TW202411711A - Electronic device and method for phase detection autofocus - Google Patents

Abstract

The present disclosure discloses a method for phase detection autofocus and an electronic device thereof. The method is applicable to an image capture device and includes the following steps: calculating a plurality of phase difference values and a plurality of defocus values of a focusing image by a lens of the image capture device, wherein each of the phase difference values corresponds to a defocus value; grouping the phase difference values into a foreground group and a background group; determining a foreground focus position by the defocus values corresponding to the phase differences in the foreground group; determining a background focus position by the defocus values corresponding to the phase differences in the background group; determining an ideal focus position based on a depth of field of the focusing image, the foreground focus position and the background focus position; and moving the lens to the ideal focus position.

Description

Electronic device and phase focusing method

The present disclosure relates to an electronic device, and more particularly to an electronic device having an image capture device and a phase focusing method.

With the popularity of smart phones, consumers have more and more opportunities to take photos with their smart phones, and their requirements for photo quality are getting higher and higher. Focus is a crucial part of the camera photography process. Faster focus speed and more accurate focus accuracy are very beneficial to improving user experience and product competitiveness.

In order to facilitate users to quickly capture clear images, the cameras on smartphones are generally equipped with an autofocus function, which can actively detect objects within the camera's field of view and automatically move the lens to focus on the object when the user activates the camera. However, the autofocus function of existing smartphones can only accurately and quickly focus when there is an object with a single depth of field within the camera's field of view. When multiple objects with different depths of field appear within the camera's field of view, the autofocus function is prone to lose focus because it cannot determine which object is the main subject.

The above “prior art” description is only to provide background technology, and does not admit that the above “prior art” description discloses the subject matter of the present disclosure, does not constitute the prior art of the present disclosure, and any description of the above “prior art” should not be regarded as any part of the present case.

The present disclosure provides a phase focusing method, which is applicable to an image capture device and includes the following steps: calculating a plurality of phase difference values and a plurality of defocus values of a picture in the focus of the lens of the image capture device, wherein each of the phase difference values corresponds to a defocus value; performing a grouping operation to group the phase difference values, thereby obtaining a near-view group and a far-view group; and using the near-view group to group the near-view group. The invention relates to a method for determining a near-view focus position according to a defocus value corresponding to the phase difference value in the far-view group; determining a far-view focus position according to the defocus value corresponding to the phase difference value in the far-view group; determining an ideal focus position of a lens of the image capture device based on a depth of field of the in-focus picture, the near-view focus position and the far-view focus position; and moving the lens of the image capture device to the ideal focus position.

The present disclosure provides an electronic device, which includes an image capture device and a processing module. The image capture device includes a lens, a lens control module and an image sensing module; the lens control module is used to control the movement of the lens, and the image sensing module generates a focused image using light entering the image capture device through the lens. The processing module is coupled to the lens control module and the image sensing module in the image capture device, and is configured to perform the following operations: calculating a plurality of phase differences and a plurality of defocus values of the lens in focus; grouping the phase difference values to obtain a near view group and a far view group; determining a near view focus position and a far view focus position from the defocus values corresponding to the phase difference values grouped into the near view group and the far view group; and determining an ideal focus position of the lens based on a depth of field of the in-focus picture, the near view focus position and the far view focus position, and causing the lens control module to move the lens according to the ideal focus position.

The following uses more specific text to describe various embodiments or examples of the present disclosure in the drawings. It should be understood that these descriptions are for illustration only and are not intended to limit the present disclosure. For those with ordinary knowledge in the field related to the present disclosure, any changes or modifications to the embodiments described, as well as further applications of the principles described in this document, are all common and customary changes. Component symbols may be reused between embodiments, but it does not mean that one or more features of one embodiment must appear in another embodiment, even if the same component symbols are used.

The terms used herein are for describing specific exemplary embodiments and are not intended to limit the inventive concepts of the present disclosure. In the present disclosure, the singular forms "a", "an" and "the" also include the plural forms unless the context clearly indicates otherwise. It should be further understood that in this specification, open terms such as "include" and "comprise" represent the presence of the features, integers, steps, operations, elements or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components or their group organizations.

