JP2000357236A - Object body discrimination system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain the noncontact type discrimination system whose communication distance between an object to be discriminated and a reader is sufficiently long and whose read speed is sufficient fast by picking up an image of the object to be discriminated and inputting the image as image data. SOLUTION: An image data capture means 21 consisting principally of a CCD camera fetches, as a piece of image data, a design of specific pattern including the discrimination information imparted on the surface of the object to be discriminated. A numerizing means 22 numerizes the inputted image data by means of quantization and discretization, and a discrimination element extracting means 23 extracts discrimination elements that the object to be discriminated has from the numerized image data and outputs them as discrimination data. A comparing means 25 compares the extracted discrimination data with the reference data stored in a reference data holding means 24, and a responding means 26 performs previously set responding operation according to the comparison result.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention identifies an object based on image data of the object captured by using optical means, and automatically performs a required response operation according to the identification result. And a non-contact type target object identification system.

[0002]

2. Description of the Related Art There is a so-called electronic lock as an application example of a conventional non-contact type object identification system. This electronic lock reads the identifier such as an identification card possessed by a resident of the house or the like by a non-contact type reader installed at a door of the house or the like, thereby locking or unlocking the door. The lock operation is performed electronically. Thus, in this electronic lock, the door is locked or unlocked only when the identification element held by the key identifier matches the reference data held by the reader.

[0003] As another application example, there is a dangerous area sensing device used to prevent an accident such as a blind person falling off a platform of a station, for example. This danger area sensing device reads the identifier embedded along the edge of the station platform (near the railroad track), for example, by a reader attached near the tip of a white cane used by the blind, and issues a warning. It is made to emit. Thus, in this danger area sensing device, when a blind person or the like approaches the edge of the platform of the station, the reader attached to the tip of the white stick reads the data of the identifier, and the user himself is now in the home. The fact that the user is near the edge is notified via sound or vibration transmitting means.

In the above-mentioned conventional non-contact type object identification system, as a means for reading identification information, an electromagnetic induction effect generated between an identifier and a reader is used.

[0005]

In a conventional non-contact type object identification system utilizing electromagnetic induction as means for reading identification information, the communication distance between the identifier and the reader is short, so that the two are sufficiently close to each other. Otherwise, a predetermined response operation is not performed. Further, there is a disadvantage that the reading speed is relatively slow, and the time until the response operation is performed is long.

An object of the present invention is to provide a non-contact type object identification system having the following advantages.

(A) The communication distance between the identification object and the reader is sufficiently long.

(B) The reading speed is high.

[0009]

Means for Solving the Problems In order to solve the above problems and achieve the object, a target object identification system according to the present invention is configured as follows.

(1) The target object identification system of the present invention comprises:
A design including an identification element on the surface is provided, and an identification target placed at a specific location, image data capture means for capturing image data by imaging the identification target, and capture by the image data capture means. Digitizing means for digitizing the image data by quantization or discretization, and identification element extraction for extracting identification elements held by the identification object from the image data digitized by the digitizing means Means, reference data holding means for storing and holding the identification elements held by the identification object as reference data in advance, and identification data extracted by the identification element extraction means, stored and held by the reference data holding means. Comparing means for comparing with the reference data
Response means for performing a set response operation based on the result of the comparison by the comparison means.

(2) The target object identification system of the present invention
The system according to (1), wherein the identification element extracting unit performs a differentiation process on the image data digitized by the digitizing unit to obtain a change point of brightness of an image. Contour extracting means for extracting a contour of the image based on a change point of brightness of the image obtained by the differential processing means; and a peak value of the contour of the image extracted by the contour extracting means. And a peak value detecting means to be obtained.

(3) The target object identification system of the present invention comprises:
The system according to (1), wherein the identification element extracting means converts the image data quantified by the quantifying means into a signal having a frequency component in polar coordinates with respect to the image data. Two-dimensional FFT processing means for performing two-dimensional FFT processing;
Frequency peak value detection means for detecting a frequency peak value from a signal having frequency components obtained by the two-dimensional FFT processing by the conversion processing means.

