JPH03224074A - Method and device for image processing - Google Patents

Abstract

PURPOSE:To shorten the time required for the binary image processing and to improve the processing accuracy by dividing the image of a model and a digital image to be processed into plural areas and using the feature values included in these areas for comparison carried out between the model and the digital image. CONSTITUTION:A signal converting part 6 converts the analog picture signal given from an image pickup part 1 into a digital picture signal. A feature value extracting part 9 extracts the feature value of the digital image data received from the part 6. Then a digital image is divided into plural small areas for extraction of the feature value of the digital image. A digital image of a model is compared with a subject digital image in the small areas of both digital images. Thus the primary position matching is performed. At the same time, those small areas are integrated into a medium area and the feature values of both digital images are compared with each other in the medium area for execution of the secondary position matching. Thus it is possible to attain the accurate position matching and also to perform the checking/identifying processes of a subject at a high speed.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to an image processing method and apparatus, and is applicable to image alignment, identification, and inspection.

More specifically, it relates to alignment, inspection, or identification processing between an image to be processed having shading and a model image.

(Prior art) In conventional image processing, first, the function of image alignment,
In other words, in many cases, the image processing apparatus itself does not have an alignment function and relies on external material handling technology for the function of aligning a model image and a target image to be compared with the model image. One example is setting up a window for positioning that corresponds to a very characteristic part in the image, searching that window in the target image, finding the characteristic part, and adjusting the position in the image by offset from that window. There is a method of searching, finding a characteristic part, and using an offset from that window to find the inspection target part in the image.

However, this method requires the existence of a characteristic part to be searched, and the alignment accuracy is not high. It is also often used in binary image processing.

Further, although the correlation method and the 5SDA method are well-known as alignment methods, these methods require a large amount of calculation time between pixels, and therefore are not often applied to general-purpose processing devices.

On the other hand, regarding inspection and identification of target images, the most common conventional image processing methods are the in-window pixel counting method, the so-called multi-window method, the SRI algorithm method,
All three template matching methods are binary image processing, so the main premise is that a good binary image can be obtained. It is not suitable for inspection.

In addition, other grayscale image processing devices and color image processing devices handle a large amount of information and are highly functional, but they require specialized professional sensors and hardware to process a large amount of data at high speed, making them extremely expensive. There is.

[Object of the Invention] In view of such conventional problems, it is an object of the present invention to improve the time required and accuracy in binary image processing.

[Summary of the invention]

In order to achieve the object, this invention adopts grayscale image processing in place of conventional binary image processing, digitizes the video signals of the model image and the image to be processed, and converts this digitized image into a predetermined image. The digital image to be processed is divided into two regions, the model and the digital image to be processed are aligned based on the feature values of each region, and the object to be processed is then inspected and identified.This enables high-speed processing. be.

〔Example〕

An embodiment of the present invention will be described below with reference to the drawings.

That is, in FIG. 1, the imaging unit, for example, a camera, images the object 2 to be processed. The processing target 2 is, for example, the printing of a stamp 4 such as the address of a manufacturing factory stamped on the body of a paper container for dairy products, as shown in FIG. The imaging device 1 is then connected to an image processing device 5. In this processing device, the signal conversion unit 6 has the function of converting the analog image signal from the imaging unit 1 into a digital image signal, and also performs preprocessing such as differentiation on the input signal, and converts the digital signal back into an analog signal. It has a function to output to the monitor 7 instead of

Further, the feature amount extracting section 9 has a function of extracting the feature amount of the digital image data from the signal converting section 6.

The frame memory 10 stores the data extracted by the feature extracting section 9. Each of these parts is connected to the calculation control section 13 and controlled by this calculation control section. The input/output section 14 has a function of receiving a trigger signal from the outside and outputting a determination result by the calculation control section 13 to the outside.

Further, a memory 15 necessary for this processing work is connected to the arithmetic control section 13.

Next, a method for processing an image to be processed will be described with reference to FIG.

First, an image of a model, that is, a non-defective sample is captured by the imaging section 1, and this image data, that is, an analog signal, is sent to the signal converting section 6 (step 1).

The image of the model input at this time, that is, the size of the area of the stamping part 4 used as the model, is the pixel size of nXm when the total image size is NXM, as shown in FIG. are N>n and M>m. Then, the image data of the model is converted into a digital signal in the signal conversion section 6. In this way, the image of the model is input.

Next, the digital image of this input model, i.e. n
The image with a size of .

