JPH01225293A - Orthogonal transformation encoder and extension reproducing device for video signal - Google Patents

Abstract

PURPOSE:To quantize, compress and encode suitably for the characteristic of a picture signal by dividing the picture signal into plural deformed blocks, orthogonally transforming blocked data, encoding and recording on a recording medium. CONSTITUTION:The picture signal from an image pickup device 14 is inputted to a block circuit 22 through an A/D converting circuit 18 and divided into the plural deformed blocks. The blocked picture data is inputted to an orthogonal transforming circuit 24, orthogonally transformed, further inputted to an encoding circuit 26 and encoded. This encoded picture data is recorded in a memory 32. The block circuit 22 divides the picture data into the plural blocks so as to have an irregularity in the boundary of the adjacent blocks.

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an orthogonal transform encoding device for image signals, and more particularly to an orthogonal transform encoding device that performs orthogonal transform and encoding of an image signal captured using a solid-state imaging device. .

For example, when storing an image signal captured by a solid-state imaging device such as a COD in a storage device such as a memory card or magnetic disk, the data of one image signal must be reduced to a small capacity in consideration of the capacity of the storage device. It is necessary to compress. Orthogonal transform encoding is known as one of such image data compression methods. This method is as follows.

First, the image represented by one image signal is divided into a predetermined number of blocks (nX
m), and the data of each pixel in each divided block is orthogonally transformed.

In the image signal, the low frequency component occupies a large component in terms of power. On the other hand, although high frequency components are not large in terms of power, they are significant in terms of information.

Moreover, the characteristics for these are also visually different.

Therefore, the image signal is converted into such low frequency components and high frequency components, and quantization appropriate for each component is performed.
encode. The encoded signal is inversely transformed on the receiving or reproducing side to obtain the original signal. In this way, efficient encoding can be performed.

In orthogonal transform encoding, a screen is divided into a plurality of blocks, each block having an appropriate number of pixels, and a numerical string consisting of sample values is orthogonally transformed for each block.

That is, linear transformation is performed using mutually independent transformation axes that match the characteristics of the original image signal. As a result, each transformed term becomes more independent (more uncorrelated) than the original sample value.

This suppresses redundant information. This method is, so to speak, an operation on the frequency axis.

As a result, power is concentrated on a specific component due to the statistical properties of the 5-image signal. Therefore, while taking visual characteristics into consideration, a large number of bits are allocated to low-frequency components with high power, and high-frequency components with low power are coarsely quantized using a small number of bits. This allows the number of bits per reblock to be reduced.

By performing orthogonal transformation and encoding in this manner, the capacity of the storage device can be reduced compared to the case where data of pixels constituting one image signal is stored.

By the way, in conventional orthogonal transform encoding, the screen is divided into blocks in such a way that the pixel data included in each block is arranged squarely. If the data of the compressed image signal is decoded and inversely transformed by the playback device after being divided into square blocks and then encoded by orthogonal transformation and stored in the storage device, the reproduced image often contains There is a problem in that square block boundaries appear, and the reproduced image is impaired by the occurrence of such block boundaries.

Purpose The present invention solves the drawbacks of the prior art, and provides orthogonal transform encoding of image signals that does not cause visual effects due to block division on the output screen when reproducing image data compressed and recorded on a recording medium. Device.

Another object of the present invention is to provide an image signal decompression/reproduction device for reproducing image data recorded by this device.

According to the present invention, an image signal captured by an imaging means is received, the image data of the image signal is divided into blocks, and then the image signal is orthogonally transformed and encoded and recorded on a recording medium. The transform encoding device includes a blocking means that blocks one image data, an orthogonal transform means that orthogonally transforms the image data reblocked by the blocking means, and an encoder that encodes the image data that has been orthogonally transformed by the orthogonal transform means. and a recording means for recording the image data encoded by the encoding means on a recording medium, and the blocking means encodes the image data so that the boundaries between adjacent blocks are uneven. It is divided into multiple blocks.

