JP2738811B2 - Automatic door opening and closing management system - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic door opening / closing management system for controlling the opening / closing of a door by identifying the characteristics of the body of a person who wants to unlock the door by image data.

[0002]

2. Description of the Related Art Conventionally, as this kind of system, for example, an "automatic door guidance system" disclosed in Japanese Patent Application Laid-Open No. 5-79232 has been proposed. The automatic guidance system includes a camera that captures a chest badge of a person approaching the door, an image data storage unit that stores and stores a plurality of chest badge image data in advance for comparison with image data from the camera, A collation determination unit that sequentially compares image data from the camera with image data from the image data storage unit to perform collation determination, a door lock mechanism that operates based on a comparison result of the collation determination unit, and a sound generator that generates sound. Section.

According to this system, by recognizing the type of a visitor's chest badge as an image, reliable door opening / closing control can be performed without malfunction, and when it is determined that entry through the door is not possible. Also, a guidance message such as "Do not enter from this door" can be generated through the voice generation unit, and appropriate guidance can be realized.

[0004]

However, in the above-described conventional system, since the door opening / closing control is performed by recognizing the type of the chest badge as an image, it is possible to enter and leave the room without the chest badge. Without the chest badge,
Management for theft prevention and the like is complicated. Conversely, if a person other than the person to whom the chest badge was assigned possesses the chest badge, it can be easily entered and exited, so there is leakage in entry and exit management in rooms etc. where importance is placed on confidentiality. The management of the chest badge must be strict and complicated, and there is a problem in securing confidentiality.

[0005] In a car or the like, the key may be forgotten in the car, or may be lost or stolen.
Can not unlock the door even though it is their own vehicles,
In addition, there is a disadvantage that the lock is unlocked by others.

SUMMARY OF THE INVENTION The present invention has been made in view of the above circumstances, and has as its object to provide a door which does not require a certification tool such as a badge, is easy to manage, and can control opening and closing of a door according to each individual. An automatic opening and closing management system is provided.

[0007]

In order to achieve the above object, according to the present invention, there is provided a door which is unlocked by inputting an opening signal;
Image capturing means for capturing at least a part of a person who wants to pass from one side of the door and a person who wants to pass from the other side of the door; storage means; and predetermined characteristics based on image data obtained by the image capturing means. And extracting the extracted data of the person who wants to pass from one side of the door in the storage means, and when the processing of storing the extracted data in the storage means by the characteristic extraction means is completed. and de stored in the extracted data and the storage means of the person wishing to pass from the other side of the door obtained by the feature extracting means
A.Compare with the extracted data of the person who wants to pass from one side , and if both data are within the pre-set matching condition range, at least if they are within the matching condition range, the open signal Is output to the lock of the door.

[0008]

According to the present invention, first, a part of the body of a person who wants to pass from one side of the door, for example, a face, is imaged.
This image data is input to the feature extraction means. In the feature extracting means, predetermined features are extracted from the input image data, and the extracted data is stored in the storage means. When the process of storing the extracted data in the storage unit by the feature extraction unit is completed, an open signal is generated in the determination unit and output to the lock of the door for a certain period of time. As a result, the door is unlocked, and a person who wants to pass from one side passes through the door and enters, for example, another room. Then, after a certain period of time, the unlocked state of the door is released and the door is closed.

When a person who has entered another room leaves the room,
The face of the person desiring to leave the room, that is, the person desiring to pass through from the other side of the door is imaged, and this image data is input to the feature extracting means. In the feature extracting unit, a predetermined feature is extracted from the input image data, and the extracted data is output to the determining unit. The determining means compares the extracted data of the person who wants to pass from the other side of the door with the extracted data obtained before passing from the one side of the door and stored in the storage means. As a result of the comparison, if the data is within the preset matching condition range, an open signal is output to the lock of the door for a fixed time. As a result, the door is unlocked, and a person who wants to pass from the other side passes through the door and leaves the room. If the result of the comparison indicates that the data is outside the preset matching condition range, no open signal is output and the door is not unlocked.

[0010]

FIG. 1 is a block diagram showing an embodiment of an automatic door opening / closing management system according to the present invention. In FIG. 1, reference numeral 10 denotes a door, 20 denotes a human detection unit, 30 denotes a CCD camera, and 40 denotes a system control unit.

