JPH09219780A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely prevent occurrence of an underflow by outputting an image output control signal to instruct a skip operation when the image data transfer speed is lower than the expansion speed that is set in an image output mode. SOLUTION: An image processing control part 206A detects the rise of a page synchronizing signal and then checks whether the encoding data on the page to be outputted next are transferred to a code buffer 204 based on the write information stored in a memory. If the transfer of the encoding data is completed, the expansion processing of the page is started. If the encoding data on the page to be outputted next are not transferred to the buffer 204 yet, an underflow of the buffer 204 may possibly occur due to a fact that the data transfer speed is lower than the page expansion speed. Thereby, the expansion processing is not carried out in that pitch and an image output control signal is outputted to an image output part to instruct a skip operation, so that the underflow can be evaded.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus, and more particularly, to reading a document of a plurality of pages, compressing and storing the image data, and then reading the image data in a specified order a specified number of times. The present invention relates to an image processing apparatus having an electronic sort function for performing expansion processing and outputting an image.

[0002]

2. Description of the Related Art Conventionally, an image processing apparatus such as a digital copying machine has an electronic sort function as one function. Such a sorting function reads a plurality of originals by an image input device such as a scanner and temporarily stores them in a storage device, reads them a specified number of times in a specified order, and forms an image according to the read image data. It is realized by doing. For example, page A,
When copying three copies of a document consisting of B and C, page A,
After reading one page at a time in the order of B and C and storing the image data, the image data is read on pages A, B, C, A,
By reading out in the order of B, C, A, B, and C, and forming an image each time, three copies of the original document are obtained in the same page order as the original document.

In the image processing apparatus having such a sorting function, when image data is input, it is compressed and stored when it is input, while when it is output, the read data is decompressed and stored. The image is returned to the image data, and image output such as image formation is performed based on the image data. The reason is that, firstly, it is necessary to store a large amount of image data in a storage device having a limited capacity, and secondly,
A hard disk (drive) having a low cost per capacity is suitable as a storage device, but since the transfer rate for such a hard disk has an upper limit, it is necessary to keep the capacity of image data to be transferred small.

As an algorithm used for such compression / expansion processing, for example, JPEG (Joint Photographic Experts Gro) combining orthogonal transformation and variable length coding is used.
There is a method for still images recommended in up). J
Since the PEG method is a variable length compression method, the generated code amount differs depending on the characteristics of the input image data. Also, since it is irreversible, if the compression rate is high, the generated code amount can be reduced, but the image deterioration when decompressing and restoring is conspicuous, and conversely, if the compression rate is low, decompression is performed. Although it is possible to suppress the image deterioration in the case of returning the image, the generated code amount increases.

On the other hand, when image input is performed for each page at a constant pitch (time interval), when the generated code amount of a page is larger than the data amount that can be transferred to the hard disk within that pitch, It becomes impossible to store the encoded data of consecutive pages in the hard disk. In such a case, there is a problem that data is stored in a state of being lost without being synchronized with the image input device. Similarly, when image output is performed at a constant pitch for each page, when the code amount of a certain page is larger than the data amount that can be transferred from the hard disk within that pitch, the encoded data of consecutive pages is recorded on the hard disk. Cannot be read from. In such a case, there is a problem that a disordered image is output without being synchronized with the image output device. In addition, it is generally a flatbed scanner or page printer that performs image input and image output at a fixed pitch, and these operations are controlled independently of the data transfer state,
This is to simplify the configuration.

As a solution to such a problem,
For example, there is a technique described in JP-A-7-107300. The technique described in this publication monitors the code amount generated for each page, and when the code amount exceeds a preset threshold value, the compression condition is changed to further increase the compression rate, The same image is read again and compressed. As a result, finally, it becomes possible to compress the image of any page to a code amount equal to or less than the threshold value and transfer the coded data of any page to the hard disk within one pitch.

The technique described in the above publication will now be described in detail. FIG. 6 is a block diagram showing a configuration of an image processing apparatus having an electronic sort function using the technique described in the above publication. In this figure, 201 is a pre-stage image processing unit that performs pre-processing of image processing on image data input from an image input device (not shown in this figure), and 202 is an image to an image output device (not shown). This is a post-stage image processing unit that performs post-processing of image processing when outputting data. 203 is
When inputting image data, the image data is decompressed to be encoded data, while when the image data is output, the encoded data is decompressed to be the original image data. It is a department. Reference numeral 204 is a code buffer for buffering encoded data. Also,
Reference numeral 205 denotes a main storage unit that stores compressed encoded data, and a hard disk is used here for the above reason. On the other hand, 206 is an image processing control unit that controls the entire image processing apparatus, and 207 is an image compression / decompression unit 2.
The code amount detecting unit detects the code amount of the image data compressed by 03 for each page.

