JP2012160978A - Imaging control device, template acquisition device, and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an imaging control device and a program that are capable of accurately estimating environmental light, and provide a template acquisition device and a program that are capable of reducing an individual difference of a success rate in viewpoint detection.SOLUTION: A control device 70 causes a light source 10 to emit light in imaging of an odd field by an imaging unit 20 for imaging a subject. A CPU 71 specifies a sunlight direction with respect to the subject based on a measurement value of an illumination environment detection device 60, and controls a light emission time of the light source 10 based on the specified sunlight direction. Then, the CPU 71 of the control device 70 performs a predetermined instruction plural times with a speaker 50 for a person as the subject, causes the imaging unit 20 to perform imaging each time being instructed to acquire subject images corresponding to the number of imaging times, compares them to thereby detect a feature area of the person in the subject images, and acquires a feature area image. The CPU 71 causes storage means to store the feature area image as a template.

Description

  The present invention relates to an imaging control device, a template acquisition device, and a program.

  An image processing technique for detecting the viewpoint of a user (mainly a driver) by a predetermined device mounted on a moving body such as a vehicle is known.

  Patent Document 1 discloses a technique for acquiring information from a solar radiation sensor (light sensor) provided on the upper surface of an instrument panel of a vehicle, calculating external light intensity based on this information, and calculating the output of camera illumination. Has been.

  Patent Document 2 discloses a technique for detecting the position of a user's eyes by performing a correlation calculation with a standard template on an image captured by an imaging apparatus.

JP 2009-96323 A Japanese Patent Laid-Open No. 06-255388

  In the technique disclosed in Patent Document 1, if only the solar radiation sensor is hidden behind a shadow such as a utility pole, it is hidden behind the shadow even though the user to be imaged is irradiated with external light such as sunlight. Since the ambient light intensity is calculated based on the data from the dark solar radiation sensor, the ambient light inside the vehicle or the like cannot be accurately determined.

  Further, the technique disclosed in Patent Document 2 has a problem in that there are individual differences in the success rate of viewpoint detection because the positions of various user eyes are detected based on a standard template.

  The present invention has been made in view of the above circumstances, and a first object thereof is to provide an imaging control apparatus and program capable of accurately estimating ambient light. A second object of the present invention is to provide a template acquisition apparatus and program that can reduce individual differences in the success rate in viewpoint detection.

In order to achieve the first object, an imaging control apparatus according to the first aspect of the present invention provides:
An imaging unit that performs imaging of a subject divided into imaging of a plurality of fields, and a light source that irradiates light to the subject that is captured by the imaging unit, and controls imaging of a predetermined part of the imaging of the plurality of fields An imaging control device that emits the light source only in
Direction specifying means for specifying the direction of sunlight with respect to the subject based on a measurement value measured by a predetermined measuring device;
Control means for controlling the light emission time of the light source based on the direction of the sunlight specified by the direction specifying means.

In order to achieve the first object, a program according to the second aspect of the present invention provides:
On the computer,
An imaging unit that performs imaging of a subject divided into imaging of a plurality of fields, and a light source that irradiates light to the subject that is captured by the imaging unit, and controls imaging of a predetermined part of the imaging of the plurality of fields A program for executing the process of causing the light source to emit light only in
A process for identifying the direction of sunlight with respect to the subject based on a measurement value measured by a predetermined measurement device;
A process for controlling a light emission time of the light source based on the identified direction of the sunlight;
Is executed.

In order to achieve the second object, the template acquisition device according to the third aspect of the present invention is:
Instruction means for giving a predetermined instruction to a person as a subject a plurality of times;
Imaging control means for causing the imaging unit to image the subject for each predetermined instruction;
Subject image acquisition means for acquiring subject images for the number of times of imaging from the imaging unit that images the subject;
A feature region image acquisition unit that detects a feature region of a person in the subject image by comparing the acquired subject images for the number of times of imaging, and acquires a feature region image indicating the detected feature region;
Storage control means for storing the characteristic area image acquired by the characteristic area image acquisition means in a storage means as a template for pattern matching in the process of detecting the viewpoint of the person.

In order to achieve the second object, a program according to the fourth aspect of the present invention provides:
On the computer,
A process of giving a predetermined instruction to a person as a subject a plurality of times;
A process of causing the imaging unit to image the subject for each predetermined instruction;
Processing for acquiring subject images for the number of times of imaging from the imaging unit that images the subject;
A process of detecting a feature region of a person in the subject image by comparing the obtained subject images for the number of times of imaging, and acquiring a feature region image indicating the detected feature region;
A process of storing the acquired feature area image in a storage unit as a pattern matching template in the process of detecting the viewpoint of the person;
Is executed.

  According to the imaging control device according to the first aspect of the present invention and the program according to the second aspect of the present invention, it is possible to accurately estimate the ambient light. Further, according to the template acquisition apparatus according to the third aspect of the present invention and the program according to the fourth aspect of the present invention, individual differences in the success rate in viewpoint detection can be reduced.

It is a block diagram which shows schematic structure of the viewpoint detection system which concerns on one Embodiment of this invention. It is a block diagram for demonstrating the function of a control apparatus. It is a figure for demonstrating the structure of a flame | frame. It is a figure for demonstrating a solar height and solar direction specific table. (A) is a figure for demonstrating a sunlight intensity specific table, (b) is a figure for demonstrating a solar orientation coefficient specific table, (c) is a measurement environment light intensity specific table. It is a figure for demonstrating. (A) is a figure for demonstrating a lighting time determination table, (b) is a figure for demonstrating a shutter time determination table. (A)-(c) is a timing chart for demonstrating the transmission timing of a light source control signal in a light source control process. It is a flowchart of a light source control process. It is a flowchart of a subject detection process. It is the figure which showed an example of the frame image. (A) is a figure showing an example of an odd field image, and (b) is a figure showing an example of an even field image. (A) is a figure showing an example of the 1st interpolation frame picture, and (b) is a figure showing an example of the 2nd interpolation frame picture. It is the figure which showed an example of the to-be-photographed object detection image. (A) And (b) is a figure for demonstrating what kind of subject detection image is obtained when a subject detection process is performed for a certain person as a subject. It is a conceptual diagram for demonstrating the viewpoint detection process in a to-be-photographed object detection process. It is a flowchart of a template acquisition process. (A)-(d) is a figure for demonstrating the specific example at the time of performing a template acquisition process by making a certain person into a subject.

  Hereinafter, an embodiment of the present invention will be described with reference to FIGS.