The door 10 is provided with a key lock mechanism 11 and a door intercom 12. The key lock mechanism 11 is unlocked only during the input period of the open signal _SOPN by the system control unit 40. The door intercom 12 is arranged at a predetermined position of the door 10, and receives the audio data D _AUD from the system control unit 40 and utters. In the present embodiment, the left side of the door 10 in the drawing is set to the outside and the right side is set to the room. On both sides of the door 10, persons HUM _IN and HUM _OUT who want to enter and leave the room stand, and defined positions POS _IN and POS _OUT for capturing a face are defined.

The person detecting section 20 is provided at a specified position PO of the door 10.
A first pyroelectric sensor 21 that is provided near S _IN and detects a person standing at a specified position POS _IN and outputs a detection signal DT _1, and is provided near a specified position POS _OUT of the door 10, and is configured to detect that a person standing in the specified location POS _OUT from the second pyroelectric sensor 22 for outputting a detection signal DT _2.

The CCD camera 30 is located near the door 10,
In order to obtain an image centered on the face of the person standing at the specified positions POS _IN and POS _OUT , for example, the height is 1.6.
It is arranged so that it can be rotated horizontally by a predetermined angle at a position of about 2.0 m. The rotation and drive control of the CCD camera 30 are performed by a camera drive unit 31. The camera driving unit 30 detects the detection signal DT _{1 from} the first pyroelectric sensor 21 and the detection signal D from the second pyroelectric sensor 22.
Upon receiving the T _2, is operated by supplying a driving power, etc. to the CCD camera 30, when the detection signal DT ₁ is inputted, the lens faces the person HUM _IN standing in the specified location POS _IN, the detection signal DT _{When 2} is input, rotation control is performed so that the lens faces the person HUM _OUT standing at the specified position POS _OUT . Note that the detection signals DT ₁ and D
When T ₂ are simultaneously input, performs rotation control based on the signal inputted to the first.

The system control unit 40 includes a pre-processing unit 41, a feature extraction unit 42, a memory unit 43, a determination unit 44, and a voice data storage unit 45. Preprocessing unit 41
Represents analog image data I obtained by the CCD camera 30.
Denoising for M, preprocesses such as binarization, and outputs the feature extraction unit 42 as a signal S _41.

The feature extraction unit 42 receives the signal S ₄₁ by the pre-processing unit 41, as shown in FIG. 2, extracts a portion excluding hair portions near of the face image as the center CNT of the face The gray value of the extracted portion is, for example, 128 × 128
To obtain a collation pattern, further divide the collation pattern into, for example, 16 to obtain a mosaic image, and average the densities in the blocks to obtain a 256-dimensional gray value vector.
When the detection signal DT _{1 from} the first pyroelectric sensor 21 of the human detection unit 20 is input, the feature extraction unit 42 stores the mosaic image data in the memory unit 43,
When the detection signal DT _{2 from} the second pyroelectric sensor 22 is input, the mosaic image data is output to the determination unit 44.

When the detection signal DT _{2 from} the second pyroelectric sensor 22 of the human detection unit 20 is input, the determination unit 44 stores mosaic image data of the face of the person who has entered the room stored in the memory unit 43, By comparing with the mosaic image data of the person who is about to leave the room by the feature extraction unit 43,
The degree of coincidence between the two is collated. If the collation degree is within a predetermined reference range, it is assumed that the person who has entered the room first exits, and an open signal _SOPN is generated to allow the person to exit the room, and is output to the key lock mechanism 11 for a certain period of time. Then, a signal S _AUD is output to the voice data storage unit 45 so as to utter a positive message, for example, “please leave the room”. In this case, the data read from the memory unit 43 and subjected to the match determination is deleted from the memory unit 43.

If the degree of collation is out of the predetermined reference range, an open signal is not generated, and a negative message, for example, a signal S _AUD is _issued so as to utter a "data cannot be matched and cannot leave the room". Output to the storage unit 45. In this case, data read from the memory unit 43 and subjected to the mismatch determination is not erased from the memory unit 43. Further, when the storing process of the mosaic image data in the memory unit 43 by the feature extracting unit 42 is completed, the determining unit 44 generates an open signal _SOPN for permitting entry into the room, and outputs the open signal _SOPN to the key lock mechanism 11 for a fixed time. At the same time, it outputs a signal S _AUD to the voice data storage unit 45 so as to utter a positive message, for example, “Please enter”.

