JP5272826B2 - Call determination apparatus, method thereof, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a phone call decision device and the like for determining whether an imaging subject person is talking on a mobile phone by analyzing image information of the imaging subject person. <P>SOLUTION: The device includes a face detection unit 320 which detects the face of an operator 101 from image information taken by a camera 120; a hand area specification unit 330 which specifies an image area of hand in a call state of mobile phone based on the detected face position; and a call decision unit 340 which determines whether the operator 101 is talking on the mobile phone 140 based on the change quantity of pixel value in the specified hand area. <P>COPYRIGHT: (C)2010,JPO&amp;INPIT

Description

  The present invention relates to a call determination device for determining whether a mobile phone is busy.

  In bank ATMs, wire fraud, which guides ATM operations with mobile phones, is a problem. Therefore, measures such as sending jamming radio waves near ATMs, prohibiting the use of mobile phones, detecting voices near ATMs, and giving warnings based on multiple keywords used when talking to criminals. Has been made.

  In addition, a technique for analyzing the shape of a person's hand and the state of operation by image processing is disclosed (for example, see Patent Document 1 or 2).

Japanese Patent No. 3863809 Special Table 2002-503856 gazette

  However, if the mobile phone cannot be used by causing jamming radio waves in the vicinity of the ATM, the surrounding mobile phone cannot be used uniformly, and people other than the ATM operator will feel inconvenience. In addition, since it is impossible to wait for reception, there is a problem that it is impossible to receive an emergency contact or the like. Also, if the voice near the ATM is detected and a warning is made based on a plurality of keywords used when talking to the criminal, it may be unclear whether the voice is from the ATM operator or privacy infringement may occur. There is a problem of having sex. Therefore, it is necessary to determine whether the mobile phone is in use only for an ATM operator.

  The technique shown in Patent Document 1 has a problem that a hand cannot be recognized when a glove is worn because a skin color portion is extracted from an image of a hand. Further, since it is assumed that a hand is captured in the input image, there is a problem that a function for detecting a hand from an image captured of the operator, such as an ATM operation, is not provided.

  The technique shown in Patent Document 2 has a problem that an assumed pattern must be registered in advance in order to perform recognition in comparison with a predetermined pattern. In addition, it recognizes the state of the hand as it is or together with the object to be held in the hand (for example, a telephone handset), but it cannot be determined whether or not the telephone is on, and the handset of the telephone is held. It may be recognized that you are making a phone call.

An accident caused by the use of a mobile phone while driving a car is also a problem.
Accordingly, the present invention has been made to solve the above-described problems, and provides a call determination device that analyzes image information of a person to be imaged and determines whether or not a mobile phone call is in progress. With the goal.

  A call determination device disclosed in the present application includes a face detection unit that detects a face of a person to be imaged from image information captured by a camera. In addition, hand region specifying means for specifying an imaging region of a hand in a call state of the mobile phone based on the face position detected by the face detecting means is provided. Furthermore, a call determination unit that determines whether or not the person to be imaged is talking on a mobile phone based on the amount of change in the pixel value in the imaging region specified by the hand region specifying unit.

  As described above, in the call determination device disclosed in the present application, the imaging region of the hand during a mobile phone call is specified based on the face position, and the call is determined based on the amount of change in the imaging region of the hand. Therefore, there is an effect that it is possible to accurately recognize that the call is in progress.

1 is a system configuration diagram of a busy detection system using a call determination device according to a first embodiment. It is a hardware block diagram of the telephone call determination apparatus which concerns on 1st Embodiment. It is a functional block diagram of the call determination device according to the first embodiment. It is a flowchart which shows operation | movement of the telephone call determination apparatus which concerns on 1st Embodiment. It is a flowchart which shows the face detection process of the telephone call determination apparatus which concerns on 1st Embodiment. It is a flowchart which shows the process which pinpoints the area | region of the hand of the telephone call determination apparatus which concerns on 1st Embodiment. It is a figure which shows the area | region of the hand specified in the telephone call determination apparatus which concerns on 1st Embodiment. It is a flowchart which shows the process which determines during the call of the mobile telephone of the call determination apparatus which concerns on 1st Embodiment. It is a figure which shows the case where the curve component, the end point of a line segment, a parallel component, and "C" shape are extracted in the telephone call determination apparatus which concerns on 1st Embodiment. It is a figure which shows the case where the difference of the color tone of C shape and the color tone of the center of C shape is extracted in the telephone call determination apparatus which concerns on 1st Embodiment.

