JPH09284541A - Picture composing device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a picture composing device reduced in the error of the correction quantity of a luminance difference on picture to be adhered, which occurs owing to the shift of coordinates themselves between the divided pictures. SOLUTION: The picture composing device has memories 201 and 202 for storing picture data of two division areas, cumulative addition parts 203 and 204 for reading and cumulatively adding picture data from a square-formed extracted areas in the overlap part of division picture data, a correction quantity calculation part 205 for dividing data from the cumulative addition part 204 by data from the cumulative addition part 203 and calculating correction quantity of the luminance difference, a multiplication part 207 for multiplying picture data from the memory 201 by correction quantity and correcting the luminance difference and an output selection part 208 for adhering and composing corrected picture data with picture data from the memory 202 and outputting it.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image synthesizing apparatus for synthesizing one image by pasting a plurality of image data.

[0002]

2. Description of the Related Art Conventionally, a photographed image is divided into a plurality of images, and an image of each divided region is read by using one or a plurality of image sensors and combined into one image to obtain a desired image. 2. Description of the Related Art An image synthesizing device is known which is adapted to obtain image reading image information. In such an image synthesizing apparatus, Japanese Patent Application Laid-Open No. 63-299457 discloses a line for an overlapping portion in which images in respective divided regions overlap with each other in order to correct a deviation in brightness of read images in respective divided regions. It is disclosed that each pixel data is extracted, the correction amount of the deviation of the brightness is obtained for each line, and the brightness of the image of each divided area is corrected.

[0003]

However, in the brightness correction means disclosed in the above publication, if a coordinate shift occurs when an image of each region divided by the image sensor is taken, the coordinate between the divided images is increased. A shift occurs, and the luminance difference caused by the shift of the extraction area is also targeted for correction, so that proper luminance correction cannot be performed. The error in the correction amount of the brightness difference caused by the deviation of the coordinates themselves between the divided images is not taken into consideration in the brightness correction means disclosed in the above publication, and therefore the brightness difference remains even after the correction of the brightness difference. was there.

The present invention has been made to solve the above problems in the conventional image synthesizing apparatus, and reduces the error in the correction amount of the brightness difference of the combined image caused by the deviation of the coordinates themselves between the divided images. It is an object of the present invention to provide an image synthesizing device that can be used.

[0005]

In order to solve the above problems, the invention according to claim 1 uses a plurality of divided image data obtained by one or a plurality of image sensors and partially overlapping in an image pickup region. In an image synthesizing device for synthesizing and combining one image, each image data of the overlapping part is partially extracted in the overlapping part of the plurality of divided image data from the image sensor, and it corresponds to the original image. Data extraction means for setting an area in which the dimension ratio in the vertical direction and the horizontal direction occupying is approximately 1: 1 and extracting each image data in the area;
A brightness difference correction means for correcting the brightness of the plurality of divided image data according to the brightness difference of each area image data extracted by the data extraction means is provided.

In the image synthesizing apparatus configured as described above, the areas for extracting the image data for obtaining the correction amount of the brightness difference of the divided images from the overlapping portion are aligned in the vertical direction and the horizontal direction in the case of being associated with the original image. Since the shape is set so that the dimensional ratio in the direction becomes approximately 1: 1, it is possible to reduce the error in the correction amount of the brightness difference due to the deviation of the coordinates of the divided images.

Next, the principle that the error of the correction amount is reduced when the area is set in such a shape will be described. FIG. 1 shows a rectangle 1 having a vertical h and a horizontal i on the Cartesian coordinates.
And a rectangle 2 in which the rectangle 1 is shifted by a shift amount α in the θ direction. Here, the rectangle 1 is an area occupied by the original image of the extracted pixel data, and the rectangle 2 is an area occupied by a coordinate difference between the read images. At this time, in the area where the rectangle 1 and the rectangle 2 do not overlap, the smaller the portion ΔS included in the rectangle 2, the more the error in the correction amount can be reduced.

Here, assuming that a circle having the same area as the rectangle 1 and a radius r is the extraction area, the deviation area ΔS caused by the expected deviation amount α and the extraction area area 2πr ² =
The smaller the ratio with h · i, the more the error can be reduced,
1/10, preferably less than
It is desirable for accurate brightness correction. Therefore, the area of the extraction area is set according to the expected shift amount α.