As shown in FIG. 1, the viewpoint detection system 1 according to the present embodiment includes a light source 10, an imaging unit 20, a display device 30, an input device 40, a speaker 50, a lighting environment detection device 60, and a control device. 70.
In the viewpoint detection system 1 having such a configuration, first, the control device 70 turns on the light source 10 (light emission time) and the shutter time of the imaging unit 20 based on various information acquired from the illumination environment detection device 60. (Time when the shutter is open). Next, the control device 70 causes the imaging unit 20 to image the subject 2 (see FIG. 2) with the determined shutter time, and turns on the light source 10 with the determined lighting time. Then, a region where the subject 2 is captured in the captured image (hereinafter referred to as “subject region”) is detected, a predetermined subject detection process is executed, and the viewpoint of the subject 2 is detected.
Hereinafter, each component with which the viewpoint detection system 1 is provided is demonstrated in detail.

  The light source 10 irradiates the subject 2 with light. The light source 10 is, for example, an infrared light source, and is turned on / off based on a light source control signal output from the control device 70.

  The imaging unit 20 images the subject 2. The imaging unit 20 is a video camera including, for example, a lens and a known solid-state imaging device such as a CCD (Charged Coupled Device), a CMOS (Complementary Metal Oxide Semiconductor), and a shutter. The imaging unit 20 forms an image of the subject 2 on the light receiving surface of the solid-state imaging element using a lens when the shutter is open, converts the intensity of the formed image light into an electric signal, and outputs the electric signal. A video signal including an image signal based on the signal is output to the control device 70.

  A case will be described in which an image (frame) A composed of pixels of 2M rows and N columns as shown in FIG. In this case, the photodiodes (arranged in correspondence with the respective pixels) arranged in a matrix that form the solid-state imaging device generate charges corresponding to the amount of light received when the shutter is open. To do. Then, after the shutter is closed, the imaging unit 20 synchronizes with the vertical synchronization signal and horizontal synchronization signal output from a synchronization signal generation circuit (not shown), and the photodiodes of the respective lines constituting the frame A The charge is read as an electrical signal.

  Here, as shown in FIG. 3, the number when the lines are counted in order from the top is referred to as a line number (that is, in a frame of 2M rows and N columns, line number 1 indicates the first row, and line number 2M indicates 2M). Shows the line). In addition, among the charges of each line constituting the frame A, a line group composed of odd-numbered lines is called an “odd field”, and a line group composed of even-numbered lines is called an “even field”. " The imaging unit 20 performs odd-field imaging and even-field imaging when imaging one frame image. In each imaging, the imaging unit 20 opens and closes a shutter and causes the light receiving element to store electric charges. When the charge of the photodiode is read, the imaging unit 20 reads each charge in the odd field in the odd field imaging, and reads each charge in the even field in the even field imaging. The imaging 20 performs such reading in synchronization with the horizontal synchronization signal. Each line in the odd field and each line in the odd field are alternately arranged.

  Specifically, i) In frame A, the imaging unit 20 starts reading the charges of the lines with odd line numbers in synchronization with the vertical synchronization signal. That is, the imaging unit 20 reads the charges on each line in synchronization with the horizontal synchronization signal in the order of line numbers 1, 3,..., 2M-1. In this way, the imaging unit 20 reads the charge representing the odd field. ii) Thereafter, the imaging unit 20 starts reading the charges of the lines with even line numbers in synchronization with the vertical synchronization signal. That is, the imaging unit 20 reads the charges on each line in the order of line numbers 2, 4,..., 2M in synchronization with the horizontal synchronization signal. In this way, the imaging unit 20 reads the charge representing the even field. That is, the imaging unit 20 performs the imaging of the subject 2 separately for the even field and the odd field.

  Such reading of charges representing odd and even fields constituting one frame is performed at intervals of 1/60 seconds, for example. Therefore, reading of charges representing one frame is performed at intervals of 1/30 seconds. Done. That is, the imaging unit 20 captures an image of 30 frames per second. Then, the imaging unit 20 outputs to the control device 70 a video signal in which an image signal, which is an electrical signal represented by the read charge, a vertical synchronization signal, and a horizontal synchronization signal are superimposed.

  Returning to FIG. 1, the display device 30 displays a subject detection image (subject detection image will be described later) output from the control device 70, and is composed of, for example, an LCD (Liquid Crystal Display).

  The input device 40 receives a user's predetermined input operation and includes, for example, a keyboard, a mouse, and a touch panel that receive a user's numerical input. Data input to the input device 40 is output to the control device 70.

  The speaker 50 is for giving a predetermined instruction to the user in a template acquisition process described later. The speaker 50 rings under the control of the control device 70 to generate a predetermined sound.

  The lighting environment detection device 60 measures various values such as the current date and time, latitude and longitude in order to calculate the estimated ambient light intensity around the subject 2 (hereinafter referred to as “estimated ambient light intensity”), This is for supplying the control device 70 with data based on these measured values. The calculation of the estimated ambient light intensity will be described in detail later. For example, when the viewpoint detection system 1 is mounted on a vehicle, the lighting environment detection device 60 supplies various data for calculating the estimated ambient light intensity near the driver's seat to the control device 70. The illumination environment detection device 60 includes a clock 61, a GPS (Global Positioning System) receiver 62, an orientation sensor 63, and a light sensor 64.

  The clock 61 is for supplying the control device 70 with data indicating the current date and time (hereinafter referred to as “date and time data”). The clock 61 is, for example, a radio clock, and receives radio waves from a predetermined transmitting station using a built-in receiver (not shown). Then, the radio signal is read and the current time is displayed. Further, the clock 61 supplies date and time data to the control device 70.

  The GPS receiver 62 receives a signal from a GPS satellite, and supplies data (hereinafter referred to as “position data”) indicating the current position (latitude / longitude) of the own apparatus to the control device 70. For example, when the viewpoint detection system 1 is mounted on a vehicle, the GPS receiver 62 and the imaging unit 20 are located in the vehicle, and the subject 2 that is an imaging target of the imaging unit 20 is a passenger such as a driver. Therefore, this position data simply indicates the latitude and longitude of the imaging device 20 (or the subject 2 that is the imaging target).

  The direction sensor 63 detects in which direction the light receiving surface of the image sensor included in the image capturing unit 20 is directed (that is, the direction in which the image capturing unit 20 is directed). The azimuth sensor 63 is mounted at a predetermined position of the imaging unit 20, for example, and supplies data indicating the azimuth facing the imaging unit 20 (hereinafter referred to as “azimuth data”) to the control device 70.

  The clock 61, the GPS receiver 62, and the direction sensor 63 constitute a measuring device in the claims. Further, the date / time information received from the GPS hygiene by the GPS receiver 62 may be used as the date / time data. As the azimuth data, for example, when the viewpoint detection system 1 is mounted on a moving body such as a vehicle, azimuth information that can be calculated along with the movement of the moving body may be used.

  The light sensor 64 detects (measures) the intensity of received light. For example, when the viewpoint detection system 1 is mounted on a vehicle, the light sensor 64 is provided on the rear portion of the rearview mirror and detects the intensity of external light incident from the windshield. The light sensor 64 detects light reception in this way and supplies a light detection signal to the control device 70.

