JPS6083182A - Information decompression method and its circuit - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to an information decompression method and a circuit thereof that can decompress various compressed information, particularly compressed image information at high speed.

Conventional configurations and their problems In recent years, editing processes such as image enlargement and reduction, which were previously performed on large computers and expensive dedicated systems, have become necessary in the field of inexpensive personal computers.

Below, conventional expansion of compressed information, which is one of the processes for enlarging/reducing images, will be explained.

Expansion is to obtain an expanded image Y according to a mask pattern P from an original image X, as shown in FIG. Each element of Y is determined as follows.

(1) Let it be i--1.

(2) Let yi-X.

(Suppose 3 rings i=1+1.

(4) If P-work = 1, y-work = X-work.

(5) If P□=Q, set y knee X knee 1.

(6) Repeat (3) to (5) until all elements are determined.

In order to carry out the above operation, a circuit as shown in FIG. 2 is devised. In FIG. 2, 1 is a shift register that stores the original image , J-1, and l. registers, and 6 is an AND gate that performs logical product.

Note that a and b are clocks, and cl a+e+f are output signals of registers 1.2 and 4.5, respectively.

A signal corresponding to the leftmost value of the shift registers 1 and 2 always appears in the output signals c and d. .

Further, as shown in FIG. 3, the clock signals a and b are signals having the same frequency but out of phase. shift register 1,
2 and 3 are entirely shifted by one bit to the left each time the shift clock signal is input manually. The shift clock signal is the clock signal a for the shift register 2, the clock signal for the shift register 3, and the output signal g of the AND gate 6 for the shift register 1.

Further, when the clock signal is inputted to the registers 4 and 5, the registers 4 and 5 perform an operation of converting the input signals C and a into new output signals e and f.

The operation of the decompression circuit configured as above will be explained below.

It is assumed that the elements of the original image X and mask pattern P shown in FIG. 1 are stored in shift registers 1 and 2 in order from the left, respectively. First, the clock signal becomes "1", and the signals (1, e) are taken into the shift register 3 and register 6, respectively.

Therefore, the signal f becomes ++ 1++. Next, when the clock signal a becomes "1", the output of the AND gate 6 also becomes "L1".
Therefore, at the same time as shift register 2 is shifted to the left, shift register 1 is also shifted, and register 4 stores the signal C just before the clock signal a becomes 11゛, and the signal e becomes x. , is output.Next, the shift register 3 is shifted and the signal e is taken in.At the same time, the signal d is taken into the register 5, and the signal f becomes ``q''.Subsequently, when the clock signal a becomes 1'', the signal f is ``q'', so the output of the A N'D gate 6 becomes Q'', and the shift register 2 is shifted. Shift register 1 is not shifted. Therefore, the value of signal C changes.

do not have. Next, the signal e is input to the shift register and the signal d is input to the register 5 in synchronization with the clock signal S. This operation is the clock signal a.

A decompressed image is generated in the shift register 3 by as many (-j) as the elements of the shift register 1 in synchronization with b.

However, in the above method, the operation is performed in synchronization with the clock signal, so if high-speed operation is to be performed, a corresponding high-speed lock signal is required.
It becomes difficult to convert to LSI.

Purpose of the Invention The present invention solves the above-mentioned conventional problems, and
It is an object of the present invention to provide an information decompression method and a circuit thereof that are suitable for SI and can perform high-speed information decompression.

Structure of the Invention The information decompression method of the present invention is based on the information to be decompressed x1゜...
・Reference mask pattern Pm (however, m
is an integer of m≧1), and the information to be expanded x1.
...xm is at least e row and e' column (however, 4,
4′ is e≧2, 4′≧2) u fiffi ) k W
When moving from the first row to the fourth row in the matrix A, the reference mask P (where 151<
a of matrix A if the first information is
l; ] (where ) is an integer of 1≦k β) moves the information k is an integer of i, which is -1 only when it is 1-1), moves the information If the information of 2 is −, then a□ of 7 trinox A,, (however, heel p is 1≦
The information X located at p≦e (an integer) in matrix A is stored in a-41. By moving to p, the above objective is achieved.