Note that in this invention, the feature amount is the characteristic value of each region obtained by dividing a digital image of a predetermined size into multiple regions with a predetermined size and accumulating the density of multiple pixels within each region. In this case, the cumulative value of the differential concentration within each small region is used. This differentiation uses the operators for each pixel 20 and 20 shown in FIG. 6 and the following equation.

f' (x, y) = If (x, y) - f (x, y - 1
)+l f(x, y)-f(x-Ly) and its cumulative value is calculated by digital differentiation in the differential value calculation circuit shown in FIG. In this figure, digital image data is input to a 1-horizontal line delay circuit 25 and a 1-pixel delay circuit 26, and is also input to difference absolute value circuits 27 and 28, and the output of the 1-horizontal line delay circuit 25 is input to a difference absolute value circuit 27. Furthermore, the outputs of the one-pixel delay circuit 26 are input to the absolute difference circuit 28, respectively. Further, the outputs of the absolute difference circuits 27 and 28 are added by an adder 29 and output as a digital differential image.

Note that a detailed explanation of this differential value calculation circuit will be omitted.

The cumulative value of the digital differential image data is calculated by the feature extracting circuit shown in FIG. This feature quantity extraction circuit has a plurality of line memories Lml to Lm5, which include a write clock generator 31, a write enable control circuit 32, a read clock generator 33, a read enable control circuit 34, and a pixel product-sum circuit. 35 are connected.

The digital differential image data has a screen having a predetermined pixel size, for example, 256×240 pixels. This screen is then divided into small areas of a predetermined number of pixels, for example, 5×5 pixels. The digital image has a size of 240
The scanning line is composed of the scanning lines of a book, and these scanning lines are written into the line memories Lml to Lm5 in the scanning order from the beginning of each line memory. Subsequently, the scanning lines from the 6th grain to the 10th grain are sequentially written into the latter half of each line memory Lml to Lm5. Next, the scanning line from the 11th grain to the 15th grain is again written to the first half of each line memory Lml to Lm5. In this way, writing is performed cyclically in the first half and second half of the five line memories in units of scanning lines (
(Figure 9a).

Next, the successive reading starts five scanning lines later than the writing, and the read clock is set to the line memories Lml, Lm2Lm3 . Lm
4. Lm5. Lml, Lm2. Lm3, . . . , the five line memories are scanned pixel by sine click.

In this way, the pixel data whose order has changed from that at the time of input (Fig. 9) is input to the 25-pixel product-sum circuit 35, and the feature amount for each small area is extracted. Finally, the small area feature amount is calculated in real time with a delay of 5 scanning lines relative to the input pixel data.

Next, as indicated by the X marks in FIG. 10, feature quantities between local cells in the 9th or 25th neighborhood of the digital image of the model are calculated (step 3).

That is, as shown in FIG. 10, a small head area 22 is formed by integrating a plurality of small head areas 21 and 21, that is, A side and @IB pieces. In this figure, there are 3 horizontal and 13 small head areas 21.
．． 21 forms one middle region. Then, the feature amount within this middle region is calculated. Furthermore, at this time, data calculation in the nearby medium area is also performed for second positioning, that is, accurate positioning. The neighboring medium area data is the first
As shown in FIG. 1, this data is based on a small head area 22 formed by integrating new small areas shifted pixel by pixel within the small head area 21 that is the source of the middle area. Note that these calculations are based on the first
This is performed by the calculation control section 13 shown in the figure.

Next, a digital image to be inspected or identified, ie, to be processed, is input. That is, a digital image to be processed is input at the full image size of NXM shown in FIG.

Next, the feature amount within the small head area 21 of the digital image to be processed is extracted. That is, in the same way as the feature amount within the small area in the digital image of the model, the feature amount within the small area in the target digital image is calculated using the same small area size. This calculation is performed by the hardware shown in FIGS. 1, 7, and 8.

Next, a first alignment is performed between the model and the digital image to be processed. That is, this alignment is a rough alignment using small areas of both images (step 6).

In general, in image processing, it is difficult to capture images of moving objects at the same timing and position using a trigger input, so software is used to align the parts of the entire image that are of interest, i.e., where to inspect or identify them. It is often necessary to do so. Therefore, in this embodiment, as shown in FIG.
It is done by law. Since this 5SDA method is normally performed pixel by pixel using the density value of the pixel, it takes an enormous amount of time to calculate. One way to shorten the calculation time is to skip pixels using coarse-fine (coarse-f 1ne).
There are some methods, but they have problems with accuracy. In this invention, the sum of feature amounts, ie, differential values, within the capitula 21 is used.

