JPS63188270A - Recording and reproducing system for digital image - Google Patents

Abstract

PURPOSE:To sharply compress a picture data without lowering reproducing picture quality, by converting picture information at a part other than an interested area to a data with a small number of bits, and recording only the interested area by using the picture data with a large number of bits. CONSTITUTION:A digital picture data is stored in an image memory 1, and also, the contour data of which is extracted at a contour extraction circuit 2, and is stored in an image memory 3. And at the time of recording the data in the image memory 1 on a recording medium 5, only the picture data in the interested area on an image designated in advance is read out, and is recorded on the recording medium 5, and as for the contour data, the contour data of whole of the image is recorded. At the time of reproducing the image, the data is read our from the recording medium 5, and is stored in the image memory 1, and the image data of the image to be displayed in the memory 1 is edited, and is read out and displayed. In such a way, the whole quantity of the picture data can be compressed sharply, and no deterioration in the reproducing picture quality is generated.

Description

DETAILED DESCRIPTION OF THE INVENTION A. Field of Industrial Application The present invention relates to an image recording and reproducing method used when converting an X-ray image for X-ray diagnosis into digital data and storing it.

It is determined by the number of bits of gradation data given to conventional technology pixels, and in high-resolution systems using high-resolution television cameras, etc., the number of pixels is 1024 x 1024, and each pixel data is about 16 bits. However, recording the image data used in such a system on a recording medium requires a large capacity recording medium. For this reason, various methods have been proposed for compressing image data to be recorded, but compression methods that do not cause deterioration of reproduced images have little data compression effect (for example, 2
The amount of data can only be reduced to about half by applying the dimensional predictive coding method.

Problems that the invention aims to solve As the resolution of imaging technology advances, it is desired that recording means for recording image data have an increasingly larger capacity, but on the other hand, it is necessary to compress image data to a greater extent without reducing image quality. This has also become an important issue. The present invention has been made in view of this situation, and is intended to significantly compress image data without degrading the reproduced image quality.

20 Means for Solving Problems In X-ray images for diagnosis and other images for various examination purposes, there are regions of interest that must be observed in detail, and the details within the regions of interest and the areas of interest within the entire screen are The relationship with the overall image, such as its position, is required as image information. Focusing on this point, the present invention specifies a region of interest on an image, extracts contour information for other regions, significantly compresses the image data, and combines all image data of the region of interest with that contour data. Only the data specifying the area is recorded on the recording medium.

Since all image data is recorded only for the area of interest on the screen, and only contour data is recorded for other parts of the screen, the amount of recorded data is greatly reduced, and when this recording is played back, the entire image represented by the contours is displayed. The region of interest is displayed in detail in the image, so the positional relationship of the region of interest in the overall image is accurately determined, and the region of interest is not subjected to any processing that would degrade the image quality, resulting in an excellent image. is reproduced.

Embodiment FIG. 1 shows an entire embodiment of the present invention. In this embodiment, the image consists of 1024x1024 pixels, and the image data for one pixel is 16 bits. A video signal output from an imaging device (not shown) is converted into digital image data by an AD converter and stored in the image memory 1 via the image bus I31, and contour data is extracted in real time by the contour extraction circuit 2. , this data is stored in the image memory 3 with one bit per pixel. 4 is an address control circuit that specifies addresses when writing and reading data in the two memories 1 and 3 mentioned above; 5 is a recording medium;
The data in the image memory 1 and the contour image memory 3 are recorded via the controller 6.

