JPH06113140A - Picture processor - Google Patents

Abstract

PURPOSE:To provide a picture processor which can efficiently compress picture data while the block distortion of a reproduced picture is reduced. CONSTITUTION:An A/D conversion means 2 converting an input picture signal into digital picture data, and a data compression means 3 compressing and outputting digital picture data converted by the A/D conversion means 2 are provided far the picture processor. A filter means 1 consisting of plural filter circuits different in characteristics removing the high frequency component of the input picture signal or digital picture data and a filter selection means 11 selecting the filter circuits which make the input picture signal or digital picture data pass through them based on digital picture data compressed by the data compression means 3 are provided.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to A / D conversion means for converting an input image signal into digital image data, and the A / D conversion means.
The present invention relates to an image processing device configured by a data compression unit that compresses digital image data converted by the conversion unit and outputs the compressed digital image data to a storage unit or the like, for example, an image such as a video conference / video phone, a color still image, or a color moving image. The present invention relates to a device having a compression / expansion function for improving data transmission and storage efficiency for data, and an image processing device used for a device contributing to AVCC (audio / visual computer / communication) fusion.

[0002]

2. Description of the Related Art A conventional image processing apparatus sets a compression rate of digital image data by a compression means to an appropriate value for the purpose of efficiently storing it in a storage means having a limited capacity and compresses it. Was configured as. As data compression means, for example, international standardization has been achieved as an image compression method for improving data transmission and storage efficiency for original images.
The orthogonal row transform coding method using DCT (discrete cosine transform) is adopted, and the orthogonal transform coding method is
A method in which an image is divided into blocks of several pixels to several tens of pixels, orthogonal transformation is applied to each block, and the transform coefficient is encoded and transmitted. The receiving side performs orthogonal inverse transformation to reproduce the image. .

[0003]

However, in the above-mentioned conventional technique, since an appropriate compression rate is set in order to store the digital image data in the storage means having a limited capacity, the reproduced image is deteriorated. However, there is a drawback in that For example, not only the above DCT but also high efficiency encoding to reduce the average number of bits per pixel or increase the compression rate causes reverse conversion and reproduction of each pixel so that each block is reproduced. The direct current component is different, which causes deterioration of reproduced image quality, but at this time, distortion occurs between adjacent blocks, resulting in a discontinuous and unsightly image. The deterioration of the image becomes remarkable in the image having many high frequency components. On the contrary, when the image quality of the reproduced image is taken into consideration, there is a drawback that the compression rate is limited, the storage capacity to be prepared in the storage means increases, and the cost rises.
An object of the present invention is to eliminate the above-mentioned conventional drawbacks.

[0004]

To achieve this object, the image processing apparatus according to the present invention is characterized in that a filter means for removing a specific frequency component of the input image signal or digital image data is provided. is there. It is preferable that the filter means removes high frequency components. In the above-mentioned configuration, it is preferable that filter adjusting means for switching the characteristics of the filter means is provided based on the digital image data compressed by the data compressing means. Further, the filter means is provided with a plurality of filter circuits having different characteristics, and a filter circuit for selecting the filter circuit for passing the input image signal or the digital image data is selected based on the digital image data compressed by the data compressing means. It is preferable that selection means is provided.

[0005]

Even if the original image does not have a specific frequency component (for example, a high frequency component) even if the compression ratio is the same, the amount of data after compression is the same as that after the original image contains the frequency component. Paying attention to the fact that the data capacity is significantly reduced compared to the above, the compression efficiency is improved by removing the specific frequency component (for example, high frequency component) from the image data before compression by the filter means in advance. At that time,
Based on the digital image data compressed by the data compression means, for example, by switching the characteristics of the filter means by the filter adjustment means so as to reduce the data volume after compression, it is possible to further increase the compression rate. Similarly, the filter means is provided with a plurality of filter circuits having different characteristics, and based on the digital image data compressed by the data compression means, for example, by the filter selection means so as to reduce the data capacity after compression. By selecting the filter circuit, the compression rate can be further increased.

[0006]

Therefore, according to the present invention, a specific frequency component (for example, high frequency component) is added to the original image under a constant compression rate.
Even when the compressed data is included, it is possible to secure the same compression ratio as when the frequency component does not exist in the original image, and when the compressed data is reproduced after such high frequency component is removed. It is now possible to provide an image processing device capable of reducing block distortion.

[0007]

EXAMPLES Examples will be described below. As shown in Figure 1,
An image processing apparatus that handles a color still image includes a filter unit 1 that removes high-frequency components from an input image signal that is a video signal, and A / A that converts the input image signal that has passed through the filter unit 1 into digital image data for each pixel. The D conversion means 2 and the data compression means 3 for compressing and outputting the digital image data converted by the A / D conversion means 2 are included.

The filter means 1 is a low-pass filter for removing high frequency components, and includes a plurality of filter circuits 1a such as a primary filter, a secondary filter, and a tertiary filter having different characteristics at a constant high cutoff frequency, and a plurality of these filter circuits 1a. The filter circuit 1a is configured by an output of one of the filter circuits 1a or a selection circuit 1b that selectively passes an output that has not passed through the filter circuit 1a.