  Of the lighting environment detection device 60, the clock 61, the GPS receiver 62, and the direction sensor 63 are provided to digitize information related to the direction of sunlight with respect to the subject 2 (how to digitize). Will be described in detail later). In addition to the light sensor 64 that directly detects the intensity of the received light, the reason for providing these devices for quantifying the information regarding the direction of sunlight with respect to the subject 2 is, for example, that the viewpoint detection system 1 is a vehicle or the like. When the light sensor 64 is hidden in the shadow and the subject 2 is illuminated by the external light, or conversely, the subject 2 is hidden in the shadow and the light sensor 64 is illuminated by the external light, This is because only the light sensor 64 cannot accurately detect the illumination environment in the vicinity of the subject 2 to be imaged.

  By configuring the illumination environment detection device 60 as described above, the light sensor 64 is hidden in the shadow and the subject 2 is illuminated by external light, or conversely, the subject 2 is hidden in the shadow and the light sensor 64 is outside. Even when illuminated by light, the solar light intensity calculated based on the “date and time data”, “position data”, and “azimuth data” supplied from the clock 61, the GPS receiver 62, and the direction sensor 63 By deriving the estimated environment light intensity from the measured environment light intensity calculated based on the “light detection signal” from the light sensor 64, the illumination environment in the vicinity of the subject 2 can be accurately detected. Note that the calculation of the measurement environment light intensity, the estimated environment light intensity, such as sunlight intensity, will be described in detail in the light source control process later.

  The control device 70 is composed of, for example, a microcomputer, and includes a CPU (Central Processing Unit) 71, a ROM (Read-Only Memory) 72, a general-purpose memory 73, a video memory 74, a sync separation circuit 75, and a field identification circuit. 76 and an external device interface (I / F) 77. In addition, the control apparatus 70 comprises the imaging control apparatus and template acquisition apparatus in a claim.

  The CPU 71 reads out and executes a predetermined program stored in advance in the ROM 72 for executing light source control processing, subject detection processing, template acquisition processing, and the like, which will be described later. The CPU 71 then: i) date and time data, position data, orientation data, and light detection signal acquired from the lighting environment detection device 60, ii) a video signal acquired from the imaging unit 20, iii) a calculation result after processing, an image Data and the like are temporarily stored in the general-purpose memory 73. Further, the CPU 71 transfers the output image data stored in the general-purpose memory 73 to the video memory 74, and outputs an image represented by the output image data to the display device 30 based on the transferred output image data. Further, the CPU 71 causes a timer (not shown) to appropriately time.

  The CPU 71 executes a light source control process and a subject detection process, which will be described later, and functions as a direction specifying unit and a control unit in the claims. Further, the CPU 71 executes a template acquisition process to be described later, and functions as an imaging control unit, a subject image acquisition unit, a characteristic area image acquisition unit, and a storage control unit in claims. Further, the CPU 71 functions as an instruction unit in the claims by causing the speaker 50 to ring and generating a predetermined sound.

  In the ROM 72, in addition to various data and the predetermined program, a solar altitude / solar direction specifying table T1 (FIG. 4) and a sunlight intensity specifying table T2 (FIG. a)), solar orientation coefficient specification table T3 (FIG. 5B), measurement environment light intensity specification table T4 (FIG. 5C), lighting time determination table T5 (FIG. 6A), shutter time determination table T6 (FIG. 6B) and the like are stored in advance. These various tables are tables that are referred to in order to obtain the time for lighting the light source 10 (hereinafter referred to as “lighting time T”) and the shutter time of the imaging unit 20. Hereinafter, the various tables will be described in detail.

  The solar altitude / solar direction specifying table T1 shown in FIG. 4 is used to calculate “solar altitude” and “solar direction” from position data and date / time data supplied from the lighting environment detection device 60 to the control device 70. It is a table. “Solar altitude” is the elevation angle of the sun with the horizon as 0 °. The “sun azimuth” is the azimuth angle of the sun when the north is 0 °, the east is 90 °, the south is 180 °, and the west is 270 °. The solar altitude / solar direction specifying table T1 includes a plurality of set values of “latitude / longitude”, “date”, and “time”, and “solar altitude” and “solar direction” determined according to a combination of these set values. Is stored data in advance. The CPU 71 refers to the solar altitude / solar direction specifying table T1 and displays “latitude / longitude” indicated by the position data acquired from the GPS receiver 62 and “date” and “time” indicated by the date / time data acquired from the clock 61. "Solar altitude" and "sun azimuth" corresponding to "." For example, as shown in the figure, “latitude / longitude” is “north latitude 37 degrees 26 minutes / east longitude 138 degrees 50 minutes”, “date” is “January”, and “time” is “12:00 to 13:00”. If there is, the “sun altitude” is specified as “32”, and the “sun direction” is specified as “191”.

The sunlight intensity specifying table T2 shown in FIG. 5A is a table for calculating “sunlight intensity” from “solar altitude”. “Sun intensity” is an estimated value obtained by estimating the brightness (illuminance) of sunlight reaching the subject. The sunlight intensity specifying table T2 is storage data that represents in advance a plurality of set values of “solar altitude” and “solar port intensity” determined corresponding to the plurality of set values. The CPU 71 specifies the “sunlight intensity” corresponding to the “solar height” specified based on the solar altitude / solar direction specifying table T1 with reference to the sunlight intensity specifying table T2. For example, as shown in the figure, when the “solar altitude” is “5 ° to 10 °”, the “sunlight intensity” is specified as “0.131”.
In addition, although sunlight intensity specific table T2 shown to Fig.5 (a) is comprised so that sunlight intensity | strength may increase as the value of solar altitude increases, it is not restricted to this. For example, when the solar altitude ranges from 0 ° to 45 °, the sunlight intensity increases as the solar altitude increases, and when the solar altitude ranges from 45 to 90 °, the sunlight intensity decreases as the solar altitude increases. You may comprise a solar intensity specific table. For example, the viewpoint detection system 1 is mounted on a vehicle, and the influence of a vehicle body that blocks sunlight (such as the ceiling of the vehicle) is considered.

The solar azimuth coefficient specification table T3 shown in FIG. 5B is a table for calculating the “solar azimuth coefficient” from the “angle with the sun azimuth”. Here, the “angle with the sun azimuth” refers to the “sun azimuth” calculated by the CPU 71 with reference to the solar altitude / sun azimuth specifying table T1 and the “imaging unit 20” indicated by the azimuth data acquired from the azimuth sensor 63. It is a value calculated by taking the difference from the “azimuth facing”. The “angle with the sun azimuth” calculated in this way indicates an azimuth angle from the reference azimuth to the sun position when the azimuth facing the imaging unit 20 is used as a reference. Thus, the incident direction of sunlight (that is, the incident direction of sunlight with respect to the subject to be imaged by the imaging unit 20) can be specified. The “solar azimuth coefficient” is obtained by converting the intensity of sunlight into a coefficient in consideration of the influence of the incident azimuth. The solar azimuth coefficient specification table T3 is storage data that represents in advance a plurality of set values of “angle with the sun azimuth” and “solar azimuth coefficients” determined in correspondence with the plurality of set values. The CPU 71 refers to the sun azimuth coefficient identification table T3 and identifies the “solar azimuth coefficient” corresponding to the “angle with the sun azimuth” calculated as described above. For example, as illustrated, when the “angle with the sun azimuth” is “5 ° to 10 °”, the “sun azimuth coefficient” is specified as “0.991”.
In addition, although not shown in figure, the solar orientation coefficient specific table T3 is configured, for example, such that the angle with the solar orientation is 180 ° and the solar orientation coefficient is the minimum value.

