Liveness detection apparatus and method
View Patent ↗A liveness detection apparatus and method are provided. The apparatus captures a live image of a driver. The apparatus calculates an eye open duration of the driver based on the live image within a first time interval. In response to the eye open duration exceeding a first duration threshold, the apparatus executes a first confirmation operation to notify the driver to complete a liveness detection operation. In response to the first confirmation operation has been executed, the apparatus continues to calculate the eye open duration based on the live image within a second time interval. In response to the eye open duration exceeding a second duration threshold, the apparatus executes a second confirmation operation to notify the driver to complete the liveness detection operation.
1 . A liveness detection apparatus, comprising:
a camera, configured to capture a live image of a driver; and
a processor, coupled to the camera, configured to execute the following operations:
calculating an eye open duration of the driver based on the live image within a first time interval;
in response to the eye open duration exceeding a first duration threshold, executing a first confirmation operation to notify the driver to complete a liveness detection operation;
in response to the first confirmation operation has been executed, continuing to calculate the eye open duration based on the live image within a second time interval, wherein the second time interval is later than the first time interval; and
in response to the eye open duration exceeding a second duration threshold, executing a second confirmation operation to notify the driver to complete the liveness detection operation, wherein the second duration threshold is greater than the first duration threshold.
2 . The liveness detection apparatus of claim 1 , wherein the operation of calculating the eye open duration further comprises:
calculating a first eye-opening ratio of the driver in the live image within the first time interval; and
determining whether the driver opens their eyes based on the first eye-opening ratio.
3 . The liveness detection apparatus of claim 2 , wherein the operation of calculating the first eye-opening ratio of the driver in the live image further comprises:
searching for a plurality of eye feature points of the driver in the live image; and
calculating the first eye-opening ratio of the driver in the live image based on the eye feature points.
4 . The liveness detection apparatus of claim 2 , wherein the operation of calculating the first eye-opening ratio of the driver in the live image further comprises:
calculating a first distance between an upper eyelid and a lower eyelid;
calculating a second distance between a left eye corner and a right eye corner; and
calculating a ratio of the first distance and the second distance as the first eye-opening ratio.
5 . The liveness detection apparatus of claim 1 , wherein the operation of calculating the eye open duration further comprises:
calculating a plurality of second eye-opening ratios of the driver in the live image within a third time interval;
calculating a ratio threshold based on the second eye-opening ratios; and
calculating the eye open duration within the first time interval and the second time interval based on the ratio threshold.
6 . The liveness detection apparatus of claim 1 , wherein the first confirmation operation comprises the following operation:
generating a notify signal to control an output apparatus to send a reminder.
7 . The liveness detection apparatus of claim 1 , wherein the second confirmation operation comprises the following operations:
increasing a triggered count; and
in response to the triggered count exceeding a count threshold, adjusting a driving assistance parameter, wherein the driving assistance parameter is configured to determine a level of driving assistance involvement for a vehicle drove by the driver.
8 . The liveness detection apparatus of claim 1 , wherein the processor is further configured to execute the following operation:
in response to determining the driver closing their eyes, recalculating the eye open duration of the driver in the live image.
9 . The liveness detection apparatus of claim 1 , wherein the processor is further configured to execute the following operations:
determining whether the driver is in a high-risk driving state based on one or more vehicle parameter; and
in response to the driver being in the high-risk driving state, reducing the first duration threshold and the second duration threshold.
10 . The liveness detection apparatus of claim 1 , wherein the processor is further configured to execute the following operations:
calculating a weight based on one or more vehicle parameter; and
calculating the first duration threshold and the second duration threshold based on the weight.
11 . A liveness detection method, being adapted for use in an electronic apparatus, wherein the liveness detection method comprises the following steps:
capturing a live image of a driver;
calculating an eye open duration of the driver based on the live image within a first time interval;
in response to the eye open duration exceeding a first duration threshold, executing a first confirmation operation to notify the driver to complete a liveness detection operation;
in response to the first confirmation operation has been executed, continuing to calculate the eye open duration based on the live image within a second time interval, wherein the second time interval is later than the first time interval; and
in response to the eye open duration exceeding a second duration threshold, executing a second confirmation operation to notify the driver to complete the liveness detection operation, wherein the second duration threshold is greater than the first duration threshold.
12 . The liveness detection method of claim 11 , wherein the step of calculating the eye open duration further comprises:
calculating a first eye-opening ratio of the driver in the live image within the first time interval; and
determining whether the driver opens their eyes based on the first eye-opening ratio.
13 . The liveness detection method of claim 12 , wherein the step of calculating the first eye-opening ratio of the driver in the live image further comprises:
searching for a plurality of eye feature points of the driver in the live image; and
calculating the first eye-opening ratio of the driver in the live image based on the eye feature points.
14 . The liveness detection method of claim 12 , wherein the step of calculating the first eye-opening ratio of the driver in the live image further comprises:
calculating a first distance between an upper eyelid and a lower eyelid;
calculating a second distance between a left eye corner and a right eye corner; and
calculating a ratio of the first distance and the second distance as the first eye-opening ratio.
15 . The liveness detection method of claim 11 , wherein the step of calculating the eye open duration further comprises:
calculating a plurality of second eye-opening ratios of the driver in the live image within a third time interval;
calculating a ratio threshold based on the second eye-opening ratios; and
calculating the eye open duration within the first time interval and the second time interval based on the ratio threshold.
16 . The liveness detection method of claim 11 , wherein the first confirmation operation comprises the following step:
generating a notify signal to control an output apparatus to send a reminder.
17 . The liveness detection method of claim 11 , wherein the second confirmation operation comprises the following steps:
increasing a triggered count; and
in response to the triggered count exceeding a count threshold, adjusting a driving assistance parameter, wherein the driving assistance parameter is configured to determine a level of driving assistance involvement for a vehicle drove by the driver.
18 . The liveness detection method of claim 11 , further comprising:
in response to determining the driver closing their eyes, recalculating the eye open duration of the driver in the live image.
19 . The liveness detection method of claim 11 , further comprising:
determining whether the driver is in a high-risk driving state based on one or more vehicle parameter; and
in response to the driver being in the high-risk driving state, reducing the first duration threshold and the second duration threshold.
20 . The liveness detection method of claim 11 , further comprising:
calculating a weight based on one or more vehicle parameter; and
calculating the first duration threshold and the second duration threshold based on the weight.