JPH05122606A - Method and device for synthesizing image - Google Patents

Abstract

PURPOSE:To easily and accurately synthesize plural images. CONSTITUTION:By imaging plural images 1 and 2 to be joined so that they can be partially overlapped, obtaining a luminance value or the color difference between the end part of the joined images and an area 4 and joining the images so that an area 5 whose luminance value is 0 or whose color difference takes a minimum value can be overlapped, the images are synthesized. Since it is not required to accurately move and position an imaging device and an object of imaging in order to join plural images, the images can easily be synthesized, and furthermore, since a joint point is obtained by using the luminance value or color difference between the images, the images can accurately by synthesized without losing concanate property.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image synthesizing processing method and an image synthesizing processing device for joining a plurality of images picked up by an image input device and recording them in an image recording device.

[0002]

2. Description of the Related Art In recent years, not only in the engineering field but also in the medical field,
In the field of clothing and living environment, the measurement and evaluation system by image processing has been constructed, which is an important technical field. In this image processing, the number of pixels of the image, that is, the number of divisions of the image is one of the important factors that determine the resolution and accuracy of the image processing, and as the image processing becomes more sophisticated, an image with more pixels is required. There is. For example, in high-density pattern detection such as a lattice method, a moire method, or a resist pattern that analyzes a periodic pattern image, the number of periodic patterns determines the measurement resolution, so an image having a larger number of pixels Analysis is needed. However, it is impossible to obtain a resolution higher than that of the input device, and in order to obtain more resolution, a method of dividing the same region into a plurality of images and inputting them to improve the resolution is used. ing.

Further, in the case of an object which cannot be imaged at one time, for example, an image of an object which is very long in the longitudinal direction such as an audio tape or a VTR tape, or a stereo camera which images a 360-degree view, Since it is impossible to capture the necessary images at once, it is necessary to input and combine a plurality of images in this case as well.

In order to synthesize a plurality of images, a method of detecting and synthesizing feature points of images obtained by using a plurality of image input devices (Japanese Patent Laid-Open No. 2-130404) and a single image input A method has been used in which an observation sample is moved by using an apparatus, or a plurality of images obtained by moving an image input apparatus are simply combined and combined.

[0005]

However, the above-described method and configuration have the following problems.

(1) The images to be combined are imaged so as to partially overlap each other, and the feature points included in the overlapping parts are detected,
Regarding the method of joining images from the positional relationship, if there is no detectable feature point in the image originally, the feature point must be intentionally created in the image, and at that feature point, the original image Information is damaged. Further, even if the characteristic point originally exists, the analyst must specify the position thereof, which hinders automation and causes measurement errors due to different analysts.

(2) In a method in which a measurement object is divided and imaged by a plurality of image input devices, and the images are simply joined and recorded, the plurality of images do not overlap at their boundaries, and the boundaries are The image input devices must be arranged so that they are in contact with each other, and it is very difficult to perform this positioning for each pixel of the image.

(3) In a method of simply joining a plurality of images taken by moving an image pickup object or an image pickup device by using a single image input device as an observation target, the range of images that can be picked up at one time is accurate. Since the imaging object or the imaging device must be moved to, the moving device is required to have sufficient positioning accuracy and repeatability accuracy, and when the image input device has a zoom function, the area that can be imaged differs depending on the zoom magnification. The amount of movement must be accurately calculated each time the zoom magnification is changed.

In view of the above problems, it is an object of the present invention to provide an image synthesizing processing method and an image synthesizing processing device for simply and accurately joining a plurality of images.

[0010]

In order to solve the above-mentioned problems, an image composition processing method and an image composition processing apparatus of the present invention capture a plurality of images for composition so as to partially overlap each other, and The luminance difference or the chromaticity difference between the image at the joining end of the image and the image to be joined is detected, and the place where the difference is minimized is combined as the joining place.

