Non-transitory computer readable medium, action detection apparatus, and action detection method
View Patent ↗A non-transitory computer readable medium stores a program causing a computer to execute a process. The process includes calculating a variance of acceleration in individual sections obtained by dividing, by a first predetermined time interval, acceleration information for multiple axis directions detected from a subject whose action is to be determined; extracting, as a first candidate region, a section with the largest variance of acceleration from among the sections; calculating a variance of acceleration in individual sections obtained by shifting the first candidate region forward and backward by a second predetermined time interval; selecting, as a second candidate region, a section with the largest variance of acceleration from among the first candidate region and the sections obtained by shifting the first candidate region forward and backward; calculating a feature value for the second candidate region; and determining the action of the subject based on the feature value.
1. A nodding action detection system comprising:
a sensor disposed on a lanyard configured for wearing on a neck of a user; and
a processor programmed to determine a nodding action of the user based on data transmitted from the sensor to a first communication unit of the processor, wherein
the sensor includes a second communication unit configured to transmit the data to the first communication unit of the processor, and
in response to the first communication unit receiving the data from the second communication unit of the sensor, the processor is programmed to:
calculate a variance of acceleration in first individual time sections obtained by dividing pieces of acceleration information for a plurality of axis directions by a first predetermined time interval, the pieces of acceleration information being included in the data transmitted by the second communication unit of the sensor to the first communication unit of the processor, the variance of acceleration being calculated for each of the plurality of axis directions;
extract, as a first candidate region, a time section with the largest variance of acceleration from among the first individual time sections;
calculate a variance of acceleration in second individual time sections obtained by shifting the first candidate region forward and backward by a second predetermined time interval which is shorter than the first predetermined time interval;
select, as a second candidate region, a time section with the largest variance of acceleration from among the first candidate region, the second individual time section obtained by shifting the first candidate region forward, and the second individual time section obtained by shifting the first candidate region backward;
calculate a feature value for the selected second candidate region;
determine the nodding action on the basis of the calculated feature value; and
output a determination result of the nodding action.
2. The nodding action detection system according to claim 1 , wherein
the nodding action is determined using a decision tree regarding the calculated feature value.
3. The nodding action detection system according to claim 2 , wherein
noise is determined using a component of the feature value corresponding to a forward direction of the user.
4. The nodding action detection system according to claim 1 , wherein
the sensor is an acceleration sensor.
5. The nodding action detection system according to claim 2 , wherein
the sensor is an acceleration sensor.
6. The nodding action detection system according to claim 3 , wherein
the sensor is an acceleration sensor.