JPS6026346B2 - Gradation image processing method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to a gradation image processing method for transmitting gradation images, and particularly to a method for nonlinearly converting gradation differences between adjacent pixels.

FIG. 1 is a block diagram showing a conventional gradation image processing apparatus of this type.

In FIG. 1, 10 is a subtracter, 11 is a gradation memory, 1
2 is a subtracter, 13 is an error memory, and 14 is a difference conversion ROM.
, 15 is an encoding circuit, and 16 is a transmission memory. Sa is the gradation value of the input image signal with one-part recognition, Sb is the gradation value of the preceding pixel, Sc is the gradation difference, Se is the gradation difference corrected by the error Sd in the preceding pixel, and Sf is the conversion difference. ,Sg
is the error corresponding to the conversion difference Sf, and Sh is the transmission code. FIG. 2 shows the tapered quantization code characteristics of the difference conversion ROM 14, with the horizontal axis representing the gradation difference Se and the vertical axis representing the conversion difference Sf.

This characteristic indicates that 31 differences in the gradation difference Se are non-linearly converted into 11 differences in the conversion difference Sf, reducing the number of differences and emphasizing the differences. Figure 3 shows the gradation value Sa, the gradation difference Se, and the conversion difference S.
This shows a modification of f.

As shown in FIG. 3, when a rapid gradation change continues in the gradation value Sa of the input image signal, the error in the difference conversion ROM 14 is excessively reproduced based on Sg, and the conversion difference Sf
ringing occurs. For example, when the gradation difference Se has a value of 0 as shown by Sea, the conversion difference Sf may take a value of -1 as shown by Sfa. On the other hand, the smaller the difference, the shorter the code length is assigned to the transmission code Sh. Therefore, in such image processing,
This means that the code length becomes longer due to differential conversion, which is a partial drawback.

Therefore, the present invention aims to eliminate such momme points, and achieves this by limiting the amplitude of reproduced gradations on the transmitting side to a certain predetermined range. Examples will be described below. FIG. 4 is a block diagram showing one embodiment of the present invention.

In FIG. 4, a subtracter calculates the difference between the gradation value Sa of the input image signal having 20' or 4 bits and 1 pitch and the reproduced gradation value Si of the preceding pixel, and outputs the gradation difference Sk. Yes, 2
Reference numeral 1 denotes a difference conversion ROM which outputs a conversion difference SI having the tapered code quantization characteristics shown in FIG. 2 and shown in the following table. Reference numeral 22 denotes an adder that calculates the sum of the conversion difference SI and the gradation value Si of the preceding pixel, reproduces the gradation value Sa on the transmitting side, and outputs the first reproduced gradation value Sm.

23, as shown in FIG. 5, when the reproduced gradation value Sm is 0 or less or 1 ratio higher, the gradation value Sm is clamped to a gradation value of 0 or 15, and the range of the gradation value Sa of the input image signal is This is a clamp circuit that limits the

Reference numeral 24 denotes a gradation memory that stores the clamped reproduction gradation Sn as a reproduction gradation for predicting subsequent pixels.

25 is an encoder that converts the conversion difference into a transmission code So according to the following table, and 26 is a transmission memory. Input difference Output code Code length +15 111110 6 18 111100 6 14 11100 5 12 1100 4 11 100 3 0 0 1 -1 101 3 -2 1101 4 -4 11101 5 -8 111101 6 -15 111111 6 6th The figure shows various waveforms in FIG. 4.

In FIG. 6, the waveform of the gradation difference Sk shows the difference between the gradation value $a and the reproduced gradation value Si (not shown),
The waveform of the conversion difference SI is the gradation difference S according to the characteristics shown in FIG.
This shows that k has been gradually decreased from 31 to 11 gradations, and the difference has been emphasized at large difference values. The waveform of the first reproduced gradation value Sm shows that the gradation value Sa is reproduced by the conversion difference SI and the reproduction scale j. In addition, as shown in black in this case, the first reproduction gradation value Sm is the gradation value S
This shows that there is a portion of excessive regeneration that deviates from the range of 0 to 15 in a. The waveform of the reproduced gradation value Sn is obtained by clamping the deviated portion of the primary reproduction gradation value Sm. The excessive playback part is a differential conversion ROM
This is caused by the gradual reduction in the number of gradations of the difference by 21, and the reproduced gradation value Sn from which this excessively reproduced portion is removed better approximates the gradation value Sa in the input image signal, and is also suitable for transmission. This increases efficiency, which will be explained next. The following table shows numerical examples of various signals in FIGS. 1 and 4, and Sh and So indicate the code lengths of the codes.

As is clear from these numerical examples, the method according to the present embodiment can transmit a total of 37 bits, whereas the conventional example requires 45 bits, and the present embodiment can transmit the conversion difference SI to be transmitted in this way. This has the effect of increasing the transmission efficiency by shifting the code to a shorter code length.

As is clear from the above description, since the number of encoding bits can be reduced, it can be used in facsimiles that transmit gradation, band compression for television, image storage devices, and the like.

Brief Description of the Drawings Fig. 1 is a block diagram of a conventional image processing device, Fig. 2 is a coded code quantization characteristic diagram, Fig. 3 is a waveform diagram of each part in Fig. 1, and Fig. 4 is a diagram of the present invention. FIG. 5 is a block diagram showing one embodiment, FIG. 5 is a characteristic diagram of the clamp circuit in FIG. 4, and FIG. 6 is a waveform diagram of each part in FIG. 4.

20...Subtractor, 21...Difference conversion R
OM, 22... Adder, 23... Clamp circuit, 24... Gradation memory, 25... Encoder, 26... Transmission memory, Sa...
Gradation value, Sk...... Gradation difference, SI...
Conversion difference, Sm... Reproduction gradation value, Sn...
...Reproduction tone value.

Figure 1 Figure 2 Figure 3 Figure 4 Figure 5 Figure 6

Claims (1)

[Claims] 1 The difference between the gradation value of the relevant pixel in the input image signal and the reproduced gradation value of the preceding pixel is created as a gradation difference, and the gradation difference is changed according to predetermined nonlinear characteristics. In a gradation image processing method in which a conversion difference is created and a code with a shorter code length is assigned to a smaller conversion difference, the sum of the conversion difference and the reproduced gradation value is calculated, and the sum is calculated. A tone image processing method, characterized in that it creates reproduced tone values for subsequent pixels by limiting the range of tone values in an input image signal.