TWI637210B - Image pick-up apparatus and fast focusing method thereof - Google Patents

Abstract

An image capturing device and a quick focusing method thereof. The image capturing device includes a lens, an image sensor, a lens control circuit, and a focusing circuit. In this method, the lens control circuit controls the movement of the lens to capture an image by the image sensor, and the focus circuit calculates the first and second curves of the focus value of the captured image at the first and second scales relative to the number of steps of the lens movement. Then, the focus circuit divides the focus value of the first curve and the second curve to obtain a characteristic curve, and estimates the position of the trough according to the position of the peak and the saddle point appearing in the characteristic curve. Finally, the lens control circuit controls the movement of the lens to the number of moving steps corresponding to the trough, and then moves to the number of moving steps corresponding to the peak of the first curve or the second curve to complete the focusing.

Description

Image capturing device and quick focusing method thereof

The present invention relates to an image capturing device and method, and more particularly to an image capturing device and a quick focusing method thereof.

With the advancement of image capture technology, the number of pixels in digital cameras has increased dramatically, but the size of cameras has been relatively small, and can be deployed on portable electronic devices such as mobile phones and tablet computers, allowing users to shoot images anytime, anywhere. In order to facilitate the user to quickly capture clear images, the cameras on the portable electronic devices are generally equipped with an Auto Focus (AF) function, which can actively detect the camera's field of view while the camera is enabled. Objects and automatically move the lens to focus on the object. Thereby, the time taken for the user to manually focus can be saved.

The traditional focus method relies on the Contrast Based Auto Focusing (CBAF) to determine the best focus position. There are many ways to calculate the image contrast. The Tenengrad function and wavelet are often applied. Wavelet), Gaussian derivative, laplac transform, Finite Impulse Response (FIR) filter, Infinite Impulse Response (IIR) filter, etc. It is widely used to calculate the contrast value of the image in the form of FIR/IIR, which is used as the basis for finding the best focus position.

However, no matter which image comparison calculation method is used, the calculated image contrast value cannot fully reflect the sharpness of the image because of the limitation of the reference area. At this time, multi-scale image contrast values can be used to assist image focus, that is, for different scale images, the same or different filters are applied to extract contrast values of different scale images, so that the subsequent focus module can quickly find the best. Focus position.

In general, the focus strategy is based on the climbing mechanism. At different focusing distances (positions), the strategy of climbing is determined according to the change of the image contrast value. Figure 1 is a schematic diagram of a conventional camera focusing mountain climbing mechanism. The climbing mechanism can be divided into three sections: the first section is a stride method of applying a large number of steps, and quickly reaches the foot of the mountain (the contrast value of the image near the focus position is slightly There is a change; the second stage is a mid-step climbing (the image contrast value will continue to increase, this means that the focus position is getting closer to the best focus distance); the third paragraph is after crossing the hill (the image contrast value) Suddenly getting smaller, ending the original rising trend, which means that the focus position is overdone, beyond the best focus position), you must retreat at a small pace to move to the top of the mountain, and finally reach the best focus position.

The existing climbing mechanism cannot avoid the above three-stage process, and can not effectively reduce the focusing time. In the process of gradually retreating to the top of the mountain beyond the hill, the image will appear blurred, clear, fuzzy, and clear, resulting in users. Have a bad shooting experience.

The invention provides an image capturing device and a fast focusing method thereof, which can accelerate the focusing speed of the camera and provide a good shooting experience for the user.

A fast focusing method for an image capturing device of the present invention, wherein the image capturing device comprises a lens, an image sensor, a lens control circuit and a focusing circuit. This method controls the movement of the lens by the lens control circuit to capture an image. Then, the focus circuit calculates a first curve of the focus value of the image captured by the image sensor at the first scale with respect to the number of steps of the lens movement, and a second curve of the focus value at the second scale with respect to the number of steps of the lens movement. . Then, the focus circuit divides the focus value of the first curve and the second curve to obtain a characteristic curve, and estimates the position of the trough according to the position of the peak and the saddle point appearing in the characteristic curve. Finally, the lens control circuit controls the movement of the lens to the number of moving steps corresponding to the trough, and then moves to the number of moving steps corresponding to the peak of the first curve or the second curve to complete the focusing.