Next, the operation of this image processing apparatus will be described. First, the operation at the time of image input will be described. First, in the image input device, the document is read page by page at a constant pitch, and image data is generated each time.
That is, within one pitch, image data corresponding to a document for one page is generated. Now, focusing on the period of one pitch, the image data for one page is subjected to pre-processing by the pre-stage image processing unit 201, and then the image compression / expansion unit 20.
Compressed by 3. Next, the compressed encoded data is buffered in the code buffer 204, and the code amount thereof is detected by the code amount detecting unit 207. Here, the image processing control unit 206 keeps the code amount detection unit 20 until the compression processing of the image data for one page is completed.
The magnitude of the code amount detected in 7 and the magnitude of the target code amount which is a preset threshold value are continuously compared. Here, the target code amount is determined based on the capacity that can be transferred to the hard disk within one pitch.

When the detected code amount does not exceed the target code amount at the time when the process of compressing the image data for one page is completed, the image processing control unit 206 considers that the process is completed normally, The hard disk 2 reads out the encoded data written up to that point to the buffer 204.
The process of transferring to 05 and storing is started. In parallel with this, the image processing control unit 206 commands the image input device to read the document corresponding to the next page.

On the other hand, if the detected code amount exceeds the target code amount by the end of the compression process for one page, the image processing control unit 206 regards it as an abnormal process and performs the next abnormal process. To do. First, the image processing control unit 206 determines, when the detected code amount exceeds the target code amount, the code buffer 20.
4 is stopped, and the write address to the code buffer 204 is returned to the top address of the page. As a result, the encoded data of the page buffered in the code buffer 204 is discarded.
However, the image processing control unit 206 causes the image input device and the image compression / decompression unit 203 to continue reading and compressing the page, respectively. The reason why these processes are continued is to determine the difference between the generated code amount and the target code amount of the page at the time when the compression process of the page ends, for the following reason, and these processes are stopped. This is because the difference cannot be obtained. Next, the image processing control unit 206 determines the compression rate at which the generated code amount of the page becomes equal to or less than the target code amount from the difference between the generated code amount and the target code amount at the time when the compression process of the page ends. The image compression / expansion unit 203. Then, the image processing control unit 206 instructs the image input apparatus to read the original corresponding to the same page again at the next pitch. Due to this abnormal process, the original on the same page is read again at the next pitch, but since the compression rate is changed and set, the code amount generated by the re-reading is less than the target code amount. , Ends normally. The above processing is performed for each page of the original document at each pitch. As a result, the code amount generated within one pitch is always less than or equal to the target code amount, and the problem of data loss during transfer to the hard disk 205 can be avoided.

FIG. 7 shows a specific operation in this image input. As shown in this figure, since the code amount of the originals 2, 4, and 5 exceeds the target code amount,
Re-reading (rescan) is occurring in each case. Since the transfer to the hard disk 205 is started when the reading of the original is finished (corresponding to the end of the compression), the usage amount of the code buffer 204 is increased from the start of reading the original and then the end of reading the original is read. However, when the next document reading is started, the coded data of the next document is generated and accumulated.

When the encoded data for the originals of all pages are stored in the hard disk 205, the output operation is started this time. First, at the first pitch, the encoded data corresponding to the first page is read from the hard disk 205 and transferred to the code buffer 204. Next, at the second pitch, the coded data corresponding to the second page is read out and transferred to the code buffer 204, and the already transferred coded data for the first page is decompressed to obtain the post-stage image. The image is output to the image output device through post-processing by the processing unit 202. As a result, the image output for the first page is obtained at the second pitch. Then, the electronic sort function is performed by repeating such processing in a designated page order a designated number of times.

