JPH11205776A - Image recognizing and setting system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To set the recognizing area and the threshold of an image recognizing camera by a personal computer and to simply correct set data. SOLUTION: A video signal from the image pickup part 11 of the recognizing camera 1 is digitally converted by an A/D conversion part 12 and outputted to a personal computer 2 from an input/output part 15 to display an image on a display part 22. The recognizing area and the threshold value of an abnormality judging reference are set by a setting part 23, image data and setting data are inputted and retained in a setting file through an input/output part 21 and setting data is send to the camera 1, stored in a storing part 14 through a control part 16 and read to recognize abnormality in the recognizing area of the image from the image pickup part. At the time of correcting set data, data in set file is fetched to a personal computer and corrected to be reset. In addition, data in the part 1 can be read to be transferred to the personal computer to be corrected and data in the setting file can be serially converted by a file data conversion part 3 to be used by the recognizing camera 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image recognition setting method, and more particularly to a method for setting recognition data of an image recognition camera or the like.

[0002]

2. Description of the Related Art A conventional image recognition apparatus converts an image signal from an image pickup unit 11 of an image recognition camera 30 into a digital signal by an A / D converter 12 as shown in FIG.
And input to the D / A conversion unit 31 via an interface, convert to an analog signal, input to the monitor 40, set the recognition area by operating the setting device 50 (pointing device, etc.) while watching the screen of the monitor 40, This is to set the threshold value for abnormality detection.
The setting data is stored in the storage unit (EEPROM) 14 via the control unit 32, and thereafter, the recognition unit 13 calculates the luminance difference between the frames in the recognition area of the image from the imaging unit 11 via the control unit 32. 14
If it is larger than the read threshold, it is assumed that an abnormality has occurred. In this case, a dedicated monitor 40 and a dedicated setting device 50 for setting a recognition area and a threshold are required, and when the data stored in the storage unit 14 is to be corrected, the above-described setting operation is performed from the beginning. There is an annoyance that it must be done.

[0003]

SUMMARY OF THE INVENTION In view of the foregoing, the present invention uses a personal computer (hereinafter abbreviated as a personal computer) to set an image recognition area and a threshold, and to use a dedicated setting device and monitor. It is an object of the present invention to make it unnecessary and to easily correct data once set.

[0004]

In order to achieve the above object, in an image recognition setting method according to the present invention, a video signal from an image pickup section is converted into a digital signal and output, and image recognition based on the video signal is performed. A recognition camera that outputs a signal indicating that there is an abnormality when the luminance difference between the frames of the region is equal to or greater than a required threshold value, captures an image from the recognition camera, sets a recognition region, and sets a threshold value as a reference for abnormality determination. , And a personal computer that sends the setting data of the recognition area and the threshold to the recognition camera.

The recognition camera has a storage unit for storing the data of the recognition area and the threshold value, and a recognition unit for recognizing the abnormality of the recognition area. The storage unit stores the data of the recognition area and the threshold value from the personal computer. The recognition unit recognizes an abnormality in the recognition area read from the storage unit using an image based on the video signal from the imaging unit.

A first input / output unit for inputting / outputting serial data to / from the recognition camera is provided. The first input / output unit outputs a video signal from the imaging unit converted into a digital signal to a personal computer. Data is input and stored in the storage unit.

The personal computer has a second input / output unit for inputting / outputting serial data, and inputs a video signal from a recognition camera through the second input / output unit, displays the image signal on a screen, and recognizes the image with reference to the screen. The setting of the area and the setting of the threshold value are performed, and the setting is transmitted to the recognition camera via the second input / output unit.

The data stored in the storage section of the recognition camera can be read out and input from the first input / output section to the personal computer via the second input / output section to correct and reset the recognition area or threshold. .

Then, a setting file for recording the image taken into the personal computer and the data of the recognition area and threshold set by the personal computer is provided, and the image and the setting data read from the setting file are input to the personal computer, and the recognition area or Modify the threshold so that it can be reset.

[0010]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described based on embodiments with reference to the drawings. FIG. 1 is a block diagram of a main part of an embodiment of an image recognition setting method according to the present invention. In the figure, 1 is a recognition camera, 2 is a personal computer, 3 is a file data converter, and 4 is a setting file (hard disk, floppy disk, etc.). Reference numeral 11 of the recognition camera 1 denotes an image pickup unit, which picks up an image with a CCD (Charge Coupled Device) and outputs a video signal. Reference numeral 12 denotes an A / D converter, which converts a video signal from the imaging unit 11 into a digital signal, outputs the digital signal to the personal computer 2 through the first input / output unit 15, and outputs the digital signal to the recognition unit 13. Recognition unit 13
The storage unit 14 via the control unit 16 with the image from the A / D conversion unit 12
Then, the abnormality of the read recognition area is recognized. The storage unit 14
First input / output unit 15 with serial data set on PC 2
The data of the recognition area and the data of the threshold value of the criterion of the abnormality determination input through the CPU are stored. Reference numeral 21 of the personal computer 2 is a second input / output unit for inputting image data from the A / D conversion unit 12,
The data is written to and read from the storage unit 24 via the control unit 25, and is input to the display unit 22. Reference numeral 23 denotes a setting unit that sets a recognition area with reference to the screen of the display unit 22 and sets a threshold that is used as a criterion for abnormality determination. These set data are output to the recognition camera 1 as serial data from the second input / output unit 21 via the control unit 25,
At the same time, input and save in the setting file 4.