(4) The target object identification system of the present invention comprises:
The system according to (1), wherein the identification element extracting unit divides a region on the screen of the image data quantified by the quantifying unit into a plurality of small regions having a fixed area. Means, and re-numerical means for numerically converting the numerical information of the image data in each small area divided by the area dividing means.

(5) The target object identification system of the present invention comprises:
The system according to (1), wherein the image data capturing unit uses a CCD camera as an image capturing unit.

[0015]

DESCRIPTION OF THE PREFERRED EMBODIMENTS (First Embodiment) FIG. 1 is a perspective view showing a schematic configuration of a target object identification system according to a first embodiment of the present invention. As shown in FIG. 1, the platform 1 of the station
The identification object 11 is provided along an edge on the top of the object 0. The identification target 11 has a plurality of identifiers 11a, 11b,... Each having a design including identification information on its surface arranged in a band at predetermined intervals. Each identifier 11a, 1
For example, a two-dimensional code or a bar code having a predetermined pattern is used as a design including identification information provided on the surface of 1b.

The reader 12 is compactly housed in a small case formed so as to be easily carried by a user. A CCD camera 13 capable of optically imaging the identification target 11 is attached to one end surface (lower end surface in the figure) of the case. In addition, a speaker 14 and a display 15 for notifying the user that a specific identification target has been detected are mounted on the front of the case. Further, an earphone jack 16 is provided on the side surface of the case, so that the earphone 17 can be connected as needed. A grip portion 18 is provided on the other end surface (the upper end surface in the figure) of the case so that the reader 12 can be suspended.

An electronic processing circuit (not shown in FIG. 1) for performing image processing and the like on the image information picked up by the CCD camera 13 is housed inside the case.

FIG. 2 is a block diagram showing a configuration of the electronic processing circuit 20 for performing the above-described image processing and the like. In FIG. 2, the image data capturing means 21 is mainly composed of the above-mentioned CCD camera 13 and
Is captured, a design of a predetermined pattern including identification information applied to the surface of each identifier 11a, 11b,... Is taken as one image data.

The digitizing means 22 quantizes the image data captured by the image data capturing means 21,
Digitize by discretization. The image data is a set of observed patterns (pattern space) and generally has a very high-dimensional continuous phase. Therefore, the feature space is extracted from the pattern, the information is compressed, and the dimension is reduced, thereby performing conversion into the feature space. After that, the data is divided into a finite number of areas, classified, identified, quantized, and transformed into a discrete space of a finite category (symbol) by a discontinuous mapping into a category space. By performing such processing, the image data is digitized.

The identification element extracting means 23 converts the image data digitized by the digitizing means 22 into the identification object 1.
1 to extract an identification element held by the element 1 and output it as identification data. The details of the identification element extracting means 23 will be described later.

The reference data holding means 24 stores the identification object 1
1 stores the identification element originally held in the memory as reference data in advance.

The comparing means 25 includes the identification element extracting means 2
The identification data extracted in 3 is compared with the reference data stored and held by the reference data holding means 24. That is, pattern matching is performed. There are several methods for pattern matching. For example, when the correlation method is used, data f (x,
y) and the template t (x, y) included in the reference data
Is calculated, and the matching is determined based on the similarity.

The response means 26 performs a preset response operation based on the result of comparison by the comparison means 25. For example, when it is detected that the user has approached the platform of the station due to the coincidence of the two data, the fact is notified to the user via a transmission medium such as voice, image, or vibration.

The discriminating element extracting means 23 includes a differential processing means 27, a contour extracting means 28, and a peak value detecting means 2.
It consists of nine.

The differentiating means 27 is provided by the digitizing means 22
The point of change (edge) in the brightness of the image is detected by differentiating the image data quantified by. There are several methods for detecting a change point (edge). For example, if a difference type operator is used, the image f (x, y)
The intensity and direction of the edge are obtained by the spatial differentiation processing of.

The contour extracting means 28 is provided with the differential processing means 2
7. The outline of the image is extracted on the basis of the change point of the brightness of the image obtained by Step 7.

