JP2006059230A - Entrance/exit management system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an entrance/exit management system which is reliable in spite of an inexpensive device, and can check incoming and outgoing persons at a later date. <P>SOLUTION: This system comprises an electric lock 2 provided on an opening and closing mechanism 1, an image acquisition part 5 and an identification information input part 6 provided on an unlocking instruction means 3 instructing unlocking of the electric lock 2, a face information determination processing part 10 provided on a control means 4 controlling the unlocking of the electric lock 2, an image storage part 15, an input information identification processing part 9, an output part 12 and a processing control part 8. The processing control part 8 has a function of determining, when the input information identification processing part 9 identifies input information, an image fetched to the image acquisition part 5 is a human face, and the face information determination processing part 10 determines that the ratio of flesh color in an area based on the position of the eye is a reference value or more, that the image is the human face to unlock the electric lock 2 through the output part 12, and a function of storing image data fetched from the acquisition part 5 to the image storage part 15. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an entrance / exit management system for managing a person entering / exiting a building.

Various systems for managing the entrance and exit of people to and from the building have been proposed. Among them, a system for identifying the face of a person entering and exiting the building and permitting people to enter and exit based on the identification result. Conventionally known. The reason why a person's face is identified in this way is that the person's face is the most personalized and has the highest reliability when considered as a means for identifying another person. In addition, the face is photographed in the process of identifying the face of the person going in and out so that it is possible to check what kind of person has entered and exited the building.
In addition, management of the person who goes in and out of the building is generally performed, and no prior literature search is performed.

In the conventional system as described above, a device for visualizing a face and recognizing the face is quite expensive. Therefore, there is a problem that such an expensive device cannot be introduced depending on a building.
An object of the present invention is to provide an entry / exit management system that is inexpensive and highly reliable, and that can also check an entry / exit at a later date.

  When the first invention recognizes an electric lock provided in an opening / closing mechanism such as a door, an unlocking instruction means for instructing the unlocking of the electric lock, and information transmitted from the unlocking instruction means, Control means for controlling the unlocking of the electric lock based on the information, the unlocking instruction means includes a video acquisition unit and an identification information input unit, and the control means includes a video captured by the video acquisition unit. And a face information determination processing unit having a function for determining that it is a human face and a function for determining whether the ratio of the skin color in the area based on the eye exceeds a reference value, and video acquisition A video storage unit that stores image data captured from the input unit, an input information identification processing unit that identifies personal information input from the identification information input unit, an output unit that controls unlocking of the electric lock, and the face information determination process Section, input information identification processing section and output section A processing control unit that controls the input information identification processing unit to identify the input information, and the video captured by the video acquisition unit is a human face, and the eye A function in which when the face information determination processing unit determines that the skin color ratio in the area based on the position is equal to or greater than a reference value, it is determined as a human face, and the electric lock is unlocked via the output unit; Further, the present invention is characterized in that it has a function of storing the image data captured from the video acquisition unit in the video storage unit.

  The second invention is characterized in that when a person entering / exiting an opening / closing mechanism such as a door operates the identification information input unit, the video acquisition unit is operated using the operation as a trigger.

  According to a third aspect of the present invention, the control means includes a video recording processing unit linked to the processing control unit and a video storage unit linked to the video recording processing unit. The video recording processing unit includes a video acquisition unit. The background digital image data captured in advance is compared with the image data captured each time to determine whether there is a brightness difference between the data. When there is a brightness difference, a new image is displayed before the video acquisition unit. The processing control unit has a function of determining that an object exists, and the processing control unit determines whether the object in front of the image acquisition unit is a human face shape and determines that it is a face shape. When the ratio of the skin color portion in the video reaches a reference value or more, the personal information condition identified by the input information identification processing unit, the human face shape condition determined by the face information determination processing unit, and the face information determination processing unit Of skin color recognized by When all the conditions are satisfied, the function of unlocking the electric lock via the output unit, and when the face information determination processing unit determines that the image data captured from the video acquisition unit is a human face, It is characterized in that image data is stored in the video storage unit.