  The measurement environment light intensity specification table T4 shown in FIG. 5C is a table for calculating “measurement environment light intensity” from the output of the light detection signal from the light sensor 64 (hereinafter referred to as “light sensor output”). It is. “Measured ambient light intensity” is a value obtained by estimating the brightness (illuminance) of ambient light (that is, sunlight) illuminating the subject. The measurement environment light intensity specification table T4 is stored data that represents in advance a plurality of set values of “light sensor output” and “measurement environment light intensity” determined in accordance with the plurality of set values. The CPU 71 refers to the measurement environment light intensity specification table T4 and specifies “measurement environment light intensity” corresponding to “light sensor output”. For example, as shown in the drawing, when the “light sensor output” is “5 to 10”, the “measurement environment light intensity” is calculated as “0.991”.

  The lighting time determination table T5 shown in FIG. 6A is a table for determining the lighting time T of the light source 10 from the estimated ambient light intensity. The shutter time determination table T6 shown in FIG. It is a table for determining the shutter time of the imaging unit 20 from the estimated ambient light intensity.

  Here, the estimated ambient light intensity is a value calculated by the following (Expression 1) or (Expression 2) based on the sunlight intensity, the solar orientation coefficient, and the measured environment light intensity.

As θ ₀ is a predetermined arbitrary threshold value,
[Measurement light intensity]> θ ₀ ,
[Estimated ambient light intensity] = [Solar orientation coefficient] × [Sunlight intensity] (Equation 1)
on the other hand,
[Measurement environment light intensity] ≦ θ ₀ ,
[Estimated ambient light intensity] = 0 (Expression 2)

For example, when the threshold value θ ₀ is set to 0.3 in advance, the CPU 71 calculates “sunlight intensity” as 0.131, the solar orientation coefficient as 0.991, and the measurement environment light intensity as 0.991. In this case, since the value of the measured environment light intensity is 0.991, which is larger than the value of the threshold θ ₀ = 0.3, the estimated environment light intensity is 0.991 × by (Equation 1). It is obtained by 0.131 and becomes about 0.130. The estimated ambient light intensity is an estimate of the brightness of the light that illuminates the subject 2, and takes a larger value as the estimated brightness becomes brighter, for example.

  The lighting time determination table T5 is stored data that represents a plurality of set values of “estimated ambient light intensity” and “lighting time T” determined in advance corresponding to the plurality of set values. The CPU 71 refers to the lighting time determination table T5 to determine the “lighting time T” corresponding to the “estimated ambient light intensity” calculated by the above (Expression 1) or (Expression 2). For example, as illustrated, when the “estimated ambient light intensity” is “0.01 to 0.02”, the “lighting time T” is determined to be “1/100 ms”.

  The shutter time determination table T6 is stored data that preliminarily represents a plurality of set values of “estimated ambient light intensity” and “shutter times” determined in accordance with the plurality of set values. The CPU 71 refers to the shutter time determination table T6 to determine “shutter time” corresponding to the “estimated ambient light intensity” calculated by the above (Expression 1) or (Expression 2). For example, as illustrated, when the “estimated ambient light intensity” is “0.01 to 0.02”, the “shutter time” is determined to be “1/10 s”. When the CPU 71 determines the shutter time as described above, the CPU 71 causes the imaging unit 20 to alternately perform the imaging of the odd field and the imaging of the even field with the shutter time. In addition, the CPU 71 turns on the light source 10 only in the imaging of the odd field performed by the imaging unit 20 by executing a light source control process described later.

  That is, among the various tables, the tables T1 to T4 are various parameters for calculating the estimated environment light intensity by the above (Expression 1) or (Expression 2) based on various information from the illumination environment detection device 60. It is a table for seeking. The lighting time determination table T5 is a table for determining the lighting time T from the calculated estimated ambient light intensity, and the shutter time determination table T6 is for determining the shutter time from the calculated estimated ambient light intensity. It is a table.

  Returning to FIG. 1, the synchronization separation circuit 75 is a circuit that separates the video signal from the imaging unit 20 into an image signal, a vertical synchronization signal, and a horizontal signal.

  The field identification circuit 76 generates a field identification signal indicating whether the image signal included in the video signal is an odd field or an even field from the vertical synchronization signal and the horizontal synchronization signal separated by the synchronization separation circuit 75. . As an example, the field identification signal is configured as shown in FIG. 7B. Based on such a field identification signal, the CPU 71 determines whether the image signal acquired from the imaging unit 20 is an odd field or an even field. Can be determined.

  The external device I / F 77 connects the light source 10, the imaging unit 20, the display device 30, the input device 40, the speaker 50, and the illumination environment detection device 60 to each unit of the control device 70.

(Light source control processing)
The light source control process executed by the CPU 71 of the control device 70 will be described with reference to FIG. This process is started on condition that the control device 70 is activated, for example.

  When starting the light source control process, the CPU 71 first acquires various types of information including date / time data, position data, orientation data, and light detection signals from the illumination environment detection device 60 (step S101).

  Next, the CPU 71 determines a period during which the light source 10 is turned on (hereinafter referred to as “lighting time T”) based on the various information acquired in step S101 (step S102).

Specifically, the CPU 71 determines the lighting time T according to the following procedure.
1) First, the CPU 71 determines “latitude / longitude” based on the position data acquired from the GPS receiver 62, and determines “date” and “time” based on the date / time data acquired from the clock 61. Then, the CPU 71 calculates the “solar altitude” and the “sun azimuth” with reference to the solar altitude / sun direction specifying table T1.
2) Next, the CPU 71 calculates “sunlight intensity” from the “solar altitude” calculated in 1) above with reference to the sunlight intensity specifying table T2.
3) Further, the CPU 71 calculates the difference between the “sun azimuth” calculated in the above 1) and the “azimuth of the imaging unit 20” indicated by the azimuth data acquired from the azimuth sensor 63, and the “angle with the sun azimuth” Is calculated. Then, the CPU 71 calculates the “solar azimuth coefficient” with reference to the solar azimuth coefficient specification table T3.
4) Further, the CPU 71 specifies the light sensor output, refers to the measurement environment light specification table T4, and calculates “measurement environment light intensity”.
5) Then, the CPU 71 calculates the “estimated environment” from the “sunlight intensity” obtained in the above 2) and the “solar orientation coefficient” obtained in the above 3) according to the above (formula 1) or (formula 2). Light intensity "is calculated.
6) Then, the CPU 71 determines “lighting time T” corresponding to the “estimated ambient light intensity” calculated in 5) above with reference to the estimated ambient light intensity specification table T5. In addition, the CPU 71 temporarily stores the determined lighting time T in the general-purpose memory 73. The lighting time T determined here is stored, for example, until the lighting time T is newly determined in the next processing.