Further, according to the present invention, an image signal decompression/reproduction device reads compressed and encoded image data from a recording medium recorded by the above-described image signal orthogonal transform encoding device, and decompresses and reproduces the image data, from the recording medium, a data reading means for reading compression encoded image data; an expansion decoding means for expanding and decoding the compression encoded image data read from the recording medium by the data reading means; and expansion decoding by the expansion decoding means. and a block synthesis means for synthesizing the decoded image data for each block, and synthesizes the decompressed and decoded image data for each block so that the boundaries between adjacent blocks are uneven.

DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, embodiments of an orthogonal transform encoding apparatus for image signals according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 shows an embodiment in which an image signal orthogonal transform encoding device according to the present invention is applied to a digital electronic still camera. Note that other parts of the camera that are not directly related to the description of the present invention, such as mechanisms such as a shutter, an aperture, and a film, are omitted from illustration.

This device has a master lens 12, and the master lens 1
An imaging device 14 that converts an optical image of a subject captured by the master lens 12 from an optical signal to a video signal is arranged behind the lens 2, and a color filter 1B is provided on the surface of the imaging device 14. The imaging device 14 converts an optical image of a subject from an optical signal to a video signal in response to a synchronization signal sent from the synchronization signal generation circuit 34 through a signal line 120.

The image signal obtained by the imaging device 14 is transmitted through the signal line 1
02 to the AD conversion circuit 18. The AD conversion circuit 18 converts the image signal sent from the imaging device 14 into a digital signal. Although not shown, between the imaging device 14 and the AD conversion circuit 18, gamma correction,
A means of white balance processing may be inserted, AD
The color image signal converted into a digital signal by the conversion circuit 18 is passed through the signal line 104 to the blocking circuit 22.
sent to.

The blocking circuit 22 performs blocking by dividing the image signal composed of RGB component pixels inputted from the AD conversion circuit 18 into a predetermined number of blocks. The area is divided into a plurality of areas 80a, Bob, . . . , blocks as shown. In this embodiment, data composed of pixels as shown in FIG. 3A is divided into modified blocks each including 18 pixels as shown in FIG. 38. This block is a modified block that is set so that every other row is shifted by one pixel to the left or right.

The image signal transformed into a modified block by the blocking circuit 22 is input to the orthogonal transform circuit 24 through the signal line 108. The orthogonal transformation circuit 24 performs orthogonal transformation on each block of the pixel signal that has been formed into a modified block. When the pixel signal of each block is divided into modified blocks, each pixel is It has a level value of . After replacing this with a square arrangement as shown in FIG. 4B, orthogonal transformation is performed. In the example of FIG. 4B, the topmost leftmost pixel has a level of 120 in the digital data, the pixel to its right has a level of 127, the third pixel has a level of 108, and the top two The leftmost pixel is at level 107, and the pixel to its right is at level 120.

When this is orthogonally transformed, data as shown in FIG. 4C, for example, is obtained. Orthogonal transformations include Hadamard transform,
Cosine transform, Fourier transform, etc. are known. In the orthogonally transformed data as shown in FIG. 4C, the horizontal frequency component of the original screen corresponds to the horizontal frequency component of the original screen in the horizontal axis direction, and the vertical frequency component of the original screen corresponds to the vertical axis direction. Furthermore, in data arrangement 1, data of low frequency components are arranged at the upper left, and data of high frequency components, that is, data with large change values between adjacent pixels, are arranged towards the right or downward. .

As mentioned earlier, in a general image, the low frequency component occupies a large component in terms of power, and the high frequency component occupies only a small component, so the data after orthogonal transformation as shown in Figure 4C is Larger values appear in the lateral direction, and the values become smaller toward the right and downwards.

The signal subjected to orthogonal transformation in the orthogonal transformation circuit 24 is sent to the encoding circuit 26 through the signal line 110.

The data is encoded in the encoding circuit 28 using encoding data sent from the lookup table 38.

Encoding is performed by converting matrix data as shown in FIG. 4C into each data ? For example, assign a predetermined number of bits to 6 as shown in FIG. 4D, and assign 8 bits to data 200 in FIG.
, 150 is 6 bits, data 100.90.40 .
4 bits for 70.80.

Two bits are assigned to data 50.50.10.5 and 10.80.20, respectively, and these data are encoded. No bit number is assigned to data located to the right or below these data; in other words, data located to the right or below a predetermined range in the data after orthogonal transformation is ignored.

I don't remember.

