Driving analysis device and driving analysis method for analyzing driver tendency
In a driving analysis device, an information generation unit generates a time series of detection elements including a line-of-sight direction indicating to which of a plurality of preset viewable areas a line of sight of a driver driving a vehicle is oriented and an open/closed state of driver's eyes. An acquisition unit acquires evaluation data from the time series of the detection elements using a time window having a preset time width. An extraction unit extracts a plurality of feature quantities including at least a result of summation of appearance frequencies with respect to each of the detection elements from the evaluation data. An analysis unit analyzes a driver's tendency using a function that receives, as input, at least a part of the feature quantities extracted by the extraction unit.
1. A driving analysis device comprising:
an information generation unit configured to generate a time series of detection elements including a line-of-sight direction indicating to which of a plurality of preset viewable areas a line of sight of a driver driving a vehicle is oriented and an open/closed state of driver's eyes;
an acquisition unit configured to acquire evaluation data from the time series of the detection elements using a time window having a preset time width;
an extraction unit configured to extract a plurality of feature quantities including at least a result of summation of appearance frequencies with respect to each of the detection elements from the evaluation data;
an analysis unit configured to analyze a driver's tendency using a function receiving, as input, at least a part of the feature quantities extracted by the extraction unit; and
a display configured to display the at least the part of the feature quantities extracted by the extraction unit in a form of radar chart.
2. The driving analysis device according to claim 1 , wherein
the display is configured to use, as a reference value shown in the radar chart, either a mean value of the feature quantities about an indefinite number of drivers or a mean value of the feature quantities about the driver as an analysis target of the driving analysis device acquired in past.
3. The driving analysis device according to claim 1 , wherein
the feature quantities extracted by the extraction unit include a value indicating a feature of state transition within an identical detection element or between the plurality of the detection elements.
4. The driving analysis device according to claim 1 , wherein
the analysis unit is configured to use either a support vector machine or a neural network as the function.
5. A driving analysis device comprising:
an information generation unit configured to generate a time series of detection elements including a line-of-sight direction indicating to which of a plurality of preset viewable areas a line of sight of a driver driving a vehicle is oriented and an open/closed state of driver's eyes;
an acquisition unit configured to acquire evaluation data from the time series of the detection elements using a time window having a preset time width;
an extraction unit configured to extract a plurality of feature quantities including at least a result of summation of appearance frequencies with respect to each of the detection elements from the evaluation data; and
an analysis unit configured to analyze a driver's tendency using a function receiving, as input, at least a part of the feature quantities extracted by the extraction unit, wherein
the analysis unit is configured to use, as the function, a regression function obtained using either a Random forest or Gradient Boosting.
6. The driving analysis device according to claim 5 ,
wherein the analysis unit is configured to use at least a result of computation on two or more target feature quantities selected from the feature quantities as comparative data used for branch determination at each node in a regression tree used for generating the regression function, and
wherein, for the target feature quantities, the feature quantities that are identical in unit of value are selected.
7. A driving analysis method comprising:
generating a time series of detection elements, the detection elements including a line-of-sight direction indicating to which of a plurality of preset viewable areas a line of sight of a driver driving a vehicle is oriented and an open/closed state of driver's eyes;
acquiring evaluation data from the time series of the detection elements using a time window having a preset time width;
extracting a plurality of feature quantities including at least a result of summation of appearance frequencies with respect to each of the detection elements from the evaluation data;
analyzing a driver's tendency by using a function receiving, as input, at least a part of the feature quantities extracted; and
displaying the at least the part of the feature quantities extracted in a form of radar chart.
8. A driving analysis device comprising:
a processor and a memory configured to:
generate a time series of detection elements including a line-of-sight direction indicating to which of a plurality of preset viewable areas a line of sight of a driver driving a vehicle is oriented and an open/closed state of driver's eyes;
acquire evaluation data from the time series of the detection elements using a time window having a preset time width;
extract a plurality of feature quantities including at least a result of summation of appearance frequencies with respect to each of the detection elements from the evaluation data;
analyze a driver's tendency using a function receiving, as input, at least a part of the feature quantities extracted; and
display the at least the part of the feature quantities extracted in a form of radar chart.