JPH0564832B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image processing device that handles two-dimensional binarized images, and particularly to a binarization device thereof.

[Conventional technology]

As a conventional method of this type, for example, an object observation area called an aperture is set in the background part of the imaging field of view, and the image (video) signal level therein is used to correct the threshold value. It is being

[Problem that the invention seeks to solve]

However, in the above-mentioned method: 1. If noise occurs in the background part, the threshold value cannot be corrected properly.

2 Transmitted illumination is often used as a method of illuminating objects. In this case, the background portion of the video signal captured by the imaging device may be saturated, but in such a case, it is meaningless to monitor the video signal level of the background portion.

There is a problem.

Therefore, this invention always provides stable binary values even when the object surface changes (drift) due to external light other than illumination (indoor lighting, outdoor external light) or deterioration of the lighting device. An object of the present invention is to provide a binarization device capable of obtaining a converted image.

[Means for solving problems]

an area forming means for detecting a reference position of a captured image obtained by imaging an object, and forming an object observation area (aperture) having a predetermined size and shape at a position having a predetermined relationship with the reference position; An extraction means for sampling an image signal through this observation area and extracting its average level value, and correcting a preset binarization threshold by comparing this image average level value with its set value. A correction means is provided, and the binarization threshold value is corrected every time an object is imaged, and the next binarization is performed using the corrected binarization threshold value.

[Effect]

A threshold value is appropriately set while observing a binarized image captured by placing an object to be processed within the field of view of an imaging device on, for example, a display device. At this time, an aperture is set at an appropriate position on the binary image, and the average level of the video signal within that area is measured. Furthermore, the X-axis projection and Y-axis projection of the binary image are measured, and the absolute coordinates of the binary image within the field of view are determined.
The relative phase (deviation amount) between this and the preset absolute coordinates of the aperture is determined. The threshold value T, average level value, and relative phase at this time are stored in memory, and after each binarization process, the reference position is detected by calculating the X-axis projection and Y-axis projection of the binary image. .

2 Open the aperture from this reference position to a position determined by the relative phase between the binary image stored in the memory and the aperture.

3. Measure the average level value of the video signal within the aperture area.

4. Compare the measured average level value with the level value stored in memory and correct the threshold value.

Repeat the operation. By doing this, it is possible to obtain a stable binary image even if the illuminance of the object surface changes due to various reasons, and with the addition of an aperture position tracking function, the image position within the field of view can be adjusted. A stable binary image can be obtained every time regardless of the background noise.

〔Example〕

Fig. 1 is a configuration diagram showing an embodiment of this invention;
The figure is an explanatory diagram for explaining an example of a method of detecting a reference position of an image. In FIG. 1, 1 is an imaging position of a television camera, etc., 2 is a binarization circuit, 3 is an image memory, 4 is an X- and Y-axis projection circuit,
5 is an aperture generation circuit, 6 is a level sample circuit, 7 is an analog/digital (A/D) converter, 8 is a central processing unit (CPU), 9 is a threshold latch circuit, 10 is a keyboard, 11 is a CRT, 1
2 is a memory.

First, a threshold value is appropriately set while observing a binarized image captured with the object to be processed placed within the field of view of the imaging position, for example, on a display device. At this time,
While looking at the binary image, an aperture of an appropriate size is set at an appropriate position using an input means such as a keyboard. The position and size of this aperture are determined while looking at the object to be treated so that the aperture is at a predetermined position within the area of the object to be treated. In addition to measuring the average level of the video signal within the aperture area at that time, the absolute coordinates of the binary image within the field of view are determined by measuring the X-axis projection and Y-axis projection of the binary image, and the absolute coordinates of the binary image within the field of view are determined. Find the relative phase (deviation amount) from the absolute coordinates of the aperture. Then, after storing the threshold value T, average level value, relative phase, and aperture size at this time in the memory 12, the next processing is performed on the object to be processed thereafter.