In FIG. 1, the shift area ΔS is expressed by the equation 1.

[0010]

[Equation 1]

Here, since the value of the deviation area ΔS changes depending on the angle θ, it is necessary to integrate with respect to θ. ΔS
Is integrated with respect to the angle θ from 0 to π / 2,
It is expressed as Equation 2.

[0012]

[Equation 2]

This value is obtained when the angle θ is from θ = 0 to θ = π / 2.
It is the integral value of ΔS when it changes to. Further, when this integrated value is divided by π / 2, the average value ΔT represented by Expression 3 is obtained.

[0014]

(Equation 3)

That is, ΔT is the average displacement area when the rectangle 1 is displaced by the displacement amount α. Therefore, the error of the correction amount can be compared by comparing ΔT. The smaller the ΔT, the smaller the error in the correction amount.

FIG. 2 shows the aspect ratio h / i with the area being constant.
It is a figure which shows the relationship between (DELTA) T and (alpha) when it is set to 1,2,3,4. As can be seen from FIG. 2, when h / i = 1, the value of ΔT with respect to α is the smallest, and h / i is 2.
The value of ΔT with respect to α increases as it increases to 3 and 4. Therefore, when h / i = 1, that is, when the ratio of height to width is 1: 1, the error in the correction amount is the smallest.

FIG. 3 shows a circle 11 having a radius r on the Cartesian coordinates and a circle 12 which is the circle 11 displaced by a shift amount α in the θ direction. Of the non-overlapping areas of Yen 11 and Yen 12, Yen 12
The portion (displacement area) ΔS included in is expressed by Equation 4.

[0018]

(Equation 4)

When ΔS is integrated with respect to θ in the range of 0 to π / 2, it is expressed by the following equation (5).

[0020]

(Equation 5)

By dividing the integral value represented by the equation 5 by π / 2, the average value ΔT represented by the equation 6 is obtained.

[0022]

(Equation 6)

FIG. 4 shows the relationship between the average value ΔT and α added to the graph shown in FIG. As can be seen from FIG. 4, the value of the average displacement area ΔT with respect to the displacement amount α in the circle is smaller than that in the case of h / i = 1, which is the smallest in the rectangle.

Next, h / i = 1, 2,
FIG. 5 shows the relationship between the shift area ΔS according to the equation 1 and the shift area ΔS according to the equation 4 and the shift angle θ in the case of 3, 4
Shown in However, the shift amount α is α = (1/4) (h · i /
π) ^1/2 = r / 4, which is constant. As can be seen from FIG. 5, in the case of a circle and a rectangle (square) with h / i = 1, ΔS is uniform with respect to θ, but h / i = 2, 3,
There is variation in the case of a rectangular shape with 4 in particular, θ = π
ΔS of the rectangle (h / i = 2, 3, 4) near / 2 is
It is 30% to 80% larger than the ΔS of the circle. Such variations greatly appear in the brightness when an image is displayed, and are not appropriate in obtaining the correction amount of the brightness difference. in this way,
Also in the variation of ΔS with respect to θ, a circle and h / i =
The case of a rectangle (square) of 1 is advantageous.

Therefore, the shape of the area occupied by the original image of the image data to be extracted in order to correct the brightness difference is set to be a circle, a square, a regular hexagon, a regular octagon, or even a shape of these shapes. A shape with rounded corners or a similar shape, that is, the dimension ratio in the vertical and horizontal directions is approximately 1: 1.
With such a shape, it is possible to reduce the error in the correction amount.

According to a second aspect of the invention, in the image synthesizing apparatus according to the first aspect, the shape of the area set in the overlapping portion by the data extracting means is circular. Thus, when the shape of the area is circular, as shown in FIG. 4, the average value ΔT of the displacement area is the smallest as compared with the other shapes, and therefore the error in the correction amount can be minimized. Further, as shown in FIG. 5, the deviation area ΔS is uniform with respect to the deviation of the deviation angle θ, and the error of the correction amount can be suppressed to the minimum without variations, and the optimum brightness correction can be performed. Becomes

According to a third aspect of the present invention, in the image synthesizing apparatus according to the first aspect, the shape of the area set in the overlapping portion by the data extracting means is a square. In this way, by making the shape of the area square, as shown in FIG. 4, the average value ΔT of the displaced area is the second smallest after the circle as compared with other shapes such as rectangles, and as shown in FIG. The change of the deviation area ΔS with respect to the deviation of the deviation angle θ is also uniform next to the circle, and therefore the error of the correction amount can be suppressed to be small and small.

