JPH03185571A - Image processor - Google Patents

Abstract

PURPOSE:To easily and rapidly control the reading of image data by directly using image data to be applied to an image storage part as address information corresponding to image data read out from the image storage part. CONSTITUTION:A two-dimensional image e.g. stored in the 1st image memory 1 is used as a distance image and the image data of the two-dimensional image (distance image) are used as the address information of the image data to read out from the 2nd image memory 2. Namely, the image data, i.e. respective values (picture element data) of plural picture elements constituting the two-dimensional image, of the two-dimensional image found out from the 1st image memory 1 are used as the address (coordinate) data of the 2nd image memory 2 and the image data stored in the address (coordinates) of the 2nd image memory 2 specified by the picture element data are read out. Thus, the reading control of the image data stored in the image storage part can be easily and efficiently executed and image data on the optional face of a three-dimensional image e.g. can rapidly be read out.

Description

[Detailed description of the invention]

[Object of the Invention] (Industrial Application Field) The present invention relates to an image processing device that can easily and efficiently control reading of image data from an image storage unit. (Prior art) Reading out image data stored in an image storage unit is as follows:
This is performed by sequentially specifying the coordinates (addresses) of the image storage section in which image data to be read is stored. For example, conventionally, when a two-dimensional image stored in an image storage section is read out and displayed in a film, the coordinates (addresses) of the image storage section are sequentially specified in association with the scanning method of image display. The two-dimensional image (image data) is read out. For example, when an image is displayed by rask scanning, the above-mentioned Image data constituting a dimensional image is read out in order of rask scanning lines. However, when displaying images in special display modes,
When performing various image processing on the images stored in the image storage section, it is necessary to specify the coordinates (addresses) of the image storage section in an order different from the order of rask scanning described above. When reading out such special image data, that is, reading out image data that does not follow a predetermined order (such as rask scanning), especially when the special readout mode is frequently used in image processing. It has been proposed to realize high-speed image readout by implementing a coordinate (address) specifying means for controlling the readout as dedicated hardware. This method is described in, for example, ``Proceedings of the 1985 National Conference of the Institute of Electronics and Communication Engineers, k
1264; Mitsuo Tabata et al. Image memory address calculator LSI J etc. are introduced in detail. In addition, in order to handle image readout using special coordinate control, it is possible to provide a coordinate calculation section separate from the image calculation section.
For example, it has been proposed in ``Technical Research Report IB 81-8 J of the Image Engineering Research Group of the Institute of Electronics and Communication Engineers.'' However, it is proposed in the IEICE Image Engineering Study Group Technical Research Report IB 81-8 J. In order to read out image data at high speed, it is undeniable that a coordinate (address) specifying means with a large-scale hardware configuration is inevitably required.Especially when the image to be processed is given as three-dimensional image data. In this case, the coordinate (address) designation control becomes extremely complicated, making it extremely difficult to read out image data at high speed. (Problems to be Solved by the Invention) As described above, Conventionally, when image data stored in an image storage unit is read out at high speed in a special readout order, the means for controlling the readout coordinates (addresses) becomes complicated, and when this is implemented using dedicated hardware, it is difficult to do so. There is a problem in that it is undeniably a fairly large-scale configuration.Particularly when reading out an image on an arbitrary plane of a three-dimensional image, there is a problem in that the coordinate control becomes extremely complicated. The invention was made taking these circumstances into consideration, and its purpose is to simply and efficiently control the readout of image data stored in an image storage unit, for example.
An object of the present invention is to provide a highly practical image processing device that can read out image data on any plane of a dimensional image at high speed. [Structure of the Invention] (Means for Solving the Problems) An image processing device according to the present invention converts image data given to an image storage unit that stores a digital image into coordinates of the image data to be output from the image storage unit. It is characterized in that it can be used as a. For example, two-dimensional image data given to the image storage section is used as a distance image for specifying coordinates for the three-dimensional image data stored in the image storage section, and image processing is performed on this distance image to create the three-dimensional image. The present invention is characterized in that by specifying an arbitrary surface of the three-dimensional image, image data on an arbitrary surface of the three-dimensional image can be easily and efficiently read out at high speed. (Function) According to the present invention, the image data of the two-dimensional image given to the image storage section is used as the coordinates of the image data to be output from the image storage section. For example, by simply defining as a distance image that specifies a specific plane in a three-dimensional image, the desired image can be easily, efficiently, and quickly retrieved from the image storage unit without performing complicated coordinate designation control. It becomes possible to read image data. In particular, even when reading image data on a surface consisting of a complex combination of surfaces in a three-dimensional image, a distance image is generated by image processing, and coordinates are specified using this distance image.
Image data of a desired surface in a dimensional image can be read out at high speed by simple coordinate control. (Embodiment) An image processing apparatus according to an embodiment of the present invention will be described below with reference to the drawings. FIG. 1 is a diagram showing an example of the system configuration of the embodiment device. This embodiment device basically stores a first image memory l that stores a two-dimensional image used as a distance image to be described later, a three-dimensional image to be read out to an image display device, etc. Alternatively, the second image memory 2, which is used as a table for image conversion, and the image processing section 3, which executes various arithmetic processes on image data, are connected to the bus 4.