  Subsequently, in step S103, the CPU 71 determines whether or not the falling edge of the vertical synchronization signal (see FIG. 7A) is detected. As described above, the vertical synchronization signal is output from the synchronization separation circuit 75 that separates the video signal from the imaging unit 20 into an image signal, a vertical synchronization signal, and a horizontal signal. If the CPU 71 determines that the falling edge of the vertical synchronization signal has not been detected (step S103; No), the process proceeds to step S106. On the other hand, if the CPU 71 determines that the falling edge of the vertical synchronization signal has been detected (step S103; Yes), the process proceeds to step S104.

In step S104, the CPU 71 determines whether or not the field identification signal acquired from the field identification circuit 76 indicates an odd field (see FIG. 7B). Specifically, when the rising edge of the field identification signal is detected, the CPU 71 determines that the field identification signal is an odd field.
When it is determined that the field identification signal does not indicate an odd field (that is, indicates an even field) (step S104; No), the CPU 71 returns the process to step S103. On the other hand, when it is determined that the field identification signal indicates an odd field (step S104; Yes), the CPU 71 advances the process to step S105.

  In step S105, the CPU 71 sets the time of a timer (not shown) to t = 0 and causes the timer to start measuring time.

Subsequently, in step S106, the CPU 71 determines whether or not the time t satisfies T _ON ≦ t ≦ T _OFF . Here, T _ON and T _OFF are the start of the time during which charges corresponding to the amount of light received by the photodiode corresponding to the odd field among the photodiodes constituting the solid-state imaging device of the imaging unit 20 are generated. Corresponds to the time and end time. Further, (T _OFF −T _ON ), that is, the lighting time T is a value that is determined by the process in step S <b> 102 and is temporarily stored in the general-purpose memory 73. Incidentally, T _ON is either stored in advance in ROM72 as a default value is set through the input device 40 by the user is stored.

When it is determined that the time count t satisfies T _ON ≦ t ≦ T _OFF (step S106; Yes), the CPU 71 outputs a light source control signal (see FIG. 7C) indicating lighting to the light source 10 (step S107). ). Then, the process returns to step S101. On the other hand, when it is determined that the time count t does not satisfy T _ON ≦ t ≦ T _OFF (step S106; No), the CPU 71 outputs a light source control signal indicating turning off to the light source 10 (step S108). Then, the process returns to step S101.

  For example, the CPU 71 repeats the light source control processing including the above processing until the control device 70 is powered off. With this processing, the light source 10 can be turned on when the imaging unit 20 is imaging an odd field, and the light source 10 can be turned off when the imaging unit 20 is imaging an even field.

(Viewpoint detection processing)
Next, viewpoint detection processing executed by the CPU 71 of the control device 70 will be described with reference to FIG. This process is started on condition that the control device 70 is activated, for example.

  When starting the subject detection process, the CPU 71 first receives various information including date / time data, position data, azimuth data, and a light detection signal from the illumination environment detection device 60 as in the process of step S101 in the light source control process described above. Obtain (step S201).

  Next, the CPU 71 determines the shutter time of the imaging layer 20 based on the various information acquired in step S201, and causes the imaging unit 20 to capture an image with the shutter time (step S202). Specifically, the estimated ambient light intensity is calculated by a method similar to that in step S102 in the light source control process described above, and the estimated estimated ambient light calculated with reference to the shutter time determination table T6 (FIG. 6B). The “shutter time” corresponding to the intensity is determined. Then, the imaging unit 20 is caused to capture an image with the determined shutter time.

  Subsequently, in step S203, the CPU 71 acquires a frame image represented by the image signal for one frame based on the vertical synchronization signal from the image signal included in the video signal output from the imaging unit 20 (step S203). Specifically, the CPU 71 is configured by one odd field image signal and one even field image signal from the image signal included in the video signal output from the imaging unit 20 in synchronization with the vertical synchronization signal. An image signal for one frame is acquired. For example, when the imaging unit 20 images the subject 2 as shown in FIG. 2, the frame image acquired by the CPU 71 is a frame image B as shown in FIG. During imaging by the imaging unit 20, since the CPU 71 executes the above-described light source control processing, in the frame image B, the subject image in the odd field is bright and the subject image in the even field is dark. This is because the light source 10 that illuminates the subject 2 to be imaged by the imaging unit 20 is turned on when the imaging unit 20 is imaging an odd field, and the imaging unit 20 is imaging an even field. This is because the light source 10 can be turned off. Since the background 3 shown in FIG. 2 is sufficiently away from the light source 10, the background portion of the frame image B corresponding to the background 3 is not affected by the light source 10.

  Subsequently, in step S204, the CPU 71 separates the frame image acquired in step S203 into an odd field and an even field, and acquires an odd field image and an even field image composed of each field. Specifically, for example, when the frame image B shown in FIG. 10 is acquired, the CPU 71 detects the odd field image C1 representing the odd field image shown in FIG. 11A and the even field shown in FIG. An even field image C2 representing an image is acquired. Both the odd field image C1 and the even field image C2 are composed of half the number of lines of the frame image B.

  Subsequently, in step S205, the CPU 71 acquires a first interpolation frame image and a second interpolation frame image from the odd field image and the even field image acquired in step S204 (step S205).

  Here, the first frame image is a frame image in which lines are newly interpolated between the lines of the odd field image, and the second frame image is a new line between the lines of the even field image. This is an interpolated frame image. Specifically, as shown in FIGS. 11A and 11B, the odd-numbered field image C1 and the even-numbered field image C2 configured with half the number of lines of the frame image B are the same as the frame image B, respectively. By interpolating a new line between each line so as to be the number of lines, from the odd field image C1, the first interpolation frame image D1 and the even field image C2 shown in FIG. A second interpolation frame image D2 shown in 12 (b) is acquired.

  As an example of the interpolation method, the first interpolation frame image D1 can be acquired by interpolating a line having an average value of luminance values of upper and lower pixels as luminance values between the lines of the odd field image C1. Similarly, the second interpolation frame image D2 can be acquired by interpolating a line having the average value of the luminance values of the upper and lower pixels as the luminance value between the lines of the even field image C2.