As described above, in this image processing apparatus, the code amount for one page generated within one pitch is 1
Since the compressed amount is equal to or less than the target code amount set on the basis of the transferable amount to the hard disk within the pitch, one page of encoded data is transferred from the hard disk 205 to the code buffer 204 at the time of image output. , Usually ends within one pitch. Therefore, when the coded data of the next page is transferred to the code buffer 204 at the next pitch, all the coded data of the immediately preceding page of the next page should be stored in the code buffer 204. Therefore, even if the image output of the immediately preceding page is performed at the pitch at which the encoded data of the next page is transferred, the image data will not be lost.

[0014]

However, this image processing apparatus has the following problems. That is, when the generated code amount exceeds the target code amount, the same document is read twice, so that it takes time to input the image. Therefore, from the viewpoint of shortening the time required for image input, the target code amount is increased so that the generated code amount does not exceed the target code amount, and it is desirable that the target code amount is within the upper limit of one pitch. If the transferable capacity of the hard disk 205 is set, the probability of occurrence of rescan is minimized, so that it should be possible to efficiently input an image. However, at the time of outputting an image, when the already-stored encoded data is read, some trouble may occur in the hard disk 205, for example, a seek error. In this case, since the coded data is read out in a longer time than usual, even if the generated code amount at the time of image input is less than the target code amount, all the coded data of the page within one pitch are code buffered. A situation may occur in which the data is not stored in 204. When this happens,
Since the decompression process is started in the state where all of the encoded data is not stored, underflow of the code buffer 204 may occur during the decompression process of the page as in the original 6 shown in FIG. . When this underflow occurs, the decompressed image data will be missing or disturbed.

The present invention has been made under such circumstances, and the buffer storage means is provided even when the transfer speed of the image data from the storage means (hard disk) cannot meet the expansion speed at the time of image output. An object of the present invention is to provide an image processing device capable of reliably preventing underflow in a (code buffer) and avoiding output of a missing image or a disturbed image.

[0016]

In order to solve the above-mentioned problems, the invention according to claim 1 stores a coded data obtained by variable-length-compressing original image data, and a read out from the memory means. Decompressing means for decompressing encoded data to be encoded, image output means for sequentially forming an image in page units based on the image data decompressed by the decompressing means,
Buffer storage means for temporarily storing the coded data read so as to absorb the speed difference between the read speed from the storage means and the expansion speed of the image output means, and the page to be output next by the image output means. And output control means for starting the output of the page on condition that all the encoded data of (1) are transferred from the storage means to the buffer storage means.

Further, the invention described in claim 2 is the first invention.
In the invention described in the above, the image output means continuously outputs the image data in page units at a predetermined pitch, and the output control means is such that the encoded data has a pitch section corresponding to the page. If not all have been transferred before the start, the pitch section for outputting the page is shifted to the next section.

The invention described in claim 3 is the same as that of claim 2.
In the invention described in (1), the image output unit includes a printing unit that forms an image on a sheet, and a paper feeding unit that feeds the sheet to the printing unit, and the output control unit controls the operation of the paper feeding unit. It is characterized by avoiding output by stopping.

[0019]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the drawings. A: Overall Configuration of Embodiment FIG. 1 is a front sectional view showing a configuration of a digital copying machine which is an embodiment of an image processing apparatus according to the present invention. In this figure, reference numeral 10 is an image input unit, reference numeral 20 is an image processing unit, and reference numeral 30 is an image output unit.

First, in the image input section 10, the document tray 1
The original 102 placed on 01 is ADF (Auto Documen
(t Feeder: automatic document feeder) 103 is fed one by one, and when the platen roller 104 is fed, it passes between the platen roller 104 and the reading glass 105,
The sheet is discharged to a sheet discharge tray (not shown).
It should be noted that this reading operation is executed at a fixed time interval (pitch) for each page, and the operation of each unit in the image input unit 10 is performed by the image input control unit 106.
Is controlled by the Light source unit 10
The irradiation lamp 108 and the reflection plate 109 are integrally formed with each other, and can be moved in the left-right direction in the drawing by a driving unit (not shown). On the other hand, the lens 110 and the CCD (line) sensor 111 are connected to the image input unit 10
Is fixed against

When the originals are continuously read one by one using the ADF 103, the light source unit 107 is fixed at the position shown in the figure, and the irradiation lamp 108 is fixed to the platen roller 10.
The document sent between the scanning glass 105 and the reading glass 105 is irradiated, and the reflected light at that time is reflected by the reflection plate 108.
Then, the reflected light is further imaged by the lens 110, read by the CCD sensor 111, and supplied to the image processing unit 20 as image data. That is, the original 102 is sub-scanned by feeding the platen roller 104,
By performing main scanning in the vertical direction in the figure by the CCD sensor 111, the originals 102 are read one by one and the image data thereof is output. When the ADF 103 is not used in the image input unit 10, the document 102 is placed on the reading glass 105 by the user, while the light source unit 107 moves left and right under the reading glass 105 in the figure, and the document concerned. By sub-scanning the document, the document can be read.

