Driving assistance device, driving assistance method, and recording medium
A driving assistance device includes a processor connected to a memory. The processor detects a latent recognition region that is unconsciously recognized by a human in an entire imaging region of a captured image. The captured image is captured by an external camera that is provided in a mobile body. The captured image is an image of at least a traveling direction in which the mobile body travels. The latent recognition region is detected based on a predetermined feature value included in the captured image. The processor presents predetermined driving assistance information in a region on a display device. The region is estimated in accordance with the detected latent recognition region.
1 . A driving assistance device comprising:
a hardware processor connected to a memory and configured to:
detect a latent recognition region being unconsciously recognized by a human in an entire imaging region of a captured image captured by an external camera provided in a mobile body, the captured image being an image of at least a traveling direction in which the mobile body travels, the latent recognition region being detected based on a predetermined feature value included in the captured image;
present predetermined driving assistance information in a region on a display device, the region being estimated in accordance with the detected latent recognition region;
detect a gaze of a driver; and
present the driving assistance information in a divided region where a dwell time of the detected gaze is equal to or longer than a predetermined dwell time and a prediction error is less than the threshold value.
2 . The driving assistance device according to claim 1 , wherein the hardware processor is configured to detect, as the latent recognition region, a divided region having a predetermined feature value exceeding a predetermined threshold value out of divided regions obtained by dividing the entire imaging region into a plurality of regions.
3 . The driving assistance device according to claim 2 , wherein the hardware processor is configured to present the driving assistance information separately between a period of time during which the feature value is less than the threshold value and a period of time during which the feature value exceeds the threshold value.
4 . The driving assistance device according to claim 2 , wherein the hardware processor is configured to:
present the driving assistance information while the feature value exceeds the threshold value, or
present the driving assistance information for a given period of time after the feature value exceeds the threshold value.
5 . The driving assistance device according to claim 2 , wherein the hardware processor is configured to:
detect, for each of the divided regions, a prediction error being a difference between pixels of a predicted image predicted from a past image captured in a past by the external camera and pixels of an actual captured image actually captured by the external camera at a time of the predicted image; and
detect, as the latent recognition region, a divided region having the prediction error exceeding a predetermined threshold value.
6 . The driving assistance device according to claim 2 , wherein the hardware processor is configured to:
detect an edge of an object from a captured image captured by the external camera; and
detect, as the latent recognition region, a divided region where a ratio of the edge exceeds a predetermined threshold value out of the divided regions.
7 . The driving assistance device according to claim 2 , wherein the hardware processor is configured to:
detect, for each of the divided regions, a prediction error indicating a difference between pixels of a predicted current image corresponding to a current captured image generated by prediction from a captured image captured in a past by the external camera, and pixels of a current captured image captured by the external camera;
detect a divided region including an edge of an object from the captured image captured by the external camera; and
detect, as the latent recognition region, a divided region from which a prediction error exceeding a predetermined threshold value is detected and in which the edge exceeding a predetermined threshold value is included.
8 . The driving assistance device according to claim 1 , wherein the hardware processor is configured to change one or more items in accordance with the detected latent recognition region, the items including:
presence or absence of the presentation of the driving assistance information,
a position of the presentation of the driving assistance information,
a number of the presentation of the driving assistance information, and
a type of the driving assistance information.
9 . The driving assistance device according to claim 8 , further comprising a speed detection sensor configured to detect a movement speed of the mobile body,
wherein the hardware processor is configured to change at least the type of the driving assistance information to be presented, the type being changed between a case where a movement speed of the mobile body detected by the speed detection sensor is equal to or more than a predetermined threshold value and a case where the movement speed is less than the predetermined threshold value.
10 . The driving assistance device according to claim 1 , further comprising a speed detection sensor configured to detect a movement speed of the mobile body,
wherein the hardware processor is configured to control operation of the mobile body in accordance with a movement speed of the mobile body detected by the speed detection sensor.
11 . A driving assistance method implemented by a computer, the method comprising:
detecting a latent recognition region being unconsciously recognized by a human in an entire imaging region of a captured image captured by an external camera provided in a mobile body, the captured image being an image of at least a traveling direction in which the mobile body travels, the latent recognition region being detected based on a predetermined feature value included in the captured image;
presenting predetermined driving assistance information in a region on a display device, the region being estimated in accordance with the detected latent recognition region;
detecting a gaze of a driver; and
presenting the driving assistance information in a divided region where a dwell time of the detected gaze is equal to or longer than a predetermined dwell time and a prediction error is less than the threshold value.
12 . A non-transitory computer-readable recording medium on which programmed instructions are recorded, the instructions causing a computer to execute processing, the processing comprising:
detecting a latent recognition region being unconsciously recognized by a human in an entire imaging region of a captured image captured by an external camera provided in a mobile body, the captured image being an image of at least a traveling direction in which the mobile body travels, the latent recognition region being detected based on a predetermined feature value included in the captured image;
presenting predetermined driving assistance information in a region on a display device, the region being estimated in accordance with the detected latent recognition region;
detecting a gaze of a driver; and
presenting the driving assistance information in a divided region where a dwell time of the detected gaze is equal to or longer than a predetermined dwell time and a prediction error is less than the threshold value.