In an embodiment of the invention, the step of calculating, by the focusing circuit, the first curve of the focus value of the image captured by the image sensor at the first scale relative to the moving step of the lens comprises calculating by using an image contrast algorithm The contrast value of the plurality of pixels in the image of the first scale, and the sum of the powers of the contrast values of the pixels located in the in-focus frame in the image as the focus value of the image at the first scale.

In an embodiment of the invention, the step of calculating, by the focusing circuit, the second curve of the focus value of the image captured by the image sensor at the second scale relative to the moving step of the lens comprises calculating by using an image contrast algorithm The contrast value of the plurality of pixels in the image of the second scale, and the sum of the powers of the contrast values of the pixels located in the focus frame in the image is used as the focus value of the image at the second scale.

In an embodiment of the invention, the step of estimating the position of the trough according to the position of the peak and the saddle point appearing in the characteristic curve comprises calculating a change of the slope of the characteristic curve during the movement of the lens to capture the image, when When the slope transitions to zero, it is judged that the peak of the characteristic curve is reached, and when the slope is changed from zero to a maximum value, the saddle point of the characteristic curve is judged, and the distance between the moving step of the peak corresponding to the saddle point is calculated, thereby The number of moving steps corresponding to the saddle point is forward plus the distance to estimate the position of the trough.

In an embodiment of the invention, the step of controlling, by the lens control circuit, the number of movement steps corresponding to the movement of the lens to the trough, and then moving to the peak of the first curve or the second curve to complete the focusing comprises: The lens control circuit controls the number of moving steps of the lens moving to the valley at a first speed, and the number of moving steps corresponding to the peak of the first curve or the second curve moved by the lens control circuit at the second speed, wherein the lens is controlled by the lens control circuit, wherein the second speed is Less than the first speed.

An image capturing device of the present invention includes a lens, an image sensor, a lens control circuit, and a focusing circuit. The image sensor is coupled to the lens for capturing images. The lens control circuit is coupled to the lens to control the movement of the lens to capture an image. The focusing circuit is coupled to the image sensor and the lens control circuit for calculating a first curve of the focus value of the image captured by the image sensor relative to the number of steps of the lens and a focus value at the second scale. The second curve of the number of steps of the lens is divided by the focus value of the first curve and the second curve to obtain a characteristic curve, and the position of the trough is estimated according to the position of the peak and the saddle point appearing in the characteristic curve. The lens control circuit controls the movement of the lens to the number of moving steps corresponding to the valley estimated by the focusing circuit, and then moves to the number of moving steps corresponding to the peak of the first curve or the second curve to complete the focusing.

In an embodiment of the invention, the focusing circuit includes calculating a contrast value of a plurality of pixels in the image of the first scale by using an image contrast algorithm, and calculating a power of the contrast value of the pixel located in the focus frame in the image. The sum of the squares is used as the focus value of the image at the first scale.

In an embodiment of the invention, the focusing circuit includes calculating, by using an image contrast algorithm, a contrast value of a plurality of pixels in the image of the second scale, and calculating a power of a contrast value of the pixel located in the focus frame in the image. The sum of the powers is used as the focus value of the image at the second scale.

In an embodiment of the invention, the focusing circuit includes calculating a change in a slope of a characteristic curve during a process of moving a captured image of the lens, wherein when the slope transitions to zero, determining a peak reaching the characteristic curve, and when the slope is When changing from zero to a maximum value, the saddle point reaching the characteristic curve is judged, and the distance between the moving step of the peak corresponding to the saddle point is calculated, and the number of moving steps corresponding to the saddle point is forwardly added to estimate the trough. position.

In an embodiment of the invention, the length and width of the second dimension are the length and width of the first dimension.

Based on the above, the image capturing device and the fast focusing method thereof of the present invention, by calculating a focus value curve of a scale-free image and dividing it to obtain a characteristic curve, observing the peak and the saddle point from the characteristic curve, and using the distance therebetween Equidistantly estimate the position of the trough, thereby moving the lens one step to the focus position corresponding to the trough, and finally applying a small step to complete the focus. Thereby, the focusing speed can be improved, and the image change caused by the movement of the lens back and forth can be prevented from affecting the user's shooting experience.

