JPS5851301B2 - Image input processing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image input processing device, and more particularly to an image input processing device for combining multiple images into a single large-scale image.

Generally, maps, aerial photographs, and the like are divided into small-scale regions due to various restrictions and size restrictions when creating them.

In order to save these divided images as a single continuous large-scale image in the form of a digital signal, etc., it is necessary to perform operations such as positioning to synthesize a large number of partial images when inputting the images. The information necessary for this alignment must be extracted from the content of the image itself, or additional information must be added.

The method of extraction from the image content involves extracting characteristic features on the image (for example, triangular points on a map) using pattern recognition processing, and then identifying similar partial images that should be located at the top, bottom, left, and right of the target image. There is a method of positioning by overlapping one point.

However, it has the disadvantage that a large amount of processing is required for recognition processing.

The latter method involves writing an easy-to-extract marker on the image in order to facilitate the above-mentioned recognition process.

However, there is a drawback that the image is damaged or missing because markers and the like are written on it.

In order to solve the above-mentioned drawbacks, the present invention provides a margin area around the input image separately from the image to be input so that markers for indicating feature points can be written, and the image data and the markers are combined. By making it possible to separate and read the images, we aim to simplify alignment between multiple images.
Moreover, the purpose is to prevent damage to the image to be input.

Hereinafter, a detailed explanation will be given using the drawings.

FIG. 1 shows an explanatory diagram illustrating the configuration considerations necessary for an input image used in the present invention.

In the figure, 100 indicates the input image, and 101 indicates the original image area, that is, the area within the inner frame, which includes the image that should originally be input.

102 is a blank area for writing markers, 103-a, 1
03b, 104-a, 104-b are markers used for alignment; 105 is the marker 103-a,
b, 104 At the intersection of the line segments obtained by a, b,
Indicates a reference point for aligning input images.

FIG. 2 is 103-a in FIG. An example of the markers b, 104-a and b is shown, in which 201 is the marker body, 20
2 is a code section for identifying each marker.

FIG. 3 shows an explanatory diagram illustrating a mode of obtaining an input image used in the present invention.

Reference numerals 100, 101, and 102 in the figure indicate the same components as shown in FIG.

Further, 301 is an area in the original image area 101 that overlaps when superimposed with another positionally adjacent partial image, and is called a margin area.

302-a, b. 303-a, b, 304-a
, b, 305-a.

b is a marker, and 306, 307, 308.degree. 309 indicates the intersection of each line segment obtained by the marker and is used as a reference point for alignment.

First, the constituent elements of the input image used in the present invention will be explained with reference to FIGS. 1 and 2.

The input image 100 is composed of an original image area 101 that is to be input, encoded, etc., and created as a file, and a margin area 102 in which a marker is written.

The original image area 101 includes one of the individual partial images for creating a large-scale image by the image input device of the present invention.

Assuming that this partial image has a part of the area overlapping with another positionally adjacent partial image to be combined, a reference point 105 is selected within this overlapping area,
Using this reference point, alignment with other partial images is performed and composite is performed.

The reference point 105 may be set in an overlapping area, for example between two areas, or a point that is easy for a machine to recognize.

Markers 103a, b, 104-a. b is for indicating the position of this reference point 105, and the intersection of the line segment formed by markers 103-a and 103-b and the line segment formed by markers 104-a and 104-b is designated as reference point 105.
Select a marker to match the .

Markers 103-a, b, 104 a, b are, for example, second
As shown in the figure, it is composed of a marker main body part 201 and a code part 202.

The marker main body 201 defines the position of the line segment for indicating the reference point 105 mentioned above.The center of gravity of the circle 201 shown in FIG. The reference point 105 is the starting point or ending point of a line segment.

As a method for determining the center of gravity at this time, it is possible to use a method well known in conventional image processing technology, and in the simplest example, as shown in FIG. 2, the coordinate X of the marker is determined.

, xl, yo, yl, the coordinates of the center, that is, the center of gravity (X Cl y□) are determined.

Note that the marker body 201 does not need to be limited to a circular shape as long as it has a shape that allows the center of gravity to be easily determined.

Also, although a certain area is given to the marker main body 201,
This may be considered to be to avoid the influence of noise etc. during reading.

The code section 202 is shown in the form of a bar in the figure, and for example, by changing the number and length of the bars, it is possible to individually identify multiple markers written on one human image. be.

This code portion 202 is also shown in a bar shape in FIG. 2, but other shapes may be used since the purpose is to separate and distinguish it from other markers.

Next, with reference to FIG. 3, description and alignment of an input image including the above-mentioned components will be described.

The original image 100 includes an original image area 101 to be input, a margin area 102 for writing a marker, and a marker 302-a.
Consists of b to 305 a, b.

At this time, it is assumed that the original image area 101 partially overlaps with images located adjacent to each other in position, and this area is called a margin area 301.

Reference points 306 to 309 for alignment and synthesis may be set in this margin area 301.

That is, for compositing with other images, for example, compositing the target image with another image located to the left of it, it is sufficient to use two reference points, for example, the reference points 306°, 309°.
It is possible to use .

At this time, if reference points are selected corresponding to reference points 306° and 309 on other images located on the left side of the target image, it is possible to align and synthesize these images by matching these reference points. .