FIG1 is a schematic diagram of the back of an electronic device 10 according to the present disclosure, and FIG2 is a functional block diagram of the electronic device 10 according to the present disclosure. Referring to FIG1 and FIG2 , the electronic device 10 includes a housing 110, an image capture device 120, and a processing module 130; a lens 122 of the image capture device 120 is exposed from the housing 110, and the processing module 130 is installed in the housing 110 and is used to control the image capture device 120 to capture images. Exemplarily, the electronic device 10 can be any of various types of computer systems that are mobile or portable and perform communications. Specifically, the electronic device 10 can be a smart phone, but the image capture device 120 of the present invention is not limited to the rear lens used in a smart phone. The image capture device 120 can be applied to the front lens of a smart phone or other devices that require mobile focus according to needs; for example, the image capture device 120 can be applied to a digital tablet, a digital camera or a smart wearable device. The above-mentioned electronic device 10 is only an example of the actual application of the present invention, and does not limit the application scope of the image capture device 120 of the present invention.

The image capture device 120 further includes an image sensing module 124 and a lens control module 126, both of which are disposed in the housing 110 and electrically coupled to the processing module 130. When the image capture device 120 is used to capture a target scene including a light source and an object, the reflection of the object in the target scene and the light of the light source will pass through one or more optical lenses (not shown) in the lens 122 and form an image on the image sensing module 124. The photoelectric element (e.g., photodiode) of each pixel in the image sensing module 124 performs photoelectric conversion on the sensed light to generate a corresponding electrical signal and transmit it to the processing module 130.

In the embodiment of FIG. 3 , some pixels 1242 in the image sensing module 124 may be symmetrically semi-shielded by a light shielding portion 1244 in pairs to form phase detection pixels 1245A and 1245B, which are specifically used for phase detection. The paired phase detection pixels 1245A and 1245B may be separated by a certain distance, and there may be one or more pixels 1242 not covered by the light shielding portion 1244 between the paired phase detection pixels 1245A and 1245B; however, in other embodiments, the paired phase detection pixels 1245A and 1245B may be arranged adjacent to each other. In some embodiments, the pixel 1242 may receive light from the target scene through a microlens 1246.

For the convenience of explanation, the phase detection pixel 1245A disposed on the left side of FIG. 3 is referred to as the left phase detection pixel, and the phase detection pixel 1245B disposed on the right side of FIG. 3 is referred to as the right phase detection pixel. The light blocking portion 1244 is used to shield light from a specific direction. For example, the light blocking portion 1244 of the left phase detection pixel 1245A blocks light from reaching the left side of the pixel 1242, and the light blocking portion 1244 of the right phase detection pixel 1245B blocks light from reaching the right side of the pixel 1242; therefore, the left phase detection pixel 1245A can only receive light from the right side, and the right phase detection pixel 1245B can only receive light from the left side.

In FIG. 4A , the lens 122 is at the focusing position F0, and the object 1250 is imaged on the pixel 1242 of the image sensing module 124. Therefore, the left light L and the right light R of the reflected light of the object 1250 after passing through the lens 122 will illuminate the same position, so that the light intensity distribution curve _IA sensed by the left phase detection pixel 1245A will overlap with the light intensity distribution curve _IB sensed by the right phase detection pixel 1245B.

However, in FIG4B , since the position of the lens 122 is too close to the object 1250, that is, the position F1 of the lens 122 is too far forward compared to the focus position F0, the peak value of the light intensity distribution curve _IA of the right side light R sensed by the left phase detection pixel 1245A will be to the left of the peak value of the light intensity distribution curve _IB of the left side light L sensed by the right phase detection pixel 1245B. In this case, if the right side is positive, a negative offset SF1 can be obtained by subtracting the peak coordinates of the light intensity distribution curve _IA from the peak coordinates of the light intensity distribution curve _IB .

In contrast, in FIG4C , since the position of the lens 122 is too far from the object 1250, that is, the position F2 of the lens 122 is too far back compared to the focus position F0, the peak value of the light intensity distribution curve _IA of the right side light R sensed by the left phase detection pixel 1245A will be to the right of the peak value of the light intensity distribution curve _IB of the left side light L sensed by the right phase detection pixel 1245B. In this case, if the right side is positive, a positive offset SF2 can be obtained by subtracting the peak coordinates of the light intensity distribution curve _IA from the peak coordinates of the light intensity distribution curve _IB .

