JPS61134886A - Detector of rotary angle difference between pictures - Google Patents

Abstract

PURPOSE:To attain verification in a short time by verifying a rotary angle difference through the collation of a shape histogram generated from two pictures. CONSTITUTION:A picture as shown in figure (a) is stored in the 1st picture memory 1 as a reference picture and a picture being conguruent thereto and having a rotary angle difference of alpha degrees is stored in the 2nd picture memory 2. Such a case is explained as follows; First an angle difference detector 5 detects four kinds of rotary angle differences of alpha deg., (alpha-90) deg., (alpha+180) deg., and (alpha+90) deg. and a picture corresponding to the picture rotary section 6 is obtained respectively. The 1st shape histogram generating section 13 generates a shape histogram as shown in figure (e) and the 2nd shape histogram generating section 14 generates a shape histogram corresponding to the picture obtained by the said rotary section 6.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to an image processing apparatus, and relates to an image-to-image rotation angle difference detection apparatus for determining a rotation angle difference between two congruent or similar images.

For example, in a seal matching device, which is an applied field of image processing technology, a matching seal impression, that is, a matching image, and a registered seal impression, that is, a reference image, are used as a reference for matching, using techniques such as pattern matching. A process of matching and checking the degree of similarity between the two is often used.

However, if the angle or orientation of the image to be matched is different from that of the reference image, matching is difficult as is.

Therefore, an inter-image rotation angle difference detection device is used to detect the rotation angle difference between two images and remove the rotation angle difference between the two images.

At this time, it is desirable that the rotation angle difference between the two images be detected accurately and in a short time.

[Conventional technology]

, FIG. 4 is a block diagram showing the configuration of a conventional example of an image rotation angle detection device. 1 and 2 are block diagrams showing the configuration of a conventional example of an image rotation angle detection device, and 1 and 2 indicate image data of a reference image and a matched image, respectively, in the X-axis direction and the y-axis on orthogonal xy coordinates. They are a first image memory and a second image memory that store the density of each pixel arranged at equal pitches in the direction.

3a and 3h and 4a and 4b are a first angular histogram generating unit 3 and a second angular histogram generating unit 3 that generate angular histograms in the main change direction of density for each pixel in the image data of the reference image and the matched image, respectively. 3a and 4a are components of the generation unit 4, and 3a and 4a scan the image data stored in the first image memory 1 and the second image memory 2, respectively, using, for example, a 3×3 differential operation window, and generate data for each pixel. Find the difference dx in the X direction and the difference dy in the Y direction of density.

The first angle calculation unit and the second angle calculation unit, 3b and 4b, each calculate the angle θ of the main change direction of the density for each pixel, for example, in units of 1 degree, as θ=arctan(dy/dx). The number of pixels for each angle θ in the main change direction of density obtained by the first angle calculation unit 3a and the second angle calculation unit 4a is counted and 0
They are a first histogram memory and a second histogram memory that store angles from degrees to 359 degrees as addresses.

Reference numeral 5 denotes an angular difference detection unit that detects a rotation angle difference between two images based on the angular histograms stored in the first histogram memory 3h and the second histogram memory 4b, respectively.

6 is the image to be matched stored in the second image memory 2,
Based on the rotation angle difference θ detected by the angle difference detection unit 5, for example, using a table, X'=XCO3θ+ys
It is an image rotation unit that rotates by performing the conversion of inθ y”=−ysin θ + y cos θ, and 7 is a third image memory that stores the image rotated by the image rotation unit 6.

8 and 9 are the first image memory 1 and 9, respectively.
a first projection unit and a second projection unit that create a projection onto the
Reference numeral 1 denotes a first projection memory and a second projection memory that store an X-axis projection and a y-axis projection obtained by a first projection section 8 and a second projection section 9, respectively.

Further, 12 is an angle for verifying the rotation angle difference detected by the angle difference detection unit 5 based on the distance between the projection stored in the first projection memory 10 and the projection stored in the second projection memo January 1. This is the difference verification section.