When the data in this image memory is recorded on the recording medium 5, only the image data of the region of interest on the image designated in advance is read out and recorded on the recording medium 5, and the contour data of the entire image is recorded. However, since contour data is one bit per pixel, if the image data is 16 bits per pixel, the image data is compressed to 1/16, and the area of interest is 91 squares, which is 1/16 of the screen. (In terms of screen size, the length of one side is 1/4), then the entire image data will be compressed to 1/8. In the case of image reproduction, first the recording medium 5 is
Contour data is read out from the image memory 3 and stored in the image memory 3 for contour data.Next, constant data giving an appropriate gradation on the screen is added to the data for each pixel read out from the image memory 3, and the data is stored in the image memory 3. 1 to the address corresponding to each pixel. Since the contour memory is 1-bit data, gradation data is added by adding an appropriate 15-bit code and transferring the entire data to the image memory 1 as 16-bit data. When this operation is completed, the data of the region of interest of the image is read out from the recording medium 5 via the I10 controller 6, and
In this way, the image data of the image to be displayed in the image memory 1 is edited, and this image data is read out and stored in place of the previously input contour data at the address corresponding to the region of interest. CR for display as shown
It is displayed as an image on T. The CPU is a processor that performs the operations described above.

The above-mentioned embodiments will be further described in detail below. Figure 2 is a display image,
The region of interest is specified by dividing the screen into 8 equal parts vertically and horizontally, dividing the entire screen into 64 sections, and selecting one or more sections containing the region of interest from among the sections. To actually designate a region of interest, after capturing the data of the entire image into image memory 1 through an imaging operation, the image data from image memory 1 is directly read out and displayed on the display CRT, and the region of interest is marked on the displayed image using a light pen. If you specify any point in the containing section, CP
U stores the pixel number of the pixel corresponding to the corner point of the upper left corner of the section containing that point as region of interest designation data. The image consists of 1024 x 1024 pixels, and the pixel numbers are expressed in hexadecimal as oooooo starting from the upper left corner, going to the right 1.2..., 003FF1, then returning to the left, going down one pixel to 00400, etc. , the bottom right corner is FFFFF
becomes.

Assume that the region of interest is, for example, in a section marked with diagonal lines. The pixel number in the upper left corner of this section is 801 FF. The device of this embodiment has a configuration in which 16 bits are operated as one word, the image data is 16 bits, and the image bus is 16 bits.
It is 6 bits. FIG. 3 shows the configuration of the contour extraction circuit 2 and the image memory 3 for contour data. The contour extraction operation is performed in real time during imaging, and is an operation that differentiates the video signal. The 16-bit image data sent by the image bus passes through path 2p in the figure and is delayed by several pixels in the delay circuit 21. , a subtraction circuit 22 performs subtraction with the image data directly coming from the path 2p, and a threshold circuit 24 binarizes the difference data.

During imaging operation, the switch Swl is in contact with the contact R side,
The 1-bit data output from the threshold circuit 24 is stored in the image memory 3. Image memory 3 is 16
The switch 5w2 (multiplexer) is operated by the clock of the imaging system to sequentially distribute contour data for each pixel to F1 to F16. In the addressing section 32 of the memory 3 in the circuit 4 (FIG. 1), the switch S w 3 is in contact with the contact a side, and the imaging system clock is divided by a frequency divider 321 of 1/16, and the imaging clock is 1 every 16 pulses
The pulse is input to the address counter 322, and the count output of the counter is sent to F1 through the 21-bit address bus.
~F16 as addressing data. During this contour extraction operation, image data output from the imaging system is stored in the image memory 1. In this way, 2fl of image data is transferred to the image memory 1.3 by the imaging operation.
is stored in

Next, the operation of transferring data stored in the image memories 1 and 3 to the recording medium 5 will be explained.

First, the contour data is transferred to the recording medium 5. At this time, in FIG. 3, switch S w 3 is switched to contact bm,
A recording system clock is input from the CPU to the address counter 322 and counted. This count output is image memory 3
Data at the same address is simultaneously read out from F1 to F16 as common addressing data and transferred to the recording medium 5 via the path 1p and the image bus [11] as 16-bit data. . In this way, 16 pieces of contour data become -word signals for every l pulse of the recording system clock, F1 to F1.
6 and recorded on the recording medium. When the recording of the contour data is completed, the image data of the specified region of interest in the image is selected from all the data in the image memo I) and transferred to the image bus B! is transferred to the recording medium through.