As shown in FIG. 2, the data compression means 3 stores one frame of digital image data of one frame converted by the A / D conversion means 2 as input image data and has 8 × 8 pixels. An image buffer 4 that outputs the reproduced image data in reverse while dividing and outputting for each block, and an orthogonal transformer 5 that performs DCT (discrete cosine transform) / inverse DCT (inverse discrete cosine transform) for each pixel block. , The transform output is quantized in accordance with a quantization table that presets the compression ratio, and then a zigzag scan for arranging the two-dimensional coefficients in one dimension is performed, and the inverse operation thereof is performed. A coder / decoder 7 for converting the original quantized data into a Huffman code and performing an inverse operation thereof, and a compressed data buffer 8 for storing the coded data. , It is constituted by the controller 9 to control their operation. That is, the data compression means 3
Also has a function as the data decompression means 10 for reproducing the compressed data.

Further, the filter means 1 is provided with a plurality of filter circuits 1a having different characteristics, and the input image signal is passed based on the compressed digital image data output from the compressed data buffer 8. A filter selecting means 11 for selecting 1a is provided.
More specifically, the filter selection means 11 counts the capacity of one frame of compressed digital image data, determines whether the count value is within a set capacity, and outputs the count value to the selection circuit 1b. When the selection circuit 1b receives a signal indicating that the capacity is not less than the set capacity, the selection circuit 1b selects the output of the higher-order filter circuit 1a as the compression target data.

Another embodiment of the present invention will be described below. In the above embodiment, the filter means 1 is provided before the A / D conversion means 2 for converting the input image signal, which is a video signal, into digital image data for each pixel, but the filter means 1 is A. It is also possible to provide it after the / D conversion means 2. Further, the filter circuit 1a constituting the filter means 1 is not particularly limited and can be appropriately constructed by a known technique. The image data input to the filter means can be filtered in any direction by using one-dimensional array data obtained by sampling the data in the block in any direction such as vertical, horizontal, or diagonal directions. Is.

In the above embodiment, the filter circuit 1a is composed of a primary filter, a secondary filter, a tertiary filter, etc., which have different characteristics at a constant high cutoff frequency.
The filter circuit 1a may be configured with different high cutoff frequencies.

In the above embodiment, the filter means 1 is provided with a plurality of filter circuits having different characteristics, and based on the digital image data compressed by the data compression means 3,
Although the filter selecting means 11 for selecting the filter circuit that allows the input image signal or the digital image data to pass therethrough has been described, a digital image compressed by the data compressing means 3 is obtained in order to obtain the same effect other than this. The filter adjusting means 12 for switching the characteristics of the filter means 1 may be provided and configured based on the data.

The input image signal is not particularly limited, and for example, by providing a converting means for converting into an RGB signal, it is possible to support any system such as the NTSC system and the PAL system.

In the above embodiment, an example in which the filter means for removing high frequency components is used has been described, but the filter means is not limited to this, and has a characteristic of removing frequency components in the middle range. Can be one,
The filter adjusting means and the filter selecting means may be operated so as to selectively remove an arbitrary frequency component in the middle to high frequencies.

In the above embodiment, the compression means 3 is JP
Although the example using the DCT (discrete cosine transform) which is the EG method has been described, the orthogonal transformer 3 is not limited to this, and may be configured by any transformer such as discrete Fourier transform or Hadamard transform. . Further, the present invention is not limited to a still image, but can be applied to an image processing device that compresses a moving image using the MPEG method or the like. In this case, the filter means 1 can be operated in the time axis direction for a plurality of frames.

It should be noted that reference numerals are added to the claims for convenience of comparison with the drawings, but the present invention is not limited to the configurations of the accompanying drawings by the entry.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of an image processing apparatus.

FIG. 2 is a block diagram of compression means.

FIG. 3 is a block configuration diagram of an image processing apparatus showing another embodiment.

[Explanation of symbols]

 1 Filter Means 2 A / D Conversion Means 3 Data Compression Means 11 Filter Selection Means 12 Filter Adjustment Means

Claims (4)

[Claims] 1. An A / D conversion means (2) for converting an input image signal into digital image data, and a data compression means for compressing and outputting the digital image data converted by the A / D conversion means (2). An image processing apparatus comprising (3) and a filter means (1) for removing a specific frequency component of the input image signal or digital image data. 2. The image processing apparatus according to claim 1, wherein the filter means (1) removes high frequency components. 3. The filter adjusting means (12) for switching the characteristic of the filter means (1) based on the digital image data compressed by the data compressing means (3). Image processing device. 4. The filter means (1) is provided with a plurality of filter circuits having different characteristics, and the input image signal or digital image data is obtained based on the digital image data compressed by the data compression means (3). Filter selection means (11) for selecting a filter circuit to be passed
The image processing apparatus according to claim 1 or 2, further comprising: