JPS6020950B2 - Sub-scanning control method for facsimile, etc. using redundancy suppression coding method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a method for performing smooth sub-scanning by continuously varying the sub-scanning speed in sub-scanning 3 of a redundancy suppressed facsimile or the like.

Conventionally, in a facsimile machine using redundancy reduction coding technology, the number of coded bits generated in one line is not constant; it is small in simple parts of a document, and increases in complex parts.

Therefore, when transmitting and receiving encoded signals at a constant transmission speed, the generation of encoded signals for each line fluctuates over time, so the sub-scanning is configured to be sent intermittently by 1 repetition of starting and stopping. It had been. Therefore, there are the following drawbacks: (1) It is difficult to increase the speed of sub-scanning because recording is impossible until the step response of starting and stopping is stabilized.

■ The transmission system such as gears, the transmitted document, etc. vibrate with the start and stop operations, and noise is generated.

■ Feed unevenness occurs due to disturbances in the step response of starting and stopping due to load fluctuations, etc.

■ An extremely powerful motor is required to start and stop the movable table of a stationary facsimile transmitter, etc., which has large inertia, at high speed.

Therefore, in order to eliminate these drawbacks, an N-stage sub-scanning speed encoder is used, and a continuously variable one is used to control acceleration and deceleration using the storage capacity of a buffer memory placed between the decoder and the modulation/demodulation section. A sub-scanning method was proposed and actually applied to stationary original facsimile transmitters.

However, this continuously variable sub-scanning method, which is controlled by the amount of storage in the buffer memory, had the disadvantage that the codec was required to be as fast as 5 to 1 times faster than the transmission speed, considering the situation of war. . If the codec is slow and there is only a small amount of line memory between the recording section and the codec, it will take a long time to code and decode a complex line with a large number of encoded bits. As a result, the buffer memory storage capacity does not reach the deceleration range, and before the encoding/decoding of the -line is completed, the transmitter's entire line memory is filled with image signals, and the receiver's line memory is filled with image signals. There have been cases where all the image signals have been exhausted and the sub-scanning has to be stopped urgently, resulting in large sub-scanning unevenness. The present invention eliminates the above-mentioned drawbacks, and even if a relatively slow codec is used, there is no need for an emergency stop with a small amount of line memory, and even if an emergency stop occurs, uneven feeding can be prevented. The object of the present invention is to realize continuously variable sub-scanning in which less occurrence of .

FIG. 1 is a block diagram showing an embodiment of the present invention when used in a receiving system. 1 is an input path for the redundancy reduction encoded signal, 2 is a buffer memory placed before the decoder, 3 is the decoder, 4 is a line memory placed between the recording section and the decoder, and 5 is the input path for the redundancy suppression encoded signal. Line memory amount detection section, 6 is recording section, 7
is a sub-scanning speed control section.

Also, assuming that there are N stages from V to VN as a speed sequence, V
, SV... (1≦i≦N-1).

Next, the operation of this control system will be explained using FIGS. 1 and 2. Encoded information input from signal input 1 is temporarily stored in a buffer memory 2, converted into one line of image information by a decoder 3, and stored in a line memory 4. One line of information stored in the line memory 4 is recorded by a recording circuit 6. One speed V between the current V, and VN
Suppose that sub-scanning is performed at J (1≦i JSN), and the sub-scanning speed is the line memory amount detected by the 1-line memory amount detection unit 5 (the number of line memories in which both signals are stored).
It is determined from LM (=1, 2, . . . N) and the current sub-scanning speed VJ as follows. In other words, the sub-scanning speed VNExT of the next line after Z is determined from the relationship between LM and i: If LM>j, VNExT=Vi... If LM=j, VN8xT=V'' If LM<j, VN6xT=Vi-, 2. This is a method of controlling the

Since the speed is changed step by step in this way, unevenness in feeding due to speed changes can be reduced.

In addition, since the sub-scanning speed is controlled by the amount of line memory each time one line is recorded, the sub-scanning speed Vj can be decreased in response to a decrease in the line limit amount LM, and the continuous variable speed controlled by the amount of buffer memory can be used. Even in the case of an emergency stop caused by the line memory becoming empty, which was a problem with the sub-scanning method, it is possible to stop from the lowest sub-scanning speed, which has the advantage of preventing large feed irregularities from occurring. There is. Next, changes in the line memory amount, buffer memory amount, and sub-scanning speed when controlled by this method will be explained using FIG. 2.

