JP2006267220A - Auto focus system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an auto focus (AF) system capable of setting a range for capturing a reference pattern image in pattern matching to an appropriate range by extracting the image of a target subject by the extraction of a difference between the photographing images of front and rear frames in an automatic follow-up function for following the target subject brought into focus by AF on an image plane by the pattern matching. <P>SOLUTION: A follow-up device 16 in a television camera or the like stores the image of the follow-up target subject as the reference pattern image, and detects the image matching with the reference pattern image in the newly captured photographing images of the frames by the pattern matching, thereby detecting the position of the target subject. The reference pattern image is successively updated, and the range of a reference pattern frame where the reference pattern image is newly captured in such a case is set to a range in accordance with the position and the size of the image of the target subject by the extraction of the difference between the photographing images of the front and rear frames. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an autofocus system, and more particularly to an autofocus system having an automatic tracking function for tracking a desired target subject as a focus target on a screen of a captured image of a camera.

  In an imaging system that converts an image of a subject into an electrical signal (image signal) by a photographing element (CCD or the like) like a television camera, a contrast method is generally adopted as an autofocus (hereinafter referred to as AF) method. Yes. In the contrast method, the contrast of the subject image is detected from the image signal captured by the image sensor, and the focus of the photographic lens is controlled so that the contrast becomes the highest, thereby automatically focusing on the best focus state (in-focus state). This is a method of matching.

  In AF such as a contrast method, the entire subject within the shooting range of the camera is not targeted for AF, but only the subject within a part of the shooting range is often targeted. In contrast-based AF, an image signal in a range targeted for AF is extracted from image signals captured by the entire imaging area of the image sensor, and focus is performed so that the contrast of the image signal in the extracted range is the highest. By controlling this, the target of AF is limited to only a certain range of subjects. In this specification, the range of the subject to be AF in the shooting range of the camera or the image range of the subject to be AF on the screen of the camera shot image is referred to as an AF area, and the AF area A frame indicating an outline is referred to as an AF frame.

Conventionally, when a single AF area is fixedly set at a predetermined position on the screen of a captured image, a rectangular AF area is normally set at the center of the screen. There is also known one in which the position of the area can be changed (for example, see Patent Document 1).
JP 2002-365519 A

  In the case where the AF area can be changed to a desired position as in Patent Document 1, when the target subject to be focused is an object that moves on the screen, the operator operates an operation device such as a trackball or a joystick. To move the position of the AF area in accordance with the movement of the target subject. For this reason, there has been a problem that the operation is troublesome and requires attention.

  Therefore, if the AF area is automatically tracked with respect to the moving subject, it is effective because the operator does not have to operate the position of the AF area. For example, a so-called pattern matching technique is known as a method for detecting a specific reference pattern image from a source image, and an image of a specific target subject can be detected from a captured image by applying the technique. The AF area can be tracked to the target subject by moving the AF area to the position of the detected target subject.

  On the other hand, in pattern matching, if the shape or size of the target subject on the screen of the captured image (video) changes, it will not match the reference pattern image stored in advance as the target subject image. It becomes difficult to detect an image. Therefore, if the reference pattern image is updated with the target subject image when the target subject image is detected by pattern matching, a large difference occurs between the reference pattern image and the target subject image at the next detection. Therefore, it is possible to appropriately detect the image of the target subject.

  However, even when the reference pattern image is updated, if the size of the image captured from the captured image as the reference pattern image is constant, for example, when the target subject moves far away or the camera zoom is set to the wide side. When the image of the target subject on the screen becomes small as in the case of changing to the reference pattern image, many background images other than the target subject are included in the reference pattern image. If an image other than the target subject is included in the reference pattern image, there arises a problem that it becomes difficult to detect the image of the target subject.

  The present invention has been made in view of such circumstances. In an automatic tracking function for automatically tracking a target subject to be focused by AF on a screen, pattern matching is performed in consideration of the position and size of the image of the target subject. It is an object of the present invention to provide an autofocus system capable of automatically setting a reference pattern image as an appropriate image.