  According to a fourth aspect of the present invention, a face information storage unit is connected to the face information determination processing unit, and the face information storage unit includes a reference frame and a person for specifying the coordinates of image data captured from the video acquisition unit. The face shape pattern for specifying the face shape is stored in advance, and the face information determination processing unit is stored in the face information storage unit with a function for specifying the center of the image data based on the reference frame. It is characterized in that it has a function of extracting a face shape pattern and matching the center of the face shape pattern with the center of the image data.

According to the first invention, even if the personal information input from the identification information input unit matches the personal information stored in advance, the image data acquired by the video information acquisition unit by the face information determination processing unit is human. Since it is not unlocked unless it is determined that the face is, the security accuracy is high. In addition, since it is only necessary to determine the approximate face shape of a person's face instead of determining its individuality, the entire system including the video acquisition unit can be made inexpensive.
Moreover, since the unlocking conditions are determined based on the ratio of the skin color, the unlocked person's face is not a hidden face with sunglasses or a mask, but is a real face. In addition, since this face can be stored as image data in the video storage unit, it can be used for later verification.

  According to the second aspect of the invention, since the video acquisition unit is operated using the operation of the identification information input unit as a trigger, the distance between the video acquisition unit and the face of the person who is the subject can be specified to some extent. That is, in the structure in which the video acquisition unit is operated using the operation of the identification information input unit as a trigger, the maximum distance between the video acquisition unit and the human face is the length of the person's arm. Therefore, if the focal length of the image acquisition unit is determined based on the length of the arm of an adult, the image will not be so out of focus.

  According to the third invention, when the three conditions are satisfied, it is determined that the unlocking condition is satisfied. Therefore, the accuracy is improved as compared with the case where the unlocking condition is determined only by the information for identifying the individual. It will be. In addition, those who come in and out of the opening and closing mechanism cannot satisfy the unlocking conditions unless they remove their sunglasses and masks, so they are consciously wearing sunglasses and masks to hide their faces. The person can be pressured to remove it. In addition, those who enter the room after unlocking the electric lock are stored in the video storage unit, so they can check who has entered from the opening / closing mechanism even after the fact. be able to.

According to the fourth invention, it is possible to determine whether the image data is a face shape by matching the center of the image data captured based on the reference frame with the center of the face shape pattern. Accuracy is improved.

FIG. 1 shows an embodiment of an access control system of the present invention.
As shown in the block diagram of FIG. 1, this system includes a door 1 as an opening / closing mechanism of the present invention, an electric lock 2 provided on the door 1, and an unlocking instruction means for instructing the unlocking of the electric lock. 3 and control means 4 for controlling the unlocking of the electric lock.

As shown in FIG. 1, the electric lock 2 provided on the door 1 is connected to the control means 4, but this control means 4 is also connected to the unlocking instruction means 3 installed outside the door 1. Yes.
The control means 4 is usually installed in a place away from the individual door 1 such as a building manager's room or a security company center, and one or more unlocking instruction means 3 and electric by means of a communication cable or radio. Connected to lock 2. Then, one control means 4 performs entrance / exit management of one or more doors 1.

  As shown in FIG. 1, the unlocking instruction unit 3 includes a video acquisition unit 5, an identification information input unit 6, and an audio output unit 7.

  The video acquisition unit 5 is, for example, a camera or a video, and takes in a video of the face of a person trying to enter through the door 1. It is desirable that the lens of this camera or the like is conspicuous so that it can be immediately recognized by a person outside the door. This is because the purpose of this embodiment is to have an inbound / outbound person actively stand in front of a camera or the like as a video acquisition unit and acquire the video.

  The identification information input unit 6 is, for example, a numeric keypad, and a person who wants to enter the inside from the outside of the door 1 inputs an ID or the like for identifying himself / herself. As shown in FIG. 2, the identification information input unit 6 includes numeric keys 0 to 9, a clear key C for correcting erroneous input, and an enter for transmitting input end information to the control means 4. Key E. Note that the content input from the numeric keys is transmitted to the control means 4 when the enter key E is pressed as a trigger.