It can be observed from FIGS. 4A to 4C that when the distance between the lens 122 and the object 1250 changes, the peak positions of the light intensity distribution curve _IA and the peak positions of the light intensity distribution curve _IB will also change. Therefore, the current focusing status of the lens 122 can be known based on the offsets SF1 and SF2 between the two peaks, and the distance between the lens 122 and the focus position F0 and the direction of the deviation, i.e., the defocus values DF1 and DF2, can be calculated.

In this embodiment, the processing module 130 can determine whether the image capture device 120 is in a focus state (i.e., the state shown in FIG. 4A ), a front focus state (i.e., the state shown in FIG. 4B ), or a back focus state (i.e., the state shown in FIG. 4C ) by comparing the phase difference of the light intensity distribution information presented in the electrical signals output by the paired phase detection pixels 1245A and 1245B. The lens control module 126 may include a motor that receives control from the processing module 130 to move the lens 122 along the optical axis to appropriately adjust the focus of the image captured by the image sensing module 124 (i.e., to realize the lens focusing function).

FIG5 is a flow chart of the phase focusing method of the present disclosure. Referring to FIG2 and FIG5, the phase focusing method 300 may include steps S302-S318 and may be applied to the electronic device 10. In some embodiments, the electronic device 10 enters the real-time preview mode (as shown in FIG6) after the user activates the camera function; at this time, the image of the target scene is displayed on the screen 140 of the electronic device 10 for the user to preview the image content to be captured. At the same time, the image sensing module 124 may continuously capture the real-time focusing picture 142 and convert it into a corresponding electrical signal to be transmitted to the processing module 130 to execute the automatic focusing procedure.

In the process of performing phase focusing, the processing module 130 calculates the phase difference value and the defocus value corresponding to each block 144 (as shown in FIG. 7 ) in the focused image 142 from the electrical signal provided by the image sensing module 124 (step S302). In this embodiment, the number of the paired phase detection pixels 1245A and 1245B is at least equal to the number of blocks 144 to ensure that the processing module 130 can calculate the phase difference value and the defocus value corresponding to each block 144 from the electrical signal output by each paired phase detection pixel 1245A and 1245B. The blocks 144 and their number shown in FIG. 7 are merely examples for the convenience of explanation and are not intended to limit the present application.

In this embodiment, the phase difference value can be calculated, for example, based on the offsets SF1 and SF2 between the peaks of the light intensity distribution curves _IA and _IB provided by the left phase detection pixel 1245A and the corresponding right phase detection pixel 1245B in FIG. 4B and FIG. 4C, respectively. However, the present application is not limited thereto. In some other embodiments, the processing module 130 can also calculate the phase difference according to other methods or definitions. Furthermore, the defocus value can be, for example, the distance DF1 between the position F1 of the lens 122 and the focus position F0 in FIG. 4B, and the distance DF2 between the position F1 of the lens 122 and the focus position F0 in FIG. 4C.

In addition, generally speaking, the accuracy of the phase shift is related to the content of the image. For example, if most of the objects in the picture do not have obvious boundaries or have more noise, the peak values of the light intensity sensed by the left phase detection pixel 1245A and the right phase detection pixel 1245B may be very close, so that it is impossible to obtain an accurate phase difference from the light intensity sensed by the two. Therefore, the focusing method of the present application can evaluate the calculated phase difference value of each block 144 according to specific parameters and remove unqualified phase difference values, thereby preventing the image capture device 120 from obtaining an erroneous lens focus position due to reference to an inaccurate phase difference. For example, in step S303, the processing module 130 may obtain a confidence level value corresponding to each phase difference value as a basis for determining whether the phase difference value is qualified. The confidence level value may be related to whether the phase difference value detected by the processing module 130 in the focused image is accurate when calculating the phase offset; the confidence level value may be, for example, a reference index proportional to the number of edges of the focused image that are easy to detect the phase difference value, that is, the reliability or accuracy of the phase difference value. In this embodiment, the confidence level value may be generated by the image sensing module 124 during the process of sensing the image and provided to the processing module 130, or may be estimated by the processing module 130 when calculating the phase difference value.

