Skip to main content
. 2018 Apr 5;18(4):1101. doi: 10.3390/s18041101

Figure 5.

Figure 5

Feature extraction. (top panel) Reference raw data (the subject is running, trips, and falls). (second panel) Feature J1 detects the fall as a large difference between its peak and jogging peaks. (third panel) Feature J2 has a similar shape but with a larger percentual difference. Both J1 and J2 are computed in bits for reducing computations on the embedded device. (bottom panel) J3 is formed by J1 and J2, increasing their coincidences and diminishing their differences.