  The video acquisition unit 5 is configured to exhibit a video acquisition function when the identification information input unit 6 such as a numeric keypad is operated. Then, the video captured by the video acquisition unit 5 is transmitted to the control means 4.

  The identification information input unit 6 may be in any form as long as a person can intentionally input identification information and can identify an individual based on the identification information. For example, a means for reading a magnetic card, an IC card, or the like may be used, and a means for reading biological information such as a fingerprint may be used depending on circumstances. In short, any identification information other than video may be used. And when it is set as the structure which reads a card | curd as mentioned above, what is necessary is just to transmit the information to the control means 4, when those information is read.

  The audio output unit 7 is, for example, a speaker or an electronic audio output device, and outputs a guidance message, an error message, or the like from the control unit 4.

  The control unit 4 includes a processing control unit 8, an input information identification processing unit 9, a face information determination processing unit 10, a video recording processing unit 11, and an output unit 12. Further, an identification information storage unit 13 is linked to the input information identification processing unit 9, a face information storage unit 14 is linked to the face information determination processing unit 10, and a video storage unit 15 is linked to the video recording processing unit 11. ing.

  The input information identification processing unit 9 is an ID of a person who is allowed to enter information identifying an individual such as an ID transmitted from the identification information input unit 6 of the unlocking instruction means 3 through the door 1 by this system. And so on. That is, the input information identification processing unit 9 is connected to the identification information storage unit 13 and collates whether the transmitted information matches an ID or the like stored in advance in the identification information storage unit 13.

  The face information determination processing unit 10 determines whether or not the video data transmitted from the video acquisition unit 5 of the unlocking instruction unit 3 represents a human face shape. The video captured by the video acquisition unit 5 is transmitted to the control unit 4 as digital image data. Here, the image recognition method for determining whether or not a person is a face is not particularly limited. For example, the following method is conceivable.

  The video acquisition unit 5 captures a video showing nothing other than the background, and transmits the video as background digital image data to the video recording processing unit 11 via the processing control unit 8 of the control unit 4. The recording processing unit 11 stores the background digital image data in the video storage unit 15 in advance. Then, the video recording processing unit 11 determines whether there is a luminance difference between the video captured by the video acquisition unit 5 and the previously stored background digital image data. If there is a luminance difference between them, the video recording processing unit 11 determines that there is a new object that is not included in the previously stored background digital image data before the video acquisition unit 5. In other words, since there is a difference in brightness between the picture that has not moved other than the background and the picture that has the background and the rest, the video recording processing unit 11 recognizes that another object is added to the background digital image data stored in advance based on the luminance difference, and transmits the recognition result to the processing control unit 8.

  On the other hand, the face information storage unit 14 stores a plurality of face shape patterns and a reference frame. This face pattern is prepared considering not only the face shape such as the square shape and the round shape but also the hair style for each man and woman. The reference frame is for specifying the XY coordinates in the face portion of the image data acquired by the video acquisition unit 5 and determining the center thereof.

  When the processing control unit 8 obtains the recognition result that there is a luminance difference in the image data, the processing control unit 8 operates the face information determination processing unit 10 to set the reference frame stored in the face information storage unit 14. In addition to the detection, the detected reference frame is matched with the location where the luminance difference has occurred, and the coordinates of the location where the luminance difference has occurred are specified. Further, the face information determination processing unit 10 detects the skin color area at the location where the luminance difference specified by the reference frame is specified, and obtains the center coordinate of the skin color area based on the reference frame. When the center of the skin color area is specified in this way, the face information determination processing unit 10 detects the face shape pattern stored in the face information storage unit 14 and determines the center of the face shape pattern as the video acquisition unit. Match the center of the flesh-colored area acquired in 5, and determine whether or not they are approximate. If it is approximated, the face information determination processing unit 10 determines that the image data transmitted from the video acquisition unit 5 of the unlocking instruction unit 3 represents a human face shape. Recognize that the person is standing. If they do not match, the face information determination processing unit 10 determines that the image data is not a human face, and stops the process of instructing unlocking.