The reason why only low frequency components are memorized and high frequency components are ignored is that in general images, most of the components are low frequency components, so the image can be generally reproduced even if high frequency components are ignored. be.

The image divided into blocks in the blocking circuit 22 is 1 as shown in FIG. 5A, for example, and 0 as shown in FIG. 5B. For example, the divided blocks 82a in the figure are of the same color. Since it is a picture, data of low frequency components whose pixel signal level changes little are arranged within the block, and the orthogonally transformed data is concentrated in the upper left. Therefore, in the encoding circuit 2B, for example, a large number of bits are assigned to the data concentrated in the upper left, and less bits are assigned to the data on the right and lower. Necessary data is read out from the look-up table 38 to the encoding circuit 2B, and encoding is performed accordingly.

In contrast, for example, the divided block 13 of FIG. 5B
Since 2h is a fine picture, the level of the pixel signal changes greatly within the block, data of high frequency components are arranged, and the orthogonally transformed data is distributed to the right and downward. Therefore, the data is read out from the lookup table 38 to the encoding circuit 26 so that the data is encoded by assigning a number of bits widely distributed to the right and downward.

The look-up table 38 stores data suitable for encoding according to the pattern of each divided block.
The data is read out to the encoding circuit 28, and the orthogonally transformed data is encoded according to the picture pattern of each block.

The data encoded in the encoding circuit 28 is output to the output terminal 30 through the signal line 112.
The data is stored in a memory 32 connected to. As the memory 32, a so-called memory card in which a semiconductor memory or the like is mounted on a card-shaped substrate is advantageously used, and four coded still images are stored therein. The memory 32 is preferably one that can be attached to and detached from the output terminal 3o, for example.

The synchronization signal generation circuit 34 is connected to the signal line 12B from the control unit 38.
A control signal sent through generates a synchronization signal,
signal line 120 to imaging device 14; signal line 122;
The signals are respectively output to the AD conversion circuit 18.

, the control unit 38 is a control unit that controls each functional unit of this device, and connects the signal line 128 to the synchronization signal generation circuit 34, the signal line 12 to the reblocking circuit 22, and the signal line 130 to the reblocking circuit 22.
Control signals are output to the orthogonal transform circuit 24, the signal line 132 to the encoding circuit 2B, and the signal line 13B to the lookup table 3, respectively, to control the operation of each part.

The operation of this device will be explained.

The optical image of the subject captured by the master lens 12 is converted from an optical signal to a video signal by the imaging device 14 and sent to the AD conversion circuit 18 through the signal line 102.

The video signal is converted into a digital signal in the AD conversion circuit 18 and sent to the blocking circuit 22 through the signal line 104. The image data input to the blocking circuit 22 is divided into a plurality of deformed blocks as described above in the blocking circuit 22, and sent to the orthogonal transform circuit 24 through the signal line 10B. The orthogonally transformed data is sent to the encoding circuit 26 through the signal line 110.

The orthogonally transformed data sent to the encoding circuit 2B is encoded by the encoding circuit 2B according to the pattern of each block and lock, and is written into the memory 32 from the connector 30 through the signal line 112.

The still image photographed by the electronic still camera in this manner is recorded in the memory 32 such as a memory card.

Figure 2 shows the image taken by the electronic still camera shown in Figure 1.
An example of a reproduction device that reproduces images recorded in the memory 32 is shown.

This playback device has an input terminal 40 to which a memory 32 is connected. The image data recorded in the memory 32 is input from the input terminal 40 and sent to the decoding circuit 4 through the signal line 142.
4 is input. The decoding circuit 44 decodes the input encoded data to obtain data as shown in FIG. 4B. The data decoded by the decoding circuit 44 is sent to the orthogonal inverse transform circuit 4B through the signal line 144. The orthogonal inverse transform circuit 48 performs orthogonal inverse transform on the decoded data to obtain data of each modified block as shown in FIG. 4A.