The second image memory 2 is connected to the second image memory 2 and is configured to read desired image data from the second image memory 2 while controlling the transfer of image data between them. The first and second image memories 1.2 each include memory control units 1a and 2a, and the memory control unit 1.
Writing and reading of image data are respectively controlled in response to address designations by a and 2a. Note that these first and second image memories 1.2 are 1.
In some cases, a first image storage area and a second image storage area are defined by dividing storage areas of two large-capacity image memories. Here, the feature of this embodiment device is that, for example, a two-dimensional image stored in the first image memory l is used as a distance image, and image data of this two-dimensional image (distance image) is used as a second image. The point is that it is used as address information for image data to be read from the memory 2. That is, the image data of the two-dimensional image obtained from the first image memory 1, that is, the values (pixel data) of each of the plurality of pixels constituting the two-dimensional image, are sent to the address (coordinates) of the second image memory 2. It is characterized in that the image data stored at the address (coordinates) of the second image memory 2 specified by this pixel data is read out. FIG. 2 is a diagram showing a conceptual configuration example of an embodiment device that realizes such a function, and 11 is an image memory (corresponding to the first image memory l) that stores a plurality of two-dimensional images (distance images). ), 12 is an image memory (corresponding to the second image memory 2) that stores an image to be read, for example, a three-dimensional image. The three-dimensional image stored in this image memory 12 is
For example, it consists of image data having values representing physical characteristics such as the amount of X-ray absorption in a three-dimensional space of x, y, and z, obtained using a CT (computer tomography) device. However, the memory control section for specifying the address to the image memory 12 and reading out the data of the three-dimensional image is the coordinate generation section 13. Synthesis selector section 14. and an image calculation section 15. Note that this image calculation section 15 may be realized as the image processing section 3 described above. The synthesis selector section 14 manually specifies the address of the image memory 12 using the address data generated by the coordinate generation section 13, or alternatively specifies the two-dimensional image (distance image) obtained from the image memory 11. input image data as address data to designate the address of the image memory 12, and further input image data obtained by image calculation from a plurality of two-dimensional images in the image calculation section 15 as address data. The address specifying operation described above is selectively executed in response to an instruction from a control means (not shown). That is, when sequentially reading image data from the image memory 12 in the order of rask scanning, for example, the synthesis selector section 14 takes in address data sequentially generated by the coordinate generating section 13 according to a predetermined rule, and reads the address data according to this address data. Address specification for the image memory 12 is performed. Therefore, in this case, the reading of image data from the image memory 12 is controlled according to the address data generated by the coordinate generation section 13, for example, according to the order of rask scanning. On the other hand, the synthesis selector section 14 selects the image memory 1.
1 or the image calculation unit 15
When inputting image data obtained through the image memory L2, the image data itself is used as address data for the image memory L2. As a result, the addresses (coordinates) indicated by the image data are sequentially specified in the image memory 12, and the image data stored at the coordinates are sequentially read out. The reading of image data from the image memory 12 using this input image data as address data for the image memory 12 will be described below using a specific example. , the image data stored in the image memory 12 is, for example, the third
It consists of data of a three-dimensional image forming a cube with three axes of XIYIZ as shown in Figure (a). Then, the three-dimensional image data is rotated according to the viewing direction of the three-dimensional image and stored in the image memory 12. In the example shown in Fig. 3 (a), the three-dimensional image of a cube is viewed from diagonally upward (45@ direction with respect to each axis of x, y, and z) with its corners facing in front. The state indicates that the image data is stored in the image memory 12 in a state in which the image data has been rotated by isometric projection. In other words, the vertical axis y of the three-dimensional image is along the vertical direction (Y-axis) of the paper, and the
Image data is stored in the image memory 12 after being rotated by isometric projection so that the axis and the two axes each make an angle of 30@ with respect to the horizontal direction (X-axis) of the page. In this case, a two-dimensional image forming a cross section in a plane having a depth in the z-axis direction parallel to the xy plane of the three-dimensional image is schematically represented as shown in FIG. 3(b). The horizontal direction of this cross section A is the X axis. On a sheet of paper with the vertical direction as the Y axis, the viewing direction (direction of the front and back of the sheet) is the two axes, and when expressed as depth information on the Z axis, it is shown in Figure 4 (a). It is expressed as data like this. Note that in this data, the smaller the numerical value, the closer the position is on the Z-axis, and the larger the numerical value, the farther back the position is on the Z-axis. Similarly, a two-dimensional image forming a cross section in a plane having a depth in the X-axis direction parallel to the yz plane of the above three-dimensional image is the third one.
If the cross section B is similarly expressed as depth information on the Z axis, it will be expressed as data as shown in FIG. 4(b). In addition, a two-dimensional image that forms a cross section on a plane with a depth in the y-axis direction parallel to the zx plane of the above three-dimensional image is shown in Figure 3 (d