  Embodiments of the present invention will be described below. The present invention can be implemented in many different forms. Therefore, the present invention should not be construed based only on the description of the present embodiment. Also, the same reference numerals are given to the same elements throughout the present embodiment.

  In the following embodiments, the apparatus will be mainly described. However, as is apparent to those skilled in the art, the present invention can also be implemented as a method and a program for operating a computer. In addition, the present invention can be implemented in hardware, software, or hardware and software embodiments. The program can be recorded on any computer-readable medium such as a hard disk, CD-ROM, DVD-ROM, optical storage device, or magnetic storage device. Furthermore, the program can be recorded on another computer via a network.

(First embodiment of the present invention)
A call determination device according to the present embodiment will be described with reference to FIGS.
(1. Configuration)
FIG. 1 is a system configuration diagram of a during-call detection system using the call determination device according to the present embodiment. The call detection system 100 includes a call determination device 110, a camera 120, and an ATM 130. The camera 120 captures an image of the operator 101 operating the ATM 130. The captured image information is transmitted to the call determination device 110 for image analysis. As a result of the image analysis, when the operator 101 is talking on the mobile phone 140, the fact is notified to the ATM 130 or a terminal used by the administrator 102 (not shown). Based on the notified information, the operator 101 continues a call with the mobile phone 140 or immediately hangs up the mobile phone 140. Further, the administrator 102 may request the operator 101 to interrupt the call directly.

  FIG. 2 is a hardware configuration diagram of the call determination device according to the present embodiment. The call determination device includes a CPU 210, a RAM 220, a ROM 230, a hard disk (HD) 240, a communication I / F 250, and an input / output I / F 260. The ROM 230 and the HD 240 store an operating system (referred to as an OS), various programs, and the like, which are read into the RAM 220 as necessary, and are executed by the CPU 210. The communication I / F 250 is an interface for communicating with other devices (here, the camera 120 and the ATM 130). The input / output I / F 260 is an interface for receiving input from a keyboard, a mouse, and the like, and outputting data to a printer, a display, a speaker, and the like. As this input / output I / F 260, USB, RS232C, or the like is used. Further, if necessary, a drive corresponding to a removable disk such as a magneto-optical disk, a floppy disk (registered trademark), a CD-R, or a DVD-R can be connected.

  FIG. 3 is a functional block diagram of the call determination device according to the present embodiment. The call determination device 110 includes an imaging information input unit 310, a face detection unit 320, a hand region specification unit 330, a call determination unit 340, a calling notification unit 350, an imaging information unit 360, and a determination condition information unit 370.

  The imaging information input unit 310 performs processing for inputting the imaging information 301 captured by the camera 120 and storing it in the imaging information unit 360. The face detection unit 320 detects the face area and / or face part of the operator 101 based on the imaging information stored in the imaging information unit 360.

  Based on the detected face area and / or face part, the hand area specifying unit 330 specifies the hand area when the operator 101 is talking on the mobile phone 140. When the operator 101 is talking on the mobile phone 140, the mobile phone 140 is held in his hand, his ear is placed on the earpiece, and the mouthpiece is placed downward (near the mouth). That is, the region from the vicinity of the ear of the operator 101 to the mouth is specified as the hand region.

