JP2014063275A - Image specifying device, image specifying method, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor capable of reducing a processing load for specifying an image obtained by capturing a predetermined form in the action of a desired object.SOLUTION: An image processor 1 comprises: an acquisition part 41; a setting part 48; a calculation part 49; and a specifying part 50. The acquisition part 41 acquires a plurality of image data obtained by consecutively capturing a series of actions of an object. The setting part 48 sets an evaluation area corresponding to the size and position of the object to each of the plurality of image data acquired by the acquisition part 41. The calculation part 49 calculates an evaluation value between the plurality of image data in the evaluation areas set by the setting part 48, respectively. The specifying part 50 specifies the image data of a predetermined state in a series of actions, based on the evaluation values calculated by the calculation part 49, respectively.

Description

  The present invention relates to an image specifying device, an image specifying method, and a program for specifying an image.

In recent years, a user can check his / her own swing form by looking at an image in which a swing, which is a series of actions in golf, is captured.
As a device for imaging such a golf swing form, a frame related to a series of swing motions performed by the subject is extracted as swing data from the captured subject's video, and each frame and reference image in the swing data is extracted. There has been proposed an apparatus for extracting a frame corresponding to each specific operation point during a swing motion according to the analysis result of the moving partial image in the difference image compared with (see Patent Document 1).

JP 2003-1117045 A

  However, in Patent Document 1, an image corresponding to a specific operation point of the swing form is extracted by calculating a difference between the entire reference image and the entire frame image related to the swing motion, which increases the amount of calculation. Therefore, there is a possibility that the processing load for specifying the specific operation point of the swing form becomes heavy.

  The present invention has been made in view of such a situation, and an object of the present invention is to reduce the processing burden of specifying an image in a predetermined state from images obtained by continuously capturing a series of movements of a subject.

In order to achieve the above object, an image processing apparatus according to an aspect of the present invention includes:
Acquisition means for acquiring a plurality of image data obtained by continuously imaging a series of movements of a subject;
Setting means for setting an evaluation area corresponding to the size and position of the subject in each of a plurality of image data acquired by the acquisition means;
First calculation means for calculating an evaluation value between the plurality of image data in the evaluation region set by the setting means;
Identification means for identifying image data in a predetermined state in the series of operations based on the evaluation values calculated by the first calculation means;
It is provided with.

  According to the present invention, it is possible to reduce the processing burden of specifying an image in a predetermined state from images obtained by continuously capturing a series of actions of a subject.

1 is a block diagram showing a hardware configuration of an image processing apparatus as an embodiment of an image processing apparatus according to the present invention. It is a figure explaining the outline | summary of the image specific process which the image processing apparatus of FIG. 1 performs. It is a functional block diagram which shows the functional structure for the image processing apparatus of FIG. 1 to perform an image specific process. It is a figure which shows an example of the setting screen which the image processing apparatus of FIG. 1 displays. It is a figure which shows an example of the learning sample for constructing | assembling an address detection discriminator. It is a figure explaining the functional structure which sets the difference calculation area | region in a setting part. It is a figure which shows an example of distribution of the difference evaluation value in the difference calculation area. 4 is a flowchart for explaining a flow of image specifying processing executed by the image processing apparatus of FIG. 3. 4 is a flowchart for explaining a flow of a second reference point setting process executed by the image processing apparatus of FIG. 3. 4 is a flowchart illustrating a flow of a first reference point setting process executed by the image processing apparatus of FIG. 3. 4 is a flowchart illustrating a flow of a difference calculation area 3 setting process executed by the image processing apparatus of FIG. 3.

Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
FIG. 1 is a block diagram showing a hardware configuration of an image processing apparatus 1 as an embodiment of an image processing apparatus according to the present invention. The image processing apparatus 1 can be configured by a digital camera, for example.

  The image processing apparatus 1 includes a CPU (Central Processing Unit) 11, a ROM (Read Only Memory) 12, a RAM (Random Access Memory) 13, an image processing unit 14, a bus 15, an input / output interface 16, and an imaging. A unit 17, an operation unit 18, a display unit 19, a storage unit 20, a communication unit 21, and a drive 22 are provided.

  The CPU 11 executes various processes according to a program recorded in the ROM 12 or a program loaded from the storage unit 20 to the RAM 13. The RAM 13 also stores data necessary for the CPU 11 to execute various processes as appropriate.

The image processing unit 14 is configured by a DSP (Digital Signal Processor), a VRAM (Video Random Access Memory), and the like, and performs various image processing on image data in cooperation with the CPU 11.
Here, in this embodiment, the processing unit of the image by the image processing apparatus 1 is one still image, and the still image as such a processing unit is referred to as a “frame image” in this specification. . That is, in the present embodiment, the moving image is composed of a plurality of frame images, and processing is executed for each frame image. Therefore, hereinafter, unless otherwise specified, an image means a frame image and an image data means frame image data.
For example, the image processing unit 14 performs image processing such as noise reduction, white balance, and camera shake correction on image data output from the imaging unit 17 described later.

  The CPU 11, ROM 12, RAM 13, and image processing unit 14 are connected to each other via a bus 15. An input / output interface 16 is also connected to the bus 15. An imaging unit 17, an operation unit 18, a display unit 19, a storage unit 20, a communication unit 21, and a drive 22 are connected to the input / output interface 16.

Although not shown, the imaging unit 17 includes an optical lens unit and an image sensor.
The optical lens unit is configured by a lens that collects light, for example, a focus lens or a zoom lens, in order to photograph a subject.
The focus lens is a lens that forms a subject image on the light receiving surface of the image sensor. The zoom lens is a lens that freely changes the focal length within a certain range.
The optical lens unit is also provided with a peripheral circuit for adjusting setting parameters such as focus, exposure, and white balance as necessary.

The image sensor includes a photoelectric conversion element, AFE (Analog Front End), and the like.
The photoelectric conversion element is composed of, for example, a CMOS (Complementary Metal Oxide Semiconductor) type photoelectric conversion element or the like. A subject image is incident on the photoelectric conversion element from the optical lens unit. Therefore, the photoelectric conversion element photoelectrically converts (captures) the subject image, accumulates the image signal for a predetermined time, and sequentially supplies the accumulated image signal as an analog signal to the AFE.
The AFE performs various signal processing such as A / D (Analog / Digital) conversion processing on the analog image signal. A digital signal is generated by various signal processing and output as an output signal of the imaging unit 17.
Hereinafter, the output signal of the imaging unit 17 is referred to as “captured image data”. Therefore, image data is output from the imaging unit 17 and is appropriately supplied to the CPU 11, the image processing unit 14, and the like.

The operation unit 18 includes various buttons and the like, and accepts user instruction operations.
The display unit 19 is composed of a liquid crystal display or the like and displays various images.
The storage unit 20 is configured by a DRAM (Dynamic Random Access Memory) or the like, and temporarily stores image data output from the image processing unit 14 or the like. The storage unit 20 also stores various data necessary for processing by the image processing unit 14 and the like.
The communication unit 21 controls communication with other devices (not shown) via a network including the Internet.

