JPH0583565A - Image data compressor - Google Patents

Abstract

PURPOSE:To suppress the picture quality degradation of an object due to patterns and to objectively improve picture quality by measuring a distance between the object and a camera system and executing prescribed weighting to the code amount of each block. CONSTITUTION:This device is equipped with a measuring means 25 to measure the distance between the object as a main body in a photographed optical image and the camera system, and weighting means 23 and 24 to execute the prescribed weighting to the code amounts of the respective blocks set by compressing means 16, 17 and 20-22 based on the measured result of the measuring means 25. Therefore, a weighting coefficient generation circuit 24 generates a weighting coefficient for each block by selecting a prescribed weighting coefficient table according to the output of the distance detecting circuit 25 and prescribing a block address in a table according to the output of a block address circuit 26. Then, the distance detection circuit 25 measures the distance between a lens 11 and the object and based on the outputted digital distance measurement data, one of plural weighting coefficient tables is selected.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improvement of an image data compression apparatus for compressing digitally encoded still image data and recording it on a recording medium such as a semiconductor memory.

[0002]

2. Description of the Related Art As is well known, an optical image of a photographed subject is converted into an electric still image signal by using a solid-state image pickup device, and this still image signal is converted into digital still image data, and an SRAM (static Electronic still cameras for recording in volatile semiconductor memory such as random access memory) have been developed. In such an electronic still camera, a memory card in which a semiconductor memory is built in a card-shaped case is detachably attached to the camera body so that it is equivalent to a film in a normal camera. It is designed to be handled easily.

By the way, in this kind of electronic still camera, in order to effectively use the storage capacity of the memory card, the digital still image data is usually subjected to data compression processing before being recorded in the semiconductor memory. In this case, as a data compression means of the digital still image data, for example, a method using DPCM (differential pulse code modulation), a method using orthogonal transformation such as DCT (discrete cosine transform), and the like can be considered. ing.

Nowadays, there is a strong tendency to increase the data compression rate. In this limit, DCT
It is said that the data compression means using the DCT is advantageous, but the data compression means using the DCT has a problem that the image quality is deteriorated due to the irreversible conversion. Achieving the compression rate is a major issue for the future.

FIG. 5 shows a conventional image data compression apparatus using DCT. That is, an optical image of a photographed subject is passed through the lens 11 to, for example, a CCD (charge
An image is formed on the solid-state imaging device 12 such as a coupled device) and converted into an electrical signal corresponding to the optical image of the subject. This electrical signal is supplied to the image pickup circuit 13 and converted into digital still image data that has undergone white balance adjustment processing, knee correction processing, and the like.

Then, the digital still image data is
After being once written and stored in the frame memory 15 via the memory controller 14, it is read by the DCT circuit 16 for each block of a size uniquely determined, and is again guided to the DCT circuit 16 via the memory controller 14.
The DCT circuit 16 subjects the digital still image data input in block units to discrete cosine transform processing. Then, the discrete cosine transform output is compression-encoded by the encoding circuit 17, and then supplied to the memory card 19 via the card controller 18 and recorded therein.

In this case, the photographer must know in advance how many digital still image data can be recorded in the memory card 19 currently used, that is, how many memory cards 19 should be taken. From the viewpoint that it is convenient to know and the demand from photographers will be high, with regard to the digital still image data to be recorded in the memory card 19, the data amount for one image is fixed. Is required to be compression encoded. That is, no matter what kind of subject is photographed, the digital still image data will have the specified data amount (code amount).
That is, it is necessary to record in the memory card 19.

By the way, in order to make the data amount fixed for an arbitrary image, how to distribute the code amount to the entire image becomes a big problem. For example, if you encode from the top of the screen without permission, the amount of code may be insufficient at the bottom of the screen and you may not be able to encode. Then, it becomes necessary to encode.

Generally, a block with almost no change in the picture has almost only DC (direct current) component, but a block with a drastic change in the picture has many AC (alternating current) component in addition to the DC component. Therefore, when compression-encoding, if this AC component is significantly deleted, the distortion in image quality becomes large. For this reason, a small code amount is given to a block with a small frequency change amount (a small change in the pattern), and a large code amount is given to a block with a large frequency change amount (a large change in the pattern). When compression coding is performed by distributing the code amount, the deterioration is dispersed and the distortion in image quality becomes inconspicuous.

Therefore, conventionally, a large amount of information (activity) for each block is obtained from the amount of frequency change,
The amount of code that can be encoded in each block is distributed in proportion to the amount of activity. That is, referring back to FIG. 5, the block activity detection circuit 20 calculates the activity of a block based on the discrete cosine transform output from the DCT circuit 16, and the calculated activity is calculated by the total activity calculation circuit 21 for all blocks. Integrate by minutes to find total activity.

Here, the code amount to be distributed to each block is
Since the total code amount is calculated by (total activity of each block / total activity of all blocks), each output of the block activity detection circuit 20 and the total activity calculation circuit 21 is supplied to the bit allocation circuit 22 to perform the operation of the above equation. The compression coding is performed by the coding circuit 17 according to the calculated code amount.