  The call determination unit 340 analyzes the image of the identified hand region based on the determination condition set in the determination condition information unit 370 and determines whether or not the operator 101 is talking on the mobile phone 140. Determine. The following features can be considered in the hand area as features of the mobile phone 140. (1) A change from the background occurs in an area having a predetermined size or more. (2) There are many curve components, end points of line segments, and parallel components due to the shape of the hand for holding the mobile phone 140. (3) There is a substantially circular “C” -shaped component by gripping the mobile phone. (4) The hue of the mobile phone 140 exists in the center of the “C” type. The call determination unit 340 comprehensively determines the characteristics (1) to (4) above, and determines whether the mobile phone 140 is busy.

  When the call determination unit 340 determines that the operator 101 is busy using the mobile phone 140 in the call determination unit 340, the call notification unit 350 notifies that fact or a warning. The notification is made through the display panel 302 and / or the speaker 303 of the ATM 120.

In addition, when the operator 101 cannot grasp the intention of notification by, for example, an elderly person, the administrator 102 may directly explain to the operator 101 to prevent a transfer fraud or the like in advance. .
Moreover, when the probability of a transfer fraud is high, the transfer function of the ATM 120 may be stopped.

(2. Operation)
FIG. 4 is a flowchart showing the operation of the call determination device according to the present embodiment. First, the operator 101 is imaged with the camera 120 installed in ATM120 (step S401). The imaging information input unit 310 inputs the captured imaging information 301 (step S402). Based on the input imaging information, the face detection unit 320 detects the face area and / or face part of the operator 101 (step S403).

  Here, detection of a face region and / or a face part will be described. FIG. 5 is a flowchart showing face detection processing of the call determination device according to the present embodiment. First, a face area is detected (step S501). The process of detecting the face area can be performed using a technique that has already been disclosed (for example, refer to Japanese Patent No. 4204336, WO 2006/087771, and WO 2006/087789). Description is omitted.

  When the face area is detected, the position of the nose is detected (step S502). The process of detecting the position of the nose can be performed using a technique that has already been disclosed (for example, refer to Japanese Patent No. 4204336 and WO 2006/090449), and detailed description thereof is omitted here.

  Next, the right eye and the left eye are detected (step S503). This eye detection process can be performed using a technique that has already been disclosed (for example, see Japanese Patent Application Laid-Open No. 2004-234494), and detailed description thereof is omitted here.

  Then, the face orientation (angle) is calculated based on the detected positional relationship between the face area, nose, and eyes (step S504). In the calculation of the face orientation, left and right orientation angles (Yaw angles), vertical orientation angles (Pitch angles), and tilt angles (Roll angles) are calculated. The process of calculating the face orientation can be performed using a technique that has already been disclosed (for example, Japanese Patent No. 4204336), and detailed description thereof is omitted here.

  Returning to FIG. 4, when a face area and / or a face part is detected in step S403, the hand area specifying unit 330 specifies the hand area when the operator 101 is talking on the mobile phone 140 (step S404). .

  Here, the specification of the hand region will be described. FIG. 6 is a flowchart showing processing for specifying a hand region of the call determination device according to the present embodiment. First, it is determined whether or not a face part (eye, nose, etc.) has been detected (step S601). If a face part is detected, the hand part on the left hand side is specified using the face part and the face orientation angle (step S602), and the hand part on the right hand side is further specified (step S603). If the face part is not detected, the hand region on the left hand side is specified using the face region (step S604), the hand region on the right hand side is further specified (step S605), and the hand region is specified. To finish the processing.

  FIG. 7 is a diagram illustrating the hand region specified in the call determination device according to the present embodiment. FIG. 7A is a diagram when the face part is not detected, and FIG. 7B is a diagram when the face part is detected. In FIG. 7A, since the face part is not detected, the face orientation angle cannot be calculated. Accordingly, assuming that the operator 101 is facing the front, the left and right predetermined areas (areas indicated by broken lines) of the face area are specified as the hand area.

  In FIG. 7B, since the face part is detected, the face orientation angle can be calculated. Therefore, a predetermined area (area indicated by a broken line) from around the ear to the mouth is specified as the hand area according to the face orientation angle.