), and when the cross section C is expressed as depth information on the Z axis, it is expressed as data as shown in FIG. 4(C). That is, this shows that cross sections A, B, and C in the three-dimensional image shown in FIG. 3(a) described above can be respectively expressed as depth information in the Z-axis direction. Therefore, when considering an xY plane image (two-dimensional image) whose image data is depth information in the Z-axis direction, the image data is the position (surface shape) of the plane (cross section) in the three-dimensional image. It is nothing but an expression of.゛In the present invention, a two-dimensional image expressing depth information for such a three-dimensional image is defined as a distance image, and image data of this distance image is used as address data for reading data of the three-dimensional image. It is characterized by The image memory 11 stores data on a plurality of types of distance images (two-dimensional images) a, b, and c that express arbitrary surfaces of such three-dimensional images. The composition selector section 14 in the memory control section selects the image memory 1
Selectively obtaining the distance images a, b, and c depending on which surface of the 2- to 3-dimensional image is to be read out,
Alternatively, the image calculation section 15 is activated to generate a distance image d of a desired surface, and the address of the image memory 12 is designated and controlled according to the data of the distance image. Therefore, for example, when reading out a two-dimensional image of cross-section A of the three-dimensional image described above from the image memory 12, a distance image a as shown in FIG.
The address of the image memory 12 is designated and controlled according to the image data. As a result, in this case, the depth of one line from left to right is [44, 45, 47
, 48, 50, 52, 53, 55° 56], image data for the above-mentioned section A of the three-dimensional image is read out. Similarly, when using a distance image b as shown in FIG. 4(b), the depth of one line from the left side to the right side is [
5(i, 55.53.52.50.4g, 47.4
5°44] is read out, and when a distance image C as shown in FIG. 4(C) is used, the depth of one line from the left side to the right side is The depth is constant for each individual line, but as the line changes from the top to the bottom, the depth becomes [56, 55,
53, 52, 50, 48, 47, 45, 44] is read out. That is, in this case, the distance images a, b, and c selectively obtained from the image memory 11 are used as they are as address information for controlling the reading of image data from the image memory 12. On the other hand, when a new distance image d is created by using the image calculation unit 15 and calculating the intermediate value of the three distance images a, b, and c described above, for example, as shown in FIG.
An image of a surface having a complicated surface shape (a combination of a plurality of surfaces A, B, and C) as shown in e) is obtained. That is, when the arithmetic processing for determining the intermediate value of the three distance images a, b, and c is executed pixel-by-pixel, a distance image d as shown in FIG. 4(d) is obtained. The surface in the three-dimensional image shown by this distance image d is the surface A shown by each of the three distance images described above. It is a combination of B and C, and these surfaces A and B. This indicates a surface portion that has an intermediate depth when C intersects. As a result, when three-dimensional image data is read out by performing address control on the image memory 12 using the distance image d, the image data read out from the image memory 12 is an image on the complex plane shown in FIG. 3(d). will be shown. As is clear from such image calculation processing, for example, by performing image processing on a distance image, a distance image that specifies an arbitrary surface of a three-dimensional image can be easily generated. Then, by performing address control on the three-dimensional image using the distance image, it becomes possible to selectively read only the image information about that plane very easily and efficiently. As described above, the processing concept of the present invention has been explained using the reading of image data for an arbitrary cross section of the cubic three-dimensional image shown in FIG. 3 as an example.The present invention introduces the concept of distance image, Image data of a two-dimensional image representing a distance image is directly used as address information for a three-dimensional image to control reading of the image data. Therefore, depending on which plane in a 3D image requires image data, it is possible to simply generate a distance image that can specify that plane. Reading of image data can be controlled only by image calculation processing. In particular, the present invention can be achieved by simply adding a function to a conventional in-membrane image storage device to use given image data (data of a two-dimensional image indicating a distance image) as address data. Characteristic image data readout control can be realized. As a result, it is possible to achieve great practical effects, such as being able to read out image data from any arbitrary surface very easily and at high speed without constructing dedicated hardware for address control. Note that the present invention is not limited to the embodiments described above. Although the reading of three-dimensional image data has been described here, when reading two-dimensional image data in a special format, the concept of the distance image described above can be applied as is. It is also possible to obtain the calculation results between the plurality of images by using the image memory 12 as a table and using, for example, a composite value of each image data of a plurality of two-dimensional images as address data. Specifically, in various types of image processing, it is often desirable to perform division processing for shading correction between images, or to perform functional calculations between image data. In such a case, it is undeniable that a large amount of calculation processing time is required to perform the calculation processing for each individual image data. Therefore, in such a case, it is possible to store table information in which calculated values corresponding to the image data are obtained in advance in the image memory 12, and to perform a table search using the composite value of the plurality of image data as address information. , it becomes possible to read out the calculated values easily and quickly and obtain the image calculation results. Further, in the embodiment, the case where the distance image is provided as a plane has been described as an example, but it is of course possible to provide it as a curved surface or a combination of a plurality of surfaces. In addition, the present invention can be implemented with various modifications without departing from the gist thereof.

