Electronic device and screen brightness adjustment method thereof

Abstract

This invention provides an electronic device and a method for adjusting its screen brightness. The screen brightness adjustment method is applicable to electronic devices including a display. The method includes the following steps: collecting multiple ambient brightness values ​​corresponding to multiple consecutive sampling time points using a photosensor; extracting multiple continuous feature brightness values ​​from the multiple ambient brightness values; using a trained classification model to determine whether the electronic device is in a specific light source environment or a normal light source environment based on the multiple continuous feature brightness values; when the electronic device is determined to be in a specific light source environment, determining a reference brightness value based on the multiple ambient brightness values ​​collected by the photosensor, and adjusting the screen brightness of the display based on the reference brightness value.

Claims

1. A screen brightness adjustment method, applicable to an electronic device including a display, the method comprising: A single light sensor collects multiple ambient brightness values ​​corresponding to multiple consecutive sampling time points. Multiple consecutive feature brightness values ​​are extracted from the multiple ambient brightness values; a trained classification model is used to determine whether the electronic device is in a specific light source environment or a normal light source environment based on the multiple consecutive feature brightness values; and when the electronic device is determined to be in the specific light source environment, a reference brightness value is determined based on the multiple ambient brightness values ​​collected by the light sensor, and the screen brightness of the display is adjusted based on the reference brightness value. The step of extracting the multiple consecutive feature brightness values ​​from the multiple ambient brightness values ​​includes: determining a specific brightness value from the multiple ambient brightness values ​​according to a predetermined condition; A brightness feature extraction interval is determined based on the specific brightness value corresponding to a specific sampling time point; and multiple consecutive feature brightness values ​​within the brightness feature extraction interval are extracted from the multiple ambient brightness values.

2. The screen brightness adjustment method as described in claim 1 further includes: When the electronic device is determined to be in the normal light source environment, the screen brightness of the display is adjusted according to one of the multiple ambient brightness values.

3. The screen brightness adjustment method as described in claim 1, wherein the specific brightness value is the peak brightness value among the plurality of ambient brightness values.

4. The screen brightness adjustment method as described in claim 1, wherein the step of determining whether the electronic device is in the specific light source environment or the normal light source environment based on the multiple consecutive feature brightness values ​​using the trained classification model includes: An input vector composed of the multiple consecutive feature brightness values ​​is input to the trained classification model to generate a classification result, wherein the classification result is used to determine whether the electronic device is in the specific light source situation or the normal light source situation, and the classification model includes a convolutional neural network (CNN) model.

5. The screen brightness adjustment method as described in claim 1, wherein when it is determined that the electronic device is in the specific light source environment, the step of determining the reference brightness value based on the multiple ambient brightness values ​​collected by the light sensor, and adjusting the screen brightness of the display based on the reference brightness value includes: Acquire multiple luminance pulses formed by the multiple ambient luminance values; The reference brightness value is determined based on a brightness peak value and a brightness valley value of each of the plurality of brightness pulses; And determine a target screen brightness based on the reference brightness value, so as to adjust the screen brightness of the display according to the target screen brightness.

6. The screen brightness adjustment method as described in claim 5, wherein the step of determining the reference brightness value based on the brightness peak value and the brightness valley value of each of the plurality of brightness pulses includes: Calculate the reference average brightness of each of the plurality of brightness pulses based on the brightness peak value and the brightness valley value of each of the plurality of brightness pulses; And a weighted average calculation is performed based on the reference average brightness of each of the plurality of brightness pulses to obtain the reference brightness value.

7. The screen brightness adjustment method as described in claim 6, wherein the weighting factor corresponding to the reference average brightness of each of the plurality of brightness pulses is defined by the following formula: , where represents the weighting factor corresponding to the reference average brightness of each of the plurality of brightness pulses, and i represents the sequential number of each of the plurality of brightness pulses after being sorted according to their occurrence time.

8. The screen brightness adjustment method as described in claim 1 further includes: The light sensor collects first raw light source information that is labeled as the specific light source context; The system collects second raw light source information labeled as the normal light source situation through the light sensor; extracts multiple first training feature brightness values ​​labeled as the specific light source situation from the first raw light source information, and extracts multiple second training feature brightness values ​​labeled as the normal light source situation from the second raw light source information to obtain a training data set; and performs a classification model training based on the training data set to obtain the trained classification model.

9. The screen brightness adjustment method as described in claim 8 further includes: An expanded training dataset is obtained by performing data augmentation based on the brightness values ​​of the multiple first training features. And to add the expanded training dataset to the training dataset.

10. An electronic device comprising: Light sensor; monitor; Storage device that records multiple instructions; A processor, coupled to the display and the storage device, executes the instructions and is configured to: collect multiple ambient brightness values ​​corresponding to multiple consecutive sampling time points via the light sensor; extract multiple consecutive feature brightness values ​​from the multiple ambient brightness values; determine whether the electronic device is in a specific light source environment or a normal light source environment based on the multiple consecutive feature brightness values ​​using a trained classification model; and when the electronic device is determined to be in the specific light source environment, determine a reference brightness value based on the multiple ambient brightness values ​​collected by the light sensor, and adjust the screen brightness of the display based on the reference brightness value. Extracting the multiple consecutive feature brightness values ​​from the multiple ambient brightness values ​​includes: determining a specific brightness value from the multiple ambient brightness values ​​based on a predetermined condition; A brightness feature extraction interval is determined based on the specific brightness value corresponding to a specific sampling time point; and multiple consecutive feature brightness values ​​within the brightness feature extraction interval are extracted from the multiple ambient brightness values.

Images