JPH05284485A - Picture correction device - Google Patents

Abstract

PURPOSE:To improve the picture quality by means of line interpolation by reading picture data compressed for each optional block and an error flag attended with the picture data and using an arithmetic mean value of picture elements around a picture element having an error so as to correct the erroneous picture element when any error is caused at the demodulation of the picture. CONSTITUTION:The device is provided with a data/flag read circuit 1, a compressed picture demodulation circuit 2, a flag use memory 3, a picture memory 4, a video output circuit 5, a computing element 6 and a correction arithmetic operation program 7. An erroneous picture element corresponding to an error flag is corrected by replacing the picture element with the arithmetic mean value of picture elements of front/back/left/right/oblique corner positions around the erroneous picture element as the arithmetic operation method. That is, the correction arithmetic operation program 7 retrieves the error flag memory 3 and obtains the correction value of a part P11 if the P11 is in error. This utilizes it that surrounding picture elements of a picture element have a large correlation and it is remarkable when the picture is closer to a natural picture.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image correction device which, when a compressed image has an error, corrects and outputs the image data having no error.

[0002]

2. Description of the Related Art In recent years, images have come to be recorded on a medium such as a CD (compact disc). Further, since the capacity of the medium is also limited, image data has been compressed and recorded. The compression method is also a natural image,
Various compression methods have been devised depending on the type of image such as a moving image, an animation image, and a graphic image.

When reproducing from a medium, a reading error occurs due to adhesion of dust or the like. When an error occurs, a method of replacing the corresponding line with the data of one line before the scanning line of the television is often used.

[0004]

However, the above-described method of correcting on a line-by-line basis has a problem that the quality of the image is deteriorated.

The present invention solves the above-mentioned problems, and an object of the present invention is to provide an image correction apparatus which corrects an erroneous pixel with an arithmetic mean value of pixels around an erroneous pixel.

[0006]

In order to achieve this object, the present invention provides a means for reading image data compressed for each arbitrary block and an error flag attached to the compressed image data, and compressed image data. Means for restoring the original, memory means for storing the restored image data, memory means for storing an error flag, means for outputting the restored image data from the memory as a video, and reading the data from the memory An arithmetic means for performing arithmetic operation and writing data in a memory, a memory means for storing an arithmetic method, and an arithmetic method calculates an erroneous pixel corresponding to an error flag by calculating an arithmetic average with values of front, rear, left and right around the pixel, and diagonal corners. It has to go and replace and correct.

[0007]

The present invention reads the image data compressed for each arbitrary block and the error flag attached to the image data,
When an error occurs at the time of demodulating an image, correction is performed not for each line but for each pixel, and the deterioration of the quality of the image is minimized and output.

[0008]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 shows a compressed image sequence. FIG. 3 shows an error flag string attached to the image string. One image is composed of n pixels in the horizontal direction and m pixels in the vertical direction. That is, it is a set of pixels of n columns × m rows. The compressed image sequence is compressed row by block, and one block includes pixels for n columns. An initial value is entered at the beginning of each block, and a difference value with each pixel is entered after that. As an example, 30, 35, 40, 60, 2
0 ,. ． ． Is the original pixel, the initial value is 0 = 2
If 0 is determined, the difference is 00 = 30−20 = 10. Next, difference 01 = 35-30 = 5, difference 02 = 40-35 =
5, difference 03 = 60-40 = 20, difference 04 = 20-60 =
It becomes -40. These values are performed over n columns × m rows. When an error occurs in each data when read from the storage medium in association with those image strings,
The error flag is set.