The peak value detecting means 29 includes the contour extracting means 2
8, the peak value, that is, the center of the edge defining the contour is obtained.

FIG. 3 is a diagram showing a pattern of the identification element extraction by the identification element extraction means 23. FIG. 3A shows the identification target 11 captured by the image data capturing unit 21.
FIG. 4 is a diagram showing an example of image data for one identifier, and shows an identification element held by the identifier, that is, a two-dimensional code of a predetermined pattern. When a differentiation process is performed on the image data to extract a contour, a contour as shown in FIG. 3B is obtained. The identification data obtained by performing the peak value detection on the contour obtained in this manner is compared with the reference data stored and held in the reference data holding means 24.

[Operation] The operation of the system of the present embodiment configured as described above will be described. Now, it is assumed that the user carrying the reader 12 has come to the vicinity of the identification target 11 on the platform 10 of the station. Then, the identification object 11 is imaged by the CCD camera 13, and image data for one identifier is captured. Then, a series of image processing is performed by the above-described units, and it is finally determined whether or not the identification target 11 is a target to be detected. As a result of the comparison determination, when it is determined that the identification target object 11 is the target object to be detected, the response unit 26
Informs that the user is currently near the edge on the home 10. That is, the speaker 14 or the earphone 17 notifies the user of the above situation by voice, and the display 15 also displays the situation on the screen. Therefore, the user can avoid danger based on the notification content.

[Operation] In the conventional object identification system using the electromagnetic induction effect, the communication distance between the reader and the identifier is only about several tens of centimeters. , The detection was not performed. In other words, there was a problem that it was not possible to provide sufficient “room” for ensuring safety.

In this respect, in the target object identification system according to the present embodiment, the identification target 11 is
, The communication distance, that is, the detectable distance extends to several meters. For this reason, if the user holding the reader 12 carries the CCD camera 13 in a state where the direction of the CCD camera 13 is directed obliquely downward or forward, for example, at a position considerably before the area where the identification target 11 is laid, The presence of the identification target 11 can be detected.
In this way, it is possible to provide sufficient “room” for maintaining safety, and improve safety.

(Second Embodiment) FIG. 4A is a block diagram showing a configuration of an identification element extracting means 30 in a reader of a target object identification system according to a second embodiment of the present invention. This discriminating element extraction means 30 is a two-dimensional FFT
And a frequency peak value detecting means 32.

The FFT processing means 31 performs a fast Fourier transform in order to process the image data with the concept of spatial frequency. The spatial frequency indicates the number of repetitions of a sinusoidal shading change per unit length.

The above-mentioned frequency peak detecting means 32 includes an FF
The peak value of the frequency is detected from the signal having the spatial frequency component obtained by the fast Fourier transform performed by the T processing means 31.

FIG. 5 is a diagram showing a pattern of the identification element extraction by the identification element extraction means 30. FIG. 5A shows the FFT of each of the image data having the same pattern as the image data shown in FIG. 3A at the positions A, B, C, and D.
Is executed (however, only horizontal scanning is illustrated). The change in brightness on each scanning line is one cycle at the position A, two cycles at the position B, one cycle at the position C, and two cycles at the position B. FIG. 5B shows the result obtained by replacing the result with a frequency level and showing the peak value.
Even if the identification element extracting means 30 is used, the same operation and effect as in the first embodiment can be obtained.

(Third Embodiment) FIG. 4B is a block diagram showing the configuration of the identification element extracting means 40 in the reader of the target object identification system according to the third embodiment of the present invention. This identification element extracting means 40 is provided by the area dividing means 4
1 and a re-quantification means 42 for each area.

The area dividing means 41 divides the image data quantified by the numerical means 22 into a plurality of small areas having a predetermined area.

The re-numerizing means 42 is provided by the area dividing means 4
Characterize each small area divided into areas by 1.
Re-digitization of "1" or "0" is performed.