  In order to achieve the above object, an autofocus system according to claim 1 is configured such that an image capturing unit that captures an image of a subject formed by an optical system and an image of a target subject to be tracked as a target to be focused are a reference pattern image. And a reference pattern image storage means stored as an image, and an image matching the reference pattern image stored by the reference pattern image storage means on the screen of the photographed image obtained by the imaging means is detected, and the detected image A reference pattern image position detecting means for detecting a position on the screen, an autofocus means for adjusting the focus of the optical system so as to focus on the subject at the position detected by the reference pattern image detecting means, and the imaging Difference image generation means for extracting a difference between two captured images sequentially obtained by the means and generating a difference image; Based on the difference image generated by the difference image generation means, a reference pattern image capture range setting means for setting a reference pattern image capture range for capturing a reference pattern image from a captured image obtained by the imaging means; and Reference pattern image capturing means for capturing, as a reference pattern image, an image of the reference pattern image capturing range set by the reference pattern image capturing range setting means from the captured image obtained by the imaging means, and the reference pattern image capturing And a reference pattern image setting means for setting the reference pattern image captured by the means as a reference pattern image stored by the reference pattern image storage means.

  According to the present invention, when a target object to be tracked as a focus target is a moving object, an image of only the moving object is extracted by extracting a difference between two captured images sequentially obtained by the imaging unit. . As a result, the position and size of the target subject on the screen can be known, and the position and size of the target subject are set by setting an image capture range to be stored as a reference pattern image based on the position and size. An appropriate reference pattern image corresponding to can be set.

  According to a second aspect of the present invention, in the invention according to the first aspect, the reference pattern image setting means uses the reference pattern image stored in the reference pattern image storage means as the reference pattern image capture means. It is a means to update with the reference pattern image taken in by (1).

  The present invention shows a mode in which means relating to setting of the reference pattern image of claim 1 is applied to update the reference pattern image in order.

  According to a third aspect of the present invention, in the invention according to the first or second aspect, the reference pattern image setting means stores the initial reference pattern image stored in the reference pattern image storage means as the reference pattern image capture. It is characterized in that it is set by a reference pattern image taken in by the loading means.

  The present invention shows an aspect in which means for setting a reference pattern image according to claim 1 is applied when setting an initial reference pattern image.

  According to a fourth aspect of the present invention, there is provided the autofocus system according to the first, second, or third aspect, wherein the binarization unit that binarizes the difference image generated by the difference image generation unit; For each pixel of the difference image binarized by the binarization means, a distribution of cumulative values obtained by accumulating the pixel values of each pixel in the horizontal direction and a distribution of cumulative values obtained by accumulating the pixel values of each pixel in the vertical direction are calculated. And a cumulative value calculating means, wherein the reference pattern image capturing range is set based on a distribution of cumulative values calculated by the cumulative value calculating means.

  The present invention shows an aspect in which the reference pattern image capture range is set based on the cumulative value distribution of the horizontal and vertical pixel values in the difference image.

  An autofocus system according to a fifth aspect of the present invention is the invention according to the first, second, third, or fourth aspect, wherein the autofocus means is a contrast type autofocus and is obtained by the imaging means. The focus adjustment is performed based on the contrast of an image within a predetermined AF area of the photographed image, and the position of the AF area is set to a position detected by the reference pattern image position detecting means. .

  The present invention shows a case where the autofocus means is a contrast method, and shows a mode in which the target subject is automatically tracked as a target to be focused by setting the position of the AF area that is the target range of AF as the target subject position. Yes.

  According to the autofocus system of the present invention, the reference pattern image in pattern matching can be automatically set to an appropriate image in consideration of the position and size of the image of the target subject in the automatic tracking function.

  Hereinafter, preferred embodiments of an autofocus system according to the present invention will be described in detail with reference to the accompanying drawings.

  FIG. 1 is a block diagram showing the overall configuration of an imaging system to which an autofocus system according to the present invention is applied. The image pickup system shown in FIG. 1 is an image pickup system used for shooting with, for example, a broadcast television camera. In the figure, a camera head 10 with a replaceable lens, and a shooting lens (optical) mounted on a lens mount of the camera head 10 A lens device 12, a frame operation unit 14, a tracking device 16 and the like provided with a system) are shown.