  The size of the hand region to be specified may be the number of pixels set in advance according to the resolution of the camera 120, or may be determined based on the size of the face region. For example, when the face area is large, the size of the specified hand area may be large, and when the face area is small, the size of the specified hand area may be small.

Returning to FIG. 4, when the hand region is specified in step S404, the call determination unit 340 determines whether or not the operator 101 is using the mobile phone (step S405).
Here, the determination during use of the mobile phone will be described. FIG. 8 is a flowchart showing a process of determining whether the mobile phone call is in progress in the call determination device according to the present embodiment. First, the change amount of the pixel value in the hand region specified in step S404 is extracted (step S801). If the change amount of the pixel value does not satisfy the predetermined specified value, the in-use flag indicating that the mobile phone is in use is turned off (step S809), and the determination process ends.

  When the change amount of the pixel value satisfies a predetermined specified value, the curve component and the end point of the line segment in the change region are extracted (step S803). Alternatively, the parallel component in the change region and the substantially circular “C” shape are extracted (step S804). If the “C” shape is extracted, the color tone of the “C” shape and the color tone of the center of the “C” shape are extracted. The difference is extracted (step S805).

  The curve component extracted in step S803 and / or the value indicating the degree of the end point of the line segment is defined as degree A, the parallel component extracted in step S805, and / or the color tone of “C” shape and “C” shape. A value indicating the degree of difference from the central color tone is defined as a degree B.

  The processing from step S803 to step S805 will be described in detail. FIG. 9 is a diagram illustrating a case where a curve component, an end point of a line segment, a parallel component, and a “C” shape are extracted in the call determination device according to the present embodiment. When the operator 101 makes a call using the mobile phone 140, some features are extracted from the imaging information of the hand.

  FIGS. 9A and 9B show a case where a curve component and an end point of a line segment are extracted from the shape of the hand during a call of the mobile phone 140. The broken line in FIGS. 9A and 9B indicates the extracted curve component, and the dotted line indicates the end point of the extracted line segment. As can be seen from the figure, many curve components and end points of line segments are extracted from the shape of the hand talking to the mobile phone 140. Based on the extracted information of the curve components and the end points of the line segments, it is determined whether or not the operator 101 is talking on the mobile phone 140.

  FIG. 9C shows a case where a parallel component is extracted from the shape of the hand during a call of the mobile phone 140. The dashed-dotted line in FIG. 9C shows the extracted parallel component. As can be seen from the figure, many parallel components are extracted depending on the angle in the shape of the hand that is talking on the mobile phone 140.

  FIG. 9D shows a case where the “C” shape is extracted from the shape of the hand during a call of the mobile phone 140. As can be seen from the figure, the “C” shape when the mobile phone 140 is gripped according to the angle is extracted in a similar color cluster as the shape of the hand that is talking to the mobile phone 140.

  FIG. 10 is a diagram illustrating a case where a difference between the C-shaped tone and the C-shaped center tone is extracted in the call determination device according to the present embodiment. As can be seen from the figure, when the mobile phone 140 is held, the color tone of the mobile phone 140 itself is often extracted at the center of the “C” shape. Based on the extracted information such as the color difference between the parallel component, the “C” shape, and the center of the “C” shape, it is determined whether or not the operator 101 is talking on the mobile phone 140.

  Returning to FIG. 8, when each element indicating the feature of the hand is extracted, the number of frames in which each element is continuously extracted is extracted (step S806). A value indicating the degree of continuity of the number of frames is defined as a degree C. Based on each element indicating the extracted feature of the hand and the degree of the number of frames, the determination during the call is performed from the determination condition of the determination condition information unit 370 (step S807). An example of the determination condition information is shown in Table 1 below.