This operation will be explained in relation to the configuration of the addressing section 31 of the memory 1 in the address control circuit 4 shown in FIG. The configuration of the addressing section 31 is shown in FIG. In FIG. 4, during the imaging operation, the switches Sw4, 8w5 are switched to the contact side. The CPU selects pixel number 80 in the upper left corner of the region of interest shown with diagonal lines in FIG. 2 on the address counter 311.
1FF is preset, and this pixel number data is transferred to the recording medium 5 and recorded. Thereafter, the recording system clock is input to the 128 count/128 line counter 313. This counter counts 128 clocks and returns to 0, and outputs one pulse every 128th clock. This pulse is input to the counter 312, and the counter 312 increments by one from O every 128 recording system clocks. Other address counter 3
11 is a preset value of 801 every time the recording system clock comes.
FF is added by 1, and every 128 recording system clocks, the CPU increases the count of the address counter 311 to 801FF.
Return to The multiplication circuit 314 converts the count output of the counter 312 into a hexadecimal number 400 (
= decimal number 1024), and this multiplication result and the output of the address counter 311 are added.
This is output to the address bus BA through the switch contact to designate the address of the image memory 1. With this configuration, addressing data is 801
Starting from FF, we proceed one by one, which means specifying the addresses in memory 1 corresponding to the pixels on the upper side of the region of interest in Fig. 2, starting from the pixel in the upper left corner, so that 801FF+
When reaching 7E (7E=127), the count of the address counter returns to 801FF, but the count of the counter 312 increases by 1 and 400 is added to 801FF, so the pixel is specified on the next scanning line of the top edge of the region of interest. The pixels along the line are specified, and the count of the counter 312 becomes 127.
, all the addresses on the memory 1 corresponding to all the pixels of the region of interest shown in FIG. 2 have been designated, and the image data (1
5 bits) are transferred to the recording medium 5 through the image bus Bl.

Finally, the operation when reading data from the recording medium 5 and creating an image on the display device will be described. The outline of the operation is as follows: First, the contour data is transferred from the recording medium 5 to the image memories F1 to F16, and the contour data for each pixel is transferred from Fl to F16 (
0 or 1), add certain gradation data to it, and sequentially store it in the image memory 1. Next, while specifying the address of the image memory 1 in the same manner as described above with respect to FIG. The image data of the region of interest is read out and transferred to the image memory 1. Therefore, the contour data of the region of interest in the image memory is replaced with image data. Since the image data to be displayed in the image memory 1 has been edited in this way, it is sequentially read out and displayed on a display CRT (not shown).

When transferring the outline booter from the recording medium 5 to the memories F1 to F16, the switch S w 3 in FIG.
Addressing is performed in common for F1 to F16, and contour data is input from the recording medium 5 as 16-bit data to F1 to F16 via path ip (FIG. 3). When transferring the contour data of F1 to F16 to the image memory 1, the switch Sw1 in FIG. 3 is switched to the contact P side, the switch Sw3 is switched to the contact A side, and the switches Sw4 . Sw5 is respectively switched to the contact a side. When the imaging system clock comes in this way,
The pulses output from the frequency divider circuit 21 for every 16 taro clocks are counted by the address counter 322, and the pulses are counted by the address counter 322.
On the other hand, switch Sw2 changes the designation of F1 to F16 by 1 every time the imaging system clock pulses, and the contour data in F1 to F16 is read out. Since this data is 1 bit, it is sent to register 25 (Fig. 3). ) is added with constant gradation data (15 bits) set by the CPU, and 16
The bit data is transferred to the image memory 1 through the path 3p. On the other hand, addressing of the image memory 1 is performed by inputting the imaging system clock to the address counter 311 through the contact a of the switch Sw4, and inputting the counting output of this address counter to the image memory 1 through the contact a of the switch Sw5. In this way, the contour data is distributed sequentially over the entire screen.

