JPH11205793A - Image compression device, image decompression device and digital still camera - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently compress image data by performing discrete cosine transformation of image data in each pixel block of a prescribed size, quantizing the image data and performing arithmetic coding of the quantized image data. SOLUTION: A converter(DCT) 32 performs discrete cosine transformation to a luminance signal and a color signal respectively which constitute image data in each pixel block that is, for example, 8×8 pixels in size. A quantizer 33 quantizes image data that is transformed by the converter 32. An encoder 34 performs arithmetic coding of quantized image data. Image data which is compressed by the converter 32, the quantizer 33 and the encoder 34 is written to a flash memory 19. A decoder 35 reads image data which is written in the memory 19 and undergoes arithmetic coding and performs arithmetic coding. A reencoder 36 performs Huffman coding of image data that is decoded by the decoder 35. Image data which is recoded is written on a memory card through a card driver 20.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image compression apparatus for compressing image data and an image decompression apparatus for reproducing image data from compressed data, and more particularly to compression and reproduction of image data of a digital still camera.

[0002]

2. Description of the Related Art Light from an object to be photographed is transferred to a charge-coupled device (CC).
2. Description of the Related Art A digital still camera that electronically captures an image by forming an image on D) is widely used. Image data representing the captured image is stored in a storage medium such as a flash memory or a memory card, read out at a later time, displayed on a liquid crystal display (LCD) provided in a camera, or used in another personal computer or other computer. Transferred to device.

[0003] The captured image has higher image quality as the number of pixels of the CCD increases, but as the number of pixels increases, the amount of image data to be stored increases, and the time required for input / output and transfer to a storage medium also increases. I do. Therefore, image data is compressed and stored. Various methods for compressing image data have been proposed. In a general digital still camera, JPEG (Joint Photographic image coding) is used.
Experts Group) has adopted the method proposed as a unified standard.

FIG. 4 shows a flow of a process of compressing image data according to the JPEG system. First, the image data is subjected to discrete cosine transform (DCT) for each pixel block of a predetermined size (for example, 8 × 8 pixels),
The image data mainly includes a DC component and a low-frequency AC component. Next, the converted image data is quantized, and individual components are classified into predetermined stages. Further, the quantized image data is subjected to Huffman coding to complete the compression.

Reproduction of image data is performed by performing Huffman decoding, inverse quantization, and inverse discrete cosine transform in this order. Huffman encoding is a lossless process in which information is completely lost by decoding. On the other hand, quantization is a process involving a change in value, and a component that contributes little to an image is lost, so that reproduced image data is slightly different from image data before compression. For this reason, the original image cannot be completely reproduced, but the deterioration of the image quality is slight, and a reproduced image that can sufficiently withstand viewing is obtained.

[0006] In a device that handles monochrome images such as a facsimile, JBIG (Joint Bi-level Imagecoding experts G) is used.
image data is compressed by the method proposed by Roup). In the JBIG system, compression is performed based on prediction of a pixel value, and arithmetic coding is used for coding, which is the final stage of processing.

[0007]

The Huffman coding adopted by the JPEG method is a lossless process, and is preferable because the quality of a compressed image is not reduced. However, the compression rate of Huffman coding is generally lower than that of arithmetic coding, which is also lossless processing. Further, the compression ratio of Huffman coding tends to fluctuate depending on the type of image, and image data cannot often be satisfactorily compressed.

FIG. 5 shows an example in which the compression rates of arithmetic coding, Huffman coding and other coding are compared. In FIG. 5, the vertical axis represents the ratio of the amount of image data after compression to the amount of image data before compression in percentage, and the smaller the ratio, the higher the compression ratio. (A) is CCI
TT test chart No.1 (black and white character image)
(B) is No. 1 of the test chart of GENESIS
(Color photographic image) as a black and white image composed of intermediate gradations.

MH, MR and MMR are based on the coding of Modified Huffman, Modified Read and Modified Modified Read, respectively, and JBIG-2 and JBIG-3 are JBIG-3 respectively.
This is based on arithmetic coding of two lines and three lines of BIG. In this example, for binary images, arithmetic coding has a higher compression rate than other coding based on run-length coding. On the other hand, arithmetic coding shows the effect of compression for halftone images, while other coding including Huffman coding results in a larger data amount due to coding. I have.

[0010] If the compression ratio is low, the purpose of efficiently using the storage capacity of the storage medium cannot be sufficiently achieved, and the time required for input / output of the storage medium becomes longer, thereby lowering the image processing efficiency. Will be. As a method of increasing the compression ratio, image data after Huffman encoding is converted into an LZH method or ZIP which is known as a lossless encoding process.
Although it is conceivable that the encoding is further performed by the method, the improvement of the compression ratio is about several percent or less, and only the time required for the encoding process becomes longer, and there is almost no effect.