First, as shown in Figure 1, the initial conditions are that up to 15 data are stored in the buffer memory, 1, 12, 14, 15 is the number of bits corresponding to the minimum transmission time, and 13 is about 17 of that.
It is assumed that the data of the line coming in after 15 is the same as 1. It is assumed that all N (=6) lines are filled in the line memory and sub-scanning is being performed at its maximum. When one line of the line memory 4 is recorded, one empty line occurs, and the decoder 3 starts decoding 1. Since the decoding speed is faster, it is one line worth of grave. Decoding ends before the scan ends, and decoding is stopped for the rest of the time. When I go to check the line memory just before starting the next recording, I find that all lines are correct, so the next sub-scanning speed is also set to the maximum speed V6. Similarly, decoding for 12 is performed, but in the case of 13, the decoding is not completed even after one line of sub-scanning is completed, so the line memory amount decreases by one, and the next sub-scanning speed is A one step slower speed is selected.
Thereafter, the sub-scanning speed decreases step by step as the number of line memories decreases until 13 decoding is completed. In the case of FIG. 2, decoding of line 13 is completed during sub-scanning at the lowest speed, and decoding of lines 14, 15, and subsequent lines are being performed. If decoding cannot be completed during the period of sub-scanning at this minimum speed, sub-scanning must be stopped, but since it is stopped from the lowest speed in this way, the unevenness of feed at the time of stopping can be reduced. It can be suppressed. For comparison, an example of control using the conventional buffer memory amount is shown in FIG. In this case V3
It has stopped since. The amount of buffer memory decreases while decoding is being performed, but when the line memory becomes full, decoding stops and the amount of buffer memory increases rapidly. The line memory increases and the sub-scanning speed accelerates, but since it accelerates one step at a time from the lowest speed, until the maximum speed is reached, the number of lines that can be recorded is smaller than the amount of data that enters the buffer memory. is small, and the amount of buffer memory increases. However, since acceleration is performed more quickly than in the method of controlling based on the amount of buffer memory shown in FIG. 3, controllability is improved. For example, if the transmission speed is 960 ps, the minimum transmission time is 10 seconds, and the transmission time is 200 msec, the decoding speed is 1920 ps.
Speed sequence 100, 90, 80, 70, 60, 50, 1 of 9 ps (lines persec)
Simulation results have shown that one stage is sufficient for control.

On the other hand, when controlling with the amount of buffer memory, even for normal images, if the line memory amount is about 6 lines with 11 speed steps, a decoding speed of 048,000 to 9,600 ps is required, and a decoding speed of about 1,920 ps is required. In terms of decoding speed, even if the speed step is set to about 6 steps, a line memory of 20 lines or more is required. Compared to the present method, this method of controlling using a secondary buffer memory has drawbacks such as requiring a high-speed decoder or a large amount of line memory. Although the number of line memories is set to N, a certain number of line memories may be provided as a fixed number for two-dimensional encoding/decoding, transfer buffers, or recording buffers.

In the above explanation, each time one line is captured or recorded, the number LM of lines in which image signals are stored in the line memory is determined from the current sub-scanning speed VJ to determine the next sub-scanning speed VN8xT. Although the example was explained,
Another example of the method for determining the next sub-scanning speed VNBxT is the following method. That is, the storage amount of the line memory is set as one line or an integer fraction of one line (for example, a chapter line), and m divisions LM are divided into n, (≧1) units from the one with the smallest storage amount.
and one speed V, for each of them (V, in the transmission system)
≦V..., in the receiving system, V, ZVi-,) are made to correspond, and the currently accumulated line memory amount is entered every time an integer fraction of one line, one line, or several lines is read or recorded. From the number i of division LMi and the current sub-scanning speed V', calculate the next sub-scanning speed VN8xT, if i<j then VNEx, =Vi-, if 2i=j then VN8
If xT=Vii>j, it can also be determined as VN6x'=Vi+.

By using this method, the time it takes to accelerate to the maximum speed can be shortened by 3 times compared to the time it takes to accelerate to the minimum speed, so it is possible to reduce the increase in the amount of buffer memory.
The stability of the control system can be improved.

In addition, when the unit of division is an integer fraction of one line, the control interval becomes shorter, so it is possible to control the line memory a little more. This has the advantage that the delay in the control system can be reduced and the change in buffer memory amount can be reduced by 4. Furthermore, it is also possible to combine the above-described methods for determining the sub-scanning speed.