The above described features and advantages of the invention will be apparent from the following description.

The invention compares the contrast value of the multi-scale image and converts it into the focus value, observes the change of the focus value and the ratio thereof, and recognizes the peak and the saddle point of the curve according to the slope change of the focus value ratio curve, and then equidistantly The way to estimate the valley position of the curve is the area where the best focus position is located. Thereby, the trend of the change of the contrast value can be effectively estimated, and the best focus position is indirectly estimated, and finally the purpose of fast focusing is achieved.

FIG. 2 is a block diagram of an image capture device according to an embodiment of the invention. Referring to FIG. 2, the image capturing device 20 of the embodiment is, for example, a digital camera, a digital video camera (DVC), or an electronic device such as a mobile phone, a tablet computer, a notebook computer, a navigation device, and a driving recorder. A camera on the device that provides a camera function. The image capturing device 20 includes a lens 22, an image sensor 24, a lens control circuit 26, and a focusing circuit 28. The functions of the image capturing device 20 are as follows:

The lens 22 is composed of a plurality of meniscus lenses which are driven by an actuator such as a stepping motor or a voice coil motor (VCM) to change the relative position between the lenses, thereby changing the focal length of the lens 22. The lens 22 is provided with an aperture and a shutter. The aperture is a ring-shaped opening formed by a plurality of metal blades. The aperture is opened or reduced according to the aperture value, thereby controlling the amount of light entering the lens 22, and the shutter is used. In order to control the length of time that light enters the lens 22, its combination with the aperture affects the amount of exposure of the image captured by the image sensor 24.

The image sensor 24 is connected to the lens 22 or disposed in the lens 22, and is configured with a Charge Coupled Device (CCD), a Complementary Metal-Oxide Semiconductor (CMOS) component, or other kinds of photosensitivity. The component senses the intensity of the light entering the lens 22 to produce an image.

The lens control circuit 26 is implemented, for example, as an integrated circuit (IC) for controlling the actuator in the lens 22 to drive the lens 22 to change its focal length. In the present embodiment, the lens control circuit 26 receives the focus information it provides from the focus circuit 28 and converts it to a progressive distance, thereby controlling the actuator 22 in the lens 22 to drive the lens 22.

The focusing circuit 28 is implemented, for example, by a microprocessor, a digital signal processor, a programmable controller, a special application integrated circuit, or the like, and receives an image captured by the image sensor 24 and analyzes the image. The contrast values at different scales are used to estimate the best focus position.

In detail, FIG. 3 is a flowchart of a fast focusing method of an image capturing device according to an embodiment of the invention. Referring to FIG. 2 and FIG. 3 simultaneously, the method of the present embodiment is applicable to the image capturing device 20 of FIG. 2, and the following uses the components of the image capturing device 20 of FIG. 2 to describe the fast focusing method of the embodiment. Detailed steps:

First, the lens 22 is controlled to move by the lens control circuit 26 to take an image by the image sensor 24 (step S302). The image capturing device 20 starts the image in the live view mode, for example, after the user enables the camera function. The captured image is instantly displayed on a display (not shown) of the image capturing device 20 for viewing by the user. While the image is being captured, the lens control circuit 26 controls the actuator in the lens 22 to move the lens to change the focal length of the lens 22 according to the focus position provided by the focusing circuit 28, thereby providing the image sensor 24 to shoot at different focal lengths. image.

Next, the focusing circuit 28 calculates a first curve of the focus value of the image captured by the image sensor 24 at the first scale with respect to the moving step of the lens 22 and a focus value at the second scale with respect to the lens 22 The second curve of the number of steps is moved (step S304). Wherein, the length and width of the second scale described above are one-half, one-quarter or one-quarter of the length and width of the first scale, and are not limited herein.