In order to provide image data to another device employing the JPEG system, encoding must ultimately be performed by Huffman encoding. However, it is not necessary to compress the image data of all captured images by Huffman coding, and it is desirable to perform more efficient coding for image data not provided to other devices.

The present invention has been made in view of the above problems, and has as its object to provide an image compression apparatus capable of efficiently compressing image data and a digital still camera provided with the apparatus. Another object of the present invention is to provide an image decompression device for reproducing image data from compressed data.

[0013]

In order to achieve the above object, according to the present invention, in an image compression apparatus for compressing image data representing an image, discrete cosine transform of the image data is performed for each pixel block of a predetermined size. A converter, a quantizer for quantizing the image data converted by the converter,
An encoder for arithmetically encoding the image data quantized by the quantizer. This image compression apparatus performs up to quantization in accordance with the JPEG system, and performs the final stage of encoding without using the JPEG system.

In the present invention, the image compression apparatus may further include a memory for storing the image data encoded by the encoder, a decoder for arithmetically decoding the image data stored in the memory, And a re-encoder for Huffman-encoding the image data decoded by the encoder. Arithmetic-coded image data stored in memory cannot be used by other devices that use the JPEG method, but image data that has been decoded and re-encoded by Huffman coding uses the JPEG method. It can be used by other devices that do.

The digital still camera includes an imaging lens, an imaging device that receives light transmitted through the imaging lens, and performs photoelectric conversion, and an image generation circuit that generates image data representing an image from a signal output from the imaging device. May be provided to the converter with the image data generated by the image generation circuit. Image data representing a captured image is subjected to a discrete cosine transform and a quantization process, and then arithmetically coded. With the configuration including the decoder and the re-encoder, it is possible to provide Huffman-encoded image data.

To achieve the above object, the present invention also provides a decoder for arithmetically decoding image data which has been subjected to discrete cosine transform, quantized and arithmetically encoded for each pixel block of a predetermined size. And an inverse quantizer for inversely quantizing the image data decoded by the decoder, and an inverse transformer for inverse discrete cosine transform of the image data inversely quantized by the inverse quantizer. Is configured.
This image decompression device is a device for decompressing the image data arithmetically encoded by the image compression device. An image display device may be provided by adding a display for displaying an image represented by the image data inversely transformed by the inverse transformer.

[0017]

FIG. 1 shows a schematic configuration of a digital still camera according to an embodiment of the present invention. The digital still camera (hereinafter simply referred to as a camera) 1 includes an imaging lens 11, an aperture 12, a CCD 13 serving as an imaging element, an analog signal processing unit (AGC / CDS) 14, an AD converter (A / D) 15, a timing generator. (TG)
16, a digital signal processing unit (DSP) 17, an image compression unit 18, a flash memory 19, a card driver 20, an image decompression unit 21, a display unit 22, an operation unit 23, and a control unit 24.

The imaging lens 11 converts light from the object to be photographed into CC light.
An image is formed on the light receiving surface of D13, and the aperture 12 is
The light flux from the CCD 13 to the CCD 13 is regulated to adjust the amount of light received by the CCD 13. The CCD 13 is composed of hundreds of thousands of pixels alternately arranged in a matrix of three types of pixels that are sensitive to red (R), green (G), and blue (B) light, respectively. And accumulates it, and outputs the accumulated charge as an analog signal.

The analog signal processor 14 performs automatic gain control on the output signal of the CCD 13 and performs double correlation sampling. The AD converter 15 includes the analog signal processing unit 1
4 is converted into a digital signal and output to the digital signal processing unit 17. The timing generator 16 includes the CCD 13 and the analog signal processing unit 1
4 and a drive signal for notifying the AD converter 15 of the operation timing.

The digital signal processing unit 17 applies white balance adjustment, shading, and the like to the output signal of the CCD 13 digitized by the AD converter 15.
Processing such as interpolation of three color signals of R, G, and B and gamma correction are performed to generate image data including a luminance signal and a color signal, and output the image data in a YUV 4: 1: 1 format. One set of image data generated by the digital signal processing unit 17 represents one captured image and can be displayed as it is.

The digital signal processing unit 17 also supplies a control signal for instructing the timing of outputting the drive signal to the timing generator 16, and outputs a control signal according to the progress of its own processing.
D13, the analog signal processing unit 14, and the AD converter 15 are operated. The digital signal processing unit 17 is 30
Generate frame image data.

The image compression section 18 is a digital signal processing section 1
7 is compressed. FIG. 2 shows the configuration of the image compression unit 18. The image compression unit 18 includes an image memory 31, a transformer (DCT) 32, a quantizer 33, an encoder 34, a decoder 35, and a re-encoder 36. The image memory 31 stores one frame of image data supplied from the digital signal processing unit 17. Since the compression processing of the image data requires a time equal to or longer than 1/30 seconds, which is the photographing cycle of the image, an image memory 31 is provided to hold the image data during that time.