Next, the microcephalic region 22 of the target digital image is calculated. That is, after the first alignment, calculation of the small head area 22, that is, integration of the small head area 21, is performed using the model and the digital image to be processed with the same medium area size. Note that, unlike in the case of the model, calculation of the neighborhood medium area is not performed here (step 7).

Next, a second alignment is performed using the model and the middle area 22 of the digital image to be processed. That is, more accurate positioning is performed using feature amounts in the 9 or 25 local cells of the previously calculated digital image of the model. This second alignment is performed by comparing the mid-region correspondence between the model and the target using 9 sets of mid-region data for 9 neighborhoods indicated by the x marks in Figure 10, and 25 sets of mid-region data for 25 neighbors, and finds the one that is most similar to the target. The position of the neighboring model is set as the accurate position. Specifically, for example, the ratio of the absolute value of the difference between the data to be processed and the model data in the corresponding middle region to the model value is calculated, and the cumulative calculation is performed for all the middle regions. A method can be considered in which the position with the smallest cumulative value among the 9 pairs or 25 pairs is determined as accurate alignment (step 8).
).

Next, an inspection is performed by comparing the middle region of the model and the target. That is, after the second alignment, feature amounts are compared between the model and the processing target in the corresponding region. Naturally, the model uses the data of the closest neighbors in the second alignment. The determination level can be adjusted by setting upper and lower thresholds for comparison. Further, the determination threshold value can be set to a different value for each medium region (step 9).

Although we have explained the case where the cumulative value of the differential density value is used as the feature quantity in the small head area 21 and the middle region 22, in addition to this, the cumulative value of the density value of the model and the target original image and the cumulative value of the differential value by direction are also used. value, the number of pixels having an original image density or differential density within a certain range, or a feature quantity of an original image density histogram, such as the number of peaks or maximum slope.

(Effects of the Invention) As described above, this invention divides the model into a plurality of regions to compare the digital image to be processed, and uses the feature amounts in both regions, so that it is possible to align the two. is accurate, and the process of inspecting and identifying objects can be performed at high speed.Furthermore, image processing using feature quantities is suitable for surface inspection, etc. where the contrast is not so strong, and in particular, the differential cumulative value is used as the feature quantity. For example, in a stamp inspection, the differential value of crushed stamps and missing parts of stamps is similarly small, making it easy to detect defective elements. It is.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing one embodiment of the image processing device according to the present invention, Fig. 2 is a flowchart of image processing, Fig. 3 is an external perspective view of an object to be inspected or identified, and Fig. 4 is a processing 5 is a front view showing the target stamp area, FIG. 5 is a plan view of the stamp area in FIG. 4 divided into small areas, and FIG. 6 is a 1
A parameter diagram showing the coordinates of pixels 20 and 20 involved in pixel differential processing, Figure 7 is a block diagram of the differential value calculation circuit, Figure 8 is a block diagram of the feature extraction circuit in a small area, and Figure 9 is a line diagram. An explanatory diagram showing the order of memory input and output,
Figure 10 is a relationship diagram showing the positional relationship between an arbitrary pixel of interest and a small area when calculating a 9-neighborhood medium area, Figure 11 is an explanatory diagram showing the relationship between a medium area and a small area, and Figure 12 is a diagram showing the relationship between a small area and a small area. 5S used
It is an explanatory diagram of the DA method. 1... 2... 4... 5... 6... 7... 9... 10... 13... 14... 15... 20... 2 l 25... Imaging section Processing target stamping section Image processing device Signal converting section Monitor feature extraction section Frame memory calculation control section Input/output section Memory pixel small area 1 horizontal line delay circuit 26... 1 pixel delay circuit 27 ...Difference absolute value circuit 2 days...Difference absolute value circuit 29...Adders Lml to Lm...Line memory 31...Clock generator 32, 9. Write enable control circuit 33...Read clock generator 34...Read enable control circuit 35...Pixel product-sum circuit

Claims (18)