[0011]

According to the present invention, a plurality of images can be combined easily and accurately with the above configuration.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An image synthesizing processing method according to a first embodiment of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing an image composition processing method of the present invention. In FIGS. 1 (a), (b), and (c), 1 represents one image to be joined, 2 represents the other image to be joined, and 3 was obtained by synthesizing 1 and 2. The image is shown. Reference numeral 4 is a part of 2, an area for searching a position to be combined, and 5 is an area showing a position that coincides with the area of 4 whose luminance value or chromaticity value is 1.

The image composition processing method of this embodiment will be described below with reference to FIG. Images 1 and 2 are images input by a CCD camera or the like so that the imaging targets are partially overlapped, and are divided into pixels, and each pixel is provided with a brightness level in the case of a monochrome image and color information in the case of a color image. have. Here, the total difference between the images of the region 4 at the end of the image 2 on the joining side and the region of the same size cut out from the right end of the other image 1 is obtained. The difference between the images uses the brightness value of each pixel when the image is monochrome, and the RG that each pixel has when the image is color.
The value of each of the three B colors is used.

While moving the region cut out from the image 1 pixel by pixel to the left, the region is similarly compared with the region 4 to search for a region where the sum of the luminance value or the chromaticity difference is 0 or the minimum value. In the case of the obtained region 5, the combined image 3 can be obtained by connecting the images 1 and 2 so that the regions 4 and 5 overlap.

In the first embodiment, the joining direction is the horizontal direction of the images, but it is needless to say that the same effect can be obtained with a plurality of images taken in the vertical overlapping direction.

An image synthesis processing apparatus according to a second embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 shows an image synthesis processing apparatus according to the second embodiment of the present invention. In FIG. 2, 6a, 6b, and 6c are television cameras for inputting images to be combined, which are set in parallel so that they can be imaged in the image combining direction in an overlapping manner. 7 is a TV camera 6
An image processing device for synthesizing the three images input from a, 6b, and 6c, and a display device 8 for displaying the synthesized images. Reference numeral 9 is an imaging target.

The image composition processing apparatus of this embodiment will be described below with reference to FIG. The imaging target 9 is imaged by the three television cameras 6a, 6b, 6c. The television cameras 6a, 6b, and 6c are set so that the areas to be imaged overlap by a predetermined amount. TV cameras 6a, 6
The images captured by b and 6c are stored in the image processing device 7 and are combined by using the image combining method shown in the first embodiment. The combined image is output to the display device 8. Further, in the case where the television camera has a zoom function, even if the zoom magnification changes, the television camera 6 is set in a state in which the regions to be imaged can be input in an overlapping manner.
If a, 6b and 6c are set, it is not necessary to move the television cameras 6a, 6b and 6c each time the zoom changes.

Although three television cameras are arranged in series in (FIG. 2), several television cameras may be set in parallel or over a plurality of rows at substantially equal intervals.

An image synthesis processing apparatus according to the third embodiment of the present invention will be described below with reference to the drawings.

FIG. 3 shows an image composition processing apparatus according to the third embodiment of the present invention. In FIG. 3, reference numeral 10 is a television camera for inputting a plurality of images. Reference numeral 11 is a rail for moving the television camera 10 in the longitudinal direction of the measurement object, and 12 is an imaging target. Thirteen
Is an image processing device for synthesizing images, and 14 is a television camera 1.
A controller for controlling the movement amount of 0, and 15 is a monitor television for observing an image or a composite image input to the television camera 10.

The image composition processing apparatus of this embodiment will be described below with reference to FIG. While observing the image captured by the television camera 10 on the monitor television 15 at any time, the television camera 10 is moved in the longitudinal direction of the image capturing target 12, and the captured image is sent to the image processing device 13. The moving interval of the television camera 10 is determined by the following method. First, the zoom magnification of the television camera 10 is adjusted so that a desired area can be imaged. After that, the TV camera 10
Monitor TV 15 so that the captured images may partially overlap.
Move while observing. The movement amount at this time is stored in the controller 14 as a movement step. After that, the television camera 10 is moved by the controller 14 at each movement step, and the process of capturing an image is repeated. The plurality of images obtained as described above are combined in the image processing apparatus 13 by the image combining method shown in the first embodiment.
The combined image can be observed on the monitor television 14.