  The camera head 10 is equipped with an image sensor (for example, a CCD), a required signal processing circuit, and the like, and an image formed by the photographing lens of the lens device 12 is photoelectrically converted by the image sensor and then subjected to signal processing. Necessary signal processing is performed by the circuit. A video signal in a predetermined format generated by the signal processing circuit is output from the video signal output terminal of the camera head 10 to an external device. Further, the camera head 10 is provided with a view finder (monitor) 18 for composition confirmation, etc., and a video signal from the camera head 10 is given to the view finder 18 and is photographed by the camera head 10. A real-time image (video) is displayed on the screen of the viewfinder 18. Further, information on an AF area that is a target range of autofocus (AF), for example, an AF frame, is also displayed on the screen of the viewfinder 18.

  The lens device 12 includes a photographing lens (optical system) (not shown) attached to a lens mount of the camera head 10, and an image of a subject is formed on the imaging surface of the imaging element of the camera head 10 by the photographing lens. . The photographing lens is provided with movable parts for adjusting photographing conditions such as a focus lens group, a zoom lens group, and a diaphragm as its constituent parts. These movable parts are electrically driven by a motor (servo mechanism) (not shown). It is designed to be driven. For example, the focus lens group and zoom lens group move in the optical axis direction, and the focus (subject distance) adjustment is performed by moving the focus lens group, and the focal length (zoom magnification) adjustment is performed by moving the zoom lens group. Is done. In the system relating to autofocus as in the present embodiment, it is sufficient that at least the focus lens group can be driven electrically, and other movable parts may be driven only manually. Further, when the predetermined movable part is electrically driven in accordance with the operation of the operator, it is based on a control signal output in accordance with the operation of the operator from an operation unit (not shown) (such as an operation unit of a controller connected to the lens device 12). Thus, the operation of the movable part is controlled, but details are omitted.

  As shown in the figure, the lens device 12 is equipped with a lens CPU 20 that performs overall control of the lens device 12, an AF processing unit 22 that executes autofocus (AF) processing, an AF imaging circuit 24, and the like. Yes. The AF image pickup circuit 24 is disposed in the lens device 12 to acquire a video signal for AF processing, and outputs an image pickup device (CCD or the like) or an output signal of the image pickup device as a video signal of a predetermined format. Etc. The image pickup element of the AF image pickup circuit 24 is referred to as an AF image pickup element. The video signal output from the AF imaging circuit 24 is a luminance signal.

  On the imaging surface of the AF imaging element, subject light branched from the subject light incident on the imaging element of the camera head 10 is imaged by a half mirror or the like disposed on the optical path of the photographing lens. The shooting range and subject distance (distance of the subject in focus) with respect to the imaging area of the AF imaging element are configured to match the shooting range and subject distance with respect to the imaging area of the imaging element of the camera head 10 and are used for AF. The captured image captured by the image sensor matches the captured image captured by the image sensor of the camera head 10. Note that it is not necessary for the shooting ranges of the two to be completely the same. For example, the shooting range of the AF imaging device may be a larger range including the shooting range of the imaging device of the camera head 10. Further, the image signal from the camera head 10 may be supplied to the AF processing unit 22 without providing an AF image sensor.

  The AF processing unit 22 acquires a video signal from the AF imaging circuit 24, and calculates a focus evaluation value indicating the level of contrast of the subject image based on the video signal. For example, after extracting a high-frequency component signal of a video signal obtained from an AF image sensor with a high-pass filter, a signal in a range corresponding to an AF area that is an AF target range among the high-frequency component signal Are accumulated one screen at a time. The integrated value obtained for each screen in this manner indicates the level of contrast of the subject image, and is given to the lens CPU 20 as a focus evaluation value. The AF area range is designated by the lens CPU 20 as will be described later.

  As will be described in detail later, the lens CPU 20 acquires AF frame information (AF frame information) indicating the range (contour) of the AF area from the frame operation unit 14, and the range within the AF frame designated by the AF frame information. Is designated to the AF processing unit 22 as an AF area. Then, the focus evaluation value obtained from the image (video signal) in the AF area is acquired from the AF processing unit 22, and the acquired focus evaluation value is maximum (maximum), that is, the contrast of the subject image in the AF area is maximum. The focus lens group is controlled so that For example, a hill-climbing method is generally known as a control method for a focus lens group based on a focus evaluation value, and the focus evaluation value starts to decrease by moving the focus lens group in a direction in which the focus evaluation value increases. Is detected, the focus lens group is set at that position. Thereby, the subject in the AF frame is automatically focused.