The converter 32 performs a discrete cosine transform on each of a luminance signal and a chrominance signal constituting image data, for example, for each pixel block having a size of 8 × 8 pixels.
The quantizer 33 quantizes the image data converted by the converter 32. The conversion and quantization of the image data by the converter 32 and the quantizer 33 are performed according to the JPEG system.

The encoder 34 arithmetically encodes the image data quantized by the quantizer 33. Here, a QM coder of the JBIG system is used as the encoder 34. The image data compressed by the converter 32, the quantizer 33, and the encoder 34 is written to the flash memory 19.

The decoder 35 reads the arithmetically encoded image data written in the flash memory 19 and performs arithmetic decoding. The decoding performed by the decoder 35 is a reverse process of the encoding performed by the encoder 34,
3 completely reproduces the image data quantized. The re-encoder 36 performs Huffman encoding on the image data decoded by the decoder 35 according to the JPEG method. The image data re-encoded by the re-encoder 36 is written to the memory card via the card driver 20. The re-encoded image data conforms to the JPEG standard.

The decoder 35 and the re-encoder 36
This is used to generate image data to be provided to another device employing the PEG system, and it is not necessary to decode and re-encode all of the image data generated by the digital signal processing unit 17.

The card driver 20 inputs and outputs image data to and from a removable memory card. The flash memory 19 stores the image data captured and compressed by the camera 1 as described above. The image data can be copied to a memory card by the card driver 20 and provided to other devices. Conversely, image data stored in the memory card can be copied to the flash memory 19, and the flash memory 19 also stores image data provided from other devices. Image data provided from another device includes a converter 32, a quantizer 33
And the one compressed by the same processing as that of the encoder 34, and the one compressed entirely according to the JPEG method.

The image decompression unit 21 performs decompression processing for expanding the compressed image data and reproducing the image data before compression. FIG. 3 shows the configuration of the image decompression unit 21. The image inverse compression unit 21 includes two decoders 41 and 42 and an inverse quantizer 4
3. It comprises an inverse transformer (inverse DCT) 44. Decoder 41
Reads arithmetically encoded image data stored in the flash memory 19 and performs arithmetic decoding. The decoder 41 is the same as the decoder 35 of the image compression unit 18 described above, and may be provided with only one of them, and may be shared by the image compression unit 18 and the image decompression unit 21. Absent.

Another decoder 42 reads out the Huffman-encoded image data stored in the flash memory 19 and performs Huffman decoding. Decoder 4
2 is the reverse of the process performed by the re-encoder 36 of the image compression unit 18, and the quantized image data is completely reproduced. JPEG-compressed image data provided from another device can be decoded by the decoder 42.

The inverse quantizer 43 outputs the signal to the decoder 41 or 4
2 performs inverse quantization on the image data decoded. The inverse transformer 44 performs an inverse discrete cosine transform on the image data inversely quantized by the inverse quantizer 43. The processing performed by the inverse quantizer 43 and the inverse transformer 44 corresponds to the inverse processing of the processing of the quantizer 33 and the transformer 32, respectively. However, since the information has already been slightly changed by the quantization, the reproduction is performed. The compressed image data does not completely match the image data before compression. However, the amount of change in the information is small,
The image represented by the reproduced image data has a quality comparable to the captured image.

The display unit 22 is composed of an LCD and is provided with image data generated by the digital signal processing unit 17 or image data reproduced by the image decompression unit 21, and displays the image. The operation unit 23 has operation members such as a dial and a button, and transmits an instruction regarding control given by operating the user to the control unit 24.

The control unit 24 controls the operation of the camera 1 in accordance with an instruction given via the operation unit 23. When the photographing start is instructed, the control unit 24 sends the digital signal processing unit 17
, And the digital signal processing unit 17 responds to the
D13, the analog signal processing unit 14, and the AD converter 15 start processing. Thus, the generation of the image data is started. However, the compression of the image data and the storage in the flash memory 19 are not performed until the release button provided on the operation unit 23 is operated.

When the release button is not operated, the control section 24 causes the digital signal processing section 17 to sequentially supply the generated image data to the display section 22. An image displayed on the display unit 22 based on these image data functions as a finder image. For this reason, the camera 1 is not provided with an optical finder, but may be provided with an optical finder.

When the release button is operated, the control unit 2
4 causes the digital signal processing unit 17 to supply the generated image data for one frame to the image compression unit 18. This image data is temporarily stored in the image memory 31, compressed as described above, and then written into the flash memory 19. The supply of the image data generated by the digital signal processing unit 17 to the display unit 22 is continued even during the compression processing of the image data. Therefore, the user can always observe the finder image while taking a picture.