The image processing section 20, which will be described in detail later, compresses the image data supplied from the image input section 10 at the time of input and stores it as coded data for each page, while at the time of output it is designated. The coded data of the page is read a specified number of times in a specified order, decompressed each time to restore the original image data, and supplied to the image output unit 30. The image processing unit 20 also supplies an image input control signal to the image input unit 10 to control its input operation, while supplying an image output control signal to the image output unit 30 to control its output operation. .

Next, in the image output section 30, the ROS (Rast
er Output Scanner) 301 drives the internal laser diode according to the image data supplied from the image processing unit 20. Then, the light emitted by the laser diode enters the photosensitive drum 303 by the reflecting plate 302. As a result, an electrostatic latent image is formed on the photosensitive drum 303, and toner is attached to the electrostatic latent image by the developing device 304 to develop the electrostatic latent image. On the other hand, the sheet conveyed from the sheet tray 305 is conveyed to the lower surface of the photosensitive drum 303 via a predetermined conveying path, and the toner attached to the photosensitive drum 303 is transferred when passing the sheet. The transferred sheet passes through the fixing unit 306 and is discharged to the copy tray 308. Note that the toner attached to the photosensitive drum 303 is not cleaned by the cleaner 3 until the next transfer to the sheet.
Removed by 10. The operation of each unit of the image output unit 30 is controlled by the image output control unit 311. Further, the image output control unit 311 can carry out the skip operation without stopping the operation of the xerographic engine by controlling the conveyance and the stop of the sheet for each pitch.

B: Structure of Image Processing Unit 20 Next, the image processing unit 20 will be described. FIG. 2 is a block diagram showing a detailed configuration of the image processing unit 20. In this figure, reference numeral 201 denotes a pre-stage image processing unit, which performs predetermined image processing such as MTF correction on the image data for each page supplied from the CCD 111 of the image input unit 10. On the other hand, reference numeral 202 denotes a post-stage image processing unit that performs predetermined image processing such as filter processing on the decompressed image data and outputs the decompressed image data to the ROS 301 of the image output unit 30. Reference numeral 203 is an image compression / decompression unit, which is
The image input unit 1 supplied via the pre-stage image processing unit 201
While the image data of 0 is compressed by the JPEG method and output as encoded data, the encoded data is decompressed by the same method and restored to the original image data at the time of output.

In the image compression / expansion unit 203, although not shown in the figure, two line buffers having a buffer capacity of 8 lines are inserted in parallel in the ports of the front-stage image processing unit 201 side or the rear-stage image processing unit 202 side. It has a so-called ping-pong buffer, which is
The data can be continuously compressed / decompressed and supplied. That is, the image compression / decompression unit 20
In 3, when inputting, the image data of the image input unit 10 via the pre-stage image processing unit 201 is stored in one line buffer, while the image data is read from the other line buffer and compressed. At this time, if the image data for eight lines is stored in one line buffer, the line buffer for storing and reading is switched. On the other hand, at the time of output, on the contrary, the expanded image data is stored in one line buffer, while the image data is read from the other line buffer and supplied to the image output unit 30. At this time, if the image data corresponding to eight lines is read out from the other line buffer, the line buffers for storing and reading are switched. As described above, the image input unit 10 or the image output unit 30 and the image compression / expansion unit 203 are synchronized with each other via the line buffer every eight lines, and the ping-pong buffer is switched every eight lines. Then, the image compression / expansion unit 203 starts compression processing or expansion processing for 8 lines each time the ping-pong buffer is switched, and when the compression processing or expansion processing for 8 lines is completed, It will be in a standby state until it is switched.

Here, since the image input section 10 scans and reads the original at a constant speed, the image data is continuously supplied at a constant rate. By synchronizing for each line, the compression process can be performed from the input side without waiting for the input of all the image data for one page. Similarly, the image output unit 30 causes the laser diode to emit light at a constant speed, and the photosensitive drum 3
Since it drives No. 03, it is necessary to continuously supply image data at a constant rate, but by synchronizing every 8 lines by such a ping-pong buffer, all image data for one page is expanded. Without waiting for
The image data can be supplied to the image output unit 30 from the expanded soba.