Further, the information decompression circuit of the present invention has at least e rows and l' columns (however, gB of 1.l'he≧2 + l'≧2) a q , r (however, q + r is 2
An information transmission means is provided at the position of knee q≦e + q>r (integer), and aS of matrix A, (II, l
, s, t are 1 < s q= l, s≦t≦6'
)44i,)o,1: Information selection means are provided at the positions of the first row, and the information to be expanded x1. ...xn (where n is an integer 0 where n≧1) is moved from the first row to the fourth row. is the information to be expanded according to the control signal corresponding to the reference mask pattern P.
−, (However, U is 1≦u×n, and only when u = 1 is u
Enter one of the
, t, the information selection means located at either one of 2L, )-1, t+l, aS-1, t of the matrix A according to a control signal corresponding to the reference mask pattern Ps. Further, the information transmission means located at a r of matrix A inputs the information X of either the information selection means or the information transmission means located at matrix a q-1, r. By doing so, the above purpose tS is achieved.

DESCRIPTION OF EMBODIMENTS An information decompression method according to an embodiment of the present invention will be described below.

1 original image X as xO, x, , x, , x3+x4
+x5 +x6. Let us consider the case where the reference mask pattern P is 1,0°1, O, O, 0,1, O.

As shown in Figure 4, the vertical direction is 1. Indicate the horizontal direction with a slow letter C, and each element of the matrix with an M letter. Original image
initially exists in the row 1-=Q, moves within the matrix in the direction from the 1st row to the 1+1th row, and the expanded image Y=xo, xo
, x1. x,,x,,x,,x2. x.

This is what you get.

When moving from the 1st row to the i-z row, if 5, g IIC summer mask pattern P = 1, all the elements in the i-th row are 1.
Moved to line +1. If P = d, the elements from -IS1 to 1-1 are moved to the i -4i +11th line, and the elements from j = i -1 to the right end are shifted by one element to A' and moved to the 1+1 line. Move to the eyes. This action is 1
This is performed up to the -7 row, and the expanded image Y'f: is obtained as an output from the matrix.

If we apply the actual values according to the above principle and look at the expansion process, we get the following.

(1) P in the 1-ISl row. All elements in the second row that are -1 are moved to the first row. Therefore, 1-i for i = 1
-Xolxl 1x21X31X41X51x61x
It becomes 7.

(2) In the row of i=1, P is a - member, so to ``20'', ``T'' is transferred, and to M21 to M27, M is transferred.
10-Ml7 are transferred in sequence. Therefore, the row where 1=2 is xo 1 X[11xi + x2 r X5 +
It becomes X 4 + X'5 + x6.

(In the row 3+1==2, P2-1, so all the elements in the second row are transferred to the third row. Therefore, the row 1-3 is ・xO
lxO・xl・x2+ x5+ x4+ x5・x6.

(4) By similarly processing VC1=s to 70 rows, the output from the matrix is Y=xo.

xo, Xl, Xl, Xl, xl, X2. x2 can be obtained.

Next, an information decompression circuit according to an embodiment of the present invention will be described.

FIG. 5 shows the wiring of an information decompression circuit in an embodiment of the present invention.

In FIG. 5, 10o to 107 are image information X that should be expanded. - Information input terminals for applying x7, 11o to 117
is a reference mask pattern P. - Binary il according to P7
j Control signal input terminals to which control signals are applied, 126-1
27 is an inverter. 13 is an element M,,k of a matrix of 8 rows and 8 columns (however, j is an integer and 1≦
j≦y, j>k) is provided at E1k, and is sent from above. It is an information transmission means for sending i-reports downward, and may simply be a signal line. 14 is provided at the position of matrix element M8゜t (both s and t are integers, OSS≦7.S≦t), and ib'l fMI 4t:1' input terminal 11S' (( Depending on the control signals aS and bS corresponding to the reference mask pattern PS sent through Logic element 1 as shown in FIG.
4fL, 14b, 14C.

14d. 158 to 157 are output terminals from which the expanded image Yi is obtained.