During a period in which photographing is not performed, image data stored in the flash memory 19 is read out, decompressed, and a reproduced image can be displayed on the display unit 22. At this time, the control unit 24 sets the two decoders 41 and 42 of the image decompression unit 21 according to whether the specified image data is arithmetic-coded or Huffman-coded. The decryption is performed by operating the appropriate one of them. Flash memory 1
It is also possible to decode the arithmetically coded image data stored in 9 and re-encode by Huffman coding, or to copy the image data between the flash memory 19 and the memory card.

With these functions, the digital still camera 1 can share image data with other devices conforming to the JPEG standard. Moreover, since the compressed image data used only by the camera 1 is arithmetically coded and has a small amount, it is possible to store image data of more frames in the flash memory 19, The time required for writing to and reading from the flash memory 19 is also shortened, and the processing operation of the camera 1, for example, reproduction and display, becomes faster.

If the function of performing arithmetic decoding is provided to another device in the same manner as the decoder 41 of the image decompression unit 21, the arithmetically encoded image data can be provided to another device as it is. At this time, writing of image data to the memory card and reading from the memory card by another device can be performed more quickly, and the processing efficiency is improved.

Although the image data is transmitted / received to / from another device by the memory card, the image data may be transmitted / received via a cable or by wireless communication such as infrared rays. Even in that case, by sending and receiving a small amount of data compressed by arithmetic coding,
The time required for communication is reduced.

[0039]

According to the image compression apparatus of the first aspect,
Image data can be compressed at a high compression ratio regardless of the type of image. Therefore, the utilization efficiency of the storage medium for storing the compressed image data is improved, the time required for storage input / output is reduced, and the efficiency of image processing is improved. The amount of information of the compressed image data is equivalent to the image data compressed by the JPEG method.

According to the image compression apparatus of the second aspect, it is possible to provide the compressed image data to another device adopting the JPEG system. Moreover, since the Huffman coding can be performed only when the compressed image data is provided to another device, there is no possibility that the compression ratio is reduced when the image data is not provided to another device.

According to the digital still camera of the third aspect, image data representing a captured image can be efficiently compressed. In addition, it is also possible to reproduce and display the photographed image on another device employing the JPEG system.

According to the image decompression device of the fourth aspect and the display device of the fifth aspect, an image of the same quality as that of the JPEG system is reproduced or displayed from image data compressed at a higher compression ratio. be able to.

[Brief description of the drawings]

FIG. 1 is a diagram showing a schematic configuration of a digital still camera according to an embodiment of the present invention.

FIG. 2 is a diagram showing a configuration of an image compression unit of the digital still camera.

FIG. 3 is a diagram showing a configuration of an image decompression unit of the digital still camera.

FIG. 4 is a view showing the flow of image data compression processing according to the JPEG method.

FIG. 5 is a diagram showing a comparative example of a compression ratio between arithmetic coding and another coding.

[Description of Signs] 1 Digital still camera 11 Imaging lens 12 Aperture 13 CCD 14 Analog signal processing unit 15 AD converter 16 Timing generator 17 Digital signal processing unit 18 Image compression unit 19 Flash memory 20 Card driver 21 Image decompression unit 22 Display unit Reference Signs List 23 operation unit 24 control unit 31 image memory 32 discrete cosine transformer 33 quantizer 34 arithmetic encoder 35 arithmetic decoder 36 Huffman encoder (re-encoder) 41 arithmetic decoder 42 Huffman decoder 43 Inverse quantizer 44 Inverse discrete cosine transformer

Claims (5)

[Claims] 1. An image compression apparatus for compressing image data representing an image, comprising: a converter for performing discrete cosine transform of the image data for each pixel block of a predetermined size; and quantizing the image data converted by the converter. An image compression apparatus comprising: a quantizer that performs arithmetic coding on the image data quantized by the quantizer. 2. A memory that stores image data encoded by the encoder, a decoder that arithmetically decodes the image data stored in the memory, and a memory that is decoded by the decoder. The image compression apparatus according to claim 1, further comprising: a re-encoder that performs Huffman encoding of the image data. 3. The image compression device according to claim 1, wherein: an image pickup lens; an image pickup device that receives light transmitted through the image pickup lens and performs photoelectric conversion on the light; and an image output from a signal output by the image pickup device. A digital still camera, comprising: an image generation circuit that generates image data representing the following and supplies the image data to the converter. 4. A decoder for arithmetically decoding image data which has been subjected to discrete cosine transform, quantization and arithmetic coding for each pixel block of a predetermined size, and a decoder which has been decoded by the decoder. An image decompressor comprising: an inverse quantizer for inversely quantizing image data; and an inverse transformer for inverse discrete cosine transform of the image data inversely quantized by the inverse quantizer. 5. An image display device, comprising: the image decompression device according to claim 4; and a display for displaying an image represented by the image data inversely transformed by the inverse transformer.