than 10% if the animal is visible at all times. Realize that the initial frames in which the subject was not detected yet is included in this percentage. If the number of samples (in brackets) under Subject not found does not increase or hardly increases anymore, and the percentage Subject not found decreases, detection is good. After acquisition, in the Trial list click the Show/Hide button and select Variables. Select Subject not found and check the percentage in the Subject not found column.
6.If detection quality is satisfactory, click Yes. If not, click No and continue with the Detection settings: Advanced setup.
Notes
▪If you work with the Deep learning detection technique, the Automated Setup (above) also includes the automatic adjustment of the Cutout window. This makes it less likely that you need to adjust this setting once again. See Adjust the settings for nose-tail base detection (Deep learning)
The next step
What you do next depends on how many subjects and what EthoVision XT functions you want to use.
▪One subject per arena; Center-point detection
At this point you do not need to do anything anymore. If detection is not good enough, see Detection settings: Advanced setup
▪One subject per arena; Nose-tail detection
See Detect the nose and the tail base. In the picture below you see an example of a mouse with its three body points detected correctly.
▪Social interaction (marker-assisted identification)
See Advanced detection settings: Marker-assisted identification, step 4.
If not all subjects are detected well after the Automated Setup, click No. Click Automated Setup again, and repeat the procedure when the subjects are in other locations. If this does not help, see Advanced detection settings for tracking color-marked subjects.
▪Social interaction (unmarked subjects)
See Advanced detection settings for tracking multiple unmarked subjects
▪Rat or Mouse behavior recognition
See Detection settings for Behavior recognition.
If detection is not satisfactory after the Automated Setup, continue with Detect the nose and the tail base. Then proceed with Detection settings for Behavior recognition.