【Effect of the invention】

As explained above, according to the present invention, the image data given to the image storage section is used as is as address information for the image data read out from the image storage section, so the format for reading out the image data from the image storage section is complicated. Even in special cases, the readout can be controlled very simply and at high speed without performing complicated address control, which is a great practical effect.

[Brief explanation of drawings]

The figures show an image processing device according to an embodiment of the present invention, in which FIG. 1 is a diagram showing an example of the system configuration of the embodiment device, and FIG. 2 is a diagram schematically showing the processing functions of the embodiment device. Figure 3 is a diagram showing the concept of a three-dimensional image and its cross section, I
FIG. J4 is a diagram showing a distance image for specifying the cross section of the three-dimensional image shown in FIG. 3. 1.2...first and second image memories, la. lb...Memory control unit, 3...Image processing unit, 4...
・Bus, 11... Image memory (stores distance images), 1
2... Image memory (stores three-dimensional images), 13...
Coordinate generation unit, 14... Synthesis selector unit, 15... Image calculation unit, A, B. C...Cross section, a, b, c, d...Distance image.

Claims (3)

[Claims] (1) An image processing device comprising: an image storage section that stores digital images; and means for using image data given to the image storage section as coordinates of image data to be output from the image storage section. . (2) The image storage unit includes a first storage area that functions as a two-dimensional image memory and a second storage area that functions as a three-dimensional image memory.
A storage area, and the two-dimensional distance image data stored in the first storage area is used as data specifying the coordinates of the three-dimensional image data to be output from the second storage area. The image processing device according to (1). (3) The image data used as the coordinates of the image data to be output from the image storage unit is obtained as a result of arithmetic processing between a plurality of image data.
The image processing device according to item 1).