In the present embodiment, the method of moving the television camera 10 is shown, but it goes without saying that the image pickup object 12 may be moved.

An image synthesis processing apparatus according to the fourth embodiment of the present invention will be described below with reference to the drawings.

FIG. 4 shows an image composition processing apparatus according to the fourth embodiment of the present invention. In FIG. 4, reference numeral 16 is a television camera for inputting a plurality of images. Reference numeral 17 denotes a rotating device for rotating the TV camera 16 around a direction perpendicular to the optical axis of the TV camera 16.
Reference numeral 8 is an image processing device for synthesizing images, and 19 is a monitor television for observing the image input to the television camera 16 and the synthesized image.

The image composition processing apparatus of this embodiment will be described below with reference to FIG. While observing the image captured by the TV camera 16 on the monitor TV 19,
The rotation device 17 is used to rotate to capture an image. The amount of rotation is such that the field of view before rotation and the field of view after rotation partially overlap. The captured image is input to the image processing device 18,
It is synthesized by the method shown in the first embodiment.

When the total rotation angle of the television camera 16 is 360 degrees or more, the obtained image becomes a panoramic image in which images in all directions are combined.

An image synthesis processing method according to the fifth embodiment of the present invention will be described below with reference to the drawings.

FIGS. 5 (a) and 5 (b) show an image synthesizing processing method when the image pickup object to be joined has a periodic pattern, 20 indicates an image having a periodic pattern to be joined, and 21 is a joining. 7 shows an image with another periodic pattern that is applied. Reference numeral 22 is an area for searching a position to be combined with a part of the periodic pattern 21, and 23 and 24 are images 2.
At 0, the difference between the brightness value and the color of the area 22 is 0 or the minimum value. Reference numeral 25 indicates an overlapping area where the joint location is not misjudged.

The image composition processing method of this embodiment will be described below with reference to FIG. The image to be combined is (Fig. 5)
In the case of the periodic pattern shown in 20 and 21 of
When the difference from the area 22 is obtained in the entire area of the image 20,
There are two areas, areas 23 and 24, where the difference in brightness value or color from the area 22 is minimum. Therefore, it is not possible to specify which region is to be the joining place. Therefore, it is necessary to limit the area to be searched by overlapping to some extent. If the search area is within one cycle of the periodicity pattern, a plurality of joining candidate locations will not occur. Further, in order to make the combined image as large as possible, it is necessary to minimize the overlapping image pickup area. Therefore, the overlapping area is the area 25 within one cycle pattern from the right end of the image 20.
Limited to By doing so, the area to be joined can be prevented from being erroneously determined, and the area to be joined can be made as wide as possible, which is effective.

An image synthesis processing apparatus according to the sixth embodiment of the present invention will be described below with reference to the drawings.

FIG. 6 shows an image synthesizing processing device in the case where an image pickup object to be joined has a periodic pattern. In FIG. 6, reference numeral 26 is an imaging target having a pattern having periodicity. Reference numeral 27 is a television camera for inputting a plurality of images. 28 is a rail for moving the television camera 27 in the longitudinal direction of the imaging target 26, 29 is an image processing device for synthesizing images, and 30 is the television camera 27.
The controller 31 controls the amount of movement of the monitor, and 31 is a monitor TV for observing an image input to the TV camera.

The image composition processing apparatus of this embodiment will be described below with reference to FIG. While observing the television camera 27 on the monitor television 31 as needed, the television camera 27 is moved in the longitudinal direction of the imaging target 26, and the captured image is sent to the image processing device 29.
The moving interval of the television camera 27 is determined by the following method. First, the zoom magnification of the television camera 27 is adjusted so that a desired area can be captured. Then TV camera 27
The monitor television 3 is configured so that the captured images partially overlap and the overlapping area is within one cycle of the periodic pattern.
Move while observing at 1. The movement amount at this time is stored in the controller 30 as a movement step. After that, the television camera 27 is moved by the controller 30 at each movement step, and the process of picking up an image is repeated. The plurality of images obtained as described above are combined in the image processing device 29 by the image combining method shown in the fifth embodiment.