  The frame operation unit 14 includes an operation member for the operator to specify the setting contents of the AF frame such as the position, size, and shape of the AF frame. As shown in FIG. 2, the AF frame 40 shows the outline of the AF area 42 that is the AF target range with respect to the shooting range or the shot image (screen) of the imaging element of the camera head 10. When the operation member is operated by the operator, the frame operation unit 14 sets and changes setting contents such as the position of the AF frame in accordance with the operation. The setting contents such as the position of the AF frame by the operation member of the frame operation unit 14 are set by the change amount corresponding to the operation amount of the subsequent operation member based on the setting contents of the AF frame before the operation. This is done by making changes. For example, the position of the AF frame is displaced by the amount corresponding to the direction and amount of movement of the trackball. Also, if the point that determines the position of the AF frame (the center position of the rectangular AF frame in the present embodiment) within the shooting range (screen of the captured image) is referred to as an AF point, the position of the AF frame is the position of the AF point. Is determined by setting

  On the other hand, when the operation member of the frame operation unit 14 is not operated and the operator is not instructed to change the setting contents of the AF frame, the setting contents of the AF frame are set by the AF frame information given from the tracking device 16 described later. Set and change. The frame operation unit 14 transmits AF frame information indicating the setting contents of the AF frame set or changed by the operation of the operator or the AF frame information from the tracking device 16 to the lens CPU 20 in accordance with a request from the lens CPU 20. As a result, the AF target range is set to the AF frame range set and changed in the frame operation unit 14.

  The tracking device 16 detects the position of the target object (target subject) specified by the operator on the screen of the captured image and adjusts the position of the AF area (AF frame) according to the movement of the target object on the screen. Is a device for realizing an automatic tracking function for automatically changing (automatic tracking), and when the operator does not perform manual operation for changing the AF frame, the tracking device 16 automatically tracks the AF frame. If the operator is performing a manual operation for changing the AF frame, the manual operation has priority over the tracking device 16 and the AF frame is changed according to the manual operation of the operator. It is also possible to turn off the automatic tracking function and enable only manual operation.

  The frame operation unit 14 also transmits AF frame information indicating the setting contents of the AF frame set and changed as described above to the camera head 10, and places the AF frame at a corresponding position on the screen of the viewfinder 18. Display. As a result, the operator can recognize the position of the AF frame at that time while looking at the viewfinder 18.

  The tracking device 16 includes an image processing unit 26, an image input unit 28, a serial communication interface 30, and the like. The image input unit 28 acquires the video signal (luminance signal) obtained by the AF imaging circuit 24 via the AF processing unit 22 and obtains the image obtained from the video signal in response to a request from the image processing unit 26. The image data of the image is given to the image processing unit 26.

  The image processing unit 26 detects the position on the screen of the target object designated by the operator based on the captured image acquired from the image input unit 28 described later in detail by pattern matching processing. Using the position of the object as the position of the AF frame (AF point), AF frame information indicating the position of the AF frame is transmitted to the frame operation unit 14 via the serial communication interface 30.

  Next, the tracking process in the image processing unit 26 of the tracking device 16 will be described in detail with reference to the flowchart of FIG. First, the operator operates the operation member of the frame operation unit 14 while viewing the image displayed on the screen of the viewfinder 18 and the AF frame, and the whole or a part of the object to be tracked as the focus target is displayed in the AF frame. To pay. As a result, the object is brought into focus by AF processing in the lens CPU 20.

  Subsequently, the operator presses the determination button of the frame operation unit 14. As a result, a reference pattern image setting instruction (reference pattern image storage instruction) is given from the frame operation unit 14 to the image processing unit 26.

  When an instruction to set a reference pattern image is given, the image processing unit 26 reads AF frame information indicating the setting contents of the AF frame at that time from the frame operation unit 14 via the serial communication interface 30 and also the lens device 12. The captured image (image data) for one screen (one frame) is taken in from the video signal supplied from the AF processing unit 22 to the image input unit 28 (step S10). The captured image captured at this time is referred to as an A image. Further, when the video signal given from the AF processing unit 22 to the image input unit 28 is, for example, an interlaced video signal composed of two fields of video signals for one screen (one frame). The captured image captured as one screen from the input unit 28 may be a captured image obtained from a video signal for one field, or may be a captured image obtained from a video signal for two fields. In the case of the non-interlace method, the captured image is obtained from a video signal for one frame.