FIG. 1 shows the configuration of an embodiment of the image correction apparatus according to the present invention. 1 is a data / flag reading circuit,
Reference numeral 2 is a compressed image demodulation circuit, 3 is a flag memory, 4 is an image memory, 5 is a video output circuit, 6 is a calculator, and 7 is a correction calculation program. When the image string data of FIG. 2 and the flag string of FIG. 3 enter the data / flag reading circuit 1, the image string data is sent to the compressed image demodulating circuit 2. FIG. 4 shows the configuration of the compressed image demodulation circuit 2. Reference numeral 10 is a difference circuit, 11 is an initial value circuit, and 12 is an addition circuit. The initial value at the beginning of the block is entered and stored in the initial value circuit 11. Subsequent difference values sequentially enter the difference circuit 10. First, the initial value 0 is stored in the initial value circuit 11. When the next difference 00 enters the difference circuit 10, the addition circuit 12 outputs the addition value of P00 = initial value 0 + difference 00. When the next difference 01 is input, the output of the initial value circuit 11 becomes the previous value P00,
From the adder circuit 12, P01 = P00 + difference 01. This is repeated until the difference is 0n. These demodulated image data are sequentially stored in the image memory 4 so that the image of FIG. 5 constitutes one image. On the other hand, the flag string data is sequentially stored in the flag memory 3 with the configuration of FIG. The data in the image memory 4 is output as an image from the video output circuit 5 if the read image data has no error.

If an error has occurred, the output value will be a different image. Assume that the flag F11 detects that an error has occurred in the pixel P11 as in the example of FIG.
Then, since the image sequence is a compressed image sequence, P11
After that, the image values up to P1n are random values.

The arithmetic unit 6 is usually called a CPU, and is internally equipped with an adder / subtractor, a multiplier / divider, and a memory value read / write interface. The correction calculation program 7 is a read-only memory in which a program for controlling the CPU for performing the correction calculation of the pixels is stored.

The correction calculation program performs the following processing.
Here, it is assumed that a search is performed from the error flag memory 3 and that the P11 portion has an error. Therefore, the correction value of P11 is calculated by the following calculation.

P11 = (Ka × P00 + Kb × P01 + Kc × P02 + Kd × P10 + Ke × P20 + Kf × P21 + Kg × P22) / 8 The coefficient from Ka to Kg is arranged as shown in FIG. 8 (a). Is shown in FIG.
(B) and (c) show examples of coefficient values. Where P11
When the correction value of is calculated, P12 to P1n are automatically obtained. In general, the fact that the periphery of one pixel is highly correlated is utilized, and the closer it is to a natural image, the more remarkable it is.

As described above, according to this embodiment, the error pixel is corrected by the arithmetic average value of the pixels around the error pixel, and the image quality by the line interpolation can be further improved. It is a correction device.

[0016]

As described above, according to the present invention,
Since the image quality deteriorates in the method of correcting in line units, error pixels are corrected by the arithmetic mean value in pixel units, and natural images and the like have a particularly large practical effect.

[Brief description of drawings]

FIG. 1 is a block diagram showing the configuration of an image correction apparatus according to an embodiment of the present invention.

FIG. 2 is a schematic diagram showing a compressed image sequence in the embodiment.

FIG. 3 is a schematic diagram showing an error flag string associated with an image string in the embodiment.

FIG. 4 is a block diagram showing an internal configuration of a compressed image demodulation circuit in the embodiment.

FIG. 5 is a schematic diagram showing a memory arrangement and an image of demodulated images in the embodiment.

FIG. 6 is a schematic diagram showing a memory arrangement of error flags in the embodiment.

FIG. 7 is a schematic diagram showing an example of an error occurrence position in the embodiment.

FIG. 8 is a schematic diagram showing an arrangement of image correction coefficients and an example in the same embodiment.

[Explanation of symbols]

 1 data / flag reading circuit 2 compressed image demodulation circuit 3 flag memory 4 image memory 5 video output circuit 6 calculator 7 correction calculation program 10 difference circuit 11 initial value circuit 12 adder circuit

Claims (1)

[Claims] 1. A data / flag reading means for reading image data compressed for each arbitrary block and an error flag attached to the compressed image data, and a compressed image demodulating means for restoring the compressed image data based on the compressed image data. An image memory unit for storing the restored image data; a flag memory unit for storing the error flag; a video output unit for outputting the restored image data from the image memory unit as a video; The data from the image memory means is read and calculated,
Computation means for writing data to the image memory means and memory means for storing the computation method of the computation means are provided, wherein the computation method detects an erroneous pixel corresponding to an error flag,
An image correction device characterized by performing arithmetic averaging with values of front, rear, left and right around the pixel, and values of diagonal corners for replacement and correction.