  A removable medium 31 made of a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is appropriately attached to the drive 22. The program read from the removable medium 31 by the drive 22 is installed in the storage unit 20 as necessary. The removable medium 31 can also store various data such as image data stored in the storage unit 20 in the same manner as the storage unit 20.

The image processing apparatus 1 having such a configuration starts live view imaging processing and live view display processing before and after imaging.
That is, the image processing apparatus 1 continues the imaging operation by the imaging unit 17, and during that time, image data sequentially output from the image processing unit 14 in units of frame images via the imaging unit 17 is stored in a memory (this embodiment). Then, it is temporarily stored in the storage unit 20). Such a series of processes is a “live view imaging process” referred to herein.
In addition, the image processing apparatus 1 sequentially reads out each image data temporarily recorded in units of frame images in the memory (storage unit 20 in the present embodiment) at the time of live view imaging, and displays a corresponding frame image. Displayed sequentially on the unit 19. Such a series of control processes is the “live view display process” referred to herein. The frame image displayed on the display unit 19 by the live view display process is hereinafter referred to as “live view image”.

The user can instruct to record image data by determining the composition while viewing the live view image and then pressing a shutter switch (not shown) of the operation unit 18 to the lower limit. The operation of pressing the shutter switch to the lower limit in this way is hereinafter referred to as “full press operation” or simply “full press”.
Hereinafter, in order to distinguish from imaging for live view imaging processing and live view display processing, imaging when a recording instruction is issued by a full-press operation is referred to as “recording imaging”.

  When a recording instruction is given by a full-press operation, the image processing apparatus 1 records a plurality of continuous image data obtained by capturing a series of movements of a subject in the storage unit 20 or the removable medium 31 as a recording image. Here, in the present embodiment, “a plurality of continuous image data” is image data of a moving image or a continuously shot still image.

FIG. 2 is a diagram for explaining an outline of the image specifying process executed by the image processing apparatus 1 of FIG.
The image specifying process is a process of specifying image data in a predetermined state in a series of operations from a plurality of image data obtained by continuously capturing a series of operations of a subject.
Specifically, the image specifying process is performed by, for example, halfway back as a predetermined state from a plurality of pieces of image data obtained by continuously capturing a series of operations in a swing of a user's golf club as a subject. (Back swing with the club parallel to the ground), Top (back swing with the club head at the top), Follow-through (after hitting the ball, the club parallel to the ground), etc. It is the process which specifies the image which imaged this state.
The image specifying process can also specify image data obtained by capturing a halfway down (a state in which the club is parallel to the ground by a downswing).

The image processing apparatus 1 acquires, from the storage unit 20 or the removable medium 31, a plurality of image data obtained by continuously capturing a series of actions (golf club swing) of a user who is a subject.
Then, the image processing apparatus 1 sets a plurality of difference calculation area groups corresponding to the subject in each of the acquired plurality of image data. In the present embodiment, the image processing apparatus 1 sets the difference calculation area groups 1, 2, and 3.

Specifically, the image processing apparatus 1 is based on, for example, the position of the user's head in image data obtained by imaging the user's address (a state in which the user stands in front of the ball and the golf club head is attached to the side of the ball). The size and position of the user are determined from the point A and the point B based on the position of the ball.
Then, the image processing apparatus 1 sets the difference calculation area groups 1, 2, and 3 according to the determined size and position of the user.

The difference calculation area group 1 includes a plurality of difference calculation areas (SWb1, SWb2, SWb3,..., SWbn) set in order to specify image data in which the halfway back state is captured. Note that the difference calculation area group 1 may be used to specify image data obtained by capturing a halfway down.
The difference calculation area group 2 includes a plurality of difference calculation areas (SWf1, SWf2, SWf3,..., SWfn) set in order to specify image data in which the follow-through state is captured.
The difference calculation area 3 includes a plurality of difference calculation areas (SWt1, SWt2, SWt3,..., SWtn) set in order to specify image data in which the top state is captured.
The image processing apparatus 1 first sets a primary difference calculation area group composed of a plurality of primary difference calculation areas in order to identify image data in which the top state is captured, and the plurality of primary differences. A primary difference calculation area group including a position where the movement of the club head is reversed is determined from the calculation areas. Then, the image processing apparatus 1 sets the difference calculation area group 3 in the determined primary difference calculation area group.

Next, the image processing apparatus 1 calculates the difference value of the component of the pixel in the same area | region between continuous image data for every set difference calculation area group.
Specifically, for each of the difference calculation area groups 1, 2, and 3, the image processing apparatus 1 continues for each set difference calculation area (for example, for each SWb 1, SWb 2, SWb 3,..., SWbn). The difference value of the color component of the pixel in the same area | region between the image data to perform is calculated.

  Then, the image processing apparatus 1 identifies image data in a predetermined state in a series of operations based on the calculated difference value. Specifically, the image processing apparatus 1 identifies image data in which a halfway back state is captured based on the difference value calculated in the difference calculation area 1. Further, the image processing apparatus 1 identifies image data in which the follow-through state is captured based on the difference value calculated in the difference calculation area 2. Further, the image processing apparatus 1 specifies image data in which the top state is imaged based on the difference values calculated in the difference calculation region 3 group.

  The image processing apparatus 1 captures, for example, a frame image based on image data in which the halfway back state is captured, a frame image based on image data in which the follow-through state is captured, or a top state based on a user operation. A frame image based on the image data thus displayed is displayed on the display unit 19.

FIG. 3 is a functional block diagram showing a functional configuration for the image processing apparatus 1 of FIG. 1 to execute the image specifying process.
3, only the CPU 11, the image processing unit 14, the imaging unit 17, the operation unit 18, the display unit 19, and the storage unit 20 are illustrated in the configuration of the image processing apparatus 1 in FIG. 1. ing. In the image processing apparatus 1, the CPU 11 and the image processing unit 14 cooperate to execute image specifying processing with reference to data stored in the storage unit 20.

In the CPU 11, the control unit 51 functions when the image specifying process is executed.
When the analysis mode is selected by the user in the operation unit 18, the image processing apparatus 1 displays a live view image on the display unit 19 and starts an image specifying process. In the present embodiment, the “analysis mode” is a mode in which image specifying processing is executed. Specifically, the user selects a “analysis mode” in the operation unit 18 to image a swing of a golf club as a series of operations, and is in a predetermined state of halfway back, follow-through, and top state. Can be displayed on the display unit 19.

When the analysis mode is selected by the user, the control unit 51 starts image specifying processing and performs control to display a setting screen on the display unit 19.
FIG. 4 is a diagram illustrating an example of a setting screen displayed by the image processing apparatus 1 of FIG.
On the setting screen, the control unit 51 displays, for example, a circular frame along with a display for instructing the distance from the ball to the installation position of the image processing apparatus 1 so that the ball enters the frame. Displays instructions for installing the. Such a display is a guideline for installing the image processing apparatus 1 so as to obtain an appropriate angle of view in the analysis mode. Further, the control unit 51 displays an instruction to operate the operation unit 18 (“calibration button” in the character display of FIG. 4) after setting the image processing apparatus 1 so that the ball enters the frame on the setting screen. To do.