なお、Ｉ_ｍｉｎは、フレーム画像Ｂの輝度値Ｉの最小値を表す。 Specifically, when the frame image B having the number of images of 2M rows and N columns and the luminance value of the pixel in the i row and j column represented by I _{i, j} is acquired by the frame image acquisition unit 70c, 'and _{i, j,} the luminance value I of the second interpolated frame image D2' luminance value I of the first interpolated frame image D1 _{'i, j} can be expressed by the following equation.

I _min represents the minimum value of the luminance value I of the frame image B.

  Returning to FIG. 9, when the CPU 71 obtains the first interpolation frame image and the second interpolation frame image in step S205, the process proceeds to step S206.

In step S206, the CPU 71 acquires a subject detection image based on the comparison between the first interpolation frame image and the second interpolation frame image. Specifically, the CPU 71 compares the luminance values of the pixels at the corresponding positions for the first interpolation frame image D1 and the second interpolation frame image D2 as shown in FIGS. 12 (a) and 12 (b). When the difference between the luminance values of the pixels is larger than the predetermined value θ, it is determined that the pixel at the position is a pixel in the subject area. Then, the CPU 71 performs this determination process on the entire area of the first interpolation frame image D1 and the second interpolation frame image D2, and determines the minimum luminance value of the frame image B for the pixels determined to be pixels within the subject area. A subject detection image E shown in FIG. 13 is obtained, which has the value I _min as the luminance value and the other pixels having the maximum luminance value I _max of the frame image B as the luminance value. Specifically, the luminance value I ′ ″ _{i, j} of the subject detection image E is given by the following equation.

Specifically, for example, when the first interpolation frame image D1 and the second interpolation frame image D2 as shown in FIGS. 12A and 12B are acquired in step S205, the CPU 71 acquires the first interpolation frame image. For the D1 and the second interpolated frame image D2, the luminance value of the pixel at the corresponding position is compared using the above (Equation 9), and the difference in the luminance value of the pixel is larger than the predetermined value θ. The pixel at that position is determined to be a pixel in the subject area. Then, the luminance value of the pixel at the position determined as the subject region is the minimum value I _min of the luminance value of the frame image B, and the luminance value of the other pixels is the maximum value I _max of the luminance value of the frame image B. A subject detection image E (FIG. 13) is acquired.

  For example, in this process, when a person who has boarded the vehicle is imaged by the imaging unit 20 as shown in FIG. 14A, the subject detection image acquired by the CPU 71 is as shown in FIG. 14B. Become. Thus, according to the subject detection process, only an image of a person who is an imaging target can be obtained as a subject detection image.

  Subsequently, in step S207, the CPU 71 executes viewpoint detection processing to detect eye coordinates on the subject detection image obtained in step S206. This viewpoint detection processing is realized by pattern matching, which is a well-known technique, as conceptually shown in FIG. Specifically, the CPU 71 reads the template TE stored in the general-purpose memory 73 and detects the coordinates of the eyes by searching the image for the region closest to the template TE. The template TE is acquired by the CPU 71 by executing a template acquisition process, which will be described in detail later, and is stored in the general-purpose memory 73.

  Subsequently, in step S208, the CPU 71 causes the display device 30 to display the subject detection image E. By this display, the user can confirm the detection result.

  The CPU 71 repeats the subject detection process including the above processes until, for example, the control device 70 is powered off.

(Template acquisition process)
Next, template acquisition processing executed by the CPU 71 of the control device 70 will be described with reference to FIGS. 16 and 17. This process is started with a signal that the control device 70 has acquired a predetermined command signal, for example, by a user operation from the input device 40.

  When starting the template acquisition process, the CPU 71 first issues an eye opening instruction by ringing the speaker 50 (step S301). Specifically, the speaker 50 is sounded to generate a predetermined sound stored in advance in the ROM 72 or the like, for example, a sound “please open your eyes” to prompt the user (subject 2) to open his eyes.

  Subsequently, in step S302, the CPU 71 causes the image capturing unit 20 to capture an image and acquires a frame image (the acquired frame image is referred to as an “open eye frame image”). Specifically, when the imaging unit 20 images a person according to the eye opening instruction in step S301, the CPU 71 acquires an eye opening frame image F0 as shown in FIG.

  Subsequently, in step S303, the CPU 71 issues an eye-close instruction by ringing the speaker 50 (step S303). Specifically, the speaker 50 is sounded to generate a predetermined sound stored in advance in the ROM 72 or the like, for example, a sound “Please close your eyes” to prompt the user (subject 2) to close his eyes.

  Subsequently, in step S304, the CPU 71 causes the imaging unit 20 to capture an image and obtain a frame image (the obtained frame image is referred to as a “closed eye frame image”). Specifically, when the imaging unit 20 images a person according to the eye closing instruction in step S303, the CPU 71 acquires a closed eye frame image F1 as shown in FIG.

  Note that imaging by the imaging unit 20 in step S302 and step S304 may be started under the control of the CPU 71 on the condition that a predetermined time has elapsed from the sounding of the speaker 50, or the user who is ready for imaging can input the input device 40. A signal indicating that a predetermined input operation has been performed may be started.

  Next, the CPU 71 calculates a difference absolute value between the acquired open eye frame image and the closed eye frame image by a known image processing method (step S305). Then, the CPU 71 performs binarization processing by a known image processing method (step S306). Specifically, when the obtained absolute difference value between the obtained eye opening frame image F0 shown in FIG. 17A and the eye closing frame image F1 shown in FIG. 17B is calculated and binarization processing is executed, the CPU 71 Can obtain a binary image F2 as shown in FIG. 17C in which the region changed between the open eye frame image F0 and the closed eye frame image F1 is emphasized. In other words, the CPU 71 obtains a binary image F2 in which the eye area in the subject image is emphasized. In addition, the dotted line in FIG.17 (c) does not show a part of frame image but is an auxiliary line.

  The binary image obtained in step S306 usually includes various noises in addition to the eye region due to the influence of the background and the movement of the subject. Therefore, in step S307, the CPU 71 calculates the area of the target region (a preset white or black region) in the binary image in order to remove the region other than the eyes. Then, in step S307, the CPU 71 selects the top two of the areas calculated in step S306 in the order of area size (where R1 is the area with the largest area, and the area with the next largest area). , R2), and R1 and R2 are the right eye region and the left eye region, respectively, considering the positional relationship between R1 and R2. Specifically, in the binary image F2 as shown in FIG. 17C, the CPU 71 sets R1 as the left eye region and R2 as the right eye region as illustrated.

Subsequently, in step S309, the CPU 71 calculates the center of gravity of each of R1 and R2. In step S310, the CPU 71 cuts out rectangular patterns S1 and S2 (see FIG. 17D) centered on the centers of gravity of R1 and R2 from the eye-opening frame image F0, and uses them as a template TE for general-purpose memory 73. And the process is terminated. Here, the stored template TE is a template used in the viewpoint detection process (step S207 of subject detection process).
The above is the template acquisition process.