FIG. 6 is a diagram showing a pattern of identification element extraction by the identification element extraction means 40. FIG. 6A shows a situation in which image data having the same pattern as the image data shown in FIG. 3A is divided into a plurality of small areas by a grid G having a predetermined area. Is shown. Here, the resolution can be changed by variably setting the size of the divided area of the grid G.
When the numerical values are re-quantized for each of the divided areas, the result is as shown in FIG. The same operation and effect as in the first embodiment can be obtained even when the identification element extracting means 40 is used.

(Modifications) The target object identification system shown in the above embodiment includes the following modifications. A device provided with a reader having a structure in which only the CCD camera 13 is attached to the tip of a white cane, a frame of glasses, a part of shoes, or the like.

The reader 12 can be detachably mounted in a predetermined posture on a belt worn on the waist of the user.

[0042]

According to the present invention, the object to be identified is, for example, C
Since the object to be identified is identified by capturing it as an image with a CD camera or the like and electronically processing the image, the communication distance between the object to be identified and the reader is sufficiently long.
In addition, a non-contact type target object identification system with a high reading speed can be provided.

[Brief description of the drawings]

FIG. 1 is a perspective view showing a schematic configuration of a target object identification system according to a first embodiment of the present invention.

FIG. 2 is a block diagram showing a configuration of an electronic processing circuit for performing image processing and the like of the target object identification system according to the first embodiment of the present invention.

FIG. 3 is a diagram showing a pattern of identification element extraction by an identification element extraction unit of the target object identification system according to the first embodiment of the present invention.

FIG. 4A is a block diagram showing a configuration of an identification element extracting means in a reader of a target object identification system according to a second embodiment of the present invention, and FIG. 4B is an object according to a third embodiment of the present invention; It is a block diagram showing composition of an identification element extraction means in a reader of an object identification system.

FIG. 5 is a diagram showing a pattern of identification element extraction by an identification element extraction unit of the target object identification system according to the second embodiment of the present invention.

FIG. 6 is a diagram showing a pattern of identification element extraction by an identification element extraction unit of a target object identification system according to a third embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Station platform 11 ... Identification object 11a, 11b ... A plurality of identifiers 12 ... Reader 13 ... CCD camera 14 ... Speaker 15 ... Display 16 ... Earphone jack 17 ... Earphone 18 ... Grip part

Claims (5)

[Claims] 1. A design including an identification element on a surface,
An identification object installed at a specific location, image data capturing means for capturing image data by imaging the identification object, and quantizing or discretizing the image data captured by the image data capturing means. Digitizing means for digitizing the image data, identification element extracting means for extracting an identification element held by the identification object from the image data digitized by the digitizing means, Identification element, reference data holding means for storing and holding in advance as reference data, identification data extracted by the identification element extraction means,
A comparison unit that compares the reference data stored and held by the reference data holding unit; and a response unit that performs a set response operation based on a comparison result by the comparison unit. Target object identification system. 2. The differentiation element extraction means: differentiation processing means for differentiating the image data quantified by the quantification means to obtain a change point in brightness of an image; Contour extracting means for extracting a contour of the image based on the obtained change point of brightness of the image; peak value detecting means for obtaining a peak value for the contour of the image extracted by the contour extracting means; The target object identification system according to claim 1, comprising: 3. The discriminating element extracting means includes: converting the image data quantified by the quantifying means into
Two-dimensional FFT processing means for performing two-dimensional FFT processing in polar coordinates on the image data so as to convert the image data into a signal having frequency components; and two-dimensional FFT processing by the two-dimensional FFT processing means From the signal containing the frequency components obtained by
The target object identification system according to claim 1, comprising frequency peak value detection means for detecting a frequency peak value. 4. The image processing apparatus according to claim 1, wherein the identification element extracting means converts an area on the screen of the image data quantified by the quantifying means,
Area dividing means for dividing into a plurality of small areas having a fixed area; and re-numerical means for re-numerizing the numerical information of the image data in each of the small areas divided by the area dividing means. The target object identification system according to claim 1, wherein 5. The target object identification system according to claim 1, wherein said image data capturing means uses a CCD camera as an image capturing means.