In this embodiment, in addition to referring to the confidence level value, the processing module 130 may selectively perform temporal filtering in step S304 to obtain the focus position average value and standard deviation, and may evaluate whether the corresponding phase difference value is qualified based on the standard deviation and the confidence level value in step S306.

When performing time domain filtering, the processing module 130 may perform step S304 at multiple time points within a given period to estimate a focus position based on each defocus value, and then obtain the focus position average and standard deviation within the given period using the multiple focus positions estimated at these time points. For example, the processing module 130 estimates the focus position using the defocus value and the current position of the lens 122 provided by the lens control module 126.

When the number of phase difference values and defocus values obtained by the lens 122 of the image capture device 120 during focusing is M and the number of time points during a given period is N , the processing module 130 performs time domain filtering operations according to equations (1) to (3): Formula 1) Formula (2) Formula (3) wherein D _i,m represents the mth defocus value obtained at time point i , F _i represents the current position of the lens 122 of the image capture device 120 at time point i , x _i,m represents the focus position estimated at time point i based on the defocus value D _i,m , represents the average value of the focus position generated by the time domain filtering operation based on the mth defocus value during the given period, and std _m represents the standard deviation generated by the time domain filtering operation based on the mth defocus value during the given period. In addition, M and N are positive integers; in this embodiment, N can be 3 to 5.

In order to improve the focusing accuracy and speed, the processing module 130 can filter out qualified phase difference values from all phase difference values obtained in step S302 based on the confidence level value and the standard deviation (step S306). For example, the processing module 130 can determine the phase difference value whose confidence level value is greater than the first threshold value and whose standard deviation is less than the second threshold value as a qualified phase difference value, and determine the phase difference value whose confidence level value is not greater than the first threshold value or whose standard deviation is not less than the second threshold value as an unqualified phase difference value; wherein the first threshold value and the second threshold value can be determined by the performance parameters of the image sensing module 124. The processing module 130 will discard the unqualified phase difference value (step S308) and will not perform grouping processing on it. It should be particularly noted that the processing module 130 may not perform the time domain filtering operation of step S304 but only use the confidence level value to filter the multiple phase difference values obtained in step S302.

In step S310, the processing module 130 may group the qualified phase difference values using a grouping algorithm to obtain a near-view group and a distant view group. In some embodiments, the processing module 130 may group the qualified phase difference values using a K-means algorithm. FIG8 is a flow chart of the grouping operation performed in the present disclosure. Referring to FIG8, in step S3102, the processing module 130 may calculate the average value of the qualified phase difference values. Then, in step S3104, the processing module 130 may perform a first grouping of all qualified phase difference values based on the average value calculated in step S3102. In some embodiments, the processing module 130 may, for example, determine whether the qualified phase difference value should be classified into a near-view group or a distant view group by comparing each qualified phase difference value with the average value; for example, when performing the first grouping, qualified phase difference values less than or equal to the average value belong to the near-view group, and qualified phase difference values greater than the average value belong to the distant view group. However, the present application is not limited thereto. In some other embodiments, the processing module 130 may also randomly select two cluster centers as the center of the distant view group and the center of the near-view group, and perform the first grouping based on the distance between each qualified phase difference value and the two cluster centers.

After the first grouping is completed, the processing module 130 calculates the average value of the qualified phase difference values belonging to the near-view group (hereinafter referred to as the near-view phase average value) and the average value of the qualified phase difference values belonging to the distant view group (hereinafter referred to as the distant view phase average value) (step S3106). Then, the processing module 130 performs a second grouping of all the qualified phase difference values with the near-view phase average value and the distant view phase average value as the group centers (step S3108).

When performing the second grouping, the processing module 130 will determine whether each phase difference value should be assigned to the near view group or the distant view group according to the difference between each qualified phase difference value and the near view phase average value and the distant view phase average value. For example, the processing module 130 can calculate a first difference between each qualified phase difference value and the near view phase average value, and a second difference between each qualified phase difference value and the distant view phase average value. When the first difference corresponding to a qualified phase difference value is less than or equal to the second difference, it means that the difference between the qualified phase difference value and the near view phase average value is smaller, so the qualified phase difference value can be assigned to the near view group; when the first difference corresponding to a qualified phase difference value is greater than the second difference, it means that the difference between the qualified phase difference value and the distant view phase average value is smaller, so the qualified phase difference value can be assigned to the distant view group. In this way, after the second grouping, the difference in phase difference between the near-view group and the far-view group can be further reduced.