  Returning to FIG. 3, when the operation unit 18 is operated by the user, in the image processing unit 14, the acquisition unit 41, the reference subject region setting unit 42, the second reference point setting unit 43, and the reference region setting unit 44, the detection unit 45, the first reference point setting unit 46, the determination unit 47, the setting unit 48, the calculation unit 49, and the specifying unit 50 function.

The acquisition unit 41 acquires reference image data obtained by capturing an image of a reference subject formed in a brighter color than the background. In the present embodiment, the “reference subject formed in a brighter color than the background” is a golf ball that is usually white. Further, in the present embodiment, “reference image data” means that the image processing unit 1 is installed so that the ball enters the frame on the setting screen, and triggered by the operation of the operation unit 18 by the user. It is the imaged image data.
Note that the reference image data is desirably image data in which a person is not imaged. Thereby, the setting precision of the 2nd reference point by the 2nd reference point setting part 43 mentioned below improves, and the precision which specifies finally the image data by which the predetermined state was imaged improves.

  The reference subject region setting unit 42 sets a region including the reference subject (hereinafter referred to as “reference subject region”) in the reference image data. Specifically, the reference subject area setting unit 42 sets the area within the frame displayed on the setting screen as the reference subject area.

The second reference point setting unit 43 calculates the center of gravity of a pixel group including pixels having a luminance equal to or higher than a predetermined threshold in the reference subject region set by the reference subject region setting unit 42, and sets the second reference point at the position of the center of gravity. Set a point.
Here, in the present embodiment, the reference image data is image data in YUV format (Y: luminance signal, U: difference between luminance signal and blue component, V: difference between luminance signal and red component). The reference image data is not limited to the YUV format, and may be image data in the RGB format (R: red, G: green, B: blue).

Specifically, the second reference point setting unit 43 measures the luminance of each pixel in the reference subject area from the reference image data in the YUV format and creates a histogram. Here, if the person is not in the reference subject area of the reference image data, the green grass on which the golf ball is placed and the white golf ball having a higher luminance than green are usually imaged.
In the histogram, the second reference point setting unit 43 sets the lower limit value of the luminance including a predetermined amount of pixels as the threshold value from the higher luminance. Then, the second reference point setting unit 43 calculates the center of gravity of the pixel group composed of pixels having a luminance equal to or higher than a predetermined threshold in the reference subject region, and sets the second reference point (for example, the middle point in FIG. 1) at the position of the center of gravity. B) is set. That is, the second reference point is the center coordinate of the golf ball in the reference image data.
If the histogram can be calculated appropriately, the center coordinates of the golf ball can be calculated without using the frame displayed on the setting screen.

  When the second reference point setting unit 43 sets the second reference point, the control unit 51 of the CPU 11 controls the display unit 19 to display a screen that prompts the user to make a golf swing and controls the imaging unit 17. Start recording imaging.

The acquisition unit 41 acquires a plurality of pieces of image data in which a golf swing, which is a series of actions of the subject, is continuously captured by the imaging unit 17.
The reference area setting unit 44 sets a reference area having a predetermined size including the subject in the image data acquired by the acquisition unit 41.
Here, in the present embodiment, the “reference area” refers to, for example, the user and golf ball in the address posture when the image processing apparatus 1 is installed at the position designated by the user on the setting screen (see FIG. 4). This is an enclosing rectangular area.

The storage unit 20 stores detection information for detecting an initial state in a series of operations.
Here, in the present embodiment, the “initial state” is a user's address posture in a golf swing that is a series of operations. In the present embodiment, “detection information” is an address detection discriminator for detecting image data of an address posture.

  The address detection discriminator uses, as learning samples, a positive sample that is an image data group of an address posture and a negative sample that is an image data group other than the address posture, for example, HOG (Histograms of Oriented Gradients) feature quantity and AdaBoost. It is constructed by machine learning by (Adaptive Boosting). Specifically, the address detection discriminator constructs a weak discriminator by calculating the HOG feature quantity that is a feature of the edge direction histogram of each image data of the positive sample and the negative sample, and weights the constructed weak discriminator. It is a strong classifier constructed by this.

FIG. 5 is a diagram illustrating an example of a learning sample for constructing an address detection discriminator. In FIG. 5, (a) and (b) show positive samples, and (c) and (d) show negative samples.
As shown in FIGS. 5A and 5B, the positive sample includes image data of address postures of various people's golf swings. Further, as shown in FIG. 5C, the negative sample includes image data of a person who is not in an address posture, as shown in FIG. 5D, in addition to image data that does not contain any person. By performing machine learning using such learning samples, it is possible to form a strong classifier that can be more accurately identified as the posture is closer to the address posture.
The rectangular frame shown in FIG. 5 is an area formed with the same number of pixels as the reference area set by the reference area setting unit 44, and the address detection discriminator is constructed using the pixels in the area as learning samples. The

The detection unit 45 uses the address detection discriminator as detection information stored in the storage unit 20 and uses the reference region set by the reference region setting unit 44 from among a plurality of image data acquired by the acquisition unit 41. Image data in an initial state having the highest feature amount in the region is detected as image data at the time of addressing.
Specifically, the detection unit 45 uses the address detection discriminator stored in the storage unit 20 for each of the plurality of image data acquired by the acquisition unit 41 in time series order to determine whether the image data of the address posture is used. The image data with the highest output is detected as the initial image data.

The first reference point setting unit 46 automatically sets a first reference point (for example, point A in FIG. 2) in the reference region of the initial image data detected by the detection unit 45.
Here, in the present embodiment, the “first reference point” is a preset position within the reference region, and the center near the upper end of the reference region, that is, the head of a person taking an address posture comes. This is the estimated position.
Here, for example, the first reference point can be automatically set by presetting a predetermined position in the image data area of the address posture detected by the address detection discriminator as the position of the head. .
Thus, the position of the subject in the image data can be determined by setting the first reference point.

  The determination unit 47 includes a first reference point set by the first reference point setting unit 46 (for example, the point A in FIG. 2) and a second reference point set by the second reference point setting unit 43 (for example, FIG. 2 middle point B), the size and position of the subject are determined.

  The setting unit 48 sets a plurality of difference calculation area groups corresponding to the subject in each of the plurality of image data acquired by the acquisition unit 41. Specifically, the setting unit 48 sets a plurality of difference calculation area groups according to the size and position of the subject determined by the determination unit 47. That is, the setting unit 48 sets a difference calculation region group based on the first reference point set by the first reference point setting unit 46.

FIG. 6 is a diagram illustrating a functional configuration for setting the difference calculation area group in the setting unit 48.
Specifically, the setting unit 48 sets rectangular difference calculation region groups 1, 2, and 3 according to the size and position of the subject determined by the determination unit 47.