  Subsequently, the image processing unit 26 extracts the image within the range from the A image by using the range of the AF frame designated by the AF frame information as the range of the reference pattern frame from the A image captured in step S10, and the reference pattern. It is set (stored) as an image (step S12).

  Here, the reference pattern image is image data used in a pattern matching process to be described later, and in this embodiment, the range of the reference pattern frame into which the initial reference pattern image is captured matches the range of the AF frame. To do. However, the initial reference pattern frame may be a range having a size (for example, slightly larger) different from the AF frame. In addition, the reference pattern frame may be displayed on the viewfinder 18 separately from the AF frame, and the range of the reference pattern frame may be set by an operation different from the AF frame.

  When the setting of the reference pattern image in step S12 is completed, the image processing unit 26 repeatedly executes the following processes of step S14 to step S32. First, a new captured image for one screen is taken from the image input unit 28 (step S14). Note that the captured image acquired in step S14 is referred to as a B image.

  Next, the image processing unit 26 executes a well-known pattern matching process to detect an image range (position) that matches the reference pattern image within the range of the B image (step S16).

  Subsequently, the image processing unit 26 determines whether or not the object has moved on the screen based on the position detected by the pattern matching process (step S18). If YES is determined, the position (AF point) of the AF area is displaced in the same direction as the moving direction by the amount of movement of the object, and AF frame information indicating the position of the AF area is displayed on the frame operation unit 14. Send to. Thereby, the setting content of the AF frame, that is, the position of the AF frame is updated (step S20). When the position of the AF frame is updated, the AF target range in the lens device 12 is updated accordingly, and the position of the AF frame displayed on the screen of the viewfinder 18 is updated. If it is determined as NO in step S18, the process of step S20 is not executed.

  Next, the image processing unit 26 generates a difference image C that is extracted based on the difference between the A image and the B image (step S22). That is, a difference between pixel values (luminance values) between corresponding pixels of the A image and the B image is obtained, and a difference image C having the absolute value of the difference as the pixel value of each pixel is generated (C = | A−B | ). For example, assume that images as shown in FIGS. 4A and 4B are obtained as an A image and a B image, respectively. In these drawings, a moving object M moving on the screen is simply shown as a target to be tracked by a circular image with a constant luminance, and a background image that does not change on the screen is omitted. For these A and B images, a difference image C is generated as an image as shown in FIG. As shown in FIG. 6C, the background image that does not change is removed from the difference image C, and only the image of the portion where the image of the moving object M in the A image and the B image does not overlap is extracted.

  Next, the image processing unit 26 sets the difference image C to 1 if the pixel value of each pixel of the difference image C generated in step S22 is equal to or greater than a predetermined threshold value, and 0 if it is less than the threshold value. Binarization is performed (step S24).

  Next, the image processing unit 26 newly takes a captured image for one screen from the image input unit 28, and updates the A image using the image as a new A image (step S26). The A image is used to generate a difference image C (step S22) in the next iterative process with respect to the iterative process of steps S14 to S32. Subsequently, the image processing unit 26 executes a binary cumulative value projection process on the binarized difference image C (step S28).

  Here, the binary cumulative value projection process will be described. The binary cumulative value projection is a horizontal cumulative value obtained by accumulating (integrating) the pixel values of the pixels arranged on the same line in the horizontal direction (lateral direction) with respect to each pixel of the binarized difference image C; The vertical direction cumulative value obtained by accumulating the pixel values of each pixel arranged on the same line in the vertical direction (vertical direction) is obtained, and the horizontal direction cumulative value and the vertical direction cumulative value of each line are assumed on the screen. Projected onto the Y axis (horizontal axis) and the X axis (horizontal axis). Thus, a binary cumulative value projection distribution indicating the distribution of the horizontal direction cumulative value and the vertical direction cumulative value is created on each of the Y axis and the X axis. For example, when binary cumulative projection processing is performed on the difference image C generated as shown in FIG. 4C, a result as shown in FIG. 5 is obtained. In FIG. 5, vertical accumulated values obtained by accumulating pixel values of pixels having the same X coordinate value (pixels arranged on the same line in the vertical direction) are projected on each coordinate point (X coordinate value) of the X axis. A horizontal accumulated value obtained by accumulating pixel values of pixels (pixels on a horizontal line) having the same Y coordinate value is projected on each coordinate point (Y coordinate value) of the Y axis.