Next, the operation of the above configuration will be described with reference to the flowcharts of FIGS. First, when the person who wants to enter the room stands at the specified position POS _IN defined on the front side of the door 10 so that the face faces the area where the CCD camera 30 is arranged, 1 (S1), and the detection signal DT
₁ is output to the camera drive unit 31 and the feature extraction unit 42 and the determination unit 44 of the system control unit 40.

In response to the input of the detection signal DT ₁ , the camera driving section 31 supplies driving power to the CCD camera 30 to operate the CCD camera 30, and the person HUM standing at the specified position POS _IN is operated.
Rotation control is performed so that the portion centered on the face of _IN faces the lens. As a result, the CCD camera 30 captures an image of a portion around the face of the person HUM _IN (S2), and the analog image data is input to the pre-processing unit 41 of the system control unit 40.

In the pre-processing section 41, pre-processing such as noise removal and binarization is performed on the analog image data IM obtained by the CCD camera 30 (S3), and the pre-processed image data is converted into a signal S _41. Is output to the feature extraction unit 41.

[0021] The feature extraction unit 42 receives the signal S ₄₁ by the pre-processing unit 41, a portion excluding a portion around hair of the face image is extracted as a center CNT of the face, shading of extracted portion The values are sampled at 128 × 128 to obtain a matching pattern. If the matching pattern is 16
The mosaic image is obtained by division (S4). here,
The densities in the block are averaged to obtain a 256-dimensional gray value vector. At this time, the feature extraction unit 42, since the detection signal DT ₁ of the first pyroelectric sensor 21 human detecting unit 20 is input, storage processing to the memory unit 43 of the mosaic image data is performed ( S5).

When the process of storing the mosaic image data in the memory unit 43 by the feature extraction unit 42 is completed (S
6), an open signal S _OPN is generated by the determination unit 44 to permit entry into the room, output to the key lock mechanism 11 for a fixed time, and the signal S _AUD is output to the voice data storage unit so as to produce a positive message. 45. As a result, the door 10 is unlocked (S7), and a message "Please enter" is uttered from the door intercom 12, and the entrance applicant HUM _IN opens the door 10 and enters the room.

Next, a person who wants to leave the room moves the face to the specified position POS _OUT defined on the rear side of the door 10.
When the user stands up to the area where the D camera 30 is arranged, the second pyroelectric sensor 22 of the human detection unit 20 detects (ST1), and the detection signal DT ₂ is output from the camera driving unit 31,
The data is output to the feature extraction unit 42 and the determination unit 44 of the system control unit 40.

In response to the input of the detection signal DT ₂ , the camera driving section 31 supplies driving power to the CCD camera 30 to operate the CCD camera 30, and the person HU standing at the specified position POS _OUT is operated.
M _{OU T} face centered portion of the rotation control so directly face the lens is performed. As a result, the CCD camera 30 captures an image of a portion around the face of the person HUM _OUT (ST).
2) The analog image data is input to the pre-processing unit 41 of the system control unit 40.

In the pre-processing section 41, pre-processing such as noise removal and binarization is performed on the analog image data IM obtained by the CCD camera 30 (ST3), and the pre-processed image data is converted into a signal S _41. Is output to the feature extraction unit 42.

[0026] The feature extraction unit 42 receives the signal S ₄₁ by the pre-processing unit 41, a portion excluding a portion around hair of the face image is extracted as a center of the face, the extracted portion of the gray values Are sampled at 128 × 128 to obtain a matching pattern. Then, the collation pattern is divided into, for example, 16 to obtain a mosaic image (ST4). Here again
The densities in the block are averaged to obtain a 256-dimensional gray value vector. At this time, since the detection signal DT _{2 from} the second pyroelectric sensor 22 of the human detection unit 20 is input to the feature extraction unit 41, the mosaic image data is
4 (ST5).

At this time, since the detection signal DT _{2 from} the second pyroelectric sensor 22 of the human detection section 20 is input to the determination section 44, the face of the entered person stored in the memory section 43 is detected. Is compared with the mosaic image data of the person who is about to leave the room by the feature extracting unit 43, and the matching between the two is checked (ST6, ST7). If the collation degree is within a predetermined reference range, it is assumed that a person who has entered the room first exits, and an open signal _SOPN is generated to permit exit, and is output to the key lock mechanism 11 for a fixed time. With a signal S to say a positive message
_{The AUD} is output to the audio data storage unit 45. As a result, the door 10 is unlocked (ST8), and a message "please leave the room" is uttered from the door intercom 12, and the exit applicant HUM _OUT opens the door 10 and exits. In this case, it is read from the memory unit 43,
The data for which the match has been determined is deleted from the memory unit 43.