  The reference frame has a constant size because it is based on the premise that the distance between the video acquisition unit 5 such as a camera and the subject is not so different. That is, since the person who is the subject operates the identification information input unit 6 such as a numeric keypad, the distance between the video acquisition unit 5 and the subject can be assumed to be the length of the person's arm at the maximum. In addition, since it is impossible to operate the identification information input unit 6 with its arms fully extended, the distance between the video acquisition unit 5 and the subject can be specified to some extent. If the distance between the video acquisition unit 5 and the subject is specified to some extent in this way, the size of the face in the video acquisition unit 5 does not change so much, so there is no problem even if the size of the reference frame is constant as described above. .

  Furthermore, in this embodiment, even if the subject is slightly out of focus, it is only necessary to roughly determine the face shape, so the reference frame does not have to exactly match the face portion of the subject. Based on this premise, the center of the image data that is roughly determined to be a face shape is specified. However, as will be described in detail later, all of the image data acquired by the video acquisition unit 5 is stored in the video storage unit 15 so that it is possible to verify who went in and out at a later date. Therefore, this image data only needs to be able to identify someone roughly, so a slight blur does not matter so much, but in any case, the captured image data may not be in focus, It is sufficient if the approximate position of the center is specified.

  When it is recognized that it is a human face shape as described above, the face information determination processing unit 10 specifies the positions of the two eyes in the face shape based on the face shape pattern, Identify the area around the eye. When the area around the eyes is specified in this way, the face information determination processing unit 10 calculates the ratio of the skin color in the area and stores the ratio of the skin color in the area in the face information storage unit 14 in advance. It is judged whether it exceeds the standard value. If the reference value is exceeded, the face information determination processing unit 10 determines that the captured image data is regular face information that does not use sunglasses or a mask. If the skin color portion falls below the reference value, it is determined that the captured image data is incorrect face information using sunglasses or a mask.

  On the other hand, all the image data captured by the video acquisition unit 5 as described above is also sent to the video recording processing unit 11, and the video recording processing unit 11 sends the sent image data to the video storage unit 15. Remember. At that time, the date and time, the ID collated by the input information identification processing unit 9 and the like may be stored together. Here, since the image data has a large data size, it is natural that the image data may be stored after being compressed.

  Note that this image data is a digital image of an image obtained by photographing the face of a person trying to enter through the door 1, and it is necessary to identify the person's face to some extent at the stage of this image data. In other words, the face information determination processing unit 10 only needs to be able to determine that it is likely to be a human face, but the image data itself is used to verify who entered and exited later, so that the human face can be identified. What was filmed to the extent is required. However, the captured image data does not need to be in focus. This is because it suffices if it is possible to identify to some extent who has entered and exited based on the image data. In that sense, the video acquisition unit 5 does not have to be so sophisticated, including the face shape determination described above. This is a factor that can provide an inexpensive access control system.

  The process control unit 8 controls various processes performed by the control unit 4. When image data is transmitted from the video acquisition unit 5 of the unlocking instruction unit 3, the processing control unit 8 causes the face information determination processing unit 10 to determine the image data, and the video recording processing unit 11 stores the data of the data. Make a record.

  Further, when an ID or the like is transmitted from the identification information input unit 6 of the unlocking instruction means 3, the process control unit 8 causes the input information identification processing unit 9 to collate the ID or the like. That is, the process control unit 8 controls all processes of the control unit 4, and the face information determination processing unit 10 and the input information identification processing unit 9 share the processing performed by the control unit 4. Become.

  Then, as a result of the determination process of the image data, the process control unit 8 cannot determine it as a face, or there is no skin color exceeding the reference value in the area centered on the second eye, or the ID is collated. As a result, when they do not coincide with each other, the voice output unit 7 of the unlocking instruction means 3 is informed to that person through the door 1. For example, when a person's face cannot be determined, the process control unit 8 performs a process of announcement via the audio output unit 7 so that the face is accurately positioned in front of the video acquisition unit 5. Also, when there is no skin color that exceeds the reference value in the area centered on the second eye, for example, it is announced that the image acquisition unit 5 should try again after removing the sunglasses. Further, when the IDs do not match, the process control unit 8 executes a process of announcing a numeric keypad input error or a card error via the audio output unit 7.