As described above, in the method of obtaining the code amount to be distributed to each block from the activity of the information amount, the digital still image data becomes the specified data amount no matter what object is photographed. This is suitable for an electronic still camera that can specify a data compression rate. In addition, making the code distribution amount to each block proportional to the information amount activity of each block makes the S / N at each position (block) on the screen uniform, resulting in positional unevenness on the screen. There is also the effect that there is no.

However, it is not always good to make the S / N uniform on the screen from the subjective evaluation point of view.
There are more cases where / N is good but subjectively it is not good. For example, in the case of an image in which the background pattern on the screen is relatively fine and the subject image exists in the center of the screen, the code amount is distributed in proportion to the information amount activity of each block. Since a relatively large amount of code is distributed, the amount of data distributed to the essential subject image is reduced.

In this case, the S / N of individual blocks on the screen
However, the distortion of the subject image, which should be the most noticeable when looking at the screen, increases by the amount of code distributed to the blocks in the background, especially at the edges of the subject image. The distortion becomes significantly large and the subjective evaluation deteriorates.

[0015]

As described above, in the conventional image data compression apparatus, the data amount is fixed while the S / N is made uniform on the screen, but it is not always good subjectively. However, there is a problem that the image quality of the subject often deteriorates depending on the design.

Therefore, the present invention has been made in consideration of the above circumstances, and is extremely good image data in which the code amount can be distributed so that the image quality deterioration of the subject due to the pattern is suppressed and the image quality is subjectively improved. An object is to provide a compression device.

[0017]

An image data compression apparatus according to the present invention comprises a conversion means for digitally encoding a photographed optical image and converting it into image data, and image data output from the conversion means in a predetermined block. A camera provided with a compression unit that is divided into units and compression-encodes with a code amount distributed according to a picture for each block, and a recording unit that digitally records the compressed data output from the compression unit on a recording medium. Intended for systems.

Then, measuring means for measuring the distance between the camera system and the main subject in the photographed optical image, and the code of each block set by the compressing means based on the measurement result of this measuring means. A weighting means for applying a predetermined weight to the quantity is provided.

[0019]

According to the above-mentioned structure, for example, the closer the distance between the subject and the camera system is, the more weight is given to the block in the central portion of the photographing screen so that the distribution of the code amount is increased. As the distance from the system increases, the weighting is uniformly applied to the entire shooting screen. Generally, when the distance to the subject is relatively short, the subject is often located in the center of the screen. Therefore, in the case of a close subject, the image quality in the central portion of the screen has a large code distribution amount and deterioration is suppressed, so that the image quality can be subjectively improved.

[0020]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described in detail below with reference to the drawings. In FIG. 1, the same parts as those in FIG. 5 are designated by the same reference numerals. That is, the activity of each block obtained by the block activity detection circuit 20 is weighted by the weighting circuit 23 based on the weighting coefficient output from the weighting coefficient generation circuit 24, and then the total activity calculation circuit 21 performs total weighting. Integration for blocks is performed to find total activity. As the weighting circuit 23, for example, a multiplier or a bit shift circuit is used.

Here, the weighting coefficient generation circuit 24 has a plurality of weighting coefficient tables, and a predetermined weighting coefficient table is selected by the output of the distance detection circuit 25 and a table is output by the output of the block address circuit 26. By designating a block address within, a weighting coefficient is generated for each block. The distance detecting circuit 25 measures the distance between the lens 11 and the subject and outputs digital distance measuring data L. Based on the digital distance measuring data L, a plurality of weighting factors in the weighting coefficient generating circuit 24 are weighted. One of the coefficient tables is selected.

That is, the plurality of weighting coefficient tables have two-dimensional characteristics corresponding to those on the screen as shown in FIG. 2A, and the characteristics of the weighting coefficients when viewed on the line H on this screen are shown in FIG. As shown in 2 (b), three modes are prepared, and these modes are selected by the digital distance measurement data L output from the distance detection circuit 25. For example, when the digital distance measurement data L is smaller than the reference distance data l1, the table having the characteristic that the peak of the weighting coefficient at the center of the screen is the largest is selected.

On the other hand, the block address circuit 26 is
By counting the number of block syncs output from the T circuit 16, the block address of the weighting coefficient table selected by the distance detection circuit 25 is generated, and the weighting coefficient is generated for each block.

Here, as the distance detection circuit 25,
An infrared distance measuring method as shown in FIG. 3 is used. That is, the infrared light emitted from the infrared light emitting element 27 is condensed by the light emitting lens 28 and applied to the subject 29, and the reflected light from the subject 29 is collected by the light receiving lens 30 and received by the light receiving sensor 31. Thus, the current ratio k corresponding to the distance from the light receiving sensor 31 to the subject 29 is obtained. Then, the output of the light receiving sensor 31 is
The digital distance measurement data L is generated by being supplied to the A / D (analog / digital) conversion circuit 32.

