JPH01240984A - Digital image processor - Google Patents

Abstract

PURPOSE: To simultaneously realize the elimination of an impulse noise and the restoration of blurring by changing an image signal to be outputted as necessary. CONSTITUTION: An image processor is composed of a storage circuit 1 storing an image signal, a decision circuit 2 deciding the necessary/unnecessary of the correction of an image and an address circuit 3 designating the address of a picture element signal to be outputted in accordance with the decision result. In the decision circuit 2, noise and a blurred part are recognized, the picture element signal of this part is not outputted as it is, the signal is changed by the most approximate correction unnecessary picture element signal and the signal is outputted. As a result, noise can be eliminated, and at the same time, the contour of a graphic can be made conspicuous.

Description

[Detailed description of the invention] [Industrial application field] The present invention relates to an apparatus for processing digital image signals.

[Prior Art] The principle of conventional image processing technology is shown in Figure 3, and the conventional method is described below. The image signal of the required area is taken in from the circuit 41 and is processed by the image processing circuit 42, and the processed signal is directly output. The basics of image processing circuits in conventional technology, as typified by two-dimensional filters, is to find the correlation between a pixel signal to be processed and its adjacent pixel signals. It is necessary to repeat the high-level calculations.

The main targets of image processing are noise removal and blur correction. Generally, filtering is effective in removing noise, but as a side effect, the resolution of the image decreases and the outline becomes blurred.

However, if you try to correct this blur and emphasize the outline, noise will also be emphasized at the same time.

[Problems to be solved by the invention] Like this conventional image processing bag! Since the method performs high-level calculations on the image signal and outputs it, the amount of calculation becomes enormous, and it is difficult to speed up the processing and implement it in real time. Also,
The drawback was that the basic processing of removing noise and restoring blur could not be performed at the same time.

[Means for Solving the Problems] In order to solve the above-mentioned problems in the high speed and functionality of conventional image processing devices, the present invention provides a storage circuit for storing image signals and correction requirements for images. A distinctive feature of the image processing apparatus is that the image processing apparatus is composed of a determination circuit that determines whether the pixel signal is unnecessary, and an address circuit that specifies the address of the pixel signal to be output according to the determination result. In particular, a distinctive feature of the present invention is that it does not require means for directly processing and outputting image signals.

[Function] As shown in FIG. 1, the image processing device according to the present invention has a memory port n1 which sends out a pixel signal designated by an address signal given from the address circuit 3 to the decision circuit 2. 9. Impulse-like noise mixed in the image signal and blurred parts are recognized as minute areas where the image signal intensity changes discontinuously. 6. The judgment circuit 2 determines whether or not it is necessary to correct the pixel signal. is binarized and given to the address circuit 3. The address circuit changes the address signal depending on the presence or absence of correction. If there is no need for correction, the address is provided to the memory circuit 1 without being changed.

On the other hand, when correction is required, the address of the pixel closest to the pixel to be corrected among the pixels that do not require correction is determined and provided to the storage circuit 1. Memory circuit 1
outputs the pixel signal specified by the address signal given from the address circuit 3.

As described above, the 1 judgment circuit 2 recognizes noise and blurred parts, and does not output the pixel signals of these parts as they are.
By replacing the pixel signal with the nearest pixel signal that does not require correction and outputting it, it is possible to remove noise and at the same time make the outline of the figure stand out.

[Embodiment] FIG. 2 shows an embodiment of a digital image processing apparatus according to the present invention. In the embodiment shown in FIG. 2, the determination circuit 2 shown in FIG.
The address circuit 3 includes a storage circuit 31 and a shift register 32 . of 7 pixels x 7 pixels. It is composed of a priority circuit 33 and a pointer circuit 34.

The shift register 21 acts as a 3 pixel x 3 pixel window for the two-dimensional image signal stored in the storage circuit 11, and the determination circuit selects the image signal A (x@, 3/II ) to determine whether correction is necessary or not.

The comparison circuit 22 consists of eight double threshold circuits, and compares the image signal A (xe, y@) and the adjacent image signal A (x@÷i, ya+j) to detect changes in signal strength. To judge whether it is continuous or not, hereafter, to simplify the notation, (xe-y@) will be expressed as r@, and (x@+i, y
@+J) is set to r. The output C(r
) is given by. Here, α is a threshold value that distinguishes between continuity and discontinuity, and is a parameter that can be given from the outside. Furthermore, 1 and j take either one of 0 and ±1. C-1 means that the change in signal strength is continuous, and C-0 means that it is discontinuous.

The output signals of the eight double threshold circuits are applied to the next stage threshold circuit 23. The output R(r+
Riha becomes. Here, β is a threshold value for determining whether correction is necessary or not, and is also a parameter that can be given from the outside.

If the value of R (r@) is 1, there is no need to correct the image signal A (rs), and if it is 0, correction is necessary.

With the above procedure, 2. The judgment result R(r@) given by the judgment circuit is sent to the memory circuit 31 that constitutes the address circuit.
is stored in 7 bits×7 bits of data are taken out from the memory circuit 31 into the shift register 32, centering on R(r@). That is, the data stored in the shift register 32 is R(xs+p, y@+q), where p and q take any one of 0, ±1, ±2, and ±3.

All contents of shift register 32 are given to priority circuit 33. The priority circuit searches for bits whose data is 1, that is, pixels that do not require correction.
Starting from the center (p, q>= (0, 0) of , q) are given to the pointer circuit 34. The pointer circuit 34 uses previously given coordinate values (XS, y
Relative coordinate values (p, ll) given from the priority circuit 33 to
The coordinate value obtained by adding q) is converted into an address signal and provided to the storage circuit 11.

The memory circuit 11 stores the image signal A (x@+p", ye+q) stored at the address specified by the pointer circuit 34.
Output. Image signal A (x@, ys
) does not require correction, R=1 when (p, q)=(0,0), so A(xs, ys) is output as is. On the other hand, if correction is required, an image signal corresponding to the nearest bit (xs+P+ys+q) with R=1 is output.

The configuration shown in the embodiment of FIG. 2 does not limit the present invention in any way, and other modifications can be easily made.

[Effects of the Invention] The digital image processing device according to the present invention exhibits the outstanding effect of being able to simultaneously remove impulse noise and restore blur by replacing output image signals as necessary. This method significantly reduces the amount of calculation compared to the conventional method of directly processing and outputting image signals, and therefore can dramatically reduce the time required for image processing. Therefore, it is extremely effective in realizing real-time processing of digital images, which has been difficult with conventional techniques.

[Brief explanation of the drawing]

The drawings are all for explaining embodiments of the digital image processing apparatus according to the present invention, and FIG. 1 is a block diagram of the digital image processing apparatus according to the present invention, and FIG. 2 is a block diagram of the digital image processing apparatus according to the present invention. FIG. 3 is a block diagram of a conventional image processing device. 1st Figure 1 Memory circuit 1. Memory circuit 2 Judgment circuit 3 Address circuit 2nd Figure 11 Memory circuit 23 Threshold circuit 3rd (2i 41 Memory circuit 42 Image processing circuit

Claims (1)

[Claims] A storage circuit that stores a digital image signal, a determination circuit that determines whether or not correction is necessary for the pixel signal that this storage circuit is going to output, and a determination circuit that the storage circuit should actually output based on the output of this determination circuit. A digital image processing device comprising an address circuit that generates an address in which a pixel signal is stored and supplies this address signal to the storage circuit.