表１に示すように、上記抽出された手の特徴の度合いから、携帯電話１４０の使用中を判定する。例えば、度合Ａ（曲線成分、及び／又は線分の端点の度合い）、度合Ｂ（平行成分、及び／又は「Ｃ」形状の中心部の色調の差異）、及び度合Ｃ（フレームの連続性の度合い）の全てが規定値を満たす場合は、携帯電話１４０を使用中であると判定する。また、度合Ａと度合Ｃ、又は度合Ｂと度合Ｃを満たした場合も携帯電話１４０を使用中であると判定する。逆に、度合Ａと度合Ｂを満たした場合や、度合Ａ、又は度合Ｂのみを満たした場合や、すべて満たさなかった場合には、携帯電話１４０を未使用であると判定する。
なお、この判定基準はあくまで一例であり、管理者１０２が任意に設定できるものとする。

As shown in Table 1, it is determined that the mobile phone 140 is in use from the degree of the extracted hand characteristic. For example, degree A (curve component and / or end point degree of line segment), degree B (parallel component and / or color difference in the center of the “C” shape), and degree C (frame continuity) When all of the degree) satisfy the specified value, it is determined that the mobile phone 140 is in use. Further, when the degree A and the degree C, or the degree B and the degree C are satisfied, it is determined that the mobile phone 140 is in use. On the contrary, when the degree A and the degree B are satisfied, only when the degree A or the degree B is satisfied, or when all the degrees A are not satisfied, the mobile phone 140 is determined to be unused.
Note that this determination criterion is merely an example, and the administrator 102 can arbitrarily set it.

  As a result of the determination in step S807, if it is determined that the mobile phone 140 is not used, the busy flag is turned off (step S809), and the determination process ends. If it is determined that the mobile phone 140 is in use, the in-use flag is set to ON (step S808), and the determination process ends.

  Here, the determination in step S803 and the processing in steps S804 and S805 are performed separately, but they may be performed continuously.

  Returning to FIG. 4, when it is determined in step S405 that the mobile phone is in use, if the in-use flag is OFF in step S406, the process returns to step S401 and the operator 101 is imaged again. If the in-use flag is ON in step S406, the busy notification unit 350 notifies that the mobile phone 140 is busy (step S407). The notification is performed through the display panel 302 and the speaker 303. Also, it is performed through a terminal used by the administrator 102. When the administrator 102 is notified that the mobile phone 140 is in a call, the administrator 102 determines whether the call action is good or bad (step S408), and if good, returns to step S401 to image the operator 101 again. . If not, a corresponding process is performed (step S409) and the process is terminated.

  The handling process may be, for example, that the administrator 102 directly calls the operator 101 to pay attention, or may stop the transfer function of the ATM 130. In addition, it is judged that the call activity is good or bad by analyzing the voice during the call and when a predetermined keyword is included (for example, “You have entered your bankbook now”). It may be determined that the call is not good.

  Furthermore, in the present embodiment, the determination during the call of the mobile phone in the ATM has been described, but the call determination device according to the present embodiment can also be applied to the determination during the call of the mobile phone while driving the car. it can.

  As described above, according to the call determination device according to the present embodiment, since the imaging region of the hand during the call of the mobile phone is specified based on the face position, it can be accurately recognized that the call is in progress. . In addition, since it is determined that the call is in progress based on the amount of change in the imaging area of the hand, not on the skin color of the hand, for example, it is determined whether or not the call is in progress even when wearing gloves. Can do.

  Further, the determination based on the characteristics of the hand during the call of the mobile phone can be surely performed based on the number of curve components, the number of parallel components, the number of end points of the line segment, and / or the presence or absence of a substantially circular component. it can.

  Furthermore, in order to determine whether a call is in progress based on the difference between the hue of the substantially circular component and the hue at the center of the substantially circular component, the call is detected by detecting the state of holding the mobile phone by hand. The inside can be reliably determined.

  Furthermore, since it is determined that the call is in progress when the call state continues for a predetermined number of frames or more, the difference between the instantaneous hand movement and the call is recognized, and the state during the call is accurately determined. Can do.

