JPH04329764A - Compression data transfer device - Google Patents

Abstract

PURPOSE:To output all data to a processing section even when a compressed data is processed in the unit of blocks. CONSTITUTION:When data transfer is finished and a data quantity of a memory 21 is a prescribed value Y(512 bytes) or below, a data is transferred by the prescribed value Y from a host computer 1 excessively or a data is outputted from a system control section 25. Thus, a difference between both counts of a write counter 22 and a read counter 24 is 512 or below and all compressed data are transferred to an expansion processing section 5. Moreover, a data is outputted by enabling a data output available signal forcibly.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a compressed data transfer apparatus, and more particularly to a compressed data transfer apparatus for decompressing encoded and compressed image data in the medical field.

0002

2. Description of the Related Art In recent years, for example, in the medical field, there has been an active movement to digitize and handle medical images including X-ray images. This digitized X-ray image has a large amount of data, and in order to store, search, or transmit this X-ray image, it is necessary to efficiently reduce the amount of data and prevent the deterioration of the X-ray image. Image compression technology that reduces the number of images is becoming extremely important. As such a compression technique, for example, DPCM (dif
ferential pulsecode module
ation) method, and DCT (di
(screte cosine transform)
, etc. In such a DPCM method, the image does not deteriorate during compression/expansion, but the compression ratio is, for example, approximately 1.
/2. Further, in the DCT method, although the image data is irreversible, the compression rate is high, for example, about 1/5 to 1/50.

When image data is compressed using the DPCM method, for example, the difference between two adjacent pixels is determined sequentially from the pixel value at the start point of a zigzag scan to the pixel value at the end point, and code data is assigned to the difference value. Although there are many types of code data for this compression/expansion, relatively good results can be obtained in terms of efficiency by using variable length code data. Examples of this variable length code include Huffman code, run length (
(hereinafter referred to as RL) are well known. Compressed data is obtained by data packing this variable length code data together with the pixel value at the starting point. Furthermore, when decompressing compressed data into image data, the process inverse to the above process may be performed.

[0004] When compressing image data using the DCT method, the image data is converted into the frequency domain by discrete cosine transformation, and after quantization processing is performed, the above-mentioned DCT method is used.
Like the PCM method, compressed data is obtained by encoding into variable length code data and packing this encoded data. In addition, when expanding this compressed data into image data, the reverse process to the above process may be performed.

In such a system that performs image data compression/decompression processing, when compressing image data, the host computer inputs and outputs the image data through an interface while temporarily storing the image data in the compression/decompression processing section. output to the image transfer unit. The compression/decompression processing section compresses the image data input to the image transfer section using a reversible compression method or an irreversible compression method, and then inputs and outputs the compressed data while temporarily storing it through a compressed data transfer device and an interface. Transfer to host computer. When compressed data is expanded to original image data, the compressed data is output from the host computer to the compressed data transfer device via the interface. The compression/decompression processing unit decompresses this compressed data and outputs the decompressed original image data to the host computer via the image transfer unit and the interface.

As mentioned above, the reversible compression method provides a completely restored image but has a low compression ratio, while the irreversible compression method has a high compression ratio but cannot obtain a completely restored image after decompression. Both functions are often required, especially in medical applications. To do this, it is necessary for the compression/expansion processing section to perform compression/expansion processing using both reversible and irreversible compression methods so that both methods can be selected. It becomes necessary to exchange image data in block units and synchronize them with a clock.

[0007]

By the way, in the case of the above-mentioned compressed data transfer device, a memory having a FIFO structure is provided, but if the amount of data read from the memory exceeds the amount of data written, Even if the data output enable signal is disabled in the middle of reading a block, the compression/decompression processing unit does not stop processing in the middle of the block, and the read is performed in an empty state, and invalid data that is not compressed data is output. It will be done. Therefore, it is conceivable to set a predetermined value Y that is greater than the read unit, and when the stored data becomes less than the predetermined value Y, disable the data output enable signal so that the data cannot be read.