  Next, the image processing unit 26 updates the reference pattern frame based on the result of the binary cumulative value projection onto the X axis and Y axis in step S28 (step S30). For example, a range that is 1 to 99 percent of the total sum (area) of the vertical cumulative values projected on the X-axis (a range that excludes a range that is 1 percent of the area from the left and right ends of the screen, respectively) ) Is the width of the reference pattern frame, and a range of 1 to 99 percent of the total sum (area) of the horizontal direction values projected on the Y-axis (values of 1 percent of the area from the top and bottom of the screen, respectively) A rectangular reference pattern frame is set with the vertical width of the reference pattern frame as a range excluding the range that becomes. For example, if a binary cumulative value projection distribution as shown in FIG. 5 is obtained, a reference pattern frame 44 as shown in FIG. 6 is set.

  When the reference pattern frame is set (updated) in this way, the image processing unit 26 then uses the newly set reference pattern frame range as the reference pattern image capture range from the B image to within the reference pattern frame. And the reference pattern image is updated with the image (step S32). And it returns to step S14 and repeats the process from step S14.

  With the above processing, the reference pattern image is updated to an appropriate image according to the position and size of the target object to be tracked as the focus target, and the target object is appropriately tracked by pattern matching. Become.

  In the above processing, the size of the AF area (AF frame) is set to a size unrelated to the update of the reference pattern frame. However, in conjunction with the size of the reference pattern frame (for example, The size of the AF area may also be changed (to match the size of the reference pattern frame).

  Further, in the above description, the case where the object moves left and right on the screen is illustrated, but this processing only changes the size of the image of the object, for example, when the object moves back and forth. It is also effective in cases.

  Further, when the reference pattern frame is set based on the binary cumulative value projection, in the above process, the range of the reference pattern frame is 1 to 99 percent for each of the sum of the horizontal direction cumulative values and the sum of the vertical direction cumulative values. The range is not limited to this. For example, other percentage ranges may be used.

  Next, when the initial reference pattern image is set in step S12 in FIG. 3, the reference pattern frame for capturing the reference pattern image is automatically set to an appropriate range in the same manner as the reference pattern frame update process in FIG. The case where it is made possible will be described with reference to the flowchart of FIG.

  As described with reference to the flowchart of FIG. 3, when an instruction for setting a reference pattern image is given, the image processing unit 26 captures a captured image (A image) for one screen from the image input unit 28 (step S10 in FIG. 3). ). And it transfers to step S12 of FIG. 3, and starts the process of the flowchart of FIG. 7 as the process of the step S12.

  When the process proceeds to the processing of the flowchart of FIG. 7, the image processing unit 26 newly takes a captured image for one screen as a B image from the image input unit 28 (step S <b> 50). Subsequently, the image processing unit 26 generates a difference image C extracted based on the difference between the A image and the B image, similarly to step S22 of FIG. 3 (step S52). Subsequently, the difference image C is binarized as in step S24 of FIG. 3 (step S54), and the area of the binarized difference image C is obtained (step S56). The obtained area is represented by a variable AREA. Next, the image processing unit 26 determines whether or not the object has moved according to whether or not the area AREA has exceeded a predetermined threshold (for example, 500) (step S58). While it is determined NO in step S58, the processes in steps S50 to S58 are repeatedly executed.

  If YES is determined in step S58, the image processing unit 26 performs a binary cumulative value projection process on the binarized difference image C as in step S28 of FIG. 3 (step S60). Subsequently, the reference pattern frame is set based on the result of the binary cumulative value projection onto the X-axis and Y-axis as in step S30 of FIG. 3 (step S62). When the reference pattern frame is set, an image in the reference pattern frame is taken from the B image, and a reference pattern image is set based on the image (step S64). When the process of step S64 ends, the process proceeds to step S14 of FIG.

  Through the above processing, the initial reference pattern image is set with an image in an appropriate range according to the position and size of the object to be tracked as the focus target.