The difference calculation area group 1 includes a plurality of difference calculation areas (SWb1,..., SWbn) set in order to specify image data in which a halfway back state is captured.
The difference calculation area group 2 includes a plurality of difference calculation areas (SWf1,..., SWfn) set in order to specify image data in which the follow-through state is captured.
The difference calculation area group 3 includes a plurality of difference calculation areas (SWt1,..., SWtn) set in order to specify image data in which the top is imaged.
The setting unit 48 first sets a primary difference calculation area group composed of a plurality of primary difference calculation areas in order to specify image data in which the top is imaged, and sets the plurality of primary difference calculation areas. A primary difference calculation area including a position where the movement of the club head is reversed is determined. Then, the setting unit 48 sets the difference calculation area group 3 in the determined primary difference calculation area.
That is, the setting unit 48 performs setting so as to change from the determined primary difference calculation area to another difference calculation area group (difference calculation area group 3).

  The setting unit 48 calculates, for example, the center coordinates (wb1xn, wb1yn) of the SWbn of the difference calculation area group 1 according to the following formulas according to the size and position of the subject determined by the determination unit 47.

(Equation 1)
wb1xn = ave_x + bs1x
wb1yn = ave_y + bs1y + N * D
ave_x = (Mx + in_head_x) / 2
ave_y = (My + in_head_y) / 2
bs1x = −float (cor_y) * P1
bs1y = −cor_y / 2
cor_y = My-in_head_y

“Ave_x” and “ave_y” in Equation 1 are the first reference point (point A in FIG. 6: center point of the human head) and the second reference point (point B in FIG. 6: center point of the golf ball). It is a human body position adjustment term calculated from
“Mx” indicates the X position of the point B. “In_head_x” indicates the X position of the point A. That is, “ave_x” indicates the center X position of the swinging person.
“My” indicates the Y position of the point B. “In_head_y” indicates the Y position of the point A. That is, “ave_y” indicates the center Y position of the person who swings.

“Bs1x” in Equation 1 indicates the distance adjustment width (lateral direction) of the center position of the difference calculation region group 1 with respect to the center position of the person who swings. “Bs1x” is set so that the position where the arm is extended becomes the right end of the difference calculation region group 1.
“Float (cor_y)” indicates the size of the subject, which is the distance from the Y position of the point B to the Y position of the point A, and is specifically a value assumed to be the height of the person who swings.
“P1” is a parameter set in advance so that only the shaft of the golf club enters the difference calculation region group 1, and is set to a value smaller than 1. By allowing only the golf club shaft to enter the difference calculation area group 1, the difference in pixel color components between successive image data in the difference calculation area group 1 becomes clear, and image data by the specifying unit 50 to be described later The specific accuracy of can be improved.

“Bs1y” in Equation 1 indicates the distance adjustment width (vertical direction) of the center position of the difference calculation region 1 with respect to the center position of the person who swings. “Bs1y” is set so that the approximate center of the head position and the ball position is the center of the difference calculation area 1.
“Cor_y” indicates the size of the subject, which is the distance from the Y position of the point B to the Y position of the point A, and is specifically a value assumed to be the height of the person who swings.

“N” in Equation 1 indicates the number of stages of a plurality of regions in the difference calculation region 1. The number of steps is set to 1 in the uppermost region in the Y-axis plus direction, for example, 2 when lowered by 1 step in the Y-axis minus direction, and 3 when lowered by 2 steps.
“D” in Equation 1 indicates a shift width. For example, the shift width is set in advance to a predetermined number of dots (for example, 5 dots).

In the present embodiment, the difference calculation area group 1 is set with nine stages of areas (SWb1 to SWb9) that are shifted by 5 dots from SWb1 in the negative Y-axis direction. The plurality of areas (SWb1 to SWb9) are set so as to partially overlap each other.
In this way, by setting a plurality of areas in the difference calculation area so as to partially overlap each other, a subtle size of the subject (for example, a swinging person) (for example, the physique of a swinging person) and a series of operations Even if there is a difference (such as swinging habits), it is possible to reliably identify image data obtained by imaging a predetermined state (half-way back, top, follow-through state) in a series of operations.

As illustrated in FIG. 6, the setting unit 48 sets the coordinate bxsn of the X-axis direction minus side end of the rectangular area SWbn to a value obtained by subtracting a predetermined value hb_down from the center coordinate wb1xn.
Further, the setting unit 48 sets the coordinate bxen of the X-axis direction plus side end of the rectangular area SWbn to a value obtained by adding the predetermined value hb_down to the center coordinate wb1xn.
In addition, the setting unit 48 sets the coordinate bysn of the Y axis direction plus side end of the rectangular area SWbn to a value obtained by adding the predetermined value vb to the center coordinate wb1yn.
In addition, the setting unit 48 sets the coordinate byen of the negative end portion in the Y-axis direction of the rectangular region SWbn to a value obtained by subtracting the predetermined value vd from the center coordinate wb1yn.

  The setting unit 48 calculates, for example, the center coordinates (wf1xn, wf1yn) of the SWfn of the difference calculation area group 2 according to the following formulas according to the size and position of the subject determined by the determination unit 47.

(Equation 2)
wf1xn = ave_x + fs1x
wf1yn = ave_y + fs1y + N * D
fs1x = cor_y * P2
fs1y = −cor_y / 2

In the following, the mathematical formula shown in Formula 2 will be described. In Formula 2, the value indicated by the same character as in Formula 1 is the same as the value in Formula 1, and the description thereof will be omitted.
“Fs1x” in Equation 2 indicates the distance adjustment width (lateral direction) of the center position of the difference calculation region group 2 with respect to the center position of the person who swings. “Fs1x” is set so that the position where the arm is extended becomes the left end of the difference calculation area group 2.
“P2” is a parameter set in advance so that only the golf club shaft enters the difference calculation region 2 and is set to a value smaller than 1. “P2” is set in order to identify the image data in which the follow-through state is captured, and thus has a value different from that of P1 for identifying the image data in which the half-way back state is captured. By allowing only the golf club shaft to enter the difference calculation area group 2, the difference in pixel color components between successive image data in the difference calculation area group 2 is clarified, and image data by the specifying unit 50 described later. The specific accuracy of can be improved.

  “Fs1y” in Equation 2 indicates the distance adjustment width (vertical direction) of the center position of the difference calculation area group 2 with respect to the center position of the person who swings. “Fs1y” is set so that the approximate center between the head position and the ball position is the center of the difference calculation area group 2.

  In this embodiment, the difference calculation area group 2 sets nine areas (SWf1 to SWf9) that are shifted by 5 dots in the Y-axis minus direction from SWf1.

As illustrated in FIG. 6, the setting unit 48 sets the coordinate fxsn of the negative end portion in the X-axis direction of the rectangular region SWfn to a value obtained by subtracting a predetermined value hb_forw from the center coordinate wf1xn.
Also, the setting unit 48 sets the coordinate fxen of the X-axis direction plus side end of the rectangular area SWfn to a value obtained by adding the predetermined value hb_forw to the center coordinate wf1xn.
Further, the setting unit 48 sets the coordinate fysn of the positive end portion in the Y-axis direction of the rectangular region SWfn to a value obtained by adding the predetermined value vb to the center coordinate wf1yn.
In addition, the setting unit 48 sets the coordinate fien of the negative end portion in the Y-axis direction of the rectangular area SWfn to a value obtained by subtracting the predetermined value vb from the center coordinate wf1yn.

