Sensor Data Structure

<< Click to Display Table of Contents >>

Navigation:  Import Sensor Data >

Sensor Data Structure

oDataView is able to open *.dtl files as well as *.bdf and *.edf files directly. Any plain text or *.csv document with a fixed column structure (one column for every single type of data) and lines (every new line represents a step further in time) can be imported.

oIf no time information is available in the data, the interval between the lines needs to be the same within one file.

oDuring the import process of a new data file, the import wizard helps you to specify your data.

oAfter the import routine, or whenever you want to store user defined layouts in one of the other file types, you need to save your data with DataView, turning it into a *.dtlx file.  That way, you do not have to import the file every time you want to use it.

Aggregate your data file

DataView was re-designed to handle high frequency data. Depending on the aim of your research, it still might be useful to aggregate your sensor data before it is transferred to DataView. Especially because some measuring equipment offers special routines to aggregate specific value types.

The rate at which your sensor data is collected, is probably much higher than the sample rate of your video. A video is usually recorded with 25 or 30 frames per second, sometimes even with just 15 or 10 frames per second. Whereas physiological recordings are often made with 500 or even 2000 hrz or more - which means that you get 500 or even 2000 values per second! (resulting in 20 to 80 different value lines per video frame)

That is why it is often advisable, to aggregate your sensor data, before you import your data into DataView. Some measuring equipment, does offer aggregated export routines, so please check the software and the documentation of your equipment.

Aggregating upfront, does not only speed up the import routine, it also reduces the file size and the number of lines assigned to one video frame. Getting 20 or more different values per single video frame might be confusing.

You may need to experiment a bit, to get the right mix of accuracy and readability for your study.