  Further, the processing control unit 8 determines that the input information identification processing unit 9 matches the input information such as the ID with the previously registered identification information, and the face information determination processing unit 10 determines that the input image is a human face. When it is determined that sunglasses are not worn, the output unit 12 is instructed to unlock the electric lock 2. In addition, the output part 12 transmits the signal etc. which control it with respect to the solenoid which is not shown in figure of the electric lock 2, for example.

  Furthermore, the processing control unit 8 instructs the output unit 12 to unlock the electric lock 2 as described above, and at the same time, the video image processing unit 11 acquires facial image data acquired by the video acquisition unit 5. Is stored in the video storage unit 15, and the video storage unit 15 stores the image data. Further, the image data stored in the video storage unit 15 in this way can be displayed on the display 16 via the processing control unit 8 and the output unit 12. In this embodiment, the electric lock 2 is satisfied if certain conditions are satisfied, that is, identification information such as an ID is regular, and a person's face is not wearing sunglasses or the like. To unlock. Therefore, it is completely different from the conventional method of accurately recognizing an individual's face and deciding whether or not to unlock the electric lock 2. That is, in this embodiment, as long as the above conditions are satisfied, it is determined that all of the legitimate persons are in and out, and thereafter, the in / out persons are managed so that the image data of the in / out persons can be checked.

Next, the operation of this embodiment will be described.
For example, a person who wants to enter the room through the door 1 provided with the electric lock 2 first operates the identification information input unit 6 to input personal identification information. When personal information is input from the identification information input unit 6 as described above, the personal information is sent from the processing control unit 8 to the input information identification processing unit 9, and the input information identification processing unit 9 receives the input personal information. Then, it is determined whether or not it matches the personal information stored in advance in the identification information storage unit 13, and the determination result is transmitted to the process control unit 8. If the result of the determination is a mismatch, the process control unit 8 makes an announcement to the effect of the mismatch via the audio output unit 7 and also makes a re-input announcement. If the personal information does not match even if the re-input is repeated a predetermined number of times or more, the unlocking operation is not performed thereafter.

  When personal information is input from the identification information input unit 6 as described above, a face determination process and a face photograph capturing process proceed in parallel with the personal information collation process described above. That is, when the identification information input unit 6 is operated as described above, the video acquisition unit 5 operates using it as a trigger. The video acquisition unit 5 sends the captured image data to the processing control unit 8, and the processing control unit 8 that has received the image data operates the video recording processing unit 11 and stores it in the video storage unit 15 in advance. The presence / absence of a luminance difference between the background digital image data and the image data sent from the video acquisition unit 5 is determined. If there is a luminance difference between them, the video recording processing unit 11 recognizes that a new object exists before the video acquisition unit 5 and transmits the recognition result to the processing control unit 8. If the recognition that a new object exists is not obtained before the video acquisition unit 5, the process for acquiring the video is stopped at that time.

  If the luminance difference cannot be determined in the video captured from the video acquisition unit 5 even though the personal information is input from the identification information input unit 6 as described above, the following cases can be considered. In other words, the first case is a case where a legitimate person has inadvertently placed his / her face away from the video acquisition unit 5 or did not know that the video acquisition unit 5 exists. The second case is a case where an unauthorized intruder such as a thief deliberately removes his face from the shooting range of the video acquisition unit 5.

  When the luminance difference cannot be determined in the video captured from the video acquisition unit 5 as described above, the process control unit 8 notifies the person in front of the unlocking instruction unit 3 via the audio output unit 7 of the video acquisition unit 5. Output a guidance message to stand in front. This guidance message prompts the legitimate person in the first case to bring a face within the shooting range of the video acquisition unit 5, and to the unauthorized intruder in the second case. By outputting the above voice message, it can be understood that the image cannot be unlocked unless a picture of his / her face is taken.