When transferring the image data of the region of interest from the recording medium 5 to a predetermined address in the image memory 1, the switch S in FIG.
w4, w5 are respectively switched to the contact side, and the image memory 1 is transferred in the same manner as described above for transferring the data in the region of interest area of the image memory to the recording medium 5.
The data (16 bits) read from the recording medium 5 by specifying the address is stored at a predetermined address in the image memory 1.

Memory 1.3 is written separately, but specifically, it is a memory with a 16-bit address, and as shown in the memory map in Figure 5, it is written in hexadecimal from address ooooooo to FF.
The area up to FFF is image memory l, and the next address 1
00000 to 1OFFFF corresponds to memory 3,
In the part corresponding to memory 3, the first bit of each address is Fl,
The second bit corresponds to F2...the 16th bit corresponds to Fl6. With such a memory configuration, 21 bits are required for address specification from O to 1000000 (hexadecimal), so the address bus BA is 21 bits, and when contour data is transferred to/from F1 to F16, the fifth Addresses F1 to F16 are specified using address 110000 of the memory shown in the figure as the first address.

When the region of interest on the image spans multiple sections in FIG. 2, the above-described operation is repeated for each section. Note that in the above embodiment, contour data is used for parts other than the region of interest, but this is just a means of binarizing the image.
In the present invention, the method of binarizing the portion outside the region of interest is arbitrary, and is not limited to the method of binarizing, that is, converting into 1-bit data, for example, the method of converting into 2-bit data into 4 levels.

Even if the background of the region of interest is expressed in terms of tone, the effect of quantitative compression of image data can be greatly obtained. In addition, in the above embodiment, the region of interest is specified by dividing the screen vertically and horizontally in advance.
Although we designate any area, the designation of the region of interest is not limited to this method. For example, on the CR7 screen, draw a line surrounding the area of interest with a light pen (and the CPU draws a line on the line) The pixel numbers are memorized, and starting from the pixel number with the smallest number, the address corresponding to the pixel between the two pixel numbers before and after corresponding to the above closed curve on one horizontal scanning line is specified on the image memory 1. You may also do so.

Effects The digital image data recording and reproducing method of the present invention uses all image data for each pixel only in the region of interest within the entire screen, and simplifies the image representation outside the region of interest by using only the outline to reduce the number of bits of image data. As a result, the necessary parts of the image can be displayed in detail, and the positional relationship of that part in the whole image can be displayed.In actual image observation, the whole image can be displayed with the same level of detail. The same value can be obtained as shown in Figure 2, and the unnecessary parts are simplified to the extent that they represent the overall relationship, so it is easier to see than displaying the whole in detail. It is significantly compressed, and there is no deterioration in playback image quality that would occur if a data compression method was uniformly applied to the entire screen.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of a device configuration according to an embodiment of the present invention, FIG. 2 is a diagram showing how the screen is divided, and FIG. 3 is a diagram showing a contour extraction circuit and an image memory for contour data; FIG. 4 is a block diagram showing an address control circuit, FIG. 4 is a block diagram of an addressing circuit for an image memory for all image data, and FIG. 5 is a memory map of each image memory of the above embodiment. ■...Image memory for all image data, 2...Contour extraction circuit, 3...Image memory for storing contour data, 4...Address control circuit, 5...Recording medium 6.
...■10 controllers. Agent: Patent Attorney Kosuke AgataFigure 4

Claims (1)

[Claims] A region of interest is set on the image, image information for other parts is converted to data with a decimal number of bits, image data with a high number of bits is used only for the region of interest, these data are recorded on a recording medium, and this recording A digital image recording and reproducing method in which the entire screen is expressed using small-bit data, and only the area of interest is displayed in detail using this expression as a background.