The data of each block obtained by the orthogonal inverse transform circuit 4B is sent to the block synthesis circuit 4 through the signal line 14B, and the data of each block is synthesized to create the original image data. In this case, the blocks are converted into modified blocks when divided into blocks in the apparatus shown in FIG. 1, and then combined. That is, blocks that have unevenness at their boundaries with adjacent blocks are combined to restore the arrangement of pixel data in the original image. The data synthesized by the block synthesis circuit 48 is sent to the OA conversion circuit 52 through a signal line 150, converted into an analog signal by the OA conversion circuit 52, outputted to the CRT 54 through the signal line 152, and the color data recorded in the memory 32 is converted into an analog signal. The image is played back on the CRT 54 screen.

The control unit 56 is a control unit that controls each functional unit of the playback device, and connects the signal line 15 to the decoding circuit 44 via the signal line 15B.
8 to the orthogonal inverse transform circuit 46, signal line 1θ0 to the reblock synthesis circuit 48, and signal line 1G4 to the DA conversion circuit 52, respectively, to control the operation of each part.

Next, the operation of the playback device will be explained.

When the memory 32 is attached to the playback device, the data recorded in the memory 32 is read out by the connector 4G and input to the decoding circuit 44 through the signal line 142. The data input to the decoding circuit 44 is decoded in the decoding circuit 44 and sent to the orthogonal inverse transform circuit 46 through the signal line 144, respectively. The data input to the orthogonal inverse transform circuit 46 is orthogonally inverse transformed to obtain data for each block. The data for each block sent to the block synthesis circuit 48 through the signal line 14B is sent to the block synthesis circuit 48.
The blocks are combined with adjacent uneven boundaries at , and sent to the DA conversion circuit 52 through the signal @150,
It is converted into an analog signal in the OA conversion circuit 52 and sent to the CRT 54 through the signal line 152.
The original still image is reproduced and displayed on the screen 54.

As described above, according to the electronic still camera shown in FIG. converted and encoded circuit 2
In B, a large storage capacity is allocated only to the data of low frequency components, and a small storage capacity is allocated to the data of high frequency components, or it is not stored. Therefore, images can be stored with a small storage capacity. images can be stored.

Furthermore, in the blocking circuit 22 of this camera.

The input pixel data is divided into non-square blocks, that is, a plurality of deformed blocks in which every other row is shifted by one pixel to the left and right. Therefore, the data of the compressed image signal stored in the memory 32 is reproduced by the reproducing device.
When outputting to the RT 54, pixel data is divided into square blocks to prevent block boundary patterns from appearing in the output image as shown in FIG. 8, for example, when the pixel data is stored and reproduced. Can be done.

Note that, as shown in FIG. 7, a method in which modified block division of an image signal is applied to a block encoding method has been proposed. In this proposal, pixel data is divided into a plurality of modified blocks, and the blocked data is directly encoded. On the other hand, in the electronic still camera shown in FIG. 1 according to the present invention, modified block division is applied to an orthogonal transform encoding device for image signals, and as described above, pixel data is divided into a plurality of transformed blocks. , blocks of data are orthogonally transformed block by block, and the orthogonally transformed data is encoded. Therefore, it is possible to perform quantization and encoding appropriate to the characteristics of the image signal, so that it is possible to perform efficient data compression encoding.

FIG. 6 shows another embodiment in which the orthogonal transform encoding device for color image signals according to the present invention is applied to a digital electronic still camera.

In this embodiment, the output of the AD conversion circuit 18 is the signal! ! ! 04 to the memory 70. memory 70
A control signal is sent from the memory control unit 72 to the memory controller 72 through the signal line 170. The digital image signal sent from the All conversion circuit is temporarily stored in the memory 70, and the memory controller 7
Based on the control signal from 2, the image data is output to the orthogonal transform circuit 24 as block data. A signal line 1 is connected to the memory control unit 72 from the control unit 3B that controls the entire device.
A control signal is sent through 74, and a synchronization signal is sent from the synchronization signal generation circuit 34 through signal line 172.

The rest of the configuration is the same as the device shown in FIG.

In the device shown in FIG. 6, the image data signal read out from the AD conversion circuit 18 is stored in the memory 70, and read out from the memory 70 in blocks according to a control signal from the memory control section 72. Therefore, memory 7
0 and the memory control unit 72 correspond to the blocking circuit 22 of the device shown in FIG. 1, and perform blocking.The other operations are the same as those of the device shown in FIG. 1, so a description thereof will be omitted.

