JPS641992B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an image display method for displaying an image signal including halftones on an image output device capable of recording a finite number of density levels.

Since devices that can accurately record halftone images are difficult to obtain, an image display method using a commonly used black-and-white binary recording device or a device capable of recording several levels is recommended. There is a lot of research going on.

A typical example of this display method is the "dither method," which is known as an extremely useful method.

The present invention attempts to obtain a high-quality recorded image by giving an idea to threshold value determination in halftone image display using the dither method.

First, the conventional dither method will be explained.
Assume that the screen is divided into 4×4 squares a as shown in FIG. Assuming that the number of recordable levels of the recording device is 2 and the recording density is 0 (white) and 1 (black), the threshold value of each pixel is preferably set to 2i - 1/32 (i = 1 to 16), and the second When the numbers determined as shown in the figure are made to correspond to the order of the threshold values, the threshold values are determined as shown in FIG. 3.

Therefore, when inputting an image signal of a certain level,
The 17 patterns that appear are shown in Figure 4. That is, if the screen is divided into blocks of n pixels and a different threshold value is assigned to each pixel within the block, (n+1) levels of gradation can be expressed when viewed over a wide range.

The above is the idea of the dither method, and the size and shape of the blocks can be arbitrary as long as they fill the plane without any gaps. Furthermore, a dither method can be considered when performing multi-value recording by giving a plurality of threshold values to each pixel.

Now, in the dither method, in order to increase the number of tones over a wide range, it is necessary to increase the block size, but if this is simply done, the equivalent resolution will drop, especially in the image signal. There was a problem that image quality deteriorated in areas with rapid changes.

The present invention provides a recorded image in which the equivalent resolution does not deteriorate as much as possible even when the number of gradations seen over a wide range is increased by adding an ingenuity to the method of determining the threshold value.

Next, the effects of the present invention will be explained by giving specific examples. FIG. 5 shows a conventional example, in which the block shape is a diamond, and the size of block b is 32 pixels. Therefore, if each pixel in block b is numbered as shown in the figure and a different threshold value is assigned to each pixel, 33 gradations can be expressed over a wide range.

FIG. 6A shows the first step in an embodiment of the present invention, in which the size of block c is set to 8 pixels, and numerical values from 1 to 8 are given regularly according to the position within block c. ing.

Figure 6B is an example in which numbers 1 to 4 are regularly assigned to block c, and Figure 6C is the final threshold value obtained from the resulting formula k=m(i-1)+j. It is an ordinal value (1-32). That is, Figure 6C
Even if the threshold order is adopted, 33 levels of gradation will be expressed over a wide range. Next, regarding the specific threshold level, the threshold order is k (k = 1 to 32).
If the range of the image signal is 0 to 1, the kth threshold value is (2k-1)/64. This is the same in both the conventional example and the present invention.

Now, as an example of an image signal that changes rapidly, the advantages of the present invention will be illustrated by taking up the case of a vertical stripe signal in which pixel signal levels 23/32 and 7/32 are repeated in units of three pixels.

Figure 7 shows the signal after binarization when such an image signal is binarized according to the value order shown in Figure 5, and the 〇 mark is the part displayed in "black". . Similarly, FIG. 8 is a result of binarization according to the threshold value order of FIG. 6C.

As is clear from comparing Figures 7 and 8,
It can be seen that the vertical and striped pattern of the original signal is better preserved when the threshold order shown in FIG. 6C is followed.

This improvement is particularly remarkable when the original signal changes rapidly, and is a remarkable manifestation of the effects of the present invention.

As described above with reference to the embodiments, according to the present invention, it is possible to provide a recorded image with a large number of pseudo gradations while maintaining a high equivalent resolution.

It goes without saying that the size and shape of the block may be other than the one embodiment of the present invention.

Further, even when using an apparatus capable of recording several levels, the present invention can be easily expanded by applying the present invention to each threshold value.

[Brief explanation of drawings]

Fig. 1 is a block diagram showing an example of dividing the screen into 4 x 4 square blocks, Fig. 2 is a block diagram showing the order of threshold values within a block, Fig. 3 is a block diagram showing threshold levels, and Fig. 4 is a block diagram showing an example of dividing the screen into 4 x 4 square blocks. The figure is a configuration diagram showing the gradation seen over a wide range, Figure 5 is a configuration diagram showing the threshold order in a conventional large diamond block, and Figure 6A is a configuration diagram showing the threshold order in a conventional small diamond block. Figure, Figure 6B
6C is a block diagram showing the order of threshold values for each block, FIG. 6C is a block diagram showing the order of threshold values according to an embodiment of the present invention, FIG. FIG. 8 is a configuration diagram showing an example of a recorded image according to an embodiment of the present invention for the same signal example. In the figure, a, b, and e indicate blocks.

Claims (1)

[Claims] 1 Divide the screen into blocks of n pixels of equal shape and area, assign a number i (i = 1 to n) to each position in the block according to the rules, and then Assign a number j (j = 1 to m), and assign k = m to the i-th pixel of the j-th block.
After calculating the threshold number k by calculating (i-1)+j, the kth value of the mn predetermined thresholds is assigned as the threshold, and the input image signal is converted to binary or multivalued using that threshold. An image display method characterized by digitizing and recording data.