JPS623385A - Picture recognizing device - Google Patents

Abstract

PURPOSE:To perform the input of pictures at a high speed and with high image resolution by using the 1st image pick-up element having a wide viewfield and low image resolution and the 2nd image pick-up element having a narrow viewfield and low image resolution. CONSTITUTION:An image forming lens 6d is set on a single side wall of an image pick-up device 6 and the 1st image pick-up element 6a containing a 2-dimensional image pick-up element is fixed to a face opposite to the lens 6d within the device 6. An optical means 6e which transmits the light partly is set at 45 deg. to the optical path of the lens 6d and the element 6a. The reflected light sent from the means 6e is made incident on the 2nd image pick-up element 6b via a magnification ratio converting lens 6f. The element 6b contains a 2-dimensional image pick-up element and set on a stage 6c which can move toward an XY axis. The pictures obtained by the element 6a with low image resolution are utilized for confirmation of a recognizing subject existing within a reading viewfield and for recognition of the position of a fine pattern. The stage 6c is driven after a reading area is decided and the pictures are supplied through the element 6b.

Description

[Detailed Description of the Invention] [Overview/Summary] The present invention is an image recognition device that uses a two-dimensional image sensor with a low resolution of 1 through a half mirror that transmits part of the input light and reflects part of it. , to a fixed first image sensor with a wide field of view.

The present invention also provides an image recognition device in which the image is guided to a second imaging device which is a slidable one-dimensional imaging device or two-dimensional imaging device with a narrow field of view and high resolution.

[Industrial application field]

The present invention relates to an image recognition device, and more particularly to an image recognition device that can simultaneously capture high-resolution images within a wide field of view.

[Conventional technology]

[Conventional technology] Recently, various robots and 1 assembly in the industrial field.

Image recognition devices are used in inspection equipment used for inspections, and these devices have built-in imaging devices for image recognition, and the imaging devices of the imaging devices include line sensors such as CODs (charge coupled devices). A device using a one-dimensional image sensor is used, such as the one shown in FIG. There is one in which a two-dimensional image is obtained by moving the one-dimensional image sensor on the imaging plane in a direction perpendicular to the electrical scanning direction of the one-dimensional image sensor. Furthermore, a CCD is installed on the two-dimensional surface as a two-dimensional image sensor.
It is also known that there are

[Problem that the invention seeks to solve]

The above-described one-dimensional imaging device generally has a high resolution, but has the disadvantage that it takes a long time to input an image because scanning is performed mechanically to scan one screen.

On the other hand, when using a two-dimensional image sensor, the image input time is approximately 33
Although it is extremely fast (m sec), it has the disadvantage of low resolution.

[Means for solving problems]

The present invention was made in view of the above-mentioned drawbacks, and its purpose is to provide a two-dimensional image sensor and a primary or two-dimensional image sensor in one imaging means to achieve high resolution and quickly input images. The purpose is to obtain an image recognition device, the means of which is to pass through an optical means that partially transmits image input light and partially reflects it, which leads to a fixed first image sensor having a wide field of view and a low resolution, and which has a narrow field of view. This is accomplished by an image recognition device characterized in that it is connected to a slidable second image pickup device with high resolution.

[For production]

In the present invention, image input light is converted into a half mirror for image recognition.
one is applied to a fixed two-dimensional image sensor and the other is applied to a movable primary or two-dimensional image sensor, and based on the imaging output recognized in a wide field of view by the fixed two-dimensional image sensor, the movable one A secondary or two-dimensional image sensor is moved in a predetermined recognition direction, and image recognition data is extracted from a movable high-resolution primary or secondary image output.

〔Example〕

An embodiment of the present invention will be described in detail below with reference to FIGS. 1 to 3. FIG. 1 is a perspective view showing the imaging means of the image recognition device of the present invention, and FIG. 2 is a diagram showing the relationship between the field of view of the first image sensor and the field of view of the second image sensor used in the image sensor of the invention. , FIG. 3 is a diagram showing a pattern produced by the imaging means of the present invention.

In FIG. 1, 6 indicates the imaging device as a whole;
An imaging lens 6d is disposed on one side wall of the imaging device 6, and a first imaging device 6a, which is a two-dimensional imaging device, is fixed to a surface facing the imaging lens 6d inside the traumatizing device. ing. Also, it transmits some light like a half mirror,
The optical means 6e that reflects part of the light is the imaging lens 6d.
It is arranged at an angle of 45 degrees to the optical path of the first image sensor, and the reflected light from the half mirror 6e is imaged on the second image sensor 6b via the magnification conversion lens 6f. The second image sensor 6b is a two-dimensional image sensor.

It is placed on a stage 6C that is movable in the XY axis directions. In the above configuration, the low-resolution image obtained from the first image sensor 6a is used to confirm that a recognition target exists within the reading field of view and to recognize the position where a thin pattern exists, and after the reading area is determined. , the stage is driven and an image is input from the two-dimensional second image sensor 6b.

At this time.

The image inputted from the two-dimensional image sensor 6b is inputted via the magnification conversion lens 6f, so it has a high magnification.

FIG. 2 shows a fixed two-dimensional first image sensor 6a.
The relationship between the field of view and the field of view of the two-dimensional second image sensor 6b on the stage 6C moving in the X-Y direction is shown. The field of view 12 of the second image sensor 6b on the stage 6C can be freely moved within the field of view 13 of the fixed first image sensor 6a.