When performing re-grouping, the processing module 130 can determine whether to terminate the grouping operation based on whether there is a qualified phase difference value that needs to be moved from one of the near-view group and the distant-view group to the other (step S3110). For example, if during re-grouping, no eligible phase difference value moves from the close-view group to the distant group, and no eligible phase difference value moves from the distant group to the close-view group, it means that the grouping operation has become stable and the grouping operation can be terminated; if any eligible phase difference value moves from the close-view group to the distant group or from the distant group to the close-view group, it means that the grouping operation may not be stable yet, and steps S3106 to S3108 must be repeated until all phase difference values no longer move between the close-view group and the distant group.

Referring to FIG. 5 again, after the grouping operation is completed, the processing module 130 determines the near-view focus position with the defocus value corresponding to the qualified phase difference value grouped in the near-view group, and determines the far-view focus position with the defocus value corresponding to the qualified phase difference value grouped in the far-view group (step S312). In some embodiments, the processing module 130 estimates a first target focus position with the defocus value corresponding to each phase difference value in the near-view group, and after obtaining the first target focus positions corresponding to all phase difference values in the near-view group, averages these first target focus positions to obtain a first average value as the near-view focus position. Similarly, the processing module 130 estimates a second target focus position using the defocus value corresponding to each phase difference value in the distant view group, and after obtaining the second target focus positions corresponding to all phase difference values in the distant view group, averages these second target focus positions to obtain a second average value as the distant view focus position.

Next, the processing module 130 calculates the hyperfocal distance, near depth of field, far depth of field and depth of field of the image capture device 120; wherein the hyperfocal distance is the shortest distance at which the lens 122 can focus while maintaining an acceptable sharpness on an object at infinity. Generally, when the lens is focused at the hyperfocal distance, all objects within a distance from half of the hyperfocal distance to infinity can be clearly imaged on the image sensing module 124, and the depth of field refers to the relatively clear imaging range before and after the focus point of the image sensing module 124. The hyperfocal distance (H), near depth of field ( _DN ), far depth of field ( _DF ) and depth of field (DOF) can be calculated as follows: Formula (4) Formula (5) Formula (6) Formula (7)

In equations (4) to (7), f is the focal length, N is the aperture value, c is the diameter of the blur circle, and s is the object distance.

After obtaining the depth of field, the processing module 130 can determine the ideal focus position of the lens 122 based on the depth of field, the near focus position and the far focus position (step S314). In some embodiments, the processing module 130 determines the ideal focus position of the lens 122 by comparing the difference between the far focus position and the foreground focus position with the depth of field. Specifically, when the distance between the distant focus position and the near focus position is greater than the depth of field, this means that in the target scene, only objects within the depth of field can be clearly imaged. Generally speaking, users usually focus on nearby objects when taking pictures. Therefore, the processing module 130 can preferentially select the near focus position as the ideal focus position (step S316). On the contrary, when the distance between the distant focus position and the near focus position is equal to or less than the depth of field, this means that all objects in the target scene can be clearly imaged. Therefore, the processing module 130 can use the average value of the near focus position and the distant focus position as the ideal focus position (step S318).

Finally, in step S320, the processing module 130 sends a control signal to move the lens 122 of the image capture device 120 to an ideal focus position to complete phase focusing.

Although the present disclosure and its advantages have been described in detail, it should be understood that various changes, substitutions or modifications may be made therein without departing from the technology of the present disclosure as defined by the appended patent applications. For example, many of the procedures discussed above may be implemented using different methodologies and may be replaced by other procedures, or a combination thereof.

Furthermore, the scope of the present application is not intended to be limited to the specific embodiments of the procedures, machines, articles, compositions, devices, methods, and steps described in this specification. As will be readily understood by those skilled in the art from the disclosure of the present invention, procedures, machines, articles, compositions, devices, methods, or steps that exist now or will be developed later and that perform substantially the same functions or achieve substantially the same results as the corresponding embodiments described herein may be used in accordance with the present invention. Therefore, the attached application is intended to include within its scope such processes, machines, manufactures, compositions of matter, means, methods, and steps.