Here, in general, in the golf swing, the position of the golf club in the top state is very different from that of the half way back and follow-through, and only the size and position of the subject determined by the determination unit 47 are present. When the difference calculation area is set based on the above, there are cases where the detection accuracy cannot be expected.
For this reason, the setting part 48 sets the difference calculation area group 3 by a functional structure below.

The setting unit 48 includes a plurality of primary difference calculation areas (frames indicated by thin dotted lines in FIG. 6) that do not overlap each other above the first reference point (point A in FIG. 6) set by the first reference point setting unit 46. ) Is set. The setting unit 48 calculates an average value of difference evaluation values between successive image data calculated by a calculation unit 49 described later. The setting unit 48 detects the primary difference calculation area immediately before the calculated average value is switched to the minimum. This detected primary difference calculation area can be estimated as a position where the head of the golf club is reversed. Then, the setting unit 48 sets the difference calculation area group 3 in the detected primary difference calculation area. The setting unit 48 sets the difference calculation area group 3 by the same processing as the difference calculation area group 1 and the difference calculation area group 2.
Thereby, the detection accuracy of the image data in which the top posture (state) is imaged is improved.

The calculation unit 49 calculates the difference value of the color components of the pixels in the same region between successive image data for each difference calculation region group set by the setting unit 48.
Specifically, the calculation unit 49 calculates a difference evaluation value between image data in each of the difference calculation region groups 1, 2, and 3.
Further, the calculation unit 49 calculates a difference evaluation value between image data in the primary difference calculation region group set by the setting unit 48.
In the present embodiment, the “difference evaluation value” is the difference absolute value of each color component between the nth (n represents an integer) image data and the (n−1) th image data, the nth image data, and the (n + 1) th image data. The sum of the number of pixels in a region where the product of the difference absolute value of each color component between the image data exceeds a predetermined threshold.
Note that the calculation unit 49 calculates the total value of the difference values of the respective color components of the pixels in the same region between successive image data, and calculates the value added according to the calculated change amount of the total value as the difference evaluation value. May be calculated as
Further, the calculation unit 49 may apply a LowPass filter to the calculated difference evaluation value to remove a value higher than a predetermined threshold value. Thereby, it can prevent that the specific | specification part 50 mentioned later specifies incorrect image data.
In the calculation of the difference evaluation value, the difference absolute value can be a difference square value. Further, in the calculation of the difference evaluation value, the multiplication value can be an addition value.

FIG. 7 is a diagram illustrating an example of a distribution of difference evaluation values in the difference calculation area 1.
In the example of FIG. 7, the difference evaluation value of the area SWb6 of the difference calculation area 1 in the 353rd image data is the highest.

First, the calculation unit 49 calculates a difference evaluation value in the difference calculation region group 3 (region for specifying image data in which the top posture is captured). Next, the calculation unit 49 calculates a difference from the image data detected as the initial image data (image data in which the address posture is imaged) by the detection unit 45 among the plurality of image data in which a series of operations has been imaged. Difference evaluation of the difference calculation area group 1 (area group for specifying the image data obtained by capturing the halfway back) in a plurality of image data with the image data having the highest difference evaluation value in the calculation area group 3 Calculate the value. Then, the calculation unit 49 next calculates a difference calculation area in a plurality of image data after the image data having the highest difference evaluation value in the difference calculation area group 3 among the plurality of image data obtained by capturing a series of operations. A difference evaluation value of group 2 (region for specifying image data in which follow is imaged) is calculated.
Thus, by limiting the range of the plurality of image data for calculating the difference evaluation value, it is possible to specify the image data in which the predetermined state is captured with a small amount of calculation and high accuracy.

Returning to FIG. 3, the specifying unit 50 specifies image data obtained by capturing a predetermined state in a series of operations based on the difference value calculated by the calculating unit 49.
Specifically, the specifying unit 50 specifies the image data having the highest difference evaluation value in each of the difference calculation region groups 1, 2, and 3.
For example, in the example of FIG. 7, the specifying unit 50 specifies the 353rd image data in the difference calculation area group 1. That is, the specifying unit 50 specifies image data in which a half-way back state is captured as a predetermined state in a series of operations.

  For example, the control unit 51 displays the image of the image data specified by the specifying unit 50 of the image processing unit 14 on the display unit 19 based on the operation of the operation unit 18 by the user.

Next, with reference to FIG. 8, the image specifying process realized by the functional configuration of FIG. 3 will be described among the processes of the image processing apparatus 1.
FIG. 8 is a flowchart for explaining the flow of the image specifying process executed by the image processing apparatus 1 of FIG.

The image specifying process in FIG. 8 is a process executed when the imaging unit 17 performs imaging.
That is, when the analysis mode is selected by the user in the operation unit 18, the image processing apparatus 1 displays a live view image on the display unit 19 and starts an image specifying process.

In step S <b> 1, the control unit 51 performs control to display a setting screen (see FIG. 4) on the display unit 19.
In step S2, the acquisition unit 41, the reference subject region setting unit 42, and the second reference point setting unit 43 execute a second reference point setting process.

FIG. 9 is a flowchart for explaining the flow of the second reference point setting process executed by the image processing apparatus 1 of FIG.
In step S <b> 21, the acquisition unit 41 acquires reference image data obtained by imaging a reference subject that is formed in a brighter color than the background.
In step S22, the reference subject region setting unit 42 sets a reference subject region in the reference image data acquired by the acquisition unit 41 in step S21.
In step S23, the second reference point setting unit 43 calculates the center of gravity of the pixel group composed of pixels having a luminance equal to or higher than a predetermined threshold in the reference subject region set by the reference subject region setting unit 42 in step S22. A second reference point (point B in FIG. 6) is set at the position of the center of gravity.

Returning to FIG. 8, in step S <b> 3, the control unit 51 controls the display unit 19 to display a screen that prompts the user to make a golf swing, and controls the imaging unit 17 to start imaging for recording.
In step S <b> 4, the acquisition unit 41 acquires a plurality of image data in which a golf swing that is a series of movements of the subject is continuously captured by the imaging unit 17.
In step S5, the reference area setting unit 44, the detection unit 45, and the first reference point setting unit 46 execute a first reference point setting process.

FIG. 10 is a flowchart for explaining the flow of the first reference point setting process executed by the image processing apparatus 1 of FIG.
In step S51, the reference area setting unit 44 sets a reference area having a predetermined size including the subject in the image data acquired by the acquisition unit 41 in step S4.
In step S52, the detection unit 45 uses the address detection discriminator stored in the storage unit 20 to select the reference area setting unit 44 in step S51 from the plurality of image data acquired by the acquisition unit 41 in step S4. The image data in the initial state having the highest feature amount in the reference region set by (1) is detected.
In step S53, the first reference point setting unit 46 sets the first reference point (point A in FIG. 6) in the reference region of the image data in the initial state detected by the detection unit 45 in step S52.