In the present embodiment, the method of moving the television camera 27 has been shown, but it goes without saying that the object imaging target 26
May be moved.

[0036]

As described above, according to the present invention, since it is not necessary to precisely move the image pickup device or the image pickup object in order to join a plurality of images, it is simple, and the brightness value or chromaticity difference between the images is simple. Since the connection position is obtained by using, it is possible to accurately combine the images without losing continuity.

[Brief description of drawings]

FIG. 1 is a diagram showing an image composition processing method according to a first embodiment of the present invention.

FIG. 2 is a diagram showing an image composition processing device according to a second embodiment of the present invention.

FIG. 3 is a diagram showing an image composition processing device according to a third embodiment of the present invention.

FIG. 4 is a diagram showing an image composition processing device according to a fourth embodiment of the present invention.

FIG. 5 is a diagram showing an image synthesis processing method according to a fifth embodiment of the present invention.

FIG. 6 is a diagram showing an image composition processing device according to a sixth embodiment of the present invention.

[Explanation of symbols]

1, 2 images 3 image after combining 4, 22 region for searching splicing position 5 region showing combining position 6a, 6b, 6c, 10, 16, 27 TV camera 7, 13, 18, 29 image processing device 8 , 15, 19, 31 Monitor TV 9, 12 Imaging target 11, 28 Rails 14, 30 Controller 17 Rotating device 20, 21 Image having periodicity 23, 24 Difference between luminance value or color and area 22 is 0 or minimum value Captured area 25 Overlapping area 26 Imaging target with periodic pattern

Claims (6)

[Claims] 1. An image processing method for joining a plurality of images picked up by an image input device, wherein the joining ends of the joining images are picked up so as to overlap a part of the joined images, and An image synthesizing method for joining images in a region where a difference in luminance or a chromaticity difference between an image at a joining end and an image to be joined is minimized. 2. An image synthesizing processing method for joining a plurality of images having a periodic pattern, wherein a part of the image where the joining ends of the images to be joined are joined is within one cycle of the periodic pattern. An image synthesizing processing method, wherein images are picked up so as to overlap in a region, and images are joined in a region where a difference in luminance or chromaticity between an image to be joined and an image to be joined is minimum. 3. An image synthesizing processing device arranged such that a part of images picked up by a plurality of image input devices overlaps with an adjacent image, and a luminance difference or a chromaticity difference between two images in an overlapping area. An image synthesizing processing device that joins a portion having the smallest value as a joining portion. 4. An image composition processing device comprising an image input device for picking up an image of an object, and a moving device for moving the object or the image input device in a direction perpendicular to an optical axis of the image input device. The device moves the object before and after the movement so as to partially overlap in the field of view of the image input device, and the luminance difference or chromaticity difference of the images in the overlapping region of the images obtained by the image input device before and after the movement is minimum. An image synthesizing processing device that joins a portion to be a joint portion. 5. An image synthesizing processing device comprising an image input device for picking up an image of an object and a rotating device for rotating the image input device around a direction perpendicular to an optical axis of the image input device. The device rotates so that the imaging regions before and after the rotation partially overlap in the field of view of the image input device, and the luminance difference or chromaticity of the images in the overlapping region of the images obtained by the image input device before and after the rotation. An image synthesizing processing device that joins a portion having the smallest difference as a joining portion. 6. An image combining processing device comprising an image input device for picking up an object having a periodic pattern, and a moving device for moving the object or the image input device in a direction perpendicular to an optical axis of the image input device. Then, the moving device moves the object before and after the movement so that the overlapping region is within one cycle of the periodic pattern in the visual field of the image input device, and the object is moved by the image input device before and after the movement. An image synthesizing processing device that joins a portion having a minimum luminance difference or chromaticity difference between neighboring images including a duplicated portion of the obtained images as a joining portion.