An image capturing device 1 captures an image of an object, and a binarization circuit 2 binarizes the image. Display the binary image on SRT11,
While observing the value image, an appropriate threshold value T is entered from the keyboard 10, set in the latch circuit 9, and stored in the memory 12. Further, the projection circuit 4 obtains an X-axis projection value P _x and a Y-axis projection value P _y from a binary image as shown by P in FIG.
_{Compare with TH}
Detected as. On the other hand, since l _x and l _y shown in FIG. 2 are stored as relative phases in the memory 12,
The aperture generating circuit 5 sets an aperture of the size stored in the memory 12 at the absolute coordinates determined by the sum of the relative phases l _x and l _y and the standard position P _s , as shown by A in FIG. , supplies the video signal within that area to the level sample circuit 6. That is, although not shown, the aperture generation circuit 5 is supplied with the scanning coordinates of the imaging device 1, and the aperture generation circuit 5 compares the scanning coordinates with absolute coordinates for generating an aperture, When the scanning of No. 1 is in the interval of absolute coordinates that generate an aperture, the gate is opened and the imaging device 1
It allows video signals from to pass through. Level sample circuit 6 measures the average level of the video signal within the area and stores it in memory 12. This average level value is obtained by further averaging the average value of the video signal for each horizontal line, for example.

The average level value of the image signal obtained in this way is
After being converted into a digital signal by the A/D converter 7, it is provided to the CPU 8. The CPU 8 compares this average level value with a preset predetermined value, and corrects the preset binarization threshold according to the result. For example, if the average level value obtained this time is L ₁ , its set value is L ₀ , and the set value of the binarization threshold is T, then the corrected binarization threshold T _i is T _i =( It is expressed as L ₁ /L ₀ )×T.

This corrected binarization threshold is given to the threshold latch circuit 9, and is used by the binarization circuit 2 as the next binarization threshold. Therefore, the next image will be binarized using this value, and after the reference position P _s of the image is determined, the operation of correcting the binarization threshold will be repeated. .

In this way, the reference position is determined from the projection value of the image, and the image (not the background) is
Since the aperture is formed at a predetermined position, it is possible to form a so-called automatic tracking aperture that is not affected by noise or illumination, and thereby always obtains a stable binarization threshold. be able to. Although the aperture is rectangular in the above description, it goes without saying that it is not limited to this shape. Further, although the binarization threshold is set while being monitored by a CRT, this can also be done by other appropriate means.

[Effects of the Invention] According to this invention, (a) It is possible to prevent external light other than illumination light (indoor lighting, outdoor external light), deterioration of lighting equipment, or objects of the same type but with different surface densities. There is no need to manually reset the binarization threshold, and a stable binary image can always be obtained.

(b) Since the aperture occurrence position is made to follow the image, it is possible to always monitor the video signal at the optimal position even if the object image changes. In other words, since the image area is observed, the influence of background noise is eliminated, and furthermore, the threshold value can be corrected even when using transmitted illumination.

Benefits such as:

[Brief explanation of drawings]

Fig. 1 is a configuration diagram showing an embodiment of this invention;
The figure is an explanatory diagram for explaining an example of a method of detecting a reference position of an image. Description of symbols, 1... Imaging device, 2... Binarization circuit, 3... Image memory, 4... X and Y axis projection circuit, 5... Aperture generation circuit, 6... Level sample circuit, 7... A/D converter, 8...CPU,
9...Threshold latch circuit, 10...Keyboard, 11...CRT, 12...Memory, P...Image, P _x , P _y ...X-axis, Y-axis projected image, _Ps ...Reference position , l _x , l _y ...X-axis and Y-axis direction deviation amount, TH _x ,
TH _y ...X-axis, Y-axis projection threshold.

Claims (1)

[Claims] 1. An area for detecting a reference position in a captured image obtained by imaging an object and forming an object observation area having a predetermined size and shape at a position in a predetermined relationship with the reference position. a forming means; an extraction means for sampling an image signal through the observation area and extracting an average level value thereof; and a binarization threshold set in advance by comparing the image average level value with the set value. A correction means for correcting the value is provided, and the binarization threshold is corrected each time an object is imaged, and the next image binarization is performed using the corrected binarization threshold. Features of image binarization device. 2. The image binarization device according to claim 1, wherein the reference position is detected based on horizontal and vertical projection values of the object image.