In detail, the focusing circuit 28 uses, for example, a Tenengrad function, a wavelet, a Gaussian derivative, a laplac transform, a Finite Impulse Response (FIR) filter, an infinite pulse. The image contrast algorithm such as Infinite Impulse Response (IIR) filter calculates the contrast value of multiple pixels in different scale images, and then uses the contrast value to calculate the focus value of different scale images. For example, the focus circuit 28 calculates the sum of the powers (eg, squares) of the contrast values of the pixels in the focus frame in the image as the focus value of the image at that scale.

For example, suppose the image of the first scale is the original image, and the image of the second scale is the image of which the original image length and width are reduced by one-half and the size is one quarter of the original image, and the embodiment has a finite impulse response ( The FIR method calculates the contrast value fv1 of the first scale image and calculates the contrast value fv2 of the second scale image by the same FIR method. Then, the sum of the squares of the contrast values fv1 of the pixels in the focus frame in the first scale image is calculated as the focus value FV1=SUM(fv1*fv1) of the first scale image, and the second scale image is also calculated. The sum of the squares of the contrast values fv2 of the pixels in the in-focus frame as the focus value FV2=SUM(fv2*fv2) of the second scale image.

Then, the focus circuit 28 divides the focus values of the first curve and the second curve to obtain a characteristic curve, and estimates the position of the valley based on the peaks appearing in the characteristic curve and the position of the saddle point (step S306). In detail, observing a large number of characteristic curves, it can be found that the best focus position area has a trend similar to the cubic curve, and the trough of the characteristic curve is also in the best focus position area, and the peak of the characteristic curve is generally located in the first curve or the second. The saddle of the curve.

Accordingly, when the focusing circuit 28 uses the first curve or the second curve as a basis for climbing (that is, moving the lens 22 such that the contrast value or the focusing value of the image captured by the image sensor 24 is gradually increased), during the mountain climbing process. In addition to observing the changes of the first curve and the second curve, the change of the characteristic curve is also observed. When reaching the foot of the first curve or the second curve, it is now near the peak of the characteristic curve. When the peak of the characteristic curve is crossed (ie, the slope of the characteristic curve changes from a negative value to a positive value or from a positive value to a negative value), the slope changes of the first curve, the second curve, and the characteristic curve can be simultaneously monitored. When the slope of the characteristic curve reaches the maximum value of the region (which may be a negative value or a positive value), it can be determined that the current focus position reaches the saddle point of the characteristic curve.

Based on the symmetry of the curve, the distance from the saddle point to the trough of the curve approximates the distance from the curve peak to the saddle point. Accordingly, in the present embodiment, when the slope of the characteristic curve reaches the maximum value of the region, the distance between the current focus position and the peak position of the characteristic curve is calculated, and the valley position of the characteristic curve is estimated by the equidistance.

For example, FIG. 4A to FIG. 4C are diagrams illustrating an example of a fast focusing method of an image capturing device according to an embodiment of the invention. 4A shows a curve in which the focus value FV1=SUM(fv1*fv1) of the first scale image changes with the number of moving steps in the foregoing embodiment, and FIG. 4B shows the focus value FV2 of the second scale image in the foregoing embodiment. =SUM(fv2*fv2) The curve that varies with the number of moving steps. FIG. 4C is a graph showing the ratio SFV=FV2/FV1 of the focus value FV2 and the focus value FV1. It can be seen that when the change trend S ₁ of the focus value in FIG. 4C is observed, when the number of steps of the lens movement reaches the foot of the curve FV1 or the left side of the curve FV2, it is near the peak P _{1 of the} characteristic curve SFV. When the peak P ₁ of the characteristic curve SFV is crossed, the slope of the characteristic curve SFV will change from a positive value to a negative value and continue to increase, and when the maximum value of the region is reached (ie, the maximum value is reached and then decreased) It is determined that the current focus position reaches the saddle point P _{2 of the} characteristic curve SFV. At this time, based on the symmetry characteristic of the characteristic curve SFV, the distance from the saddle point P ₂ to the trough P ₃ will approximate the distance d ₁ from the peak P ₁ to the saddle point P ₂ . Accordingly, the position of the valley P ₃ can be estimated by adding the distance d ₁ corresponding to the saddle point P ₂ to the distance d ₁ .