10: Electronic equipment 110: Housing 120: Image capture device 122: Lens 124: Image sensing module 126: Lens control module 130: Processing module 140: Screen 142: Focusing image 144: Block 300: Phase focusing method 1242: Pixel 1244: Light blocking part 1245A: Left phase detection pixel 1245B: Right phase detection pixel 1246: Micro lens 1250: Object L: Left side light R: Right side light S302-S320: Steps S3102-S3110: Steps SF1: Offset SF2: offset

When referring to the embodiments and the drawings in combination with the scope of the patent application, a more comprehensive understanding of the disclosure of the present application can be obtained. The same component symbols in the drawings refer to the same components. FIG. 1 is a schematic diagram of the back of an electronic device according to the disclosure. FIG. 2 is a functional block diagram of an electronic device according to the disclosure. FIG. 3 is a cross-sectional view of a portion of pixels of an image sensing module according to the disclosure. FIG. 4A is a schematic diagram of an image capture device in a focused state according to the disclosure. FIG. 4B and 4C are schematic diagrams of an image capture device in a defocused state according to the disclosure. FIG. 5 is a flow chart of a phase focusing method according to the disclosure. FIG. 6 is a schematic diagram of a focusing screen according to the disclosure. FIG. 7 is a schematic diagram of dividing the focused image into multiple blocks according to the disclosure. FIG. 8 is a flow chart of the grouping operation according to the disclosure.

S302-S320: Steps

Claims (18)