  When a recognition result indicating that a new object exists before the image acquisition unit 5 is obtained, the processing control unit 8 operates the face information determination processing unit 10 to store the reference stored in the face information storage unit 14. A frame is detected, and the detected reference frame is matched with the location where the brightness difference has occurred, and the coordinates of the location where the brightness difference has occurred are specified. Further, the face information determination processing unit 10 detects the skin color area at the location where the luminance difference specified by the reference frame is specified, and obtains the center coordinate of the skin color area based on the reference frame. When the center of the skin color area is specified in this way, the face information determination processing unit 10 detects the pattern stored in the face information storage unit 14 and acquires the center of the pattern by the video acquisition unit 5. It is made to coincide with the center of the skin color area, and it is determined whether or not they are approximate. If they are approximate, it is determined that the image data transmitted from the video acquisition unit 5 represents a human face, and it is recognized that a person is standing in front of the video acquisition unit 5. If they do not match, the face information determination processing unit 10 determines that the image data is not a human face, stops the process of instructing unlocking, and stores the image data in the face information storage unit 14. Don't let it go.

  When the face information determination processing unit 10 determines that the image data acquired from the video acquisition unit 5 as described above is human face information, the face information determination processing unit 10 uses the face shape pattern as a basis. Then, the position of the two eyes in the face shape is specified, and the area around the eyes is specified. When the area around the eyes is specified in this way, the face information determination processing unit 10 calculates the ratio of the skin color in the area and stores the ratio of the skin color in the area in the face information storage unit 14 in advance. It is judged whether it exceeds the standard value. If the reference value is exceeded, the face information determination processing unit 10 determines that the captured image data is regular face information that does not use sunglasses or a mask. If the skin color portion falls below the reference value, it is determined that the captured image data is incorrect face information using sunglasses or a mask.

  Then, the processing control unit 8 matches the personal information input from the identification information input unit 6 with the personal information stored in advance in the identification information storage unit 13, and the captured image data includes sunglasses, a mask, and the like. When it is determined that the face is not a person, the electric lock 2 is unlocked via the output unit 12. In other words, the following three conditions were satisfied: the condition that the personal information matches, the condition that the image data is a person's face, and the condition that the person's face in the image data is not wearing sunglasses or a mask. At this time, the process control unit 8 unlocks the electric lock 2. If even one of these three conditions is not satisfied, the process control unit 8 does not unlock the electric lock 2.

  However, in this embodiment, when the face information determination processing unit 10 determines that the image data captured from the video acquisition unit 5 is a human face, even if it is image data wearing sunglasses or a mask, The face information determination processing unit 10 stores all of the image data in the face information storage unit 14. Regardless of whether the image data stored in this way is legitimate or illegal, wearing sunglasses or a mask, the processing control unit 8 sends it to the display 16 via the output unit 12. It can be displayed. In this way, unauthorized image data is also stored in the face information storage unit 14 so that it can be displayed on the display. It is verified later what kind of person stood in front of the unlocking instruction means 3. This is to make it possible.

  As described above, in this embodiment, the unlock instruction unit 3 transmits two types of information, that is, identification information and image data, to the control unit 4, and the control unit 4 receives the received two types of information. A verification process or a determination process is performed, and based on the result, it is determined whether the door 1 that is the opening / closing mechanism can be unlocked. That is, as described above, there are three conditions: the condition that the personal information is the same, the condition that the image data is a human face, and the condition that the person's face in the image data is not wearing sunglasses or a mask. When the condition is satisfied, the process control unit 8 unlocks the electric lock 2. If even one of these three conditions is not satisfied, the process control unit 8 does not unlock the electric lock 2. Therefore, since it is determined whether or not unlocking is possible by combining the right and wrong of the three types of conditions, the reliability in terms of security is higher than that of a system that uses only the right and wrong of one type of information such as an input ID as the unlocking condition. improves.

  Further, at the level for determining whether or not unlocking is possible, the video acquisition unit 5 and the face information determination processing unit 10 recognize that the subject is a human face and recognize that the subject is not wearing sunglasses or a mask. If possible, the entire system is much cheaper than a device that identifies a particular person's face.