  The control device 70 according to the present embodiment includes an imaging unit 20 that performs imaging of the subject 2 (person) separately for odd field / even field imaging, and a light source 10 that emits light to the subject 2 captured by the imaging unit 20. The light source 10 is caused to emit light only in the imaging of the odd field, and the CPU 71 determines the direction of sunlight with respect to the subject 2 based on the measurement values measured by the illumination environment detection device 60 (based on various data and light detection signals). ) Specify (here, the specification includes calculating the direction of sunlight with respect to the subject 2 based on various data and light detection signals), and based on the specified direction of sunlight, the light source 10 The lighting time T (light emission time) is controlled. The control device 70 (CPU 71) having such a configuration includes date and time data, position data, and direction data acquired from each of the clock 61, the GPS receiver 62, and the direction sensor 63, and a light detection signal acquired from the light sensor 64. Based on the above, the estimated ambient light intensity around the subject 2 is calculated. Thus, according to the control device 70 according to the present embodiment, it is possible to accurately estimate the ambient light. In particular, for example, when the viewpoint detection system 1 is mounted on a moving body such as a car, environmental light around the subject 2 to be imaged is captured even when a part of the interior of the car is hidden behind a shadow such as a utility pole. It can be estimated with high accuracy. For example, when the viewpoint detection system 1 is mounted on a vehicle, at least the clock 61 and the GPS receiver 62 among the components constituting the illumination environment detection device 60 are generally mounted devices on the vehicle. The detection system 1 can be configured easily.

  Further, the CPU 71 of the control device 70 estimates the brightness of light that illuminates the subject 2 based on the identified direction of sunlight (calculates the estimated ambient light intensity). The lighting time T of 10 is controlled to be long. In this way, for example, even when the ambient light around the subject 2 is too bright, the lighting time T of the light source 10 can be controlled to be long, and the first interpolation frame image and the second interpolation frame image can be controlled. A difference in brightness can be obtained. Thereby, a subject detection image can be acquired with high accuracy.

  Further, the control device 70 calculates the lighting time T (light emission time) of the light source 10 and the shutter time of the imaging unit 20 based on the estimated ambient light intensity calculated as described above, and based on these, the light source 10 and the imaging are calculated. The drive of the unit 20 is controlled. Thereby, the control apparatus 70 can acquire the subject detection image which is not easily influenced by the background, as illustrated in FIGS. 13 and 14B. That is, since the control device 70 can accurately acquire a subject detection image such as a person's face for the viewpoint detection process, it is possible to reduce the variation in the detection result in the viewpoint detection.

  In addition, the CPU 71 of the control device 70 measures the current date and time, the latitude and longitude of the imaging unit, and the orientation that the imaging unit faces (date and time data, position data, and orientation data) measured by the lighting environment detection device 60. ), The direction of sunlight with respect to the subject 2 is specified, so that the influence of external light around the subject 2 can be accurately estimated. Furthermore, the control device 70 in the present embodiment is also based on the illuminance of the ambient light around the subject 2 measured by the light sensor 64 (measured ambient light intensity calculated based on the light detection signal acquired from the light sensor 64). Since the lighting time T (light emission time) of the light source 10 and the shutter time of the imaging device 20 are determined, the lighting environment around the subject 2 is accurately estimated, and the subject 2 is set with a setting suitable for the lighting environment. An image can be taken.

  Further, the CPU 71 of the control device 70 according to the present embodiment executes a template acquisition process, issues an eye opening instruction and an eye closing instruction to the person who is the subject, and causes the imaging unit 20 to image the subject 2 for each instruction. Then, subject images corresponding to the number of times of imaging are obtained from the imaging unit 20, and the obtained subject images are compared to detect a region near the human eye in the subject image, and an image indicating the detected region is obtained. The acquired image is stored in the storage unit as a pattern matching template in the viewpoint detection process for detecting the viewpoint of the person. As a result, pattern matching can be performed based on a template that captures the characteristics of each person who is the target of viewpoint detection instead of a standard template, so that individual differences in the success rate in viewpoint detection can be reduced.

  In addition, the control device 70 according to the present embodiment uses the light source 10 based on the vertical synchronization signal included in the video signal output from the imaging unit 20 and the light source 10 when the imaging unit 20 is imaging an odd field. The light source 10 can be turned off and the light source 10 can be turned off when the imaging unit 20 is imaging an even field. Then, in the captured image represented by the image signal included in the video signal output from the imaging unit 20, the subject area is detected by utilizing the fact that the luminance values of the pixels of the subject image in the odd field and the even field are different. Therefore, the control device 70 can detect a region where an object is present in an image with a simple configuration without requiring a camera capable of high-speed shooting.

(Modification)
In addition, this invention is not limited to the above embodiment, A various deformation | transformation and application are possible. In the above embodiment, the lighting time T of the light source 10 and the shutter time of the imaging unit 20 are determined based on various information from the lighting environment detection device 60, but only the lighting time T may be determined. In this case, the shutter time may be a preset value, or may be a value that can be appropriately set by a user operation from the input device 40. Thus, even if only the lighting time T is determined, the control device 70 can cause the imaging unit 20 to image the subject 2 with settings suitable for the illumination environment around the subject.

  Moreover, although the illumination environment detection apparatus 60 showed the example provided with the light sensor 64 in the above embodiment, it is not restricted to this. The control device may calculate the estimated environment light intensity based on information supplied from an illumination environment detection device including a clock, a GPS receiver, and a direction sensor. In this case, it is only necessary to cause the control device to calculate the estimated ambient light intensity by the above (Equation 1).

  The lighting environment detection device may be, for example, a mobile terminal (a mobile phone, a smartphone, or the like) having at least a GPS reception function and a direction sensor function.

  Further, in the above embodiment, the direction sensor 63 detects which direction the light receiving surface of the imaging unit 20 is oriented, but is not limited thereto. The direction sensor may not detect the normal direction of the light receiving surface, and may detect the direction parallel to the light receiving surface, for example. The direction to be detected is arbitrary as long as the positional relationship between the sun and the imaging unit 20 (or subject 2) can be specified.

  In the above embodiment, when the control device 70 (CPU 71) detects the falling edge of the vertical synchronization signal and the field identification signal indicates an odd field, the timer is set at time t = 0. However, the control device 70 may set the timer at time t = 0 when the rising edge of the vertical synchronization signal is detected and the field identification signal indicates an odd field. Also in this case, the control apparatus 70 can control the light source 10 similarly to said embodiment.

  Further, the control device 70 according to the above embodiment turns on the light source 10 when the imaging unit 20 is imaging an odd field, and turns off the light source 10 when the imaging unit 20 is imaging an even field. However, it is not limited to this. The light source 10 may be turned on when the imaging unit 20 is imaging an even field, and the light source 10 may be turned off when the imaging unit 20 is imaging an odd field.