Further, for the correction processing of the brightness difference in the brightness difference correction means, the data extraction means associates the data for calculating the correction amount from the image data read in the bit map memory with the original image. When extracting from an area whose shape occupies a square, the bit map memory address control means can be realized by a small-scale circuit, whereas when extracting data from a circular area Must have a look-up table for address control, and the control means is a large-scale circuit. Therefore, by making the shape of the area of the extracted image data occupying when corresponding to the original image into a square, the error of the correction amount can be small and small with a small circuit.

[0029]

Next, an embodiment will be described. FIG. 6 is a block diagram showing the first embodiment of the image synthesizing apparatus according to the present invention. In FIG. 6, 101
Reference numeral 102 is an image sensor, 102 is a data processing device as an image synthesizing processing means, 103 is an output device such as a monitor, a printer, etc. apparatus
The composite processing is performed in 102, and the output device 10
By 3, the image after combining is output.

Next, the operation of the embodiment shown in FIG. 6 will be described. In FIG. 7, the captured image 121 is captured by the image sensor 101.
It is the figure which showed the area divided into two and read by horizontal direction by. An area M ₁ surrounded by a solid line is an area read by the image sensor 101 for the first time, an area M ₂ surrounded by a broken line is an area read by the image sensor 101 for the second time, and M ₃ Indicates an area read in duplicate. Area M ₁ and Area M
_{In 2} , the luminance of the image data is different due to the optical system and the like, and the luminance of the area M ₃ that is read redundantly is also different between the first reading and the second reading. In area M ₁ and area M ₂ , the image sensor
There is a coordinate shift due to the mounting accuracy of 101.

The data processing device 102 corrects the brightness difference between the image data in the areas M ₁ and M ₂ and performs processing for combining the data, and a specific configuration example thereof is shown in FIG. In FIG. 8, 201 is a bitmap memory for storing the first read image data, 202 is a bitmap memory for storing the second read image data, and 203 is a memory 20.
1 Overlapping part Cumulative addition part of read data, 204 is memory
202 is a cumulative addition unit of read data of the overlapped portion, 205 is a luminance difference correction amount calculation unit, 206 is a latch for holding the correction amount, 20
7 is a multiplication unit for multiplying and correcting the first read image data from the memory 201 by a correction amount, 208 is an output selection unit at the time of image combination, 209 is a control circuit, and 301 is an image sensor 101.
Input terminal Din, 302 of the image data output by the output device
The output terminal Dout of the combined image data to 103.

The image data of the area M ₁ read by the image sensor 101 is written in the memory 201,
_{The second} image data is written in the memory 202. Memory 20
The image data extracted from 1 and 202 is the data of the extraction area M ₄ in the area M _{3 shown} in FIG. In this configuration example, the extraction area M
The data of ₄ is 30 pixels in the vertical direction on the memories 201 and 202,
This is data of a square area having 30 pixels in the horizontal direction. Area M _{3 of the} image data written in the memory 201
The data in the extraction area M ₄ therein is read out by sequentially designating the row and column addresses by the control circuit 209 and output to the cumulative addition unit 203. Similarly, out of the image data written in the memory 202, the data of the extraction area M ₄ in the area M ₃ is extracted and read, and is output to the cumulative addition unit 204.

The extraction area M is calculated by the cumulative addition units 203 and 204.
The data of ₄ is cumulatively added, and the correction amount calculation unit 205 divides the data from the cumulative addition unit 204 by the data from the cumulative addition unit 203 to calculate the correction amount of the brightness difference. Then, the latch 206 holds the calculated correction amount. Next, the image data of the area M ₁ written in the memory 201 is read, the brightness difference is corrected by the multiplication unit 207, and the result is output to the output terminal 302 through the output selection unit 208. Next, the area M written in the memory 202
_{The second} image data is read out and output to the output terminal 302 through the output selection unit 208. As described above, images are combined with each other in which the brightness difference between the combined images is reduced.