In other words, areas are divided based on line memory raw storage amount or sub-scanning speed, etc. For example, 5, 1 stop, 201 ps is 1
/ Control every 4 lines, 30, 4ri 5ri 60, 701
ps per line, and 8 stop 9 ulo dish ps is 2
By controlling each line, it is possible to reduce the time required for sub-scan control during high-speed sub-scanning and increase the time used for decoding processing, as well as change line memory during low-speed sub-scanning. It is possible to improve the responsiveness to, and the applicability of control can be improved. Also,
Hysteresis characteristics may be provided during acceleration and deceleration. When controlling sub-scanning, in order to provide a condition of 0 that allows stable operation with limited buffer memory even under the worst conditions, the sub-scanning speed sequence and memory storage amount monitoring cycle settings are as follows. It is better to do it like this.

That is, by encoding and decoding a line string in which there is one (≧1) line whose code bit number corresponds to the assumed maximum transmission time among consecutive lines with the minimum number of bits, The amount of change in buffer memory storage amount that changes from decelerating from the maximum sub-scanning speed state to returning to the original maximum sub-scanning speed state △BM (=buffer memory storage amount after returning to the maximum sub-scanning speed state - The buffer memory storage amount before deceleration) is up to the length corresponding to the maximum transmission time assumed by the number of code bits, in the transmission system, the buffer memory storage amount before deceleration is The value of 1 is set in advance so that if there is sufficient free space, △BMZO, and in the receiving system, if there is sufficient storage in the buffer memory before 5 decelerations, △BM≦0. For this, the sub-scanning speed sequence and the monitoring cycle of the line memory storage amount are set. Taking a receiver as an example, the explanation is as follows.

Line 1 with the longest number of bits in the line string with the minimum number of bits
If books are stored consecutively, decoding takes time, so the line memory is reduced, and the sub-scanning speed and buffer memory amount are also reduced. After that, when decoding of the line with the minimum number of bits is started, the line memory quickly becomes full because one line can be decoded in a short time. While decoding, the amount of buffer memory decreases, but once the line memory is full, the sub-scanning speed is slow and the number of lines that can be recorded per unit time is small, so the buffer memory is used until it is accelerated to the maximum speed. The amount increases. If the total buffer memory amount change is positive even if there is sufficient buffer memory storage before deceleration and the buffer memory does not become empty during the process, if such a line repeatedly enters, the buffer memory amount will increase as much as possible. increases, and the finite buffer memory will overflow, so this is not a better method for controlling sub-scanning. Therefore, this buffer memory amount change is set to be negative or zero. Here, consider the value of 1. The value of 1 is 1
If the number increases, the decoding time becomes longer, and the amount of buffer memory decreases increases.

On the other hand, the maximum amount of increase in buffer memory is determined by the timing in sub-scanning at which decoding of a long line ends, so in the worst case, this change value decreases, and increasing by 1 tends to loosen the condition. It is in. In other words, if there is a high probability that lines with long code lengths are consecutive in the transmitted original, and if there are many consecutive lines, the value of 1 can be equivalently increased, and the restrictions on the sub-scanning speed etc. can be relaxed. You will be able to do so. For example, the example given in the explanation of the above-mentioned embodiment satisfies the condition of 1=1, and the buffer memory base will not overflow.
One effective method for setting the sub-scanning speed sequence is as follows:
The receiver has been explained above as an example, but if the lower sub-scanning speed is applied to the side with the least amount of line memory storage, then the lower sub-scanning speed is applied starting from the side with the least amount of free space in the line memory, and control is performed using the amount of free space in the line memory. , one speed sequence is defined for each LM, and the same can be realized in a transmitter in which this condition is not satisfied.