However, with this configuration, when the transfer of compressed data from the host computer is completed, the amount of data in the memory finally falls below the predetermined value Y, and processing stops, causing all data in the memory to be lost. There is a possibility that it will not be possible to read out the data. The present invention has been made in view of such conventional problems, and an object of the present invention is to provide a compressed data transfer device that can output all data even when compressed data is processed block by block.

[0009]

[Means for Solving the Problems] Therefore, the present invention has the following features:
As shown in the figure, a series of compressed data output from a host computer is input, and based on the read signal of the decompression processing unit that decompresses the compressed data, the series of compressed data is sent to the decompression processing unit in the order in which it was input as unit data. A compressed data transfer device that outputs compressed data in units of amounts, storage means for inputting and temporarily storing compressed data from the host computer, write counting means for counting the total number of writes of compressed data from the host computer to the storage means; a read count means for counting the total number of reads of data stored in the storage means to the decompression processing unit; and a read count means for counting the total number of reads of data stored in the storage means; The amount of data is monitored, and when the amount of data exceeds a predetermined value set to a value greater than or equal to the data read unit, the data output enable signal to be output to the storage means is enabled, and the data in the decompression processing section is control signal output means for enabling readout and disabling a data output enable signal to prohibit data readout when the data output enable signal is below a predetermined value; A read end control means for outputting all data to the decompression processing section is provided.

[0010] Furthermore, the read end control means transfers extra data exceeding the predetermined value from the host computer in addition to the series of compressed data at the end of data transfer, so that all the series of compressed data are output as data. The data may be configured to be read out to the decompression processing unit in the enabled state of the enable signal. Alternatively, the read end control means may be configured to transfer extra data equal to or greater than the predetermined value from the compressed data transfer device in addition to the series of compressed data when the data transfer ends.

Alternatively, the read end control means may be configured to enable a data output enable signal output from the control signal output means after the transfer of all series of compressed data is completed.

0012

According to the above structure, when a series of compressed data is input to the storage means from the host computer, the total number of written compressed data is counted. Further, when the stored data is read out in blocks by the decompression processing section, the number of reads is counted. The control signal output means monitors the amount of data stored in the storage means based on the count values of both the write counter and the read counter, and when the data amount becomes less than a predetermined value, outputs the data output enable signal. It is disabled and data reading is prohibited. When the transfer of a series of compressed data from the host computer is completed, the predetermined value at which the data output enable signal is disabled is set to be greater than the data read unit, so the amount of data in the storage means is less than the predetermined value. Although some data may remain, all data stored in the storage means is output to the decompression processing section by the read end control means.

[0013] Furthermore, in a device in which extra data exceeding a predetermined value is transferred from the host computer, when the difference between the count values of the write count means and the read count means becomes less than the predetermined value in the last read, The remaining data in the storage means is extra data transferred from the host computer, which means that all compressed data has been transferred to the decompression processing section.

In addition, if the remaining data below a predetermined value is transferred from the control signal output means to the data above the predetermined value, all data is similarly output to the decompression processing section. In addition, in the case where the data output enable signal output from the control signal output means is forcibly enabled,
Similarly, all data is output to the decompression processing section.

[0015]

Embodiment An embodiment of the present invention will be described below with reference to FIGS. 2 to 8. In FIG. 2 showing a system for compressing/expanding image data, when compressing image data, the host computer 1 outputs a series of image data to the image transfer unit 3 via the interface 2, and converts the compressed data into the original image. When decompressing data, a series of compressed data is output from the host computer 1 to the compressed data transfer device 4 via the interface 2. This interface 2 employs, for example, a SCSI interface that allows commands to be created freely. The image data compression/expansion device is
The above-mentioned image transfer unit 3 that inputs and outputs image data while temporarily storing it, and the above-mentioned compressed data transfer device 4 that inputs and outputs code data obtained by compressing the image data while temporarily storing it. It is composed of a compression processing section and an expansion processing section 5 that perform compression/expansion processing of image data, and a system control section 6 that controls these operations. Here, the compression processing unit and expansion processing unit 5 include an encoding unit 7 that encodes image data during compression processing, a decoding unit 8 that decodes coded data into original data during coded data expansion processing, and and a packing/depacking circuit 9 for packing and depacking the packed data.

