JPS63198420A - Picture compressing device - Google Patents

Abstract

PURPOSE:To prevent the appearance of a block joint of a restored picture by providing a 1st orthogonal conversion section applying orthogonal conversion to an original picture at each block, a 2nd orthogonal conversion section rearranging its output into the block at every same frequency component and applying orthogonal conversion at every block and a quantization section for the output of orthogonal conversion. CONSTITUTION:An original picture written in a memory 1 is split into blocks by an address control 2 and subjected to 2-dimensional high speed cosine conversion 3. The result is subjected to address control 8 and rearranged 7 into a block at every same frequency component. Then the result is subjected to 2-dimensional or linear high speed cosine conversion 10 and quantized (4). By such a constitution, the picture power is concentrated to decrease the error caused by the quantization. Thus, in restoring the compressed picture information generated by the compressor, the number of joints of blocks caused by the quantizing error is decreased, the S/N is improved and the excellent picture is obtained.

Description

[Detailed description of the invention] [Purpose of the invention] (Industrial application field) The present invention is an image compression device, especially for encoding images.

The present invention relates to an image compression device that compresses the apparent amount of data while preserving the amount of information by performing projection or conversion processing.

(Prior Art) When transmitting and receiving a digital image through a communication line, when storing it in a storage device, or when reducing the load on a communication line or a storage device, there is a method of expressing the original image with fewer words, a so-called method. Image compression is required, and there are two types of image compression methods: reversible compression and irreversible compression.

With reversible compression, the compressed image can be completely restored to the original image by using the run-length UDE ink method or the Huffman coding method, which is compressed using predictive coding, whereas non-reversible compression methods such as KLT and cosine transformation can completely restore the original image. Reversible compression allows a higher compression rate than lossless compression, but it cannot completely restore the original image, and the image quality deteriorates in proportion to the compression rate, so the image quality is not noticeable to the human eye. It is necessary to limit the compression ratio to a level that does not make it noticeable.

Orthogonal transform, especially block orthogonal transform, is well known for lossy compression, and among these, block cosine transform is considered the most practical and promising.

FIG. 5 shows a conventional image compression device that employs the block cosine transform method. As shown in the figure, this device includes a memory 1 that temporarily stores the captured original image, an address controller 2 that controls the address of this memory 1 and outputs a part of the image for each block, and a memory 1. It has an orthogonal transformation unit 3 that performs orthogonal transformation (for example, two-dimensional high-speed cosine transformation) on the original image in each block, and a quantization unit 4 that quantizes the output of this transformation. The quantizer 4 includes a bit allocation table 5 and a quantizer 6 that converts the transform output of the orthogonal transformer 3 based on the table information, and the output of the quantizer 6 is compressed. The image data is sent to an external device such as a storage device.

(Problems to be Solved by the Invention) However, when a compressed image by a conventional device that employs the block orthogonal transformation method is restored, block seams appear in the restored image, resulting in a decrease in image quality and diagnostic ability. points are occurring. This phenomenon is caused by a quantization error in low frequency components of the image, and becomes more significant as the compression rate increases.

The present invention has been made in view of the above circumstances, and its purpose is to create compressed image data that does not cause development of block seams or the like in the restored image, despite using the block orthogonal transformation method. An object of the present invention is to provide an image compression device that can generate images.

[Structure of the Invention] (Means for Solving the Problems) The present invention includes a first orthogonal transform unit that orthogonally transforms an original image block by block, and a first orthogonal transform unit that orthogonally transforms the orthogonal transform output into blocks of the same frequency component. It has a rearrangement section for arranging, a second orthogonal transform section that performs orthogonal transformation for each rearranged block, and a quantization section that converts the orthogonal transform output into a matrix.

(Function) The original image is orthogonally transformed block by block through the first orthogonal transformation, and the output of this transformation is rearranged by the rearrangement unit into blocks of the same frequency component, and each rearranged block is Orthogonal transformation is performed by the second orthogonal transformation unit. The power of the image is further concentrated by the rearrangement by the rearrangement unit and the orthogonal transformation by the second orthogonal transform unit, thereby reducing the quantization error in the mother-child generator. Therefore, it is possible to generate compressed image data that does not cause phenomena such as block seams in the restored image.

(Example) Hereinafter, the present invention will be specifically explained with reference to Examples.

FIG. 1 is a block diagram of an image compression VL system according to an embodiment of the present invention. The device of this embodiment includes a memory 1, a first address controller 2. First orthogonal transformation unit 3° rearrangement unit 7. Second orthogonal transformation unit 10. It has a quantization section 4.

The memory 1 temporarily stores the original image, the first address controller 2 outputs the original image block by block by controlling the address of the memory 1, and the first orthogonal transform unit 3 The original image output block by block from the memory 1 is orthogonally transformed (for example, two-dimensional high speed cosine transform) for each block, and both have the same functions as shown in FIG.