Similarly, by observing the change trend S ₂ of the focus value in FIG. 4C, when the number of steps of the lens movement reaches the foot of the curve FV1 or the curve FV2, it is about the peak P _{4 of the} characteristic curve SFV. When the peak P ₄ of the characteristic curve SFV is crossed, the slope of the characteristic curve SFV will change from a negative value to a positive value and continue to increase, and when the maximum value of the region is reached (ie, the maximum value is reached and then decreased) It is determined that the current focus position reaches the saddle point P _{5 of the} characteristic curve SFV. At this time, based on the symmetry characteristic of the characteristic curve SFV, the distance from the saddle point P ₅ to the trough P ₃ will approximate the distance d ₂ from the peak P ₄ to the saddle point P ₅ . Accordingly, the position of the valley P ₃ can be estimated by adding the distance d ₂ corresponding to the saddle point P ₅ to the distance d ₂ .

Returning to the flow of FIG. 3, finally, the lens control circuit 26 controls the lens 22 to move to the number of moving steps corresponding to the aforementioned trough, and then moves to the number of moving steps corresponding to the peak of the first curve or the second curve to complete the focusing (step S308). . In detail, the lens control circuit 26 controls, for example, the lens 22 to move to the number of moving steps corresponding to the trough at the first speed, and then moves to the number of moving steps corresponding to the peak of the first curve or the second curve at the second speed to Focusing is completed, wherein the second speed is less than the first speed. Wherein, the lens control circuit 26 can, for example, control the actuator of the lens 22 to move the lens 22 to the above-mentioned moving step or asymptotically move the lens 22 to the above-mentioned moving step, and then perform a small step search. And observe the change of the first curve or the second curve, and when the maximum value of the first or second curve is reached, the focus is completed.

By the above method, the number of steps of the second stage and the third stage of the part of the climbing mechanism can be omitted or reduced, thereby improving the focusing speed.

In summary, the image capturing device and the fast focusing method thereof of the present invention calculate the contrast value of different scale images and analyze the changing trend to estimate the region where the best focus position is located in the focusing process. Thus, the lens is controlled to move in one step or asymptotically to the vicinity of the best focus position, and finally a small step search is performed to complete the focus. Thereby, the focusing speed can be improved, and the image change caused by the movement of the lens back and forth can be prevented from affecting the user's shooting experience.

Although the present invention has been disclosed in the above embodiments, it is not intended to limit the present invention, and any one of ordinary skill in the art can make some changes and refinements without departing from the spirit and scope of the present invention. The scope of the invention is defined by the scope of the appended claims.

20‧‧‧Image capture device

22‧‧‧ lens

24‧‧‧Image Sensor

26‧‧‧Lens control circuit

28‧‧‧ Focusing circuit

FV1, FV2‧‧‧ focus value curve

SFV‧‧‧focus value ratio curve

P ₁ , P ₄ ‧ ‧ crest

P ₂ , P ₅ ‧‧‧ saddle point

P3‧‧‧ trough

S ₁ , S ₂ ‧ ‧ the change trend of the focus value

d ₁ , d ₂ ‧‧‧ distance

S302~S308‧‧‧ steps of a fast focusing method of an image capturing device according to an embodiment of the present invention

FIG. 1 is a schematic diagram of a conventional camera focus climbing mechanism. FIG. 2 is a block diagram of an image capture device according to an embodiment of the invention. FIG. 3 is a flow chart of a method of fast focusing of an image capturing device according to an embodiment of the invention. 4A-4C are diagrams illustrating an example of a fast focus method of an image capturing device according to an embodiment of the invention.