In FIG. 3 showing the compressed data transfer device 4 of this embodiment, the memory 21 serving as a storage means has, for example, 64K memory.
A memory buffer with a memory capacity of bytes. A write counter 22, which is a write counting means, stores compressed data output from the host computer 1 in the memory 2.
For example, it is a 16-bit counter that counts write pulses to 1. Further, the read counter 23, which is read counting means, is, for example, a 16-bit counter that counts read pulses when reading compressed data. The memory 21, write counter 22, and read counter 23 constitute a FIFO, which stores compressed data sent from the host computer 1 and sequentially outputs the compressed data in the order in which they are sent. When compressed data is expanded, the data stored in the memory 21 is output based on read signals from the compression processing section and the expansion processing section 5. The compression processing section and the expansion processing section 5 process the amount of data in units of blocks, such that when converted to image data, it becomes one block of 16 x 16 x 10 bits. Therefore, for example, when compressed data is expanded, the compressed data stored in the memory 21 is read out continuously until one block is expanded. The above-mentioned interface 2 is, for example, a SCSI interface with a bit size of 1 byte. The host computer 1 outputs, for example, compressed image data to the memory 21 via the interface 2. The data output enable signal control unit 24, which is a control signal output means, inputs the count value of the write counter 22 and the count value of the read counter 23, and sets the predetermined value Y to, for example, 512 bytes as shown in FIG.
When the amount of data within is less than a predetermined value Y, the data output enable signal is disabled, and when it exceeds 512 bytes, it is enabled and output is controlled. The aforementioned system control unit 6
is the interface 2 in the compressed data transfer device 4.
, data output enable signal control section 24, write counter 2
2. Controls the operation of the read counter 23.

Next, the operation of the data output enable signal control section 24 will be explained based on the flowchart of FIG. In step 1 (denoted as "S" in the figure, the same applies hereinafter), when compressed data is written from the host computer 1 to the memory 21 through the interface 2,
This routine starts, the number of write pulses is counted by the write counter 22, and this count value is sent to the data output enable signal control section 24.

In step 2, when the compressed data stored in the memory 21 is read from the decompression processing unit 5, the read counter 23 counts the number of read pulses;
This count value is sent to the data output enable signal control section 24. Note that the count value of the read counter 23 remains 0 when writing to the memory 21 is started. In step 3,
The difference between the count values of the write counter 22 and the read counter 23 is calculated, and the upper limit of the difference is set to, for example, (64
k-512) When the number of bytes exceeds the upper limit value, the process proceeds to step 4, sets the (FULL-Y) flag and outputs it to the system control unit 6; if it is less than the upper limit value, the process proceeds to step 5. The system control unit 6 is this (FULL-Y)
When the flag is set, a command to interrupt the data transfer is output to the host computer 1, and the memory 2
Prevent the data in 1 from being overwritten.

In step 5, it is checked whether the amount of data stored in the memory 21 exceeds a predetermined value Y. This predetermined value Y is expressed by . When the amount of compressed data in the memory becomes less than or equal to a predetermined value Y, the data output enable signal control section 24 disables the data output enable signal. However, since the decompression processing unit 5 including the packing/depacking circuit 9 does not stop operating until one block is completely processed, compressed data is read even if the data output signal is disabled during processing of one block. I end up. Therefore, if the predetermined value Y is 0, no data remains in the memory, so if the data output enable signal is disabled in the middle of a block, incorrect data will be read. Therefore, if the data output enable signal is controlled so that the amount of data always exists in the memory by a predetermined value Y, invalid data will not be read out. This predetermined amount Y has a processing unit of n×
When processing is performed in units of n blocks, for example, the maximum number of bits per pixel of the image to be compressed/expanded is b bits (n x n x b) or more, and the processing unit is L lines (L is 1 (above), the maximum number of bits per pixel of the image to be compressed/expanded is b bits,
If the number of columns per line is m, the predetermined amount Y is
It is set to be more than (L×m×b) bits, and is set to about 512 bytes, for example. When the amount of data exceeds the predetermined value Y, the process proceeds to step 6, where the data output enable signal control unit 24 enables the data output enable signal and returns to step 1, and when it is less than the predetermined value Y, the process proceeds to step 7. .