Next, the operation of the image recognition setting method according to the present invention will be described. The image to be recognized is captured by the imaging unit 11 of the recognition camera 1, converted into a digital signal by the A / D converter 12, and input to the personal computer 2 from the first input / output unit 15 via the second input / output unit 21. Then, the data is written to the storage unit 24 via the control unit 25, read out, and input to the display unit 22 to display the image shown in FIG. While viewing this screen, the setting unit 23 sets a recognition area (for example, a rectangular area from the coordinates x1, y1 to the coordinates x2, y2 in the figure), and sets a threshold as a reference for abnormality determination.
These set data are temporarily stored in the storage unit via the control unit 25.
24, read these together with the recorded image data, input them to the setting file 4 as serial data via the second input / output unit 21, and file them, and at the same time, set the recognition area and threshold setting data to the recognition camera 1. To send to. Data input from the first input / output unit 15 is written to the storage unit 14 via the control unit 16, and thereafter, an image captured by the imaging unit 11 and input via the A / D conversion unit 12 (the image in FIG. 2). At the recognition unit
13, the recognition area (coordinates x1, y1
, A luminance difference between frames in a rectangular area from (x2, y2) to the coordinates x2, y2) is calculated. If this value is larger than the threshold value read from the storage unit 14, a recognition signal indicating abnormality is output.

When correcting the set data, the image data and the setting data are read out from the setting file 4, taken into the personal computer 2 through the second input / output unit 21, and once written into the storage unit 24 via the control unit 25, respectively. The image data is read out, displayed on the display unit 22, the setting data is read out, and the setting unit 23 is read out.
Correct and reset with. The reset data is stored again in the setting file 4 together with the image data, and the setting data is sent to the recognition camera 1. Alternatively, the setting data stored in the storage unit 14 of the recognition camera 1 may be corrected. That is, data in the storage unit 14 is read out via the control unit 16, input to the personal computer 2 from the first input / output unit 15 via the second input / output unit 21, written in the storage unit 24, read out, and set by the setting unit 23. Make corrections and reset.

The setting data stored in the setting file 4 may be directly taken into the recognition camera 1. That is, the setting data read from the setting file 4 is converted into serial data by the file data conversion unit 3, input to the recognition camera 1 from the first input / output unit 15, written into the storage unit 14 via the control unit 16, and The reading and recognition unit 13 performs the recognition process.

[0014]

As described above, according to the image recognition setting method of the present invention, an image from a recognition camera is taken into a personal computer, and a recognition area and a threshold are set by operating the personal computer. Costs can be saved because no dedicated setting device and monitor are used for setting, and the captured image and setting data are saved in the setting file, so only the necessary parts are corrected using the previous setting data. This makes it possible to save time and effort as compared with the case of resetting everything.

[Brief description of the drawings]

FIG. 1 is a main block diagram of an embodiment of an image recognition setting method according to the present invention.

FIG. 2 is an example of setting a recognition area.

FIG. 3 is a block diagram of a main part of an example of a conventional image recognition setting method.

[Explanation of symbols]

 1, 30 Image recognition camera 2 Personal computer 3 File data conversion unit 4 Setting file 11 Imaging unit 12 A / D conversion unit 13 Recognition unit 14, 24 Storage unit 15, 21 Input / output unit 16, 25, 32 Control unit 22 Display unit 23 Setting unit 31 D / A conversion unit 40 Monitor 50 Setting device

Claims (6)

[Claims] 1. A video signal from an imaging unit is converted into a digital signal and output, and a signal indicating an abnormality is output when a luminance difference between frames of an image recognition area based on the video signal is equal to or larger than a required threshold. And a personal computer that captures an image from the recognition camera, sets a recognition area, sets a threshold that is a reference for abnormality determination, and sends setting data for the recognition area and the threshold to the recognition camera. Image recognition setting method. 2. The recognition camera according to claim 1, further comprising: a storage unit configured to store the data of the recognition area and the threshold value; and a recognition unit configured to recognize an abnormality in the recognition area. 2. The image recognition setting method according to claim 1, wherein data of a threshold value and a threshold value are stored, and the recognition unit recognizes an abnormality in a recognition area read from the storage unit using an image based on a video signal from the imaging unit. A first input / output unit for inputting / outputting serial data to / from the recognition camera; outputting a video signal from the imaging unit converted into a digital signal to the personal computer; 3. The image recognition setting method according to claim 2, wherein data of an area and a threshold is input and stored in said storage unit. 4. The personal computer has a second input / output unit for inputting / outputting serial data, inputs a video signal from the recognition camera via the second input / output unit, displays the video signal on a screen, and displays the screen. 4. The image recognition setting method according to claim 1, wherein the setting of the recognition area and the setting of the threshold value are performed with reference to the setting, and the setting is transmitted to the recognition camera via a second input / output unit. 5. The data stored in the storage unit of the recognition camera can be read out and input from the first input / output unit to the personal computer via the second input / output unit to correct and reset the recognition area or threshold. 5. The image recognition setting method according to claim 4, wherein: 6. A setting file for storing an image captured by the personal computer and data of a recognition area and a threshold set by the personal computer is provided, and the image and setting data read from the setting file are input to the personal computer. 6. The image recognition setting method according to claim 1, wherein a recognition area or a threshold value can be corrected and reset.