The code buffer 204 is, here, RA.
It is composed of pure electric elements such as M and FIFO. Also,
The hard disk 205 is used for the reason described in the related art. On the other hand, reference numeral 206A is an image processing control unit that controls the entire image processing unit 20, but since the operation is different from that shown in FIG. 2, the subscript A is added.
Reference numeral 208 is an overflow detection unit that detects an overflow of encoded data in the code buffer 204.

Now, a method of buffering encoded data in the code buffer 204 and a method of detecting overflow by the overflow detection unit 208 will be described. At the time of image input, the code buffer 204 sequentially advances the write address to write the compressed encoded data, while when transferring to the hard disk 205, the read address follows the write address. The encoded data is buffered by reading the encoded data in a stepwise manner. Writing and reading of encoded data are usually performed in a time division manner. Here, regarding the step of both addresses, for example, if the capacity of the code buffer 204 is 1 Mbyte, the address is “00000”.
If you start stepping from and start stepping to "FFFFF",
It will be in the form of starting from "00000" again. That is,
Considering a circle having one round of the capacity of the code buffer, the read address rotates in the same direction as the rotation direction of the write address and makes a circular motion of following the write address.

At the time of image input, the supply of the encoded data from the image compression / expansion unit 203 is faster than the transfer to the hard disk 205, so that the write address advances faster than the read address. For this reason, when the coded data is written to a certain write address and a certain amount of time elapses, different coded data is written to the same address. However, if the read address catches up by that time, Since the encoded data of is read, the write address is empty and there is an area to be written. Conversely, if the read address has not caught up and read after the encoded data has been written to a certain write address, the write address is not empty and there is no area to be written. The data will be lost and spilled. This is an overflow. Therefore, the overflow detection unit 2
08 monitors the write address and the read address, and detects that an overflow has occurred when the write address overtakes the read address which should originally be stepping backward at the time of image input. Such detection is performed by the code buffer 2
When using a general RAM for 04, the FIFO
Even with a buffer of the form, overflow can be detected by monitoring the read address and the write address.

On the other hand, at the time of image output, the code buffer 204 sequentially advances the write address to write the encoded data read from the hard disk 205, while supplying it to the image compression / decompression unit 203. , The read address follows the write address, and the encoded data is read out to buffer the encoded data. However, when outputting an image, the image is read from the hard disk 205 by the image compression unit 20.
Since it is slower than the supply to 3, the step of the write address is slower than that of the read address, unlike when the image is input.
Therefore, at the time of image output, the read address may overtake the write address, contrary to the time of image input. This is called underflow. In this case, the encoded data will not be buffered normally, so
In this embodiment, in order to avoid this underflow, output control described later is performed.

C: Operation of Embodiment Next, the operation of the embodiment will be described. First, the operation at the time of image input will be described. Regarding the control of each part,
The image processing control unit 206A performs this. First, the image input unit 10 reads a document at a constant pitch page by page, and image data is generated each time the document is read.
The image data is compressed by the image compression / expansion unit 203 after preprocessing by 01 and buffered in the code buffer 204. Then, when the image is input, the code buffer 204
The overflow detection unit 208 detects whether or not an overflow has occurred.

At this time, if a certain amount of coded data is stored in the code buffer 204, the coded data is read out and transferred to the hard disk 205. That is, unlike the related art, the transfer start timing of encoded data is not the time when the process of compressing one page of image data is completed, but the time when a certain amount of encoded data is stored. Here, the certain amount may be, for example, one block defined by the transfer protocol standard of the hard disk 205, or corresponds to the case where the image data having the above-mentioned line buffer capacity (eight lines) is compressed. It may be the code amount to be executed. Thus, the reason for accelerating the transfer start timing of the encoded data is to end the transfer to the hard disk as soon as possible and to keep the capacity required for the code buffer 204 low.

If no overflow has occurred, the image processing control unit 206A instructs the image input unit 10 to read the original on the next page at the next pitch, and the image data obtained by reading the next page is read. Similarly, compression, buffering and transfer are performed. in this case,
In the code buffer 204, reading of the encoded data of the previous page and writing of the encoded data of the next page may be performed in parallel.