  Note that the initial setting process of the reference pattern image shown in FIG. 7 can be applied even when the reference pattern image is not updated as shown in FIG.

  As described above, in the above embodiment, the lens device 12, the frame operation unit 14, and the tracking device 16 are illustrated as separate devices, but any two or all of them may be integrated devices. In the above embodiment, an operation member (for example, a determination switch for determining an object) related to the processing of the tracking device 16 is also provided in the frame operation unit 14, but may be provided in the tracking device 16.

  In the above-described embodiment, the image signal is acquired for the AF by using an image sensor dedicated for AF that is different from the image sensor for the camera head 10, but the image obtained by the image sensor for the camera head 10 for AF is used. A signal may be used. However, if the AF video signal is obtained from the AF-dedicated imaging device as in the above embodiment, the HD signal is obtained when the camera head 10 is compatible with a high-definition (HD) television system. Since it is possible to perform AF without using a lens, it is advantageous for miniaturization and power saving.

  In the above embodiment, the contrast type AF is used. However, the present invention can be applied even when an AF other than the contrast type is used.

FIG. 1 is a block diagram showing the overall configuration of an imaging system to which an autofocus system according to the present invention is applied. FIG. 2 is a diagram illustrating an AF frame. FIG. 3 is a flowchart showing the procedure of the tracking process in the tracking device. FIG. 4 is an explanatory diagram used for explaining the difference image. FIG. 5 is an explanatory diagram used to explain the binary cumulative value projection process. FIG. 6 is an explanatory diagram used for explaining the setting (updating) of the reference pattern frame. FIG. 7 is a flowchart showing an aspect of processing related to setting of an initial reference pattern image.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Camera head, 12 ... Lens apparatus, 14 ... Frame operation part, 16 ... Tracking apparatus, 18 ... Viewfinder, 20 ... Lens CPU, 22 ... AF process part, 24 ... Imaging circuit for AF, 26 ... Image process part, 28 ... Image input unit, 30 ... Serial communication interface, 40 ... AF frame, 42 ... AF area, 44 ... Reference pattern frame

Claims (5)

Imaging means for imaging a subject image formed by an optical system;
Reference pattern image storage means for storing an image of a subject to be tracked as a target to be focused as a reference pattern image;
A reference pattern that detects an image that matches the reference pattern image stored in the reference pattern image storage means on the screen of the captured image obtained by the imaging means, and detects the position of the detected image on the screen. Image position detecting means;
Auto focus means for adjusting the focus of the optical system so that the subject at the position detected by the reference pattern image detection means is in focus;
Difference image generating means for extracting a difference between two captured images sequentially obtained by the imaging means and generating a difference image;
Based on the difference image generated by the difference image generation means, a reference pattern image capture range setting means for setting a reference pattern image capture range for capturing a reference pattern image from a captured image obtained by the imaging means;
A reference pattern image capturing unit that captures, as a reference pattern image, an image of the reference pattern image capturing range set by the reference pattern image capturing range setting unit from the captured image obtained by the imaging unit;
A reference pattern image setting means for setting a reference pattern image captured by the reference pattern image capturing means as a reference pattern image stored by the reference pattern image storage means;
An autofocus system characterized by comprising   The reference pattern image setting means is means for updating the reference pattern image stored by the reference pattern image storage means with the reference pattern image captured by the reference pattern image capture means. 1 autofocus system.   3. The reference pattern image setting means sets an initial reference pattern image stored by the reference pattern image storage means based on a reference pattern image captured by the reference pattern image capture means. Autofocus system.   The reference pattern image capturing range setting unit is configured to binarize the difference image generated by the difference image generation unit and each pixel of the difference image binarized by the binarization unit. A cumulative value calculating means for calculating a cumulative value distribution in which the pixel values of each pixel are accumulated in the horizontal direction and a cumulative value distribution in which the pixel values of each pixel are accumulated in the vertical direction; 4. The autofocus system according to claim 1, wherein the reference pattern image capture range is set based on the distribution of the accumulated values.   The autofocus means is contrast-type autofocus, and performs focus adjustment based on the contrast of an image within a predetermined AF area of the captured image obtained by the imaging means, and the reference pattern image 5. The autofocus system according to claim 1, wherein the position of the AF area is set at a position detected by a position detecting means.

Images