As explained above, according to the present invention, the sub-scanning speed is selected and controlled based on the amount of image signals stored in the line memory, so even if a relatively slow decoder is used, the amount of line memory is small. It is possible to control the sub-scanning speed, and even in the event of an emergency stop, it is possible to stop from a low speed, which has the advantage of reducing the occurrence of uneven feeding and realizing smooth sub-scanning control.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a receiver implementing the method of the present invention for controlling the sub-scanning speed by the amount of line memory, and FIG. A diagram that graphically expresses changes in scanning speed, line memory amount, and buffer memory amount,
Figure 3 is a diagram schematically showing changes in the sub-scanning speed line memory amount and buffer memory amount in the conventional method of controlling the sub-scanning speed using the buffer memory amount (M is a standard value for acceleration and deceleration). It is. 1...Input path of redundancy suppression encoded image signal;
2...Buffer memory, 3...Decoder, 4...Fin memory, 5...Line memory amount detection unit, 6... Recording section, 7... Sub-scanning speed control section. Figure l Figure 2 Figure 3

Claims (1)

[Claims] 1. In a transmission system such as a facsimile using a redundancy reduction coding method, N line memories are installed between a reading section and an encoding section, and in a receiving system, between a decoding section and a recording section. established,
For each number LM of line memories in which image signals are stored, the minimum sub-scanning speed V_m corresponding to the maximum transmission time of one line or the transmission time that is practically considered to be approximately the maximum.
In the transmission system, V
_m_a_x=V_1≧V_2≧……≧V_N=V_m
In the _i_n receiving system, V_m_a_x=V_N
≧V_N_-_1≧...≧V_1=V_m_i_n, and each time one line is read or recorded, the number of line memories LM in which image signals are stored and the current sub-scanning speed V_j (1≦j≦N ) if LM>j then V_N_
E_X_T=V_j_+_1 If LM=j then V_N_E
_X_T=V_j If LM>j then V_N_E_X_T=
Next sub-scanning speed V_N_E_X as V_j_-_1
A sub-scanning control method for facsimile and the like using a redundancy suppression coding method characterized by determining and controlling _T. 2 In a transmission system such as a facsimile that uses a redundancy reduction coding method, a line memory is provided between the reading section and the encoding section, and in a receiving system, between the decoding section and the recording section, and the images stored there are provided. The minimum sub-scanning speed V_m_i_n corresponding to the maximum transmission time or the transmission time that is practically considered to be almost the maximum depending on the amount of signal accumulation, and the sub-scanning speed corresponding to the minimum transmission time or the maximum sub-scanning speed V_m_a_x which is faster than that. In a sub-scan control method for facsimile, etc., which uses a redundancy suppression coding method to select a speed between 1 and 2 and control sub-scan,
N divisions L of n_i (≧1) units each, starting from the one with the smallest accumulated amount, using 1/1 line or 1 line as a unit
M_i, each with one speed V_i (V_i≦V_i_−_1 in the transmitting system, V_i≧V_ in the receiving system
i_-_1), and each time an integer fraction of one line, one line, or several lines is read or recorded, a section LM_ containing the currently accumulated line memory amount is created.
From the number i of i and the current sub-scanning speed V_j, calculate the next sub-scanning speed V_N_E_X_T. If i<j, then V_N_E_
X_T=V_j_-_1 If i=j, then V_N_E_X_
T=V_j If i>j then V_N_E_X_T=V_j_
1. A sub-scanning control method for facsimile, etc., using a redundancy suppression coding method characterized by determining and controlling as +_1. 3. In a transmission system such as a facsimile that uses a redundancy reduction coding method, a line memory is provided between the reading section and the encoding section, and a buffer memory is provided after the encoding section, and in the receiving system, a line memory is provided between the reading section and the encoding section, and a buffer memory is provided after the encoding section. A line memory is provided between the sections, and a buffer memory is provided before the decoding section, and the transmission time corresponds to the maximum transmission time or the transmission time that is practically considered to be approximately the maximum, depending on the amount of image signals stored in the line memory. The minimum sub-scanning speed V_m_i_n and the sub-scanning speed corresponding to the minimum transmission time, or the maximum sub-scanning speed V which is faster than that
In a sub-scanning control method such as facsimile that uses a redundancy suppression coding method that selects a speed between A line string containing one (≧1) continuous line with a length corresponding to The amount of change ΔBM in the buffer memory storage amount (=buffer memory storage amount after returning to the state of maximum sub-scanning speed - buffer memory storage amount before deceleration) increases until the code bit number corresponds to the maximum transmission time assumed. For any line length, in the transmitting system, if there is sufficient space in the buffer memory before deceleration, ΔBM ≥ 0, and in the receiving system, if there is sufficient storage in the buffer memory before deceleration. 1. A sub-scanning control method, characterized in that a sub-scanning speed sequence and a monitoring cycle of a line memory storage amount are set for the preset value of 1 so that ΔBM≦0.