By the way, at the time of image input, when image data of a page having a low compression rate and a large generated code amount is read at a continuous pitch, or when the hard disk 205 is used.
If some trouble occurs in the code buffer 204, the code amount supplied from the image compression processing unit 203 to the code buffer becomes larger than the code amount transferred from the code buffer 204 to the hard disk 205, and overflow may occur in the code buffer 204. Will be more likely. When such an overflow occurs, it means that the coded data of the page that has been read and compressed at that time has been lost, so the image processing control unit 206A performs the next abnormality process.

That is, the image processing control unit 206A causes the image compression / expansion unit 203 to stop the compression process currently being performed, returns the write address of the code buffer 204 to the head of the page, and further An image input control signal is output to the input unit 10 to instruct to read the document corresponding to the same page at the next pitch. As a result, the image compression / decompression unit 203 is in a standby state until the image data of the same page is input at the next pitch, and the code buffer 204 stores the coded data of the page which is to be read / compressed when an overflow occurs. It will be discarded. Here, in the case where the encoded data of the page that is read / compressed when an overflow occurs is partly transferred to the hard disk 205 before being discarded from the code buffer 204 due to the earlier transfer start timing. There is. In this case, the image processing control unit 206A returns the write address of the hard disk 205 to the head of the page. As a result, the coded data of the page that is read and compressed when an overflow occurs is also discarded from the hard disk 205. In addition, depending on the generation state of the code amount, there may be coded data in the code buffer 204 before the page for reading / compressing when the overflow occurs. In this case, the image processing control unit 206A continuously executes the process of transferring the encoded data to the hard disk 205. As a result, the code buffer 20 is not ready until the image data of the same page is input at the next pitch.
The empty area of 4 will be expanded from the time when the overflow occurs. Therefore, the next time an original of the same page is read, even if the compression rate is not increased and the generated code amount is not reduced, the overflow will not occur again, and the image quality will not deteriorate when decompressed. Such processing is executed until the image data of the originals of all pages is stored in the hard disk 205.

FIG. 3 shows a specific operation in image input according to this embodiment. As shown in this figure, the originals 4 and 5 each have a large generated code amount and are continuous pages, so that an overflow occurs in the reading pitch of the subsequent original 5, so that the image processing control unit 206A generates. Based on the image input control signal, the original document 5 is rescanned. In this example, at the time when the document 5 overflows, part of the encoded data has already been transferred to the hard disk 205. Therefore, in this case, as described above, the code buffer 2
In addition to 04, the encoded data corresponding to the original 5 transferred to the hard disk 205 is also discarded. Further, in this figure, in the originals 1, 2, 4, and 5, the buffer usage state sharply rises in the first part, and then the inclination becomes slightly gentle because the transfer start timing of the encoded data is This is due to the point in time when a certain amount of encoded data is stored in the code buffer 204. That is, up to this point, the encoded data is only supplied from the image compression / expansion unit 203, and the hard disk 20.
5, the usage amount of the code buffer 204 rises sharply, but after that point, the hard disk 2 is not used.
This is because the transfer is also performed in parallel with 05, and the inclination becomes gentle accordingly. Therefore, since the originals 3 and 6 other than these originals are stored in a form of being accumulated in the previous encoded data, such a thing does not occur.

Such an image processing unit 20 has the following advantages when inputting an image. In the present embodiment, it is not necessary to set the target code amount in the first place. Further, even if the compression rate for the image data of one page becomes small and the generated code amount exceeds the transferable amount to the hard disk 205 within the period of one pitch, rescanning is performed unless the code buffer 204 overflows. It never happens. Normally, manuscript 1
It is rare that the compression rate for image data of a page becomes small, and it is also rare that such a document is input in continuous pages, so that the possibility of overflow is extremely low. For this reason, the number of times rescanning occurs at the time of image input can be suppressed as much as possible, and the overall processing speed can be improved. It should be noted that in the above-described example of FIG. 3, the originals 4 and 5 each have a large generated code amount and are continuous pages.
The overflow occurred only at the reading pitch of. Even if an overflow occurs, it is not necessary to perform control for increasing the compression rate at the time of rescanning, so image quality does not deteriorate. Even if some kind of unpredictable trouble occurs in the hard disk and the transfer rate is substantially reduced, the data loss can be dealt with by the processing at the time of image input.