Returning to FIG. 8, in step S6, the determining unit 47 sets the first reference point (point A in FIG. 6) set by the first reference point setting unit 46 in step S53 and the second reference point setting in step S23. Based on the second reference point (point B in FIG. 6) set by the unit 43, the size and position of the subject are determined.
In step S7, the setting unit 48 adds a plurality of difference calculation areas corresponding to the size and position of the subject determined by the determination unit 47 in step S6 to each of the plurality of image data acquired by the acquisition unit 41 in step S4. Groups 1 and 2 are set.
In step S8, the setting unit 48 and the calculation unit 49 execute a difference calculation area group 3 setting process.

FIG. 11 is a flowchart for explaining the flow of the difference calculation area 3 setting process executed by the image processing apparatus 1 of FIG.
In step S81, the setting unit 48 includes a plurality of primary difference calculation areas (FIG. 6) that do not overlap each other above the first reference point (point A in FIG. 6) set by the first reference point setting unit 46 in step S53. Set the frame indicated by the thin dotted line inside.
In step S82, the calculation unit 49 calculates a difference evaluation value between image data in the primary difference calculation region set by the setting unit 48 in step S81.
In step S83, the setting unit 48 calculates the average value of the difference evaluation values between successive image data calculated by the calculation unit 49 in step S82. Then, the setting unit 48 detects the primary difference calculation area immediately before the calculated average value switches to the minimum, and sets the difference calculation area 3 group so as to overlap the position of the detected primary difference calculation area. .

Returning to FIG. 8, in step S <b> 9, the calculation unit 49 calculates the difference in the same region between successive image data for each of the difference calculation region groups 1, 2, and 3 set by the step S <b> 7 and step S <b> 83 setting unit 48. An evaluation value is calculated.
In step S10, the specifying unit 50 specifies image data obtained by capturing a predetermined state in a series of operations based on the difference value calculated by the calculating unit 49 in step S9.
In step S <b> 11, the control unit 51 displays an image of the image data specified by the image processing unit 14 on the display unit 19.

As described above, the image processing apparatus 1 according to the present embodiment includes the acquisition unit 41, the setting unit 48, the calculation unit 49, and the specifying unit 50.
The acquisition unit 41 acquires a plurality of pieces of image data obtained by continuously capturing a series of actions of the subject.
The setting unit 48 sets an evaluation area corresponding to the size and position of the subject in each of the plurality of image data acquired by the acquisition unit 41.
The calculation unit 49 calculates an evaluation value between a plurality of image data in the evaluation region set by the setting unit 48.
The specifying unit 50 specifies image data in a predetermined state in a series of operations based on the evaluation values calculated by the calculating unit 49.

Thereby, in the process of specifying an image in which a predetermined state is captured from a plurality of image data, the difference value of the same region between successive image data can be obtained only by a plurality of difference calculation region groups corresponding to the subject. calculate. Therefore, the processing load of the image processing apparatus 1 can be reduced as compared with the case where the difference value between successive image data is calculated in the entire image data.
Therefore, it is possible to reduce the processing load for specifying an image obtained by capturing a predetermined state in the operation of a desired subject.
In addition, by reducing the processing load in this way, it is possible to specify a captured image of a predetermined state even in a portable terminal having a CPU with relatively low processing capability.

The image processing apparatus 1 according to the present embodiment further includes a reference area setting unit 44, a storage unit 20, a detection unit 45, and a first reference point setting unit 46.
The reference area setting unit 44 sets a reference area having a predetermined size including the subject in the image data acquired by the acquisition unit 41.
The storage unit 20 stores detection information for detecting an initial state in a series of operations.
The detection unit 45 uses the detection information stored in the storage unit 20, and the feature amount in the reference region set by the reference region setting unit 44 is the largest among the plurality of image data acquired by the acquisition unit 41. Detect high initial image data.
The first reference point setting unit 46 sets the first reference point in the reference region of the image data in the initial state detected by the detection unit 45.
The setting unit 48 sets a difference calculation region group based on the first reference point set by the first reference point setting unit 46.

  Thereby, it is possible to detect image data in an initial state in which an initial state in a series of operations (for example, an address posture state in a swing of a golf club) is captured. Then, in the reference area of the initial state image data, a first reference point (for example, the position of the head of the person performing the swing) is set as a reference for determining the size and position of the subject (for example, the person performing the swing). Thus, it is possible to improve the accuracy of processing for specifying an image in which a predetermined state of the subject is captured.

The image processing apparatus 1 according to the present embodiment further includes a reference subject region setting unit 42, a second reference point setting unit 43, and a determination unit 47.
The acquisition unit 41 acquires reference image data obtained by capturing an image of a reference subject formed in a brighter color than the background.
The reference subject area setting unit 42 sets a reference subject area in the reference image data.
The second reference point setting unit 43 calculates the center of gravity of a pixel group including pixels having a luminance equal to or higher than a predetermined threshold in the reference subject region set by the reference subject region setting unit 42, and sets the second reference point at the position of the center of gravity. Set a point.
The determination unit 47 determines the size and position of the subject based on the first reference point set by the first reference point setting unit 46 and the second reference point set by the second reference point setting unit 43.
The setting unit 48 sets a plurality of difference calculation area groups according to the size and position of the subject determined by the determination unit 47.

As a result, the center of gravity of the reference subject (eg, golf ball) can be calculated based on the luminance in the reference subject region, and this center of gravity can be set as the second reference point (eg, the center position of the golf ball). Therefore, by limiting the detection range of the reference subject to the reference subject region, the processing speed for setting the second reference point can be improved as compared with the case where the detection range is the entire image data.
Also, by determining the size and position of the subject based on two points, the first reference point and the second reference point, which serve as a reference for assuming the size and position of the subject, an image in which a predetermined state of the subject is captured can be obtained. The accuracy of the specified process can be further improved.

  In addition, this invention is not limited to the above-mentioned embodiment, The deformation | transformation in the range which can achieve the objective of this invention, improvement, etc. are included in this invention.

  In the above-described embodiment, the series of operations that are the target of the image specifying process executed by the image processing apparatus 1 is a golf club swing, but is not limited to this, for example, a baseball bat swing, various sports and dances. It may be a form or the like.

  In the above-described embodiment, the control unit 51 displays a circular frame on the setting screen displayed on the display unit 19 and displays an instruction to install the image processing apparatus 1 so that the ball enters the frame. However, the present invention is not limited to this. For example, the display unit 19 is configured by a touch panel that can accept a user operation, and the control unit 51 can accept the position of the frame on the setting screen by a user touch panel operation. .

  In the above-described embodiment, the control unit 51 displays the setting screen on the display unit 19. However, the present invention is not limited to this. For example, the control unit 51 stores the position of the first hit ball in the storage unit 20 and the second and subsequent shots. Then, the display of the setting screen may be omitted, and when it is determined that the ball is installed at the position of the ball stored in the storage unit 20, the imaging unit 17 may be controlled to start recording for recording.

  In the above-described embodiment, the determination unit 47 is set by the first reference point (the position of the person's head that swings) set by the first reference point setting unit 46 and the second reference point setting unit 43. The size and position of the subject are determined based on the second reference point (the center of the ball). However, the present invention is not limited to this, and for example, the display unit 19 is configured with a touch panel that can accept a user operation. The size and position of the subject may be determined based on the first reference point and / or the second reference point received by the user's touch panel operation.