The square shape of the extraction area M ₄ is shown in FIG.
As can be seen from the characteristic diagram shown in FIG. 1, it is not necessary to set the number of vertical and horizontal pixels to be exactly 1: 1 and it is sufficient if it is approximately 1: 1. Further, in the present embodiment, a single image sensor captures two divided images, but a plurality of image sensors may capture individual divided images. This also applies to the second embodiment and the like described next.

Next, a second embodiment of the present invention will be described. This embodiment is different from the first embodiment in that in the data processing device shown in FIG.
Image data extraction area M ₄ extracted to calculate the brightness correction amount from the image data read in 1 202
That is, the shape occupied by the original image is circular instead of square. Specifically, the control circuit 209 obtains an address included in a circle on the memory 201, for example, with a radius of 17 pixels, using a lookup table, specifies the address in the memory 201, and reads it from the memory 201. The output image data is output to the cumulative addition unit 203. Similarly, the address of the circle on the memory 202 is designated in the memory 202, and the read image data is output to the cumulative addition unit 204. Subsequent processing is performed in the same manner as in the first embodiment, and images with reduced brightness difference of the combined image are combined.

[0036]

As described above based on the embodiments, according to the invention described in claim 1, since the correction amount of the brightness difference of each divided image is obtained, the pixel extracted from the overlapping portion of each divided image is obtained. Since the dimensional ratio of the shape occupied by the data in the case of being associated with the original image is approximately 1: 1 in the vertical direction and the horizontal direction,
It is possible to reduce the error in the correction amount due to the coordinate shift between the divided images. According to the second aspect of the invention, since the shape occupied by the extracted image data is circular, the error of the correction amount can be suppressed to the minimum without variation.
According to the third aspect of the present invention, since the shape occupied by the extracted image data is a square, it is possible to reduce the variation in the correction amount error with a small-scale circuit and to keep it small.

[Brief description of drawings]

FIG. 1 is an explanatory diagram for explaining a principle of error reduction of a brightness correction amount in an image synthesizing apparatus according to the present invention.

FIG. 2 is a diagram showing a relationship between an average shift area and a shift amount when an area occupied by extracted image data is rectangular.

FIG. 3 is another explanatory diagram for explaining the principle of error reduction of the brightness correction amount in the image synthesizing apparatus according to the present invention.

FIG. 4 is a diagram showing a relationship between an average shift area and a shift amount when an area occupied by extracted image data is a rectangle or a circle.

FIG. 5 is a diagram showing a relationship between a shift area and a shift angle when an area occupied by extracted image data is a rectangle or a circle.

FIG. 6 is a block configuration diagram showing a first embodiment of an image synthesizing apparatus according to the present invention.

FIG. 7 is a diagram showing divided areas of a captured image read by the image sensor of the embodiment shown in FIG.

FIG. 8 is a block configuration diagram showing a configuration example of a data processing device in the embodiment shown in FIG.

FIG. 9 is a diagram showing an extraction area in an area that is read redundantly.

[Explanation of symbols]

 1 rectangular area 2 offset rectangular area 11 circular area 12 offset circular area 101 image sensor 102 data processing device 103 output device 121 captured image 201,202 bitmap memory 203,204 cumulative addition unit 205 correction amount calculation unit 206 latch 207 multiplication unit 208 output selection 209 Control circuit 301 Input terminal 302 Output terminal

Claims (3)

[Claims] 1. An image synthesizing device for synthesizing one image by pasting together a plurality of divided image data obtained by one or a plurality of image sensors and having overlapping imaging regions, In the overlapping part of the divided image data, an area is set in which each image data of the overlapping part is partially extracted, and the size ratio in the vertical direction and the horizontal direction when occupying the original image is approximately 1: 1. Data extraction means for extracting each image data in the area, and brightness difference correction means for correcting the brightness of the plurality of divided image data according to the brightness difference of each area image data extracted by the data extraction means. An image synthesizing device characterized by being provided. 2. The image synthesizing apparatus according to claim 1, wherein the shape of the area set in the overlapping portion in the data extracting means is circular. 3. The image synthesizing apparatus according to claim 1, wherein the shape of the area set in the overlapping portion in the data extracting means is a square.