Next, as a reference image and an image to be matched, FIG. 5(a)
To explain the case where congruent images with a rotation angle difference of 45 degrees are stored in the first image memory 1 and the second image memory 2, respectively, as illustrated in FIGS. The generation unit 3 and the second angular histogram generation unit 4 create the image shown in FIG.
Angular histograms like 1) and (d) are obtained.

The angular difference detection unit 5 compares the angular histograms of C1 and (dl in the angular axis direction (horizontal direction in the figure)) to determine the difference between the images in FIG. 5(a) and -). The rotation angle difference between them (with respect to the reference image, counterclockwise rotation is positive and clockwise rotation is negative) is detected as 45 degrees or 1135 degrees.

Figures 6 (al and b) are the images in Figure 5 (bl) rotated by -45 degrees and 135 degrees, respectively, and their angular histograms are shown in Figure 5 (e).
) is exactly the same as that of

That is, the rotation angle difference between two images cannot be determined using only the angle histogram.

Therefore, the projection determined by the second projection section 9 from the image to be matched by 145 degrees or 135 degrees rotated by the one-image rotation section 6, and the projection determined from the reference image by the first projection section 8 are calculated by changing the angle. This is the station to be verified by the difference verification unit 12.

A projection as shown in Fig. 7 (Ml) can be obtained from both the image rotated by 145 degrees and the reference image, and the projection shown in Fig. 7 (b) when the image to be compared is rotated by 135 degrees is obtained. The projection is changed as shown, and as a result, it is verified that the rotation angle difference between the reference image and the matched image is 45 degrees.

Note that in the case of the images illustrated in FIG.
Two rotation angle differences of degrees and -135 degrees were obtained,
In particular, when dealing with kanji, including seal verification, 9
In many cases, four types of rotation angle differences having a difference of 0 degrees are detected.

[Problem that the invention seeks to solve]

In the image-to-image rotation angle difference detection device having the above configuration, for example, four types of rotation angle differences may be detected as described above, and in such a case, it is necessary to perform verification four times, which takes a long time. There was a problem that it took time.

[Means for solving problems]

The inter-image rotation angle difference detection device according to the present invention includes an angular histogram generation unit that generates an angular histogram by counting the main change direction of the density at each pixel in image data expressed as the density of each pixel. , an angular difference detection unit that detects a rotation angle difference between two images based on the generated angle histogram, and rotates at least one of the two image data so as to remove the detected rotation angle difference. an image rotation unit; a shape histogram generation unit that generates a shape histogram by counting the number of occurrences of each of a plurality of predetermined shape patterns in each partial region in the image data for each pattern; and the generated shape. The above-mentioned problem is solved by including an angle difference verification section that verifies the detected rotation angle difference using a histogram.

[Effect]

That is, in the present invention, the rotation angle difference detected by the angle difference detection section is verified by comparing the shape histograms generated from each of the two images from which the rotation angle difference has been removed. Even if four types of rotation angle differences having a difference of 90 degrees are detected, the correct rotation angle difference can be determined only by checking one of these types.
1 [Example] The gist of the present invention will be specifically explained below with reference to Examples.

FIG. 1 is a block diagram showing the configuration of an embodiment of an image-to-image rotation angle difference detecting device according to the present invention, and the same reference numerals as those in the conventional example in FIG. 4 refer to the same objects.

In addition, 13a and 13b and 14a and 14b indicate 1 in each partial area of 03×3 pixels in the image data stored in the first image memory 1 and the third image memory 7, respectively.
For example, a first shape histogram generation section 13 and a second shape histogram generation section 14 count the number of appearances of each of 12 types of shape patterns as shown in FIG. 2 and generate a shape histogram. 13a and 14a scan the image data stored in the first image memory 1 and the third image memory 7, respectively, using a 3×3 pattern detection window, as shown in FIG. These are a first pattern extraction section and a second pattern extraction section that extract patterns of 12 types of shapes from ■ to O for example.