In the above-described embodiment, the difference calculation area 1 is used as an area for specifying image data obtained by capturing a halfway back. However, the present invention is not limited to this, and the difference calculation area 1 is in a halfway down state. You may use as an area | region for specifying the imaged image data. Thereby, the image data in which the halfway down is imaged as the predetermined state can be specified.
In addition to the above-described embodiment, image data in which an impact (the moment when the golf club head hits the ball) may be specified. In this case, the setting unit 48 sets the difference calculation area group based on the second reference point (the center of the ball at the time of address) set by the second reference point setting unit 43. The calculation unit 49 analyzes the difference value between successive image data in the difference calculation area group set by the setting unit 48. Specifically, in this analysis, a difference in presence or absence of a ball is detected. Then, the specifying unit 50 specifies the last image data in which the ball is imaged as image data in which the impact is imaged.
In the above case, in the difference calculation area group 3, the calculation unit 49 analyzes a difference value between a plurality of pieces of image data after the image data having the highest difference evaluation value, and detects a difference in presence or absence of the ball. Next, in a plurality of pieces of image data between the image data having the highest difference evaluation value in the difference calculation area group 3 and the last image data in which the ball is detected, the difference calculation area group 1 (halfway down is imaged). The difference evaluation value of the area for specifying the image data obtained is calculated. Then, the calculation unit 49 next performs difference evaluation of the difference calculation region group 2 (region for specifying image data in which the follow is captured) in a plurality of image data after the last image data in which the ball is detected. Calculate the value.
As described above, even when there are a plurality of predetermined states to be specified, by limiting the range of the plurality of image data for calculating the difference evaluation value, an image in which the predetermined state is captured with a small amount of calculation and high accuracy. Data can be specified.

  In the above-described embodiment, the image processing apparatus to which the present invention is applied has been described as an example configured as an image processing apparatus such as a digital camera. However, the present invention is not particularly limited to an image processing apparatus, and in general, an electronic apparatus that can execute the above-described image processing regardless of the presence or absence of an imaging function (data of captured images may be acquired from another apparatus). Can be applied. Specifically, for example, the present invention can be applied to a personal computer, a video camera, a portable navigation device, a portable game machine, and the like.

The series of processes described above can be executed by hardware or can be executed by software.
In other words, the functional configuration of FIG. 3 is merely an example, and is not particularly limited. That is, it is sufficient that the image processing apparatus 1 has a function capable of executing the above-described series of processes as a whole, and what functional blocks are used to realize this function is not particularly limited to the example of FIG. .
In addition, one functional block may be constituted by hardware alone, software alone, or a combination thereof.

When a series of processing is executed by software, a program constituting the software is installed on a computer or the like from a network or a recording medium.
The computer may be a computer incorporated in dedicated hardware. The computer may be a computer capable of executing various functions by installing various programs, for example, a general-purpose personal computer.

  The recording medium including such a program is not only constituted by the removable medium 31 of FIG. 1 distributed separately from the apparatus main body in order to provide the program to the user, but also in a state of being incorporated in the apparatus main body in advance. It is comprised with the recording medium etc. which are provided in this. The removable medium 31 is composed of, for example, a magnetic disk (including a floppy disk), an optical disk, a magneto-optical disk, or the like. The optical disk is composed of, for example, a CD-ROM (Compact Disk-Read Only Memory), a DVD (Digital Versatile Disk), or the like. The magneto-optical disk is configured by an MD (Mini-Disk) or the like. In addition, the recording medium provided to the user in a state of being preliminarily incorporated in the apparatus main body includes, for example, the ROM 12 in FIG. 1 in which a program is recorded, the hard disk included in the storage unit 20 in FIG.

  In the present specification, the step of describing the program recorded on the recording medium is not limited to the processing performed in time series along the order, but is not necessarily performed in time series, either in parallel or individually. The process to be executed is also included.

  As mentioned above, although embodiment of this invention was described, this embodiment is only an illustration and does not limit the technical scope of this invention. The present invention can take other various embodiments, and various modifications such as omission and replacement can be made without departing from the gist of the present invention. These embodiments and modifications thereof are included in the scope and gist of the invention described in this specification and the like, and are included in the invention described in the claims and the equivalents thereof.

The invention described in the scope of claims at the beginning of the filing of the present application will be appended.
[Appendix 1]
Acquisition means for acquiring a plurality of image data obtained by continuously imaging a series of movements of a subject;
Setting means for setting an evaluation area corresponding to the size and position of the subject in each of a plurality of image data acquired by the acquisition means;
First calculation means for calculating an evaluation value between the plurality of image data in the evaluation region set by the setting means;
Identification means for identifying image data in a predetermined state in the series of operations based on the evaluation values calculated by the first calculation means;
An image specifying apparatus comprising:
[Appendix 2]
Subject determination means for determining the subject based on the image data;
The setting means sets an evaluation area according to the size and position of the subject determined by the subject determination means;
The image specifying device according to Supplementary Note 1, wherein
[Appendix 3]
A first reference point setting means for setting a first reference point in the image data;
The subject determining means determines the position of the subject based on the first reference point set by the first reference point setting means;
The image specifying device according to Supplementary Note 2, wherein
[Appendix 4]
Reference area setting means for setting an area of a predetermined size including the subject in the image data acquired by the acquisition means;
Storage means for storing detection information for detecting a specific state in the series of operations;
Using the detection information stored in the storage unit, the image data having a high feature amount in the region set by the reference region setting unit is identified from the plurality of image data acquired by the acquiring unit. Detecting means for detecting image data in the state of
Further comprising
The first reference point setting means sets the first reference point in an area set by the reference area setting means in the image data detected by the detection means;
The image processing apparatus according to appendix 3, characterized in that:
[Appendix 5]
A second reference point setting means for setting a second reference point in the image data;
The subject determination unit is configured to determine the size and position of the subject based on the first reference point set by the first reference point setting unit and the second reference point set by the second reference point setting unit. To decide,
The image processing apparatus according to appendix 3 or 4, characterized in that:
[Appendix 6]
The acquisition means acquires image data of a reference subject,
Area setting means for setting an area including the reference subject in the image data obtained by the obtaining means;
Centroid calculating means for calculating the centroid of a specific pixel group in the area set by the area setting means;
Further comprising
The second reference point setting means sets a second reference point at the position of the center of gravity calculated by the center of gravity calculating means;
The subject determining unit is configured to determine the size of the subject based on the first reference point set by the first reference point setting unit and the second reference point calculated by the second reference point setting unit. And determine the position,
The image processing apparatus according to appendix 5, characterized in that:
[Appendix 7]
The setting means sets a plurality of evaluation areas according to the subject,
The first calculation unit calculates an evaluation value between the plurality of image data in the plurality of evaluation regions set by the setting unit, respectively.
The image specifying device according to any one of supplementary notes 1 to 6, wherein:
[Appendix 8]
The setting means sets a plurality of evaluation areas corresponding to the subject such that each evaluation area partially overlaps,
The image specifying device according to any one of supplementary notes 1 to 7, characterized in that:
[Appendix 9]
The setting means sets a plurality of evaluation areas close to each other,
The first calculation unit calculates an evaluation value between the plurality of image data in the plurality of evaluation regions set by the setting unit, respectively.
Based on the evaluation values respectively calculated by the first calculation means, area changing means for changing from a plurality of evaluation areas set by the setting means to a plurality of overlapping evaluation areas;
Second calculation means for calculating evaluation values between the plurality of image data in the plurality of evaluation areas changed by the area changing means;
Further comprising
The specifying means specifies the image data in the predetermined state based on the evaluation values calculated by the second calculating means,
The image specifying device according to any one of supplementary notes 1 to 8, characterized in that:
[Appendix 10]
The first calculation means calculates, as the evaluation value, a pixel difference value between the plurality of image data in the evaluation region.
The image processing apparatus according to any one of supplementary notes 1 to 9, wherein
[Appendix 11]
An image processing method executed by an image processing apparatus that identifies image data obtained by capturing a predetermined state in a series of movements of a subject,
An acquisition step of acquiring a plurality of image data obtained by continuously imaging the series of operations;
A setting step of setting a plurality of evaluation areas according to the size and position of the subject in each of the plurality of image data acquired in the acquisition step;
A calculation step for calculating an evaluation value between the plurality of image data in the evaluation region set in the setting step;
A specifying step of specifying the image data in the predetermined state based on the evaluation values calculated in the calculating step;
An image processing method comprising:
[Appendix 12]
A computer that controls image processing for identifying image data in which a predetermined state in a series of movements of a subject is captured
Acquisition means for acquiring a plurality of image data in which a series of actions of a subject are continuously captured;
Setting means for setting an evaluation area corresponding to the size and position of the subject in each of a plurality of image data acquired by the acquisition means;
Calculating means for calculating evaluation values between the plurality of image data in the evaluation area set by the setting means;
Identification means for identifying the image data in the predetermined state based on the evaluation values respectively calculated by the calculation means;
A program characterized by functioning as