  Furthermore, when it is determined that the person to be imaged is talking on a mobile phone, to prevent this, damages due to bank transfer fraud at ATMs and accidents caused by using a mobile phone while driving a car are prevented. can do.

Although the present invention has been described with the above embodiment, the technical scope of the present invention is not limited to the scope described in the embodiment, and various modifications or improvements can be added to this embodiment. And embodiment which added such a change or improvement is also contained in the technical scope of the present invention. This is apparent from the claims and the means for solving the problems.
The following additional notes will be given regarding the embodiment.

  (Appendix 1) Face detection means for detecting the face of the person to be imaged from the imaging information captured by the camera, and the imaging area of the hand in the call state of the mobile phone based on the face position detected by the face detection means A hand region specifying means for specifying, and a call determining means for determining whether or not the person to be imaged is talking on a mobile phone based on a change amount of a pixel value in the imaging region specified by the hand region specifying means. A call determination device provided.

  (Supplementary note 2) In the call determination device according to supplementary note 1, the call determination unit includes: the number of curve components, the number of parallel components, the number of end points of the line segment in the imaging region specified by the hand region specifying unit; A call determination device that determines whether or not the person to be imaged is talking on a mobile phone based on whether or not a substantially circular component is detected.

  (Supplementary note 3) In the call determination device according to supplementary note 2, when the call determination unit detects a substantially circular component, a difference between a hue of the substantially circular component and a hue of a central portion of the substantially circular component A call determination device that determines whether or not the person to be imaged is talking on a mobile phone based on the above.

  (Supplementary note 4) In the call determination device according to any one of supplementary notes 1 to 3, the call determination unit may be configured to carry the imaging target person when the imaging target person is in a conversation state for a predetermined number of frames or more. A call determination device that determines that a telephone call is in progress.

  (Supplementary Note 5) In the call determination device according to any one of Supplementary Notes 1 to 4, when the call determination unit determines that the person to be imaged is talking on a mobile phone, a notification unit that notifies that fact A call determination device comprising:

  (Appendix 6) A face detection step in which the computer detects the face of the person to be imaged from the imaging information captured by the camera, and the hand in the call state of the mobile phone based on the position of the face detected by the face detection step Call area determination step for determining whether or not the person to be imaged is talking on a mobile phone based on a hand area specifying step for specifying an imaging area and a change amount of a pixel value in the imaging area specified by the hand area specifying step A call determination method for executing steps.

  (Supplementary note 7) In the call determination method according to supplementary note 6, the call determination step includes the number of curve components, the number of parallel components, the number of end points of a line segment in the imaging region specified in the hand region specifying step, and A call determination method for determining whether or not the person to be imaged is talking on a mobile phone based on whether or not a substantially circular component is detected.

  (Supplementary note 8) In the call determination method according to supplementary note 7, when the call determination step detects a substantially circular component, a difference between a hue of the substantially circular component and a hue of a central portion of the substantially circular component A call determination method for determining whether or not the person to be imaged is talking on a mobile phone based on the above.

  (Supplementary note 9) In the call determination method according to any one of supplementary notes 6 to 8, the call determination step is performed when the imaging target person carries the imaging target person when the imaging target person's busy state continues for a predetermined number of frames or more. A call determination method for determining that a telephone call is in progress.

  (Supplementary note 10) In the call determination method according to any one of supplementary notes 6 to 9, when the call determination method determines that the person to be imaged is busy on a mobile phone, a notification step of notifying that effect Call determination method including

  (Supplementary Note 11) Face detection means for detecting the face of the person to be imaged from the imaging information captured by the camera, and specifying the imaging area of the hand in the call state of the mobile phone based on the face position detected by the face detection means The computer functions as a call determination unit that determines whether the person to be imaged is talking on a mobile phone based on the amount of change in the pixel value in the imaging region specified by the hand region specification unit Call judgment program to make.