Hereinafter, the configuration of the 4':'t''1jlQ selection means 14 will be explained in more detail with reference to FIG.

As shown in FIG. 7, the information selection means 14
When the control signal a3 sent through the ff1-1f signal input terminal 11s is "0" or the control signal b3 is 1'', the support for (a8.bs)-(0,1) is determined by the matrix. Input information C located at "s-+, t-+" of "-force control 1 + signal aS is 1""control signal bs is" o
”, that is, (2L3+bs)-(1+on), information d located at Ms-+, t of the matrix is selectively input.

The operation of the above configuration will be explained below.

Note that the reference mask pattern P is 1.0, 1.0°0.0,
1, O, and finally the decompression information X. +xQ +X,
,X, ,X, ,Xi,X2. The assumption that X2 is obtained is the same as the explanation shown in FIG.

As shown in Figure 8 e), the information input terminals 10° to 1
0. The original image information X is sent to the information selection means 14o to 147 via. - Send X7. At that time, information selector 1, 14
6 to 147 are control signal input terminals 11o as control signals.
(2Lo,bo)-(1,0) is applied through i-
By doing so, the information selection means 14o to 147 are operated as human input signals. Since the information on the ~D7 side is selected, the original image information is X. -x71 respectively.

Next, as shown in FIG. 8(b), the information selection means 14° selects the original picture information X. is transmitted from the information transmission means 13oVc. On the other hand, the information selection means 148 to 1414 input (a, +b1) via the control signal input terminal 111 as a control signal.
)-(o, 1) is applied, so that the information on the lj c o - c 6 side is selected, so the original image opening information X. - Enter x6.

Next, as shown in FIG. 8(C), the original iI! of the information transmission means 13° and the information selection means 148! ! Izuka information teeth,
1) Send the information to the information transmission means 131 and 132. On the other hand, the information selection means 1415 to 142o input (a2.b2)-(1
By applying ゜0), the lines D8~
Since information on the D14 side is selected, the original image shield't12x o
-x, enter all.

Next, as shown in FIG. 8(d), information transmission means 131 to
132 and the original image 1-j of the information selection means 1415
The information xo,x, is the information transmission means 13. ~135 will be tsummed. On the other hand, the information selection means 1421 to 1425 select the control signal input terminal 11. via (a,.

By applying b3)=(0,1), the information on the glands 07 to C++ is selected, so the original image information x1 to x5 is inputted, respectively.

Next, as shown in FIG. 8 (6), the information transmission means 133 to
135, and the original image Xo of the information transmission means 1421,
x1 is information transmission means 136-13. C is sent to C. On the other hand, the information selection means 1426 to 1429 input (a4°b4)=(0
,1) is applied, so that n Z n,
Since the information on the f+LI C+ 2 to 015 side is selected, the original image information x1 to x4 are respectively input.

Next, as shown in FIG. 8(f), the information transmission means 136 to
132. and the original image information x of the information transmission means 1426
,,x1 are the information transmitting means 131o to 131. n are sent to l, respectively. On the other hand, information selection means 145o to 1432
is sent as a control signal via the control signal input terminal 11 (a
By applying s, bs) - (o, 1), the information on the lines C16 to C18 is selected, and is input to the original image information x1 to X3, respectively.

Next, as shown in FIG. 8(g), the information transmission means 13i (
, ~1314 and the original image information Xo, Xj of the information transmitting means 143o are sent to the information transmitting means 1315-132°. On the other hand, the information selection means 143511434 receives a control signal via the control signal input terminal 116.

Since (a6.b6)-(1,O) is applied, the line J5+"+6'1 is selected for the 1st rule.
Since the length of each image is large, input each into the original image information x2iX3.

Finally, as shown in FIG. 8 (11), the image information X of the newsletter transmission means 135 to 1316 and the information selection means 1435. ~x2 is the signal transmission means 1321~1゛3.
．． 7 respectively. On the other hand, the information selection means 14 receives a control signal via the control signal input terminal 117 ('
71b7)-(o+') is applied, information on the line J9 side is selected, so the original image 1
Input °＾information x2. And output island + 1f child 156 ~
157 all through 2 and original image information x. ,x, ,x, +x
The decompression information Y'(i-) can be obtained by extracting it as the final output to l +Xj +X21X2.