  Further, in the above embodiment, the video acquisition unit 5 is assumed to be a normal camera, but it is also conceivable that a human and a doll cannot be distinguished only by a normal camera. Therefore, if an infrared detector is attached to the camera and it is determined whether or not a temperature difference on the surface of a person is detected, it can be distinguished from a human being and an inorganic substance such as a doll having no body temperature.

  Furthermore, in the above-described embodiment, a sound output unit such as a speaker is used as the sound output unit 7. However, in consideration of a person with hearing impairment, it may be used in combination with a display that displays a message. The voice output unit 7 is a means for transmitting a message from the control means 4 and may be anything that can achieve its purpose.

  In the above embodiment, the control means 4 is installed at a location remote from the individual door 1 and the unlocking instruction means 3, but the control means 4 is installed in the vicinity of the door 1 and one unlocking means is provided. It may correspond to the lock instruction means 3 and one electric lock 2.

It is a block diagram which shows the whole structure of embodiment. It is a conceptual diagram which shows the unlocking instruction | indication means of embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Opening / closing mechanism 2 Electric lock 3 Unlocking instruction means 4 Control means 5 Image | video acquisition part 6 Identification information input part 7 Audio | voice output part 8 Process control part 9 Input information identification process part 10 Face information determination process part 11 Video recording process part 12 Output Part 15 Video storage part

Claims (4)

  When an electric lock provided in an opening / closing mechanism such as a door, an unlocking instruction means for instructing to unlock the electric lock, and information transmitted from the unlocking instruction means are recognized, the electric lock is based on this information. The unlocking instruction means is provided with a video acquisition unit and an identification information input unit, and the control means includes a human face from the video captured by the video acquisition unit. A face information determination processing unit having a function for determining whether or not the ratio of the skin color in the area based on the eye exceeds the reference value, and image data captured from the video acquisition unit A video storage unit for storing information, an input information identification processing unit for identifying personal information input from the identification information input unit, an output unit for controlling the unlocking of the electric lock, the face information determination processing unit, and the input information identification processing To control the output section and output section The processing control unit identifies the input information by the input information identification processing unit, and the video captured by the video acquisition unit is a human face, and the position of the eye is used as a reference. When the face information determination processing unit determines that the skin color ratio in the selected area is greater than or equal to the reference value, the face is determined to be a human face, and the electric lock is unlocked via the output unit, and video acquisition Access control system having a function of storing image data captured from a video storage unit in a video storage unit.   The entrance / exit management system according to claim 1, wherein when a person entering / exiting an opening / closing mechanism such as a door operates the identification information input section, the operation is triggered by the operation.   The control means includes a video recording processing unit linked to the processing control unit, and a video storage unit linked to the video recording processing unit, the video recording processing unit is a background captured in advance via the video acquisition unit The digital image data is compared with the captured image data each time to determine whether there is a luminance difference between the data. When there is a luminance difference, it is determined that there is a new object in front of the video acquisition unit. A processing control unit that determines whether the object in front of the image acquisition unit is a human face shape and a skin color portion in the image that is determined to be a face shape The ratio of the personal information identified by the input information identification processing unit, the human face condition determined by the face information determination processing unit, and the skin color ratio recognized by the face information determination processing unit All of the conditions When adding the function to unlock the electric lock via the output unit and the face information determination processing unit determines that the image data captured from the video acquisition unit is a human face, the image data The entrance / exit management system according to claim 1 or 2 configured to be stored in a storage unit.   A face information storage unit is connected to the face information determination processing unit, and a reference frame and a human face shape for specifying the coordinates of the image data captured from the video acquisition unit are specified in the face information storage unit. The face shape pattern is stored in advance, and the face information determination processing unit extracts the function of specifying the center of the image data based on the reference frame and the face shape pattern stored in the face information storage unit. The entrance / exit management system according to claim 1, further comprising a function of matching a center of the face shape pattern with a center of the image data.

Images