  DESCRIPTION OF SYMBOLS 1 ... Image processing apparatus, 11 ... CPU, 12 ... ROM, 13 ... RAM, 14 ... Image processing part, 17 ... Imaging part, 17 ... Operation part, 19 * -Display unit, 20 ... Storage unit, 21 ... Communication unit, 22 ... Drive, 31 ... Removable media, 41 ... Acquisition unit, 42 ... Reference subject area setting unit, 43 ... 2nd reference point setting part, 44 ... Reference area setting part, 45 ... Detection part, 46 ... 1st reference point setting part, 47 ... Determination part, 48 ... Setting part 49 ... calculation unit, 50 ... specification unit

Claims (12)

Acquisition means for acquiring a plurality of image data obtained by continuously imaging a series of movements of a subject;
Setting means for setting an evaluation area corresponding to the size and position of the subject in each of a plurality of image data acquired by the acquisition means;
First calculation means for calculating an evaluation value between the plurality of image data in the evaluation region set by the setting means;
Identification means for identifying image data in a predetermined state in the series of operations based on the evaluation values calculated by the first calculation means;
An image specifying apparatus comprising: Subject determination means for determining the subject based on the image data;
The setting means sets an evaluation area according to the size and position of the subject determined by the subject determination means;
The image specifying device according to claim 1. A first reference point setting means for setting a first reference point in the image data;
The subject determining means determines the position of the subject based on the first reference point set by the first reference point setting means;
The image specifying device according to claim 2. Reference area setting means for setting an area of a predetermined size including the subject in the image data acquired by the acquisition means;
Storage means for storing detection information for detecting a specific state in the series of operations;
Using the detection information stored in the storage unit, the image data having a high feature amount in the region set by the reference region setting unit is identified from the plurality of image data acquired by the acquiring unit. Detecting means for detecting image data in the state of
Further comprising
The first reference point setting means sets the first reference point in an area set by the reference area setting means in the image data detected by the detection means;
The image processing apparatus according to claim 3. A second reference point setting means for setting a second reference point in the image data;
The subject determination unit is configured to determine the size and position of the subject based on the first reference point set by the first reference point setting unit and the second reference point set by the second reference point setting unit. To decide,
The image processing apparatus according to claim 3, wherein the image processing apparatus is an image processing apparatus. The acquisition means acquires image data of a reference subject,
Area setting means for setting an area including the reference subject in the image data obtained by the obtaining means;
Centroid calculating means for calculating the centroid of a specific pixel group in the area set by the area setting means;
Further comprising
The second reference point setting means sets a second reference point at the position of the center of gravity calculated by the center of gravity calculating means;
The subject determining unit is configured to determine the size of the subject based on the first reference point set by the first reference point setting unit and the second reference point calculated by the second reference point setting unit. And determine the position,
The image processing apparatus according to claim 5. The setting means sets a plurality of evaluation areas according to the subject,
The first calculation unit calculates an evaluation value between the plurality of image data in the plurality of evaluation regions set by the setting unit, respectively.
The image specifying device according to claim 1, wherein the image specifying device is one of the following. The setting means sets a plurality of evaluation areas corresponding to the subject such that each evaluation area partially overlaps,
The image specifying device according to claim 1, wherein the image specifying device is an image specifying device. The setting means sets a plurality of evaluation areas close to each other,
The first calculation unit calculates an evaluation value between the plurality of image data in the plurality of evaluation regions set by the setting unit, respectively.
Based on the evaluation values respectively calculated by the first calculation means, area changing means for changing from a plurality of evaluation areas set by the setting means to a plurality of overlapping evaluation areas;
Second calculation means for calculating evaluation values between the plurality of image data in the plurality of evaluation areas changed by the area changing means;
Further comprising
The specifying means specifies the image data in the predetermined state based on the evaluation values calculated by the second calculating means,
The image specifying device according to claim 1, wherein the image specifying device is one of the following. The first calculation means calculates, as the evaluation value, a pixel difference value between the plurality of image data in the evaluation region.
The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus. An image processing method executed by an image processing apparatus that identifies image data obtained by capturing a predetermined state in a series of movements of a subject,
An acquisition step of acquiring a plurality of image data obtained by continuously imaging the series of operations;
A setting step of setting a plurality of evaluation areas according to the size and position of the subject in each of the plurality of image data acquired in the acquisition step;
A calculation step for calculating an evaluation value between the plurality of image data in the evaluation region set in the setting step;
A specifying step of specifying the image data in the predetermined state based on the evaluation values calculated in the calculating step;
An image processing method comprising: A computer that controls image processing for identifying image data in which a predetermined state in a series of movements of a subject is captured;
Acquisition means for acquiring a plurality of image data in which a series of actions of a subject are continuously captured;
Setting means for setting an evaluation area corresponding to the size and position of the subject in each of a plurality of image data acquired by the acquisition means;
Calculating means for calculating evaluation values between the plurality of image data in the evaluation area set by the setting means;
Identification means for identifying the image data in the predetermined state based on the evaluation values respectively calculated by the calculation means;
A program characterized by functioning as

Images