As described above, according to this embodiment, the information transmitting means 13 and the information selecting means 14 have a circuit configuration arranged in a matrix, so that even when rapid decompression is required, the clock It is suitable for LSI implementation because it does not require a circuit and has a regular circuit configuration. Ushida control signal input terminal 11
Other expansion information @Y can be easily obtained by simply changing the control signal applied to .

In this embodiment, for convenience of explanation, the information transmitting means 13 is provided, or since the information transmitting means 13 may be a simple wiring as described above, the circuit shown in FIG. 5 may be replaced with the circuit shown in FIG. 9. good.

Although only image information has been described in this embodiment, other information may be applicable to the present invention.

Effects of the Invention The present invention is capable of decompressing information at high speed without using a clock by processing the information to be decompressed using a matrix. It is also suitable for LSI implementation, and its value is invaluable.

[Brief explanation of drawings]

Figure 1 is a diagram showing image decompression, Figure 2 is a block diagram of a conventional image decompression circuit, Figure 3 is a clock signal waveform diagram of the same circuit, and Figure 4 is an uninvented one. 5 is a wiring diagram of an information decompression circuit in an embodiment of the present invention, FIG. 6 is a circuit diagram of the information selection means, and FIG. 7 is a circuit diagram of the information selection means. 8 is a wiring diagram showing the operation of the information expansion circuit, and FIG. 9 is a wiring diagram of the information expansion circuit in another embodiment of the present invention. 13... Information transmission means, 14... Information selector ↓
-ri1゜Name of agent: Patent attorney Satoshi Nakao Li3 and 1 other personFig. ``l''ノ, l, χ2. χ3. i4. χ5. χl, χ7
p:f, θ,! , θ, ρ7 θ,! , θY=Z', μD
, ZL, χ1. 'tr, χ5. Ru. 12g+j 2nd figure 3 mouth lαfu (A+ 'b 4th figure 5th figure!-v 1.2,)'・ □□□□-rei'J //, ζ36 figure P. Figure 7

Claims (3)

[Claims] (1) Information to be expanded x4.・・・・・・xn (However,
has a reference mask pattern PIIl consisting of at least binary information (where m is an integer of m≧1), and the information to be expanded x1...x,
The reference mask P □ If (however, an integer of 1≦1×e) is the first information, then Ma) IJ Nox A's a,] (however, j is an integer of iS×4) is located at I! I-information X of matrix A'i-N. ]+1 and move matrix A
a□, k (where k is an integer of i, -1 only when 1-1) information X located at a of matrix A +
On the other hand, if the reference mask P is the second information, a, p of matrix A (however,
An information decompression method characterized in that information is moved to 'i-N,p within an Xf matrix person, where P is an integer of 1sP≦e). (2) Su<Su< , 1! : MoIll row (I' column ((B L,, a, eVic=2, integer of e'≧2)
aq of matrix A, □− (however, q+ r is 2≦
An information transmission means is provided at a position where q≦(1+q > an integer of r), and 's, t of matrix A (however,
s and t are integers satisfying i<s≦l r ”≦t≦4′) and an information selection means is provided at the position of the first row to select the information x to be decompressed.
1. ......When moving xn (where n is an integer of n≧1) from the first row to the fourth row, the information selection means provided at the position of the first row is Corresponding to the mask pattern P1: 1ill III The information to be expanded according to the signal Xu lxu+l (however, U is 1≦
When u<n and u-1, the information selection means inputs one of u4-1=1 only when u-1, and the information selection means located at 2Ls, t of the matrix person responds to the control signal corresponding to the reference mask pattern Ps. te, Madosokusu A's &s-+,
t-+ + as-i, the information x2 of the information selection means located at either one of t, and the information transmission means located at hq, r of matrix A are located at matrix aQ'+r. an information expansion circuit that inputs information X from either the information selection means or the information transmission means; (3) The information decompression circuit according to claim 2, wherein the information transmission means is a signal line.