The rearrangement unit 7 rearranges the transform output of the first orthogonal transform unit 3 into blocks of the same frequency component, and includes a memory 9 and a second address controller that controls the address of this memory 9. 8. The second orthogonal transformation unit 10 performs orthogonal transformation for each block rearranged by the rearrangement unit 7. Note that the second orthogonal transform unit 8 performs two-dimensional (2-D) or one-dimensional (1-D>) high-speed cosine transform.

Then, the mother-child converting section 4 converts the transformed output of the second orthogonal transform section 8 into a mother-child, and as shown in FIG.
It comprises a bit allocation table 5 and a mother/son generator 6.

Next, the operation of the above configuration will be explained according to the flowchart of FIG. 2.

First, the original image written in the memory 1 under the control of the first address controller 2 (see FIG. 3(a)) is again written to FIG. 3(b) under the control of the first address controller 2. As shown in FIG. is orthogonally transformed (two-dimensional fast cosine transform) into (step 32>,
This converted output is sent to the rearrangement section 7. This rearrangement section 7 rearranges blocks for each same frequency component under the control of the second address controller 8 (step 3).
3). A more detailed representation of the orthogonal transform output in FIG. 3(C) is shown in FIG. 3(d), and FIG. 3(e) shows this rearranged into blocks for each frequency component. still,
In the figure, the difference in circles means the difference in frequency.

Then, for each rearranged block, the first orthogonal transform unit 1
0, an orthogonal transform (two-dimensional or one-dimensional fast cosine transform) is performed (step S4), and the transform output is sent to the quantizer 4, where it is quantized (step S35).
). FIG. 3(f) shows the transform output from the second orthogonal transform section 10.

In this way, in the device of this embodiment, the original image is orthogonally transformed block by block, the transform output is rearranged into blocks of the same frequency component, orthogonal transform is performed for each rearranged block, and this The conversion output is quantized,
Unlike conventional methods, the original image is transformed block by block and quantized immediately after orthogonal transformation, so it is possible to concentrate the power of the image more than before, and reduce errors caused by quantization. can. Therefore, when compressed image data generated by the device of this embodiment is restored,
"Block seams" caused by quantization errors are reduced, and images with excellent diagnostic performance can be obtained by improving the S/N ratio. In addition, at the same compression rate, quantization can be performed using fewer bits and with smaller errors than conventional devices, so a large number of bits can be allocated to high frequency components, which improves the spatial resolution and density resolution of the restored image. It has the advantage that it can also increase Roughly speaking, since a restored image of the same quality as before can be obtained with a smaller number of bits, it is also possible to increase the compression rate compared to the conventional method.
It is possible to shorten the transfer time of compressed image data and save storage capacity.

Although one embodiment of the present invention has been described above, it goes without saying that the present invention is not limited to the above embodiment, and various modifications can be made.

For example, in the above embodiment, the second orthogonal transformation section 10 performs orthogonal transformation on all blocks rearranged by the rearrangement section 7, but the orthogonal transformation is not required for the rearranged blocks. Since there are also blocks, the second orthogonal transform unit 1
It is preferable to send the signal to the quantization unit 4 without passing through 0.

This will be more advantageous than the above embodiments in terms of image quality and processing time reduction. An example of the configuration in this case is shown in FIG. In the figure, a rearrangement section 7 and a second orthogonal transformation section 10 are shown.
A block selector 11 is disposed between the block selector 11 and the block selector 11 to divide the blocks into blocks that require orthogonal transformation by the second orthogonal transform unit 10 and blocks that do not. Whether or not orthogonal transformation is required can be clearly determined from the address of the rearranged block, so block selection can be easily achieved by controlling the address. Additionally, the pit allocation table of the slave unit is also switched as necessary.

[Effects of the Invention] As detailed above, according to the present invention, compressed image data that does not cause development of block joints etc. in the restored image can be generated, even though the block orthogonal transformation method is adopted. It is possible to provide an image compression device that can perform

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a flowchart for explaining the operation of the device of this embodiment,
3(a) to 3(f) are explanatory diagrams of various processes in the apparatus of this embodiment, FIG. 4 is a block diagram showing another embodiment of the present invention, and FIG. 5 is a block diagram of a conventional example. 3... First orthogonal transformation unit, 4... Quantization unit, 7...
- Relocation unit, 10... second orthogonal transformation unit.

Claims (1)

[Claims] An image compression device that compresses an original image includes a first orthogonal transformation unit that orthogonally transforms the original image block by block, a relocation unit that rearranges the orthogonal transformation output into blocks of the same frequency component, and a rearrangement unit that rearranges the orthogonal transformation output into blocks of the same frequency component. The second step is to perform orthogonal transformation for each block.
An image compression device comprising: an orthogonal transform unit; and a quantizer that quantizes the output of the orthogonal transform.