  (Additional remark 12) In the call determination program according to additional remark 11, the call determination means includes the number of curve components, the number of parallel components, the number of end points of the line segment in the imaging area specified by the hand area specifying means, and A call determination program that determines whether or not the person to be imaged is talking on a mobile phone based on whether or not a substantially circular component is detected.

  (Supplementary note 13) In the call determination program according to supplementary note 12, when the call determination unit detects a substantially circular component, a difference between a hue of the substantially circular component and a hue of a central portion of the substantially circular component A call determination program for determining whether or not the person to be imaged is talking on a mobile phone based on the above.

  (Supplementary note 14) In the call determination program according to any one of supplementary notes 11 to 13, the call determination unit may be configured to carry the imaging target person when the imaging target person is in a telephone conversation state for a predetermined number of frames or more. A call determination program for determining that a telephone call is in progress.

  (Supplementary Note 15) In the call determination program according to any one of Supplementary Notes 11 to 14, when the call determination unit determines that the person to be imaged is talking on a mobile phone, a notification unit that notifies that fact A call determination program that causes a computer to function as a computer.

DESCRIPTION OF SYMBOLS 100 In-call detection system 101 Operator 102 Administrator 110 Call determination apparatus 120 Camera 130 ATM
140 mobile phone 210 CPU
220 RAM
230 ROM
240 HD
250 Communication I / F
260 Input / output I / F
301 Imaging Information 302 Display Panel 303 Speaker 310 Imaging Information Input Unit 320 Face Detection Unit 330 Hand Area Identification Unit 340 Call Determination Unit 350 In-call Notification Unit 360 Imaging Information Unit 370 Determination Condition Information Unit

Claims (7)

Face detection means for detecting the face of the person to be imaged from imaging information captured by the camera;
Means for detecting a face orientation angle based on the face position detected by the face detection means;
Based on the face orientation angle detected by the means for detecting the face orientation angle, hand region specifying means for specifying the imaging region of the hand in the call state of the mobile phone;
A call determination device comprising: call determination means for determining whether or not the person to be imaged is talking on a mobile phone based on a change amount of a pixel value in an imaging area specified by the hand area specifying means. The call determination device according to claim 1,
Based on the number of curve components, the number of parallel components, the number of end points of the line segment, and / or the presence or absence of detection of a substantially circular component in the imaging area specified by the hand area specifying means, A call determination device that determines whether or not the person to be imaged is talking on a mobile phone. The call determination device according to claim 2, wherein
When the call determination unit detects a substantially circular component, the imaging subject is talking on the mobile phone based on the difference between the hue of the approximately circular component and the hue of the central portion of the approximately circular component. A call determination device that determines whether or not In the call determination device according to any one of claims 1 to 3,
A call determination device in which the call determination unit determines that the imaging target person is talking on a mobile phone when the imaging target person is in a call state for a predetermined number of frames or more. In the call determination device according to any one of claims 1 to 4,
When the call determination unit determines that the person to be imaged is in a call on a mobile phone, the call determination device includes a notification unit that notifies that fact. Computer
A face detection step for detecting the face of the person to be imaged from imaging information captured by the camera;
Detecting a face orientation angle based on the face position detected by the face detection step;
Based on the face orientation angle detected by the step of detecting the face orientation angle, a hand region specifying step for specifying an imaging region of a hand in a call state of the mobile phone;
A call determination method for executing a call determination step of determining whether or not the person to be imaged is talking on a mobile phone based on a change amount of a pixel value in an imaging region specified by the hand region specifying step. Face detection means for detecting the face of the person to be imaged from imaging information captured by a camera;
Means for detecting a face orientation angle based on the face position detected by the face detection means;
Hand region specifying means for specifying an imaging region of a hand in a call state of the mobile phone based on the face orientation angle detected by the means for detecting the face orientation angle ;
A call determination program for causing a computer to function as a call determination unit for determining whether or not the person to be imaged is talking on a mobile phone based on a change amount of a pixel value in an imaging region specified by the hand region specifying unit.

Images