Also, menu marker 3 indicates the position of reference points 306 and 309.
02-a, b, 304-a.

b, 305-a, b and use them.

Furthermore, for aligning and compositing the target image with the images located above, below, and to the left of it, it is sufficient to similarly set two reference points; in addition to the reference points 306 and 309, the reference point 30
It is clear that this can be achieved by using 7,308.

Also, markers 302a, b to 305-a.

b can distinguish these by each having a code section 202, and it is possible to easily detect which reference point the marker indicates, and it is possible to set a certain rule between the marker position and the code regardless of the image. It will be obvious that this will simplify the work during synthesis.

Also, the area actually recorded in the file etc. is the reference point 306
Needless to say, this applies only to the inside of 309.

Next, one embodiment of the present invention will be described using FIG. 4.

In the figure, reference numeral 401 is an image input section, 402 is an A/D conversion section, 403 is an input buffer section, 404 is a designated position detection section,
405 is an encoding unit, 406 is a composition processing unit, and 407 is an image file storage unit.

This operation will be explained below.

An input image as shown in FIG. 3 is scanned by an image input unit 401 and converted into an electrical signal.

The image input unit 401 can use an existing input device such as a TV camera, a flying spot scanner, or a facsimile.

Next, the obtained image signal is subjected to analog-to-digital conversion in an A/D conversion section 402 and then sent to an input buffer section 403.

The input buffer unit 403 has a memory for temporarily accumulating image signals and a function of separating the original image area 101 and blank area 102 in the accumulated image signal.

This separation involves distinguishing between a blank space and an original image area, and can be easily realized using existing processing techniques.

The separated signals are sent to indicated position detection section 404 and encoding section 405, respectively.

The encoding unit 405 performs band mapping processing such as predictive encoding on the input image signal, and is provided for the purpose of reducing the amount of files stored.

The encoded image signal is output to the image synthesis processing section 406.

Further, the indicated position detection unit 404 extracts the written marker and detects the position where the marker is attached.

To extract the marker, it is sufficient to utilize the contrast difference between the margin and the marker, and finding the center of gravity of the marker is nothing but finding the position of the center of gravity of the group of pixels expressing the marker.

Furthermore, the position of the marker can be determined from the address in the input buffer section 403 to be stored, and the reference point can be easily determined from this marker position by simple calculation.

The determined position of the reference point is then output to the synthesis processing section 406.

The synthesis processing unit 406 extracts a region to be input to the image file storage unit 407 from these encoded image signals and reference point positions, and writes the area to a file.

Needless to say, at this time, in order to create a large-scale image by the image file storage unit 407, the writing position of the image in the file is also calculated.

To do this, it is possible to specify the position between 2 and 3, but all the partial images to be input should be numbered in a series, and these numbers should be written in the margin area 102 in the same way as markers, so that they can be read. It is clear that this can be easily achieved.

If the apparatus of the present invention described above is used for a group of photographic data divided into small regions, such as an aerial photograph, the regional data can be handled as a unified large-scale file.

In particular, in general, all photos are taken seamlessly so that no information is missing, and adjacent photos always have overlapping parts.

Further, there is a margin area for writing photographic data, etc., and if these are used as the margin area 301 and the margin area 102 described in the explanation of the present invention, the current photographic data can be easily used.

In addition, if the original image has longitude lines and latitude lines written on it like a map, it will be easier to set markers etc. by using these longitude lines and latitude lines. It is also clear that it will be easier to position the system.

Also, in the above explanation, we mainly described a method for creating a large-scale image file by stitching and compositing multiple images using a reference point and a marker to indicate the reference point. Since it is possible to specify an arbitrary position within an image, it goes without saying that it can also be used to specify a cropping area of an image, a feature point, etc., without causing damage to the image content.

As explained above, according to the present invention, it is possible to easily create a single large-scale image from image data that is divided into small areas, such as an aerial photograph or a map, without damaging the image data itself. There is virtually no break in the image on the file, and it is possible to access and display any area.

[Brief explanation of the drawing]

FIG. 1 is an explanatory diagram illustrating the components necessary for the input image used in the present invention, FIG. 2 is an example of a marker used in the present invention, and FIG. 3 is an explanation of a mode for obtaining the input image used in the present invention. An explanatory diagram, FIG. 4, shows the configuration of one embodiment of the present invention. 100... Input image, 101... Original image area, 102---- Margin area, 103-a,
b, 104-a. b------Marker, 105---Reference point, 2
01... Marker body part, 202... Code part, 301 , Margin area, 302-a
, b to 305-a. b... Marker, LaO2 without 306...
・Reference point, 401... Image input section, 402...
...A/D conversion unit, 403...Input buffer unit, 404...-Instruction position detection unit, 405...
...Encoding section, 406...-Composition processing section, 4
07... Image file storage section.

Claims (1)

[Claims] 1. An image input unit that scans an image and converts the image information into an electrical signal and extracts it, an original image area of the input image, and a marker for indicating a reference point outside the original image area frame. an input buffer section that separates the blank space for
an encoding unit that encodes an image signal in the original image area;
an indicated position detection section that detects a marker for indicating the reference point written in the blank area to determine the reference point position; and a indicated position detection section that detects the image signal encoded by the encoding section. An image input processing device characterized in that the image input processing device is intercepted from a compositing processing unit that aligns the reference point positions obtained and uses them as an input image code.