JPH0687263B2 - Image binarization method - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to an image processing apparatus that processes an image input through an image input apparatus such as a television camera and inspects and confirms the image, and particularly, its binary value. Regarding the conversion method.

[Conventional technology]

Conventionally, as this type of binarization method, for example, a method using a density histogram of an image signal digitalized in pixel units is known.

Fig. 5 is for explaining such a method.
Is called the P-tile method, and the same figure (b) is called the modal method. That is, the former is the position where the area ratio S ₂ / (S ₁ + S ₂ ) is R (%), where S _{1 is} the area of the background part and S ₂ is the area of the graphic part in the density histogram shown in FIG. The binarization threshold value TH is set to, and it is premised that the area ratio is known in advance.

On the other hand, the latter is to detect the valley of the density histogram and set the threshold TH at this position.

[Problems to be solved by the invention]

However, the above-mentioned methods cannot be applied to those for which the area ratio is unknown or for which the valleys in the histogram cannot be detected, and thus the target is limited, and in reality, there are many such targets. There is.

By the way, the density histogram is ideally shown in FIG.
It is desirable to obtain It should be noted that even if there are 5 types of images, each of them is separated as K ₁ , K ₂ , K ₃ , K ₄ and K ₅ ,
It is an extracted example. However, in reality, when the image is input, the density level “dulls” due to shadows and blurring in the illumination system,
For such "sagging" of the density level due to the spatial frequency characteristics of the sensor system (image input device), the concentration histogram as FIG. 6 (b), block of K ₁ ~K ₅ is connected without being separated It often happens. As a result, for example, when there are two images Ki and Kj as shown in FIG. 7, their true contours Zi and Zj
Due to the above influence, it spreads to Zi 'and Zj', so that it becomes impossible to set the binarization threshold at the position of the true contour.

In other words, even such a problem cannot be dealt with by the above-mentioned conventional methods, and the advent of appropriate means has been desired.

Therefore, it is an object of the present invention to provide a binarization method capable of easily setting a threshold level and binarizing any density histogram.

[Means for solving problems]

Separation means for separating into a first part and a second part other than that, the relationship between the density level range and the frequency in the density histogram is substantially constant, and only the first part based on the output from this separation means. And a histogram creating means for creating another density histogram including the density histogram.

[Action]

The reason why the binarization threshold level cannot be determined by the density histogram as shown in FIG. 6B is that there is a sloped portion (second portion) as G _{1 to} G _{6 in} FIG. The separation means performs the separation, and the histogram creation means creates a density histogram as shown in FIG. 6 (a) excluding this portion, so that the binarization threshold level can be easily determined.

〔Example〕

FIG. 1 is a block diagram showing an embodiment of the present invention. In the figure, 1 is an analog / digital (A / D) converter, 2
Is an image memory, 3 is a control signal generator, 4 and 6 are density histogram creating circuits, 5 is a differentiation processing section, 7 is a binarization threshold value generating circuit, 8 is a comparing circuit, A is an input image signal, and B is a Indicates a binarized image signal.

Image signal A obtained via image input means (not shown)
Is first converted into a digital value pixel by pixel by the A / D converter 1 and stored in the image memory 2 in synchronization with the control signal output from the signal generator 3. To be

As shown in FIG. 2, the density histogram creating circuit 4 is specifically composed of a density level number setting section 41 and a density level range setting section 4
2, a counter 43, a density histogram output unit 44, etc. The number of density levels (N) is appropriately initialized in the setting unit 41, and the setting unit 42 sets the density level accordingly,
Since the density level range Li that falls within the lower limit values Lmax and Lmin is determined, the counter 43 counts the number of pixels that fall within each density level range to create a density histogram Hf (i), and outputs the output unit 44. Output through. The density histogram is as FIG. 6 (b), by connexion each concentration block the inclined portion G ₁ ~ ₆ are connected to each other, not gives off into a form suitable for determining the binarizing threshold Is as described above. Therefore, in this embodiment, the differential processing unit 5 is provided to perform the differential processing in order to remove such an inclined portion.

This differential processing operation is shown in FIG. this is, Is obtained as shown in (b) of the figure, and positive and negative thresholds TH _P and TH _N are set for this value to be binarized as shown in (b) of the figure, and the result is the high level "H". If it is, it is a slope, and if it is a low level "L", it is a signal part.

Based on the output from the differentiation processing unit 5, the density histogram creation circuit 6 removes the above-described sloped part from the density histogram Hf (i) shown in, for example, FIG. Create a histogram Hg (i).
With respect to the thus obtained histogram as shown in FIG. 4, density levels α ₁ and α _{2 at} which the frequency becomes P _α for the image K ₄ are obtained by the threshold value generating circuit 7 as shown in FIG. _The binarization threshold level can be determined from the values TH _L and TH _R that allow for a certain margin for ₁ and α ₂ . In addition,
By performing the same processing on the other images K _{1 to} K ₃ and K ₅ , it is possible to obtain a total of 5 sets of binarization threshold values in this example. If the binarization threshold value is thus determined, it is guided to the comparison circuit 8 and compared with the image data read from the image memory 2, whereby the binarized image signal B can be obtained.

〔The invention's effect〕

According to the present invention, the density histogram is differentiated so that the slope portion is removed from the density histogram including the shadow of the illumination system, the blurring, the noise due to the spatial frequency characteristic of the sensor system, and the like to create a true density histogram. I did so, 2
The threshold value can be determined easily and accurately, and as a result, it is possible to perform image processing with high accuracy and stability.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram showing a concrete example of a density histogram creating circuit, and FIG.
FIG. 4 is an explanatory diagram for explaining a differential process, FIG. 4 is an explanatory diagram for explaining a method of determining a binarization threshold value, and FIG. 5 is a conventional example of a binarization method using a density histogram. FIG. 6 is a graph showing an example of the density histogram, and FIG. 7 is an explanatory diagram for explaining the relationship between the image and the binarization threshold value. Explanation of code 1 …… Analog / digital (A / D) converter, 2 …… Image memory, 3 …… Control signal generator, 4,6 …… Density histogram creation circuit, 5 …… Differentiation processing section, 7 …… Threshold generator circuit, 8 ... Comparison circuit, 41 ... Density level number setting unit, 42 ... Density level range setting unit, 43 ... Counter, 44 ... Density histogram output unit, A ... Input image signal , B ... Binary image signal.

Claims (1)

[Claims] 1. An image signal obtained by picking up an image of an object is digitized pixel by pixel according to its density, and a density histogram showing a relationship between each density level range set in advance and the number of pixels belonging to each range is created. A binarization method in which a binarization threshold value level is created based on the density histogram and the image is binarized, wherein the created density histogram has a substantially constant relationship between the density level and the frequency. And a histogram creating means for creating another density histogram containing only the first portion based on the output from the separating means. , A binarization method for an image, wherein a binarization threshold level is determined from the other histogram.