[Claims] (1) Input a model image to an imaging unit, convert the output from this imaging unit from analog to digital to obtain a digital image of the model, and integrate this digital image with a predetermined number of pixels. Divide the image into small regions, extract the feature amount within each small region, input the image to be processed into the imaging device, convert the output from analog to digital, and obtain the digital image to be processed. This digital image is divided into a plurality of small regions that are integrated with a predetermined number of pixels, the feature amount within each small region is extracted, and the digital image of the above model and the above target digital image are combined into two digital images. The first alignment is performed by comparing the small areas, and the small areas of the digital image of the model and the target digital image are further integrated to form a middle area, and the An alignment method in image processing characterized by extracting respective feature quantities and comparing the feature quantities in middle regions of both digital images to perform secondary alignment. (2) The alignment method in image processing according to claim 1, wherein the feature amounts in the digital image of the model and the digital image to be processed are based on the density of pixels in each small area. (3) The alignment method in image processing according to claim 1, wherein the feature amounts in the digital image of the model and the digital image to be processed are based on the sum of density of pixels in each small area. (4) An alignment method in image processing according to claim 1, wherein the feature amounts in the digital image of the model and the digital image to be processed are based on the differential value of the density of pixels in each small area. . (5) The image processing method according to claim 1, wherein the feature amounts in the digital image of the model and the digital image to be processed are based on the sum of differential values of the density of pixels in each small area. (6) The feature amount in the digital image of the model and the digital image to be processed is the cumulative value of the differential value of each direction of pixels in each small area. Image processing method. (7) The feature amount in the digital image of the model and the digital image to be processed is based on the cumulative density value of the model and the original image to be processed. Image processing method described. (8) The feature amounts in the digital image of the model and the digital image to be processed are based on the number of pixels having a predetermined density within a predetermined area in the original image of the model and the original image to be processed. An image processing method according to claim 1. (9) The feature amounts in the digital image of the model and the digital image to be processed are based on the number of pixels having a predetermined differential density within a predetermined area in the original image of the model and the original image to be processed. An image processing method according to claim 1. (10) The image according to claim 1, wherein the feature amounts in the digital image of the model and the digital image to be processed are based on density histograms of the model and the original image to be processed. Processing method. (11) A patent claim characterized in that the middle region of the digital image to be processed by the model is defined at a point where the feature amount reaches a predetermined cumulative value in the direction in which the feature amount of the digital image is read. The image processing method according to item 1. (12) The alignment of the digital image of the model and the digital image to be processed is performed by comparing the digital image of the model within the digital image to be processed. Image processing method. (13) The image processing method according to claim 1, wherein the area of the digital image of the model is smaller than the area of the digital image to be processed. (14) The first alignment involves calculating the residual cumulative value between the small area of the digital image of the model and the small area of the digital image to be treated, and determining the point with the smallest value as the optimal position. , and in the second alignment, the residual cumulative value between the middle region of the digital image of the model and the middle region of the digital image to be processed is calculated, and the point with the smallest value is set as the optimal position. An image processing method according to claim 1, characterized in that: (15) Input the model image to the imaging unit, convert the output from the imaging unit from analog to digital to obtain a digital image of the model, and combine the digital images with a predetermined number of pixels. Divide the image into small regions, extract the feature amount within each small region, input the image to be processed into the imaging device, convert the output from analog to digital, and obtain the digital image to be processed. This digital image is divided into small regions of the same size and number as the digital image of the above model, and the feature amount within each small region is extracted, and the digital image of the above model and the above target digital image are combined. The first alignment is performed by comparing the small regions of both digital images, and the small regions of the digital image of the model and the target digital image are further integrated to form a middle region, and this Extracting the feature amounts in the middle region, performing secondary alignment by comparing the feature amounts in the middle region of both digital images, and further comparing the digital image to be processed with the digital image of the model. An image processing method characterized by inspecting or identifying the image to be processed. (16) An imaging unit that captures a model image or an image to be processed, a signal conversion unit that converts an analog image signal from the imaging unit into a digital image signal, and the digital image of the model or processing target, a feature amount extraction unit that extracts feature amounts within a region divided into a plurality of regions so as to integrate this image with a predetermined number of pixels; An image processing device that includes an arithmetic control unit that aligns both digital images by comparing both areas. (17) An imaging unit that captures a model image or an image to be processed, a signal conversion unit that converts an analog image signal from the imaging unit into a digital image signal, and the model or digital image to be processed, a feature amount extraction unit that extracts feature amounts within a region divided into a plurality of regions so as to integrate this image with a predetermined number of pixels; An image processing device comprising an arithmetic control unit that aligns both digital images by comparing both areas, and further inspects or identifies the processing target. (18) The arithmetic control unit divides the model and the digital image to be processed into a plurality of small regions that are integrated with predetermined pixels, and compares the feature amounts of the small regions of both digital images to obtain the first image. Align the
Further, the small areas of both digital images are further integrated to form a middle area, and the feature amounts in the middle areas of both digital images are compared to perform secondary alignment. The image processing device according to item 16 or 17.