Therefore, as shown in FIG. 3(a), the first image sensor 6a
When a character pattern 142 or the like with a wide field of view is recognized, the second image sensor 6b is moved in the X-axis direction A.

'It is necessary to move it as shown by B and C. Also, Figure 3 (
B) As shown in FIG.
If recognition is made at point a, the stage 6c on which the second image sensor 6b is mounted is moved in the X and Y axis directions as shown at A, B, C, and D.

Another embodiment of the imaging device of the present invention is shown in FIG.

The light that has passed through the imaging lens 6d and the half mirror 6e forms an image on the two-dimensional first imaging device 6a.

Further, the light reflected by the half mirror 6e passes through a magnification conversion lens 6f for increasing the imaging magnification.

An image is formed on a one-dimensional second image sensor 6b' provided on a uniaxial stage 6c that moves only in the X-axis direction.

The one-dimensional second image sensor 6b' has a higher resolution per line than the two-dimensional first image sensor 6a.
By moving the dimensional second image sensor 6b' in a direction perpendicular to the main scanning direction of the image sensor,
The entire field of view of the low-resolution two-dimensional second image sensor 6a can be covered.

In the above two embodiments, a half mirror was used as an optical means for transmitting part of the light and reflecting part of the light. In order to transmit the light from the imaging lens 6d onto the first image sensor 6a, the mirror is mechanically lowered and in order to enter the second image sensor 6b, 6b', the mirror is moved upward and the light is reflected by the mirror. You may also do so.

FIG. 5 shows a block diagram of an example of an image recognition apparatus according to the present invention corresponding to FIG. 4. The image formed on the first imaging element 6a in the imaging device 6 is outputted as an analog video signal e by the control signal d from the control circuit 1, and converted into a digital video signal f by the analog-digital conversion circuit 2. It is stored in the frame memory 3. Frame memory 3
The image data g stored in the dictionary memory 4 and the dictionary data h stored in advance in the dictionary memory 4 are compared and verified by the recognition circuit 5, and the recognition result i is sent to the control circuit 1.

In the control circuit 1, when the information that the object to be recognized has entered the field of view or the position information of the place to be recognized is obtained, a stage drive start signal C is sent to the stage drive circuit 7, which moves only in the X-axis direction. One-dimensional second image sensor 6b'
Sends an imaging start signal to. The uniaxial stage 6C driven by the stage drive circuit 7 starts moving in a direction perpendicular to the electrical main scanning direction of the one-dimensional second image sensor 6b'. The analog video signal output from the one-dimensional second image sensor 6b'' is converted into a digital video signal β by the analog-digital conversion circuit 8 and stored in the frame memory 9. The data m is compared with the dictionary data n stored in the dictionary memory 10 in advance in the recognition circuit 11 to which the recognition circuit control signal is applied from the control circuit 1.

It is configured to output 0 as a result of character or pattern matching, reading, and inspection.

〔Effect of the invention〕

Since the present invention is configured as described above, it is possible to obtain an imaging device that combines the advantages of an image sensor with a wide field of view and low resolution and an image sensor with a narrow field of view and high resolution.
The present invention has the feature that an imaging device having both the advantages of a dimensional imaging device and a two-dimensional imaging device can be obtained.

[Brief explanation of drawings]

FIG. 1 is a perspective view showing the optical means of the imaging device of the present invention. FIG. 2 is a diagram showing the relationship between the field of view of the first image sensor and the field of view of the second image sensor used in the imaging means of the present invention. FIGS. 3(a) and 3(b) are diagrams showing patterns produced by the imaging means of the present invention. FIG. 4 is a perspective view of an optical means of an imaging device showing another embodiment of the present invention. FIG. 5 is a system diagram of the image recognition device of the present invention. 1... Control circuit. 2.8...Analog-digital conversion circuit. 3.9...Frame memory. 5.11...Recognition circuit. 6...Imaging device. 6a...first image sensor. 6b... second image sensor. 6e Half mirror - Figure 1 Figure 2

Claims (9)

[Claims] (1) Image input light is guided through an optical means that partially transmits and partially reflects to a fixed first image sensor with a wide field of view and low resolution; An image recognition device characterized in that the image recognition device is configured to guide the image to a second image pickup device. (2) The image recognition device according to claim 1, wherein the first and second image sensors are two-dimensional image sensors. (3) The image recognition device according to claim 1, wherein the first image sensor is a two-dimensional image sensor, and the second image sensor is a one-dimensional image sensor. (4) The image recognition device according to claim 2, wherein the second image sensor is slidable in the X and Y axis directions. (5) The image recognition device according to claim 3, wherein the second image sensor is slidable in a direction orthogonal to a scanning direction of the image sensor. (6) The image recognition device according to claim 1, wherein the optical means that partially transmits the image input light and partially reflects the image input light is a half mirror. (7) In order to partially transmit the image input light and partially reflect it, a mirror is moved upward to guide it to the second image pickup device, and by moving the mirror down, it is guided to the first image pickup device. An image recognition device according to claim 1, characterized in that the image recognition device is configured as follows. (8) The analog video signal output from the second image sensor is digitized, stored in a frame memory, and compared with a dictionary memory to output a recognition result. Image recognition device. (9) A drive device that digitizes the analog video signal from the first image sensor, stores it in a frame memory, compares it with a dictionary memory, adds the recognition result to the control circuit, and slides the second image sensor. The image recognition device according to claim 1, characterized in that the image recognition device is driven by a.