A phase focusing method is applicable to an image capture device, and the phase focusing method includes: Calculating a plurality of phase difference values and a plurality of defocus values of a focus image of a lens of the image capture device, wherein each of the phase difference values corresponds to a defocus value; Performing a grouping operation to group the phase difference values, thereby obtaining a near-view group and a far-view group; Determining a near-view focus position by the defocus value corresponding to the phase difference value in the near-view group; Determining a far-view focus position by the defocus value corresponding to the phase difference value in the far-view group; Determining an ideal focus position of the lens of the image capture device based on a depth of field of the focus image, the near-view focus position and the far-view focus position; and Move the lens of the image capture device to the ideal focus position. The method as described in claim 1, wherein: When the distance between the distant focus position and the near focus position is greater than the depth of field, the near focus position is selected as the ideal focus position, and When the distance between the distant focus position and the near focus position is equal to or less than the depth of field, an average value of the foreground focus position and the distant focus position is selected as the ideal focus position. The method as described in claim 1, wherein: Determining the near-view focus position by the defocus value corresponding to the phase difference value in the near-view group includes: Estimating a plurality of first target focus positions by the defocus value corresponding to each phase difference value in the near-view group; and Calculating a first average value of the first target focus positions, and using the first average value as the near-view focus position; and Determining the far-view focus position by the defocus value corresponding to the phase difference value in the far-view group includes: Estimating a plurality of second target focus positions by the defocus value corresponding to each phase difference value in the far-view group; and Calculating a second average value of the second target focus positions, and using the second average value as the far-view focus position. As described in claim 1, when calculating the phase difference values, a confidence level value associated with each of the phase difference values is obtained, and in a screening operation before performing a grouping operation, the confidence level value is used to determine whether the associated phase difference value is qualified. The method as described in claim 4 further comprises: Before performing the grouping operation, discarding unqualified phase difference values whose confidence level is lower than a first threshold. The method as described in claim 5 further comprises: At a plurality of time points within a given period, a focus position is estimated according to each of the defocus values, and then a focus position average value and a standard deviation are calculated for the given period using the plurality of focus positions estimated at the time points; wherein, in the screening operation, the confidence level value and the standard deviation are used to determine whether the phase difference value is qualified. The method as described in claim 6 further comprises: Before performing the grouping operation, discarding unqualified phase difference values whose confidence level value is lower than a first threshold and whose standard deviation is greater than a second threshold. The method as claimed in claim 6, wherein the number of the phase difference values and the defocus values obtained in the focused image is M , M is a positive integer, and the M phase difference values correspond to the M defocus values one by one, and the method further comprises: performing a time domain filtering operation: Wherein, the number of time points in the predetermined period is N , D _i,m represents the mth defocus value among the defocus values obtained at one of the time points i , F _i represents a current position of the lens of the image capture device at the time point i , x _i,m represents a focus position estimated at the time point i according to the defocus value D _i,m , and std _m respectively represent a focus position average value and a standard deviation generated according to the m -th defocus value during the given period through the time domain filtering operation. The method as described in claim 1, wherein the grouping operation comprises: (a) calculating an average value of the phase difference values; (b) determining whether each phase difference value should be assigned to the near view group or the distant view group by comparing each phase difference value with the average value; (c) calculating a near view phase average value and a distant view phase average value, wherein the near view phase average value is the average value of the phase difference values in the near view group, and the distant view phase average value is the average value of the phase difference values in the distant view group; (d) determining whether each phase difference value should be assigned to the near view group or the distant view group by the difference between each phase difference value and the near view phase average value and the distant view phase average value; and (f) repeating steps (c) to (d) until the phase difference values no longer move between the near view group and the distant view group. A method as described in claim 9, wherein if the difference between the phase difference value and the foreground phase average value is smaller than the difference between the phase difference value and the foreground phase average value, the phase difference value belongs to the foreground group, and if the difference between the phase difference value and the foreground phase average value is larger than the difference between the phase difference value and the foreground phase average value, the phase difference value belongs to the foreground group. The method as described in claim 1, wherein the clustering operation is performed using a K-means algorithm. An electronic device, comprising: An image capture device, comprising: A lens; A lens control module for controlling the movement of the lens; and An image sensing module for generating a focused image with light entering the image capture device through the lens; and A processing module, coupled to the lens control module and the image sensing module in the image capture device, and configured to perform the following operations: Calculating a plurality of phase differences and a plurality of defocus values of the focused image of the lens; Grouping the phase difference values to obtain a close-up group and a distant group; Determine a near-view focus position and a far-view focus position from the defocus values corresponding to the phase difference values grouped into the near-view group and the far-view group; and Determine an ideal focus position of the lens based on a depth of field of the image in focus, the near-view focus position and the far-view focus position, and enable the lens control module to move the lens according to the ideal focus position. The electronic device as described in claim 12, wherein: The processing module selects the near-view focus position as the ideal focus position when the distance between the far-view focus position and the near-view focus position is greater than the depth of field; and The processing module selects an average value of the near-view focus position and the far-view focus position as the ideal focus position when the distance between the far-view focus position and the near-view focus position is equal to or less than the depth of field. An electronic device as described in claim 12, wherein: The processing module estimates a plurality of first target focus positions from the defocus values corresponding to the phase difference values grouped as the near-view group, and calculates a first average value of the first target focus positions as the near-view focus position; and The processing module estimates a plurality of second target focus positions from the defocus values corresponding to the phase differences grouped as the far-view group, and calculates a second average value of the second target focus positions as the far-view focus position. An electronic device as described in claim 12, wherein the processing module is further configured to perform the following operations: When calculating the phase difference values, obtain a confidence level value associated with each of the phase difference values; Determine whether the associated phase difference value is qualified based on the confidence level value; and Before grouping the phase difference values, discard unqualified phase difference values. An electronic device as described in claim 15, wherein the processing module is further configured to perform the following operations: At a plurality of time points within a given period, estimate a focus position according to each of the defocus values, and then obtain a focus position and a standard deviation at the given time using the plurality of focus positions estimated at the time points; and Determine whether the relevant phase difference value is qualified using the confidence level value and the standard deviation; and Discard unqualified phase difference values before grouping the phase difference values. The electronic device as claimed in claim 12, wherein the number of the phase difference values and the defocus values obtained by the lens in focus is M , M is a positive integer, the M phase difference values correspond to the M defocus values one by one, and the processing module is further configured to perform a time domain filtering operation: Wherein, the number of time points in the predetermined period is N , D _i,m represents the mth defocus value among the defocus values obtained at one of the time points i , F _i represents a current position of the lens of the image capture device at the time point i , x _i,m represents a focus position estimated at the time point i according to the defocus value D _i,m , and std _m respectively represent a focus position average value and a standard deviation generated according to the m -th defocus value during the given period through the time domain filtering operation. The electronic device as described in claim 14, wherein the processing module utilizes a K-average algorithm to divide the phase difference values into the near-view group and the far-view group.

Images