Claims (8)

A fast focusing method for an image capturing device, the image capturing device comprising a lens, an image sensor, a lens control circuit and a focusing circuit, the method comprising the steps of: controlling, by the lens control circuit, the lens movement to be sensed by the image The camera captures an image; the focus circuit calculates a first curve of the captured image at a first scale relative to a moving step of the lens and a focus value at the second scale relative to the lens a second curve of the number of moving steps, wherein a length and a width of the second dimension are fractions of a length and a width of the first dimension; and the focusing value of the second curve is divided by the first curve by the focusing circuit The focus value is obtained to obtain a characteristic curve, and the position of the trough is estimated according to the position of the peak and the saddle point appearing in the characteristic curve, wherein the position of the trough is estimated according to the position of the peak and the saddle point appearing in the characteristic curve. The step includes: calculating a change in a slope of the characteristic curve during the moving of the lens to capture the image; and when the slope transitions to zero, determining the arrival of the characteristic curve a peak; when the slope changes from zero to a maximum value, determining the saddle point reaching the characteristic curve, and calculating a distance between the wave step and the number of moving steps corresponding to the saddle point, the movement corresponding to the saddle point The step is forwardly added to estimate the position of the trough; and the lens control circuit controls the movement of the lens to the moving step corresponding to the trough And moving to the first curve or the peak of the second curve corresponding to the number of moving steps to complete the focus. The method of claim 1, wherein the focusing circuit calculates the first of the focus values of the image captured by the image sensor at the first scale relative to the number of steps of the lens The step of the curve includes: calculating, by the image contrast algorithm, a comparison value of the plurality of pixels in the image of the first scale; and calculating a sum of powers of the contrast values of the pixels in the focus frame in the image as the The focus value of the image at the first scale. The method of claim 1, wherein the focusing circuit calculates the second of the focus value of the image captured by the image sensor at the second scale relative to the number of steps of the lens The step of the curve includes: calculating, by the image contrast algorithm, a comparison value of the plurality of pixels in the image of the second scale; and calculating a sum of powers of the contrast values of the pixels in the focus frame in the image as the The focus value of the image at the second scale. The method of claim 1, wherein the lens control circuit controls the movement of the lens to the number of movement steps corresponding to the trough, and then moves to the movement corresponding to the peak of the first curve or the second curve. The step of completing the focusing includes: controlling, by the lens control circuit, the lens to move to the moving step corresponding to the trough at a first speed; Controlling, by the lens control circuit, the lens is moved to a second speed to the number of movement steps corresponding to the peak of the first curve or the second curve, wherein the second speed is less than the first speed. An image capturing device includes a lens; an image sensor coupled to the lens to capture an image; a lens control circuit coupled to the lens to control movement of the lens to capture the image; and a focusing circuit coupled to the image sensor And the lens control circuit, calculating a first curve of the focus value of the image captured by the image sensor at the first scale relative to the moving step of the lens, and the focus value at the second scale relative to the a second curve of the moving step of the lens, wherein a length and a width of the second dimension are a fraction of a length and a width of the first dimension, and the focus value of the second curve is divided by the focus of the first curve The value is obtained to obtain a characteristic curve, and the position of the trough is estimated according to the position of the peak and the saddle point appearing in the characteristic curve, wherein the focusing circuit includes calculating a slope of the characteristic curve during the movement of the lens to capture the image. a change, wherein when the slope transitions to zero, the peak reaching the characteristic curve is judged, and when the slope is changed from zero to a maximum value, the saddle point reaching the characteristic curve is judged, and The distance between the peak of the moving step corresponding to the saddle point, the moving step corresponding to the saddle point is forwardly added to estimate the position of the trough; the lens control circuit controls the lens to move to The number of moving steps corresponding to the valley estimated by the focusing circuit is further moved to the moving step corresponding to the peak of the first curve or the second curve to complete focusing. The image capturing device of claim 5, wherein the focusing circuit comprises: using a image contrast algorithm to calculate a contrast value of the plurality of pixels in the image of the first scale, and calculating the image in the focus frame The sum of the powers of the contrast values of the pixels is used as the focus value of the image at the first scale. The image capturing device of claim 5, wherein the focusing circuit comprises: using a image contrast algorithm to calculate a contrast value of the plurality of pixels in the image of the second scale, and calculating the image in the focus frame The sum of the powers of the contrast values of the pixels is used as the focus value of the image at the second scale. The image capturing device of claim 5, wherein the lens control circuit comprises: controlling the lens to move to the wave corresponding to the trough at a first speed, and then controlling the lens to move to the second speed The number of movement steps corresponding to the peak of the first curve or the second curve, wherein the second speed is less than the first speed.