  In the template acquisition process according to the above embodiment, the CPU 71 stores the rectangular patterns S1 and S2 cut out from the eye-opening frame image F0 as the template TE in the general-purpose memory 73, but is not limited thereto. The rectangular patterns S1 and S2 may be cut out from the closed eye frame image F1 and stored as a template. Further, the CPU 71 may store the template cut out from the open eye frame image F0 and the template cut out from the closed eye frame image F1 in the general-purpose memory 73 or the like. In this case, when detecting the viewpoint of the person (subject) in the viewpoint detection processing (subject detection processing step S207), the CPU 71 is not limited to detecting the viewpoint with reference to the template TE from the eye-opened frame image F0. The viewpoint may be detected by referring to the template from the closed eye frame image F1, or the viewpoint may be detected by referring to the templates related to both the opened eye frame image F0 and the closed eye frame image F1. Further, the image that is cut out from the open eye frame image F0 and the closed eye frame image F1 and used as a template may not be a rectangular pattern.

  Further, in the template acquisition process according to the above-described embodiment, an image near the eyes of a person who is a subject is acquired by instructing eye opening and closing, but the present invention is not limited thereto. For example, an image of the vicinity of the person's mouth that is the subject may be acquired by instructing the person to open and close, and the image may be used as a template for pattern matching. In this case, by estimating from the position of the mouth, it is possible to detect the direction of the viewpoint of the person who is the subject and the face of the person.

  Further, in the template acquisition process according to the above-described embodiment, instructions for opening and closing eyes are given to a person by voice, but the present invention is not limited to this. The instruction may be given by electrical stimulation, physical stimulation by vibration, or the like.

  In addition, the control apparatus 70 which concerns on embodiment of this invention is realizable using a normal computer system irrespective of a dedicated apparatus. For example, the above-described processing is executed by installing the program from a medium (flexible disk, CD-ROM, DVD-ROM, etc.) storing the program for executing the above-described processing in a computer equipped with a network card or the like. A control device can be configured.

  Any method for supplying the program to the computer is arbitrary. For example, you may supply via a communication line, a communication network, a communication system, etc. For example, the above-described processing can be executed by posting the program on a bulletin board (BBS) of a communication network and distributing it on a carrier wave via the network.

DESCRIPTION OF SYMBOLS 1 Viewpoint detection system 10 Light source 20 Imaging part 30 Display apparatus 40 Input apparatus 50 Speaker 60 Lighting environment detection apparatus 61 Clock 62 GPS receiver 63 Direction sensor 64 Light sensor 70 Control apparatus 71 CPU
72 ROM
73 General-purpose memory 74 Video memory 75 Sync separation circuit 76 Field identification circuit 77 External device I / F
2 Subject 3 Background

Claims (11)

An imaging unit that performs imaging of a subject divided into imaging of a plurality of fields, and a light source that irradiates light to the subject that is captured by the imaging unit, and controls imaging of a predetermined part of the imaging of the plurality of fields An imaging control device that emits the light source only in
Direction specifying means for specifying the direction of sunlight with respect to the subject based on a measurement value measured by a predetermined measuring device;
Control means for controlling the light emission time of the light source based on the direction of the sunlight specified by the direction specifying means,
An imaging control apparatus characterized by that. The control means includes
Based on the direction of the sunlight specified by the direction specifying means, the brightness of the light that illuminates the subject is estimated, and as the estimated brightness is brighter, the emission time of the light source is controlled longer.
The imaging control apparatus according to claim 1. The control means includes
Based on the direction of sunlight specified by the direction specifying means, the light emission time of the light source and the shutter time of the imaging unit are controlled.
The imaging control apparatus according to claim 1, wherein the imaging control apparatus is configured as described above. The control means includes
Based on the direction of the sunlight specified by the direction specifying means, the brightness of the light that illuminates the subject is estimated, and as the estimated brightness is brighter, the shutter time of the imaging unit is controlled longer.
The imaging control apparatus according to claim 3. The direction specifying means includes
Specifying the direction of sunlight with respect to the subject based on the current date and time measured by the predetermined measuring device, the latitude / longitude of the imaging unit, and the direction in which the imaging unit faces;
The imaging control apparatus according to any one of claims 1 to 4, wherein the imaging control apparatus is characterized in that: The control means includes
Controlling the emission time of the light source based on the direction of the sunlight specified by the direction specifying means and the illuminance of the ambient light around the subject measured by a light sensor;
The imaging control apparatus according to any one of claims 1 to 5, wherein The control means includes
Controlling the shutter time of the imaging unit based on the direction of the sunlight specified by the direction specifying means and the illuminance of ambient light around the subject measured by a light sensor;
The imaging control apparatus according to claim 3, wherein the imaging control apparatus is characterized. On the computer,
An imaging unit that performs imaging of a subject divided into imaging of a plurality of fields, and a light source that irradiates light to the subject that is captured by the imaging unit, and controls imaging of a predetermined part of the imaging of the plurality of fields A program for executing the process of causing the light source to emit light only in
A process for identifying the direction of sunlight with respect to the subject based on a measurement value measured by a predetermined measurement device;
A process for controlling a light emission time of the light source based on the identified direction of the sunlight;
A program that executes Instruction means for giving a predetermined instruction to a person as a subject a plurality of times;
Imaging control means for causing the imaging unit to image the subject for each predetermined instruction;
Subject image acquisition means for acquiring subject images for the number of times of imaging from the imaging unit that images the subject;
A feature region image acquisition unit that detects a feature region of a person in the subject image by comparing the acquired subject images for the number of times of imaging, and acquires a feature region image indicating the detected feature region;
Storage control means for storing the characteristic area image acquired by the characteristic area image acquisition means in a storage means as a template for pattern matching in the process of detecting the viewpoint of the person,
A template acquisition apparatus characterized by that. The plurality of instructions performed by the instruction means includes an eye opening instruction and an eye closing instruction,
The subject image acquisition means includes
Obtaining a first subject image corresponding to an eye opening instruction and a second subject image corresponding to an eye closing instruction;
The feature region image acquisition means includes
By comparing the first subject image and the second subject image, a region near the eyes of the person in the subject image is detected, and an image showing the region near the eyes is obtained,
The storage control means
Storing the acquired image showing the area near the eyes of the person in the storage means as a pattern matching template in the process of detecting the viewpoint of the person;
The template acquisition apparatus according to claim 9. On the computer,
A process of giving a predetermined instruction to a person as a subject a plurality of times;
A process of causing the imaging unit to image the subject for each predetermined instruction;
Processing for acquiring subject images for the number of times of imaging from the imaging unit that images the subject;
A process of detecting a feature region of a person in the subject image by comparing the obtained subject images for the number of times of imaging, and acquiring a feature region image indicating the detected feature region;
A process of storing the acquired feature area image in a storage unit as a pattern matching template in the process of detecting the viewpoint of the person;
A program that executes

Images