JPH0724066B2 - Binary level automatic correction method - Google Patents

Description

The present invention relates to a method of resetting a binarization level in an image processing apparatus.

[Conventional technology]

For example, in automobile parts inspection lines and the like, lines are automated using a binary image processing device.
In this case, if the binarization level of the binary image processing device is fixed, stable image acquisition may not be possible due to deterioration of the illumination device or change in illuminance of the illumination device due to power supply fluctuation.

FIG. 3 is an example showing a grayscale image (FIG. 3 (a)) and its histogram (FIG. 3 (b)) when the object is a metal. Since the surface of the object is uniform, the distribution range of the brightness level of the histogram for the object becomes very narrow.

When the binarization level is reset for such an object, good results can be obtained by using an automatic threshold selection method based on discrimination and the least squares criterion (for example, the method proposed by Otsu). I know. However, this method is a statistical method, and since the processing time is long, it cannot be applied to parts inspection lines. Therefore, P-
The tile method may be applied.

[Problems to be solved by the invention]

However, this P-tile method has the following problems.

It is assumed that the number of pixels is added from the highest brightness level of the histogram, the brightness level at which the total number of pixels becomes 100% of the reference area is obtained, and the level is set to the binarization level. Then, as shown in FIG. 4 (a),
Since the variation in the distribution width of the brightness level of the target histogram is small, the binarization level approaches the target histogram. When the illuminating device is a commercial fluorescent lamp, the detected brightness level at the time of inputting a new image may be lowered as shown in FIG. 4 (b) due to the flicker or the influence of illuminance change due to power supply fluctuation. In this case, the target image will have an area of only 60%, for example, and a component determination error will occur.

In addition, even in the case of the same type of parts, there are variations in the manufacture of individual objects. Therefore, in the case of an object slightly smaller than the reference area, when the lightness level at which the number of pixels is 100% is obtained, it may enter the level on the background side as shown in FIG. 5 (a). When binarization is performed at this level, when the detection level rises due to the same detection level fluctuation as above, as shown in FIG. 5 (b), the background significantly enters the target, the area increases, and a component determination error occurs. It will be.

For this reason, it is difficult to reset the binarization level, and for the deterioration of the lighting device, the binarization level should be manually reset periodically, or the lighting level may be changed due to power fluctuation. Was dealing with the AVR settings etc.

The present invention has been made in view of such conventional problems, and when the inspection target is a metal or the like, an optimum binary value is provided in response to deterioration of the illumination device, illumination change due to power supply fluctuation, and the like. The aim is to provide an easy way to determine the level of activation.

[Means for solving problems]

In order to achieve this object, according to the binarization level automatic correction method of the present invention, the brightness level of the histogram of the image is divided into two levels, that is, the target and the background, and the variation in the brightness of the histogram corresponding to each target and the background is small. When a binary image is processed on an object to be inspected such as metal, the image is captured and analyzed using the previously obtained binarization level, and when it is confirmed that the object is the target object, the image The maximum or minimum level of brightness of the histogram of, that is, the number of pixels of each level is taken out in order from the side corresponding to the target and added, and the brightness level at which the total value becomes a predetermined ratio of the target number of reference pixels is set. Obtained from that level, only a certain level, low brightness level or high brightness direction,
That is, the feature is that the level moved to the side corresponding to the background is set as the next binarization level.

[Action]

In the present invention, as shown in FIG. 1, the brightness level at which the total number of pixels is 70% of the target reference area by the P-tile method.
The LEV is obtained, and the level moved by the constant value Δ1 in the direction in which the target area increases is set as the next binarization level.

The LEV obtained in this way is within the target histogram even if the area variation of the target object is, for example, about ± 20%, and as shown in the problem point, does not enter the background side level. .

The level that is Δ1 away from the LEV (this time THR in Fig. 1)
By setting the binarization level to, it is possible to solve the influence of the instantaneous fluctuation of the illumination due to the flicker of the illumination as shown in the problem.

〔Example〕

Hereinafter, the present invention will be specifically described based on the embodiments shown in the drawings. FIG. 2 shows the arrangement of parts on a truck when the present invention is applied to an assembly line. On the dolly,
As shown in, a set of fixed parts is carried on a conveyor. Therefore, the binary image processing device checks whether or not the component is placed and the type of the component.

Since this component is a metal such as a gear whose surface is uniform, its histogram is as shown in FIG.

Further, the lightness detection levels are different because the materials (colors) are different for each part and the truck is considerably large. Therefore, a window is set for each component and each component has a binarization level.

In this example, after capturing an image, determining a component, and confirming that the component is correctly placed, the image is retaken and the binarization level for the next image processing is obtained by the present invention. At this time, the processing is repeated by reading out the histogram data, adding it, and comparing it with the preparation value, and the processing time is short.

It should be noted that if a shadow or the like enters at the time of capturing the histogram, the binarization level becomes an unexpected value. Therefore, as shown in FIG. 1, the binarization level obtained this time and the previous two
Compared with the digitization level (previous THR), if the brightness difference is within a certain value (± Δ2), the value is used for the next image processing.
The value level is used.

By doing so, the above-mentioned adverse effects can be eliminated.

Further, by setting the lower limit LEV-L (see FIG. 1) of the binarization level and issuing an alarm when the binarization level obtained this time becomes equal to or lower than this LEV-L, It is possible to know when maintenance is required due to deterioration of the lighting device.

By resetting the binarization level by the above method,
The above equipment can capture a binary image that is stable with respect to the illuminance change level.

〔The invention's effect〕

As described above, in the present invention, the level at which the total number of pixels becomes a preset ratio of the number of target pixels is obtained, and the level obtained by moving a certain level from that level is set as the next binarization level. ing. Therefore, even if the object to be inspected is a metal or the like, the binarization level can be optimally reset, and it is possible to cope with the secular change of the lighting device or the illuminance change due to the power supply fluctuation. Can be simplified.

[Brief description of drawings]

FIG. 1 is an explanatory view showing how to obtain a binarization level according to the present invention,
2 is an explanatory view of an embodiment of the present invention, FIG. 3 is an explanatory view of an embodiment of the present invention, FIG. 4 is an example of a histogram of a metal object,
FIG. 5 is an explanatory diagram showing a problem when the binarization level is set by the conventional method.

Claims (1)

[Claims] 1. When performing binary image processing on an object to be inspected, such as a metal, in which the brightness level of the histogram of the image is divided into two levels, that is, the object and the background, and the brightness of the histogram corresponding to each of the object and the background is small. , As a result of capturing and analyzing the image using the binarization level obtained last time, when it is confirmed that the target object is the target object, each level in order from the highest level or the lowest level of the brightness of the histogram of the image. The number of pixels of is taken out and added, the lightness level at which the total value becomes a predetermined ratio of the target reference number of pixels is obtained, and a certain level is moved from that level toward the low or high lightness level direction. A binarization level automatic correction method, characterized in that the level is set to the next binarization level.