The relationship between the distance to the object 29 and the current ratio k in the distance detection circuit 25 is inversely proportional as shown in FIG. 4, and the digital measurement output from the A / D conversion circuit 32 is performed. The distance data L is output to the weighting coefficient generation circuit 24 as a signal indicating each region (mode) in which the current ratio k in FIG. 4 is divided into three regions. That is, this digital distance measurement data L is, for example, L
It is output to the weighting coefficient generation circuit 24 as a mode signal (2-bit “00”, “01”, “10”) corresponding to <l1, l1 ≦ L ≦ l2, 12 <L.

With the above configuration, for example, when the digital distance measuring data L is L <l1, that is, when the subject 29 is relatively close, as is apparent from FIG. The weighting coefficient table with the largest weight in the section is selected, and the code amount distributed to the blocks around the center of the screen increases, and conversely, the code amount distributed to the blocks at the edge of the screen decreases.

Therefore, although the image quality is deteriorated to some extent at the edge portion of the screen, the image quality deterioration at the center portion of the screen is suppressed to that extent. When the subject 29 is relatively close to the subject 29, the subject 29 is often located in the central portion of the screen. Therefore, suppressing the image quality deterioration in the central portion of the screen is irreversible. 29, the image quality deterioration is suppressed and the image quality is subjectively improved.

When the digital distance measurement data L is 12 <L, the image is a distant view such as a landscape when the subject 29 is distant. Therefore, as is apparent from FIG. 2B, a weighting coefficient table that allows uniform weighting to the entire screen is selected, and the entire screen has the conventional code amount distribution to provide a screen that is not affected by weighting. It will be.

By the way, in the above embodiment, the distance detecting circuit 25 is described as an infrared distance measuring system, but the present invention is not limited to this, and a known means such as an ultrasonic system or an SST system can be used as appropriate. Further, manual focusing may be performed to digitalize the distance to the subject 29 at that time. Alternatively, the distance may be measured subjectively by the photographer, that is, the electronic still camera may be provided with a selector for setting the near view, the distant view, and the middle thereof, and the mode selected by the photographer may be digitized. Further, the A / D conversion circuit 32 does not have to output 2 bits, and any number of bits may be used depending on the balance between performance and cost. The present invention is not limited to the above-described embodiments, but can be variously modified and implemented without departing from the scope of the invention.

[0030]

As described in detail above, according to the present invention,
It is possible to provide an extremely good image data compression apparatus that can distribute the code amount so as to subjectively improve the image quality by suppressing the image quality deterioration of the subject due to the pattern.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of an image data compression device according to the present invention.

FIG. 2 is a diagram for explaining characteristics of a weighting coefficient table according to the first embodiment.

FIG. 3 is a block configuration diagram showing a specific example of a distance detection circuit of the same embodiment.

FIG. 4 is a characteristic diagram showing a relationship between a measured distance and an output of the distance detection circuit.

FIG. 5 is a block configuration diagram showing a conventional image data compression device.

[Explanation of symbols]

11 ... Lens, 12 ... Solid-state imaging device, 13 ... Imaging circuit,
14 ... Memory controller, 15 ... Frame memory, 1
6 ... DCT circuit, 17 ... Encoding circuit, 18 ... Card controller, 19 ... Memory card, 20 ... Block activity detection circuit, 21 ... Total activity calculation circuit,
22 ... Bit distribution circuit, 23 ... Weighting circuit, 24 ... Weighting coefficient generation circuit, 25 ... Distance detection circuit, 26 ... Block address circuit, 27 ... Infrared light emitting element, 28 ... Light emitting lens, 29 ... Subject, 30 ... Light receiving lens, 31 ... Light receiving sensor, 32 ... A / D conversion circuit.

Claims (3)

[Claims] 1. A conversion unit that digitally encodes a captured optical image and converts it into image data, and image data output from this conversion unit is divided into predetermined blocks, and each block is distributed according to a pattern. In a camera system provided with a compression means for compression-encoding with the generated code amount and a recording means for digitally recording the compressed data output from the compression means on a recording medium, the camera system is a main part of the captured optical image. Measuring means for measuring the distance between the subject and the camera system, and weighting means for applying a predetermined weighting to the code amount of each block set by the compression means based on the measurement result of this measuring means. An image data compression apparatus comprising: 2. The weighting means weights the block in the central portion of the photographing screen so that the code amount is distributed more as the distance between the subject and the camera system becomes shorter. The image data compression apparatus according to claim 1, wherein the image data compression apparatus is configured such that as the distance from the camera system increases, more uniform weighting is applied to the entire photographing screen. 3. The weighting means selects a plurality of types of weighting tables according to the measurement result of the measuring means,
The image data compression apparatus according to claim 2, wherein the image data compression apparatus is configured to perform weighting based on the selected weighting table.