In step 7, since the amount of data stored in the memory 21 has become less than the predetermined value Y, the data output enable signal control section 24 generates a data output enable signal so that data cannot be read from the memory 21. Disable and return. Next, the read end control routine of the first embodiment will be explained based on the flowchart of FIG. In the first embodiment, the read end control routine is executed by the host computer 1, and causes the host computer 1 to output extra data and output all the compressed data.

In step 21, it is determined whether a series of compressed data transfers have been completed. This predetermined value Y is, for example, 512 bytes for one block. Then, the process waits until the data transfer is completed, and when the data transfer is completed, the process advances to step 22. In step 22, since the transfer of a series of compressed data has been completed, extra data by a predetermined value Y (for example, 512 bytes for one block) is transferred. As a result, when the difference between the count values of the write counter 22 and the read counter 23 becomes less than the predetermined value Y in the last read, the data remaining in the memory 21 is the extra data transferred from the host computer 1. Yes, all compressed data has been transferred to the decompression processing section 5. Then, this step is executed to end the execution of this routine.

Note that this read end control routine corresponds to read end control means. According to this configuration, when the data transfer from the host computer 1 is completed, the extra data is transferred from the host computer 1 to the memory 21 by a predetermined value Y for one block, which is the data read unit, so that the last data transfer is completed. When the difference between the count values of the write counter 22 and the read counter 23 during reading becomes equal to or less than a predetermined value, it means that all the compressed data has been transferred to the decompression processing section 5. Furthermore, all compressed data can be output with such a simple circuit configuration.

Next, a second embodiment will be explained. In the second embodiment, the system control unit 6 executes a read end routine, outputs extra data from the system control unit 6, and reads all compressed data. The read end routine of the second embodiment will be explained based on the flowchart of FIG.

In step 31, the host computer 1
Detects whether the transfer of a series of compressed data from has finished. To detect the end of compressed data transfer, when the host computer 1 transfers the last compressed data,
Either the final flag is set and the system control unit 6 receives the final flag, or the host computer 1 sends a transfer end command after completing the transfer of all compressed data, and the system control unit 6 receives this transfer end command. Alternatively, the total compressed data amount is notified from the host computer 1 in advance, and the system control unit 6 checks the total write count amount. If the transfer of the compressed data has not yet been completed, the process proceeds to step 1, and if the end of the data transfer is detected, the process proceeds to step 32.

In step 32, a predetermined value Y (for example, 512 bytes for one block) of extra data is written from the system control unit 6 to the memory 21. As a result, the write counter 22 and read counter 23 are
When the difference between the two count values becomes less than a predetermined value Y, the data remaining in the memory 21 is transferred to the host computer 1.
This means that all the compressed data has been transferred to the decompression processing section 5. Then, this step is executed to end the execution of this routine. still,
Even if the system control unit 6 increments the count value of the write counter 22 by an extra 512 instead of writing 512 bytes of data, it will function in the same way and all the compressed data can be transferred to the decompression processing unit 5. becomes.

According to this configuration, the data remaining in the memory 21 can be outputted by the system control section 6 without using the host computer 1. Next, a third embodiment will be described. In the third embodiment, the system control unit 6 executes a read end routine and forcibly outputs an enable signal.

The read end routine of the third embodiment will be explained based on the flowchart of FIG. In step 41, it is detected whether the transfer of compressed data from the host computer 1 has been completed. The detection method is the same as step 31. In step 42, the system control section 6 outputs a mask signal to the data output enable signal control section 24, and ends this routine.