13b and 14b count the number of each of the 12 types of shape patterns extracted by the first pattern extraction unit 13a and the second pattern extraction unit 14a, respectively, and store them for each address from 1 to 12. They are a third histogram memory and a fourth histogram memory.

15 is by comparing the shape histogram stored in the third projection memo 3b with the shape histogram stored in the fourth histogram memory 14b.

This is an angular difference verification section that verifies the rotation angle difference detected by the angular difference detection section 5.

The 12 types of patterns illustrated in Figure 2 are basically ■
・There are three types, ■ and ■, and the ones that are rotated by 190 degrees are ■, ■ and [F], +
The ones given a 180 degree rotation are ■・■ and ■, and +
Those given a 90 degree rotation are [phase], ◎ and @.

Hereinafter, the first image memory 1 is stored as a reference image as shown in FIG.
A case will be explained in which an image such as a) is stored, and an image congruent with this and having a rotation angle difference of α degrees is stored in the second image memory 2.

First, the angle difference detection unit 5 detects four types of rotation angle differences: α degree, (α-90) degree, (α+180) degree, or (α+90) degree, and the image rotation unit 6 detects each rotation angle difference as shown in FIG. images such as al・mountain)・(C1 and (dl) are obtained.

In the first shape histogram generation unit 13, as shown in FIG.
In the second shape histogram generation unit 14, the shape histograms as shown in FIG. 3 (Ql, ffl,
A shape histogram such as (along or (hl) is generated.

For example, if the rotation angle difference detected by the angle difference detection unit 5 is α degree, the first shape histogram generation unit 13 obtains a shape histogram as shown in FIG. The histogram generation unit 14 also obtains a shape histogram of the same shape.

The angle difference verification unit 15 compares these and determines the rotation angle difference α detected by the angle difference detection unit 5 as a correct conflict.

Further, suppose that the rotation angle difference detected by the angle difference detection unit 5 is 1, for example (α-90) degrees.

In the angle difference verification unit 15, the image shown in FIG. 3 obtained from the reference image (
By comparing the shape histogram shown in FIG. 81 with the shape histogram shown in FIG. It is verified that there is an error of 190 degrees, so the correct rotation angle difference is determined as α+90 degrees.

In the above embodiment, the shape histogram is obtained by counting the frequency of appearance of patterns of 3 x 4 types for each partial area of 3 x 3 pixels, but the size of one partial area and the pattern The numbers are not limited to those listed above.

〔Effect of the invention〕

As explained above, according to the present invention, it is not necessary to verify all four types of rotational angle differences obtained by the angle difference detection unit, each having a difference of 90 degrees.
Verification can be done in a short time.

4. Figure ka. Easy, bright
1 FIG. 1 is a block diagram of one embodiment of the present invention.

Figures 2 - @ and Figure 3 (al - (h) are explanatory diagrams of the same embodiment.

FIG. 4 is a block diagram of a conventional example.

FIG. 5+111 to (dl) FIG. 6 (al (bl) and FIG. 7(a) (bl is an explanatory diagram of a conventional example.

In the figure.

3a is the first angle calculation section + 4a is the second angle calculation section.

3b is the first histogram memory.

4b is a second histogram memory.

5 is an angle difference detection unit, and 6 is an image rotation unit.

13a is a first pattern extraction section.

14a is a second pattern extraction section.

13a is a first histogram memory.

14b is a second histogram memory.

15 is an angle difference verification section.

Evil Nko to the snout × mouth to pine death) Rotation

Claims (1)

[Claims] an angular histogram generation unit that generates an angular histogram by counting the main direction of change in density at each pixel in image data expressed as the density of each pixel for each angle; an angular difference detection unit that detects a rotation angle difference between the images, an image rotation unit that rotates at least one of the two image data so as to remove the detected rotation angle difference, and all parts of the image data. a shape histogram generation unit that generates a shape histogram by counting the number of patterns each of a plurality of predetermined shapes appear in a region for each pattern, and verifying the detected rotation angle difference using the generated shape histogram. An inter-image rotation angle difference detection device, comprising: an angle difference verification section.