Next, the operation of outputting an image in the embodiment will be described. In this case also, the image processing control unit 206
A controls each part. First, at the first pitch, the coded data of the first page, which is the first page to be subjected to image output, is read from the hard disk 205, transferred to the code buffer 204, and stored therein. Then, at the next pitch, the encoded data of the first page stored in the code buffer 204 is decompressed by the image compression / decompression unit 203, subjected to post-processing by the post-stage image processing unit 202, and then the image output unit 30. Is output to. This allows
The image output of the first page is performed at the next pitch.

Here, the transfer start timing from the hard disk 205 to the code buffer 204 does not necessarily match that of image output. That is, the image processing control unit 206A determines the code amount of each page, the transferred data amount of the page during data transfer, and the code buffer 204.
If the free space of one block is managed and the code buffer 204 has a free area of one block or more,
One block of data is transferred. At the end of transfer of each block, an interrupt signal is supplied from the hard disk 205 to the image processing control unit 206A, and the image processing control unit 206A detects the end of data transfer by this interrupt signal. When the transfer of the encoded data of each page is completed in this way, the image processing control unit 206A
Writes information indicating the end in a memory (not shown).

By the way, focusing on the storage state of the code buffer 204 at a certain pitch, there are cases where encoded data for a plurality of pages has already been stored, but only encoded data for less than one page is stored. There may be cases. On the other hand, regarding the transfer speed, the data supply from the code buffer 204 for the decompression process to the image compression / decompression unit 203 is faster than the data supply from the hard disk 205 to the code buffer 204 for the transfer process. Therefore, when continuously outputting pages corresponding to encoded data with a large amount of generated code, there is a possibility that an underflow may occur in the code buffer 204, in which there is no encoded data to be stored.

However, as long as no underflow occurs, as described above, while the image of the previous page is output at a certain pitch, the coded data of the previous page and subsequent pages are continuously output from the hard disk 205 to the code buffer 204.
The image output unit 30 transfers the
Image output for pages is normally obtained.

However, when an underflow occurs, decompression processing is performed in a state where the encoded data is lost, and as a result, a missing image is formed in the image output unit 30 and mixed with a normal image. become.

Also, when an error occurs in the hard disk 205 and the data transfer speed is reduced, the same problem as described above occurs. The error of the hard disk 205 is a seek error, a data read error, a sector ID read error, or the like. When such an error occurs, the hard disk 205 retries several times and tries to recover from the error state. During that time, the data read / write operation is suspended between the disk and the head, and the data transfer rate is reduced. For example, when a data read error occurs and a retry operation is performed 16 times, waiting for 16 rotations occurs. When the disk rotation speed is 5400 rpm, the waiting time is 177.8 ms, so
The amount of data transferred per second is reduced by about 18%.
Further, when normal data reading cannot be performed even after a predetermined number of retries, the sector is regarded as a defective sector, and its replacement sector is created in another area of the disk. If the alternative sector is not on the same track, an extra rotation wait or seek will occur, and the data transfer rate will decrease. When such an error of the hard disk 205 occurs and the data transfer rate decreases, underflow of the code buffer 204 may occur during the decompression process even if the compression rate of the page being transferred is higher than the target compression rate. There is a nature.

In this embodiment, the following control is performed in order to avoid such underflow. That is, the image processing control unit 206A is supplied with a page synchronization signal (one cycle corresponds to one pitch) for page synchronization from the image output unit 30.
A detects the rising edge of this page sync signal. Upon detecting the rise of the page sync signal, the image processing control unit 206A checks whether or not the encoded data of the page to be output next has been transferred to the code buffer 204 by referring to the write information in the memory (not shown). .

If the transfer is completed, the decompression process for the page is started. Since all the coded data of the page has been transferred to the code buffer 204, the code buffer 204 is in the middle of decompression processing of the page.
Underflow is unlikely to occur.

On the other hand, if the encoded data of the page to be output next is not completely transferred to the code buffer 204, the underflow of the code buffer 204 may occur due to the fact that the data transfer is not in time for the expansion speed. is there. Therefore, in this case, in order to avoid the occurrence of underflow, the image output control signal for instructing the skip operation is output to the image output unit 30 without performing the expansion process at that pitch.