When this mask signal is input to the data output enable signal control section 24, the data output enable signal is forcibly enabled, and all compressed data in the memory 21 can be output to the decompression processing section 5. Become. When the difference between the count values of the write counter 22 and the read counter 23 becomes 0, all the compressed data is transferred to the decompression processing unit 5.
Therefore, the data output enable signal control unit 24 disables the data output enable signal. still,
It is also possible for the decompression processing section 5 to count the image size when the mask signal is output, and to stop reading when the processing of one image is completed.

In the first to third embodiments, the transfer with the host computer 1 was controlled by the (FULL-Y) flag, but since this transfer is not in block units, even if it is controlled by the EMPTY and FULL flags. good. At this time, the flags sent to the compression processing unit and decompression processing unit 5 and the flags sent to the interface 2 will have different flags.

[0030]

As described above, according to the present invention, even when reading data from the decompression processing unit is processed in read units, the read end control means All data can be output to the decompression processing section.

Further, after the transfer of a series of compressed data from the host computer is completed, by transferring extra data from the host computer, both the count values of the write means and the read count means become less than a predetermined value, and all the compressed data are can be read. Further, after the transfer of a series of compressed data from the host computer is completed, all the compressed data can be read by outputting extra data from the control signal output means.

Further, after the transfer of a series of compressed data from the host computer is completed, all the compressed data can be read by forcibly enabling the data output enable signal.

[Brief explanation of drawings]

[Figure 1] Claim correspondence diagram of the present invention

[Figure 2] System configuration diagram of the present invention

FIG. 3 is a block circuit diagram showing the configuration of an embodiment of the present invention.

[
Figure 4 is an explanatory diagram of the memory structure of the memory in Figure 3.

[Figure 5] Figure 3
Flowchart showing the first embodiment of

FIG. 6 is a flowchart showing the read end control routine in FIG. 5;

FIG. 7 is a flowchart showing the second embodiment of FIG. 3.

[Figure 8
] Flowchart showing the third embodiment of FIG. 3

[Explanation of symbols]

1 Host computer 2 Interface 4 Compressed data transfer device 5 Decompression processing section 6 System control section 21 Memory 22 Write counter 23 Read counter

Claims (4)

[Claims] Claims: 1. A series of compressed data outputted from a host computer is inputted, and based on a read signal of an expansion processing unit that decompresses the compressed data, a series of compressed data is sent to the expansion processing unit in the order of input as unit data. A compressed data transfer device that outputs compressed data one by one, a storage means for inputting and temporarily storing compressed data from the host computer;
write counting means for counting the total number of writes of compressed data from the host computer to the storage means;
read counting means for counting the total number of reads of data stored in the storage means to the decompression processing unit;
The amount of data stored in the storage means is monitored based on the count values of both the write counter and the read counter, and the data amount is set to a predetermined value equal to or larger than the data read unit. When the data exceeds the limit, enable a data output enable signal to be outputted to the storage means to enable reading of data from the decompression processing section;
control signal output means for disabling the data output enable signal and prohibiting data reading when the value is below a predetermined value;
1. A compressed data transfer device comprising: readout end control means for outputting all data stored in the storage means to a decompression processing section after the transfer of a series of compressed data from a host computer is completed. 2. The read end control means transfers extra data of a predetermined value or more from the host computer in addition to the series of compressed data at the end of data transfer, so that all the series of compressed data are 2. The compressed data transfer device according to claim 1, wherein the compressed data transfer device is configured to be read out by the decompression processing unit in an enabled state of the output enable signal. 3. The read end control means is configured to transfer extra data exceeding the predetermined value from the compressed data transfer device in addition to the series of compressed data when the data transfer ends. The compressed data transfer device according to item 1. 4. The read end control means is configured to enable a data output enable signal output from the control signal output means after the transfer of all series of compressed data is completed. The compressed data transfer device according to item 1.