Also in the apparatus of this embodiment, captured image data is divided into blocks and encoded using look-up table data corresponding to the picture pattern, so that it is possible to increase the data compression rate and improve the encoding efficiency. Moreover, in the blocking circuit 22, the pixel data is divided into a plurality of deformed blocks that are not square but have uneven boundaries, so when the stored compressed image signal data is reproduced and output to the CRT 54, the output image is It is possible to prevent patterns from occurring at block boundaries due to block division.

In this example, each block in the block formation of one pixel data is made up of 18 pixels, and a modified block in which a part of the block is shifted is used.The shape of the modified block is, for example, as shown in FIG. 7. Blocks with a staircase shape such as
It is also possible to set the number of pixels in one block to a number different from that in this embodiment.

In the above embodiment, a device in which an orthogonal transform encoding device for one image value code is applied to an electronic still camera has been described, but the present invention is not limited to electronic still cameras. It can be applied to various devices that require orthogonal transform encoding.

Effects According to the present invention, one image value code is divided into a plurality of deformed blocks, the block data is orthogonally transformed, and the data after orthogonal transformation is encoded and recorded on a recording medium, so that the characteristics of one image value code are improved. Quantization and compression encoding suitable for Moreover, when dividing the image signal into blocks, the image signal is divided into a plurality of deformed blocks that are different from squares and have uneven boundaries, so when playing back compressed image data recorded on a recording medium, the output image is It is possible to prevent visual distortion caused by

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment in which an image signal orthogonal transform encoding device according to the present invention is applied to a digital electronic still camera, and FIG. 2 is a decoding of data encoded and recorded by the device shown in FIG. 3A is a block diagram showing an example of the pixel state of one screen before being divided into blocks; FIG. 3B is a block diagram showing an example of the screen in FIG. 3A being divided into blocks; 4A is a diagram showing an example of pixel data of modified block 1 in FIG. 3B, and FIG. 4B is a diagram showing an example of data after the pixel data of the modified block in FIG. 4A is replaced with a square arrangement. FIG. 4C is a diagram showing an example of data obtained by orthogonally transforming the pixel data in FIG. 4B, FIG. 4D is a diagram showing an example of the number of bits allocated in encoding the data in FIG. 4C, and FIG. FIG. FIG. 5B is a diagram showing an example in which the modified block of FIG. 3B is applied to the image of FIG. 5A. FIG. 6 is a block diagram showing another embodiment in which the image signal orthogonal transform encoding device according to the present invention is applied to a digital electronic still camera. FIG. 7 is a diagram showing another example of dividing the screen shown in FIG. 3A into blocks, and FIG. 8 is a diagram showing an example of visual distortion caused in an output image during playback due to conventional blocking. Explanation of symbols of main parts 14. Imaging device +8. , , , AD conversion circuit 221921. ,Blocking circuit 24, ,Orthogonal transform circuit 2B, ,Encoding circuit 32, ,Memory 70, ,Memory 72, ,Memory control section Patent applicant: Fuji Photo Film Co., Ltd. Representative: Katori Takao Maruyama Takao Isamu 4AI! I Ice Figure 48 Figure 4C Figure Tail lD Figure 5A Figure 8 Fu 5B Kukou Figure 7

Claims (1)

[Claims] 1. Orthogonal transform encoding of the image signal, which receives an image signal captured by an imaging means, blocks the image data of the image signal, and then orthogonally transforms and encodes the block and records it on a recording medium. In the apparatus, the apparatus includes: blocking means for blocking the image data; orthogonal transformation means for orthogonally transforming the image data reblocked by the blocking means; It has an encoding means for encoding the image data, and a recording means for recording the image data encoded by the encoding means on the recording medium, and the blocking means has a boundary between adjacent blocks. An orthogonal transform encoding device for an image signal, characterized in that the image data is divided into a plurality of blocks so as to have unevenness. 2. data reading means for reading out the compressed and encoded image data from the recording medium recorded by the apparatus according to claim 1; It has an expansion decoding means that expands and decodes compression-encoded image data, and a block synthesis means that combines the image data of each block expanded and decoded by the expansion decoding means, the block synthesis means . An image signal expansion/reproduction device characterized in that the expanded and decoded image data of each block is combined so that boundaries between adjacent blocks form the unevenness.