On the other hand, if the degree of collation is out of the predetermined reference range, no open signal is generated and the signal S _AUD is output to the voice data storage unit 45 so as to produce a negative message. . As a result, a message saying "cannot exit due to data mismatch" is uttered, and exit requester HUM _OUT is refused to exit (ST9). In this case, data read from the memory unit 43 and subjected to the mismatch determination is not erased from the memory unit 43.

As described above, according to the present embodiment,
A door 10 having a key lock mechanism which is unlocked by input of an opening signal, a person who wants to enter the room from the door, and a door 1
Feature extraction that creates mosaic image data from a CCD camera 30 that captures a face image of a person who wants to leave the room from 0 and image data obtained by the camera 30 and stores the mosaic image data of the HUM _IN of the person who wants to enter the memory unit 43 Part 42,
When the process of storing the data in the memory unit 43 by the feature extracting unit 42 is completed, and the mosaic image data of the exit applicant HUM _OUT obtained by the feature extracting unit 42 and the memory unit 4
3. The determination unit 44 compares the mosaic image data stored in the key lock mechanism 11 with the mosaic image data stored in the key lock mechanism 11 and outputs the open signal _SOPN to the key lock mechanism 11 for a certain period of time when both data are within the preset matching condition range. The automatic door management system, which does not require certification tools such as badges, does not require complicated work to prevent loss or theft, is easy to manage, and can control door opening and closing according to each individual Can be realized.

Although this embodiment has been described by taking as an example the case of entering and leaving a certain room, the present invention can also be applied to a door lock system of an automobile or the like. For example, a switch is provided on the door, the camera is set to operate when the switch is pressed, the driver gets out of the car, closes the door, presses the switch to obtain a face image, and performs the same feature extraction as described above. To be stored in the memory unit. Then, when the data storage is completed, the door is in a lockable state, and the door can be locked by pressing a switch provided separately. To unlock the door, press the camera switch to obtain an image of the face, compare the image data with the extracted features with the data stored in the memory, and release the lock if they match. The configuration is as follows.

By adopting such a structure, there is an advantage that the key can be locked without having the key of the car and that only the person who locked the car can unlock it, so that the loss of the key or theft of the car can be prevented.

The feature extraction processing of face data includes:
It goes without saying that the present invention is not limited to the creation of mosaic image data. In addition, it is a matter of course that, instead of the face data, an image of an index finger or the like may be imaged to control the entrance and exit.

[0033]

As described above, according to the present invention,
Proof tools such as badges are not required, no complicated work for preventing loss or theft is required, management is easy, and door opening and closing control can be performed according to each individual.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an embodiment of an automatic door opening / closing system according to the present invention.

FIG. 2 is a diagram illustrating an example of a feature extraction process according to the present invention.

FIG. 3 is a flowchart for explaining an operation at the time of entering a room.

FIG. 4 is a flowchart for explaining the operation at the time of leaving the room.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Door 11 ... Key lock mechanism 12 ... Door intercom 20 ... Human detection sensor 21 ... 1st pyroelectric sensor 22 ... 2nd pyroelectric sensor 30 ... CCD camera 31 ... Camera drive part 40 ... System control part 41 ... Front Processing unit 42 Feature extraction unit 43 Memory unit 44 Judgment unit 45 Voice data storage unit

Claims (1)

(57) [Claims] 1. An image pickup means for picking up an image of a door to be unlocked by input of an open signal, a person who wants to pass from one side of the door, and a person who wants to pass from the other side of the door. Storage means; a feature extraction means for extracting a predetermined feature from image data obtained by the imaging means and storing extracted data of a person who wants to pass from one side of the door in the storage means; door according to when the process of storing the extracted data into said storage means is completed, and stored in the extracted data with the memory means for a person who wishes to pass from the other side of the door obtained by the feature extracting means From one side of
The extracted signal is compared with the extracted data of the person who wants to pass, and the open signal is sent to the lock of the door when both data are within the matching condition range set in advance, at least when they are within the matching condition range. A door automatic opening / closing management system, comprising: a judging means for outputting.