Here, the skip operation will be described.
In the normal image forming process in the image output section 30, the copy latent image formed on the photosensitive drum 303 is transferred to the sheet conveyed to the transfer position, and then the fixing section 30.
It is discharged to the copy tray 308 through the sheet 6. In this case, the sheet is conveyed to the transfer position with the timing adjusted so as to coincide with the beginning of the copy latent image on the rotating photosensitive drum 303. Here, if the image output control signal is supplied from the image processing unit 20 before the conveyance of the sheet to the transfer position is started, the image output control unit 311 stops the conveyance of the sheet until the next pitch. During this period, the rotation of the photosensitive drum 303 is continued, but the transfer onto the sheet and the discharge of the sheet are not performed.

On the other hand, the coded data is continuously transferred from the hard disk 205 to the code buffer 204 even during the one-pitch skip operation. If the transfer of the coded data is not completed at the next rising edge of the page sync signal after the one-pitch skip operation, the same operation as described above is repeated until the data transfer is completed.

Here, a concrete operation in the image output of this embodiment is shown in FIGS. In the example shown in FIG.
Since the generated code amounts of the 4th and 5th pages are continuously large, the data transfer of the 5th page is not completed at the start of the pitch of the 5th page (that is, at the rising edge of the page synchronization signal), and the skip operation is performed. Is occurring. Even during this skip operation, the transfer of the encoded data of the fifth page is continued, and the data transfer of the fifth page is completed at the start of the next pitch. Done.

Further, in the example shown in FIG. 5, the skip operation is caused by an error in the hard disk 205. That is, the data transfer is delayed due to an error in the hard disk 205 during the data transfer of the fourth page, the data transfer is not completed at the start of the next pitch, and the skip operation occurs. 4 even during this skip operation
The data transfer of the page continues, and at the start of the next pitch, since the data transfer of the fourth page is completed, the expansion processing and image output of the page are performed.

Electronic sorting is performed by repeating the above-described processing for each page in a designated page order a designated number of times.

[0052]

As described above, according to the present invention, when the image compression rate is low, the generated code amount is large, or when the data transfer rate is lowered due to an abnormal operation of the storage means (hard disk), etc. Even when the transfer rate of the image data cannot meet the decompression rate at the time of image output as described above, the occurrence of underflow in the buffer storage means (code buffer) can be reliably prevented, and a missing image or a disturbed image can be obtained. Can be output.

[Brief description of drawings]

FIG. 1 is a front sectional view showing the configuration of a digital copying machine which is an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a block diagram showing an electrical configuration of an image processing unit in the same embodiment.

FIG. 3 is a diagram for explaining an example of an image input operation in the same embodiment.

FIG. 4 is a diagram for explaining a first example of an image output operation in the same embodiment.

FIG. 5 is a diagram for explaining a second example of the image output operation in the same embodiment.

FIG. 6 is a block diagram showing an electrical configuration of a conventional image processing apparatus.

FIG. 7 is a diagram for explaining an example of an image input operation in the image processing apparatus.

FIG. 8 is a diagram for explaining an example of an image output operation in the image processing apparatus.

[Explanation of symbols]

 10 image input unit 20 image processing unit 30 image output unit 201 pre-stage image processing unit 202 post-stage image processing unit 203 image compression / expansion unit 204 code buffer 205 hard disk 206A image processing control unit 207 code amount detection unit 208 overflow detection unit

Claims (3)

[Claims] 1. A storage unit for storing encoded data obtained by variable-length-compressing original image data, a unit for decompressing the encoded data read from the storage unit, and a page based on the image data decompressed by the decompression unit. Image output means for sequentially forming images in units, and buffer storage means for temporarily storing the coded data read to absorb the speed difference between the read speed from the storage means and the expansion speed of the image output means. And output control means for starting the output of the page provided that the encoded data of the page to be output next by the image output means is all transferred from the storage means to the buffer storage means. An image processing device characterized by: 2. The image output unit continuously outputs image data in page units at a predetermined pitch, and the output control unit starts a pitch section in which encoded data corresponds to the page. 2. The image processing apparatus according to claim 1, wherein the pitch section for outputting the page is shifted to the next section if not all have been transferred before being processed. 3. The image output unit includes a printing unit that forms an image on a sheet, and a paper feeding unit that feeds the sheet to the printing unit, and the output control unit stops the operation of the paper feeding unit. The image processing apparatus according to claim 2, wherein the output is avoided by causing the output.