Image signal processing device

• US 5,497,246 A
• Filed: 07/15/1994
• Issued: 03/05/1996
• Est. Priority Date: 07/15/1993
• Status: Expired due to Term

• Current Assignee: PENTAX RICOH IMAGING COMPANY, LTD.
• Original Assignee: Asahi Kogaku Kogyo Kabushiki Kaisha
• Inventors: Abe, Nobuaki
• Primary Examiner(s): Rogers, Scott A.
• Application Number:
  US08/275,357
• Time in Patent Office: 599 Days
• Field of Search: 358/426, 358/261.2, 358/430, 358/432, 348/403, 348/405, 348/404, 348/390, 348/384, 382/56, 382/252, 382/251, 382/250, 382/248
• US Class Current: 382/252
• CPC Class Codes: H04N 11/044 involving transform coding
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/126 Details of normalisation or weighting functions, e.g. normalisation matrices or variable uniform quantisers
• CPC Class Codes: H04N 19/13 Adaptive entropy coding, e.g. adaptive variable length coding [AVLC] or context adaptive binary arithmetic coding [CABAC]
• CPC Class Codes: H04N 19/186 the unit being a colour or a chrominance component
• CPC Class Codes: H04N 19/30 using hierarchical techniques, e.g. scalability H04N19/63 takes precedence
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/91 Entropy coding, e.g. variable length coding [VLC] or arithmetic coding

Abstract

A DCT processing circuit outputs DCT coefficients based on image data stored in an image memory. The DCT coefficients are quantized by a quantization processing circuit, then input to a Huffman encoding processing circuit and an inverse quantization processing circuit. In the inverse quantization processing circuit, the DCT coefficients are reproduced. The differences between the reproduced DCT coefficients and the DCT coefficients output from the DCT processing circuit are determined as the quantized error data by subtractors. An error data synthesizing unit stores the error data in a fourth component of a JPEG format and inputs the error data to the Huffman encoding processing circuit. The output signals of the quantization processing circuit and the output signals of the error data synthesizing unit are Huffman encoded and recorded to an IC memory card.

Claims

1. An image signal processing device, comprising:

orthogonal transformation means for applying an orthogonal transformation to original image data comprising at least one image component to obtain orthogonal transformation coefficients corresponding to predetermined spatial frequencies for each image component, said original image data comprising a first, a second, and a third image component;

quantization means for quantizing said orthogonal transformation coefficients, comprising quantization tables including quantization coefficients to obtain quantized orthogonal transformation coefficients;

first inverse quantization means for inversely quantizing said quantized orthogonal transformation coefficients to obtain restored orthogonal transformation coefficients;

error data producing means for obtaining quantized error data of a difference between said restored orthogonal transformation coefficients and said orthogonal transformation coefficients for each image component;

encoding means for encoding said quantized error component and said quantized orthogonal transformation coefficients to form a compressed image data;

error data synthesizing means for arranging said quantized error data into unit data of a fixed length for each same spatial frequency of components so as to synthesize a fourth component, said encoding means encodes said fourth component and said quantized orthogonal transformation coefficients to form compressed image data; and

recording means for recording said compressed image data to a recording medium.

2. An image signal processing device according to claim 1, wherein said error data synthesizing means comprises bit number allocating means for allocating a number of bits comprising said quantized error data stored in said fourth component so that the component among said first, second, and third components which has the greatest effect on the visual characteristics of the reproduced image is lengthened.

3. An image signal processing device according to claim 2, wherein said error data synthesizing means comprises error data quantization means for requantizing said quantized error data when the number of bits allocated by said bit allocating means is less than said quantized error data.

4. An image signal processing device according to claim 3, wherein said error data quantization means uses a maximum value of said quantization coefficients of said quantization tables so as to requantize all components of the quantized error data.

5. An image signal processing device as set forth in claim 3, wherein said error data quantization means uses the requantization coefficients corresponding to said quantization coefficients of the quantization tables to requantize predetermined constituent elements of the quantized error data.

6. An image signal processing device according to claim 5, wherein when the number of bits of any of said quantized error data of a requantized first, second, and third component is smaller than said allocated number of bits, said error data quantization means lengthens the number of bits of the quantized error data of the other components by the amount corresponding to said fixed length.

7. An image signal processing device according to claim 2, further comprising sign adjusting means for multiplying by "-1" said quantized error data corresponding to an orthogonal transformation coefficient outputted from the orthogonal transformation means when the orthogonal transformation coefficient is negative.

8. An image signal processing device according to claim 7, further comprising scaling means for obtaining encoded error data by allocating code words to quantized error data, adjusted in sign by said sign adjusting means, which are shorter in length, the higher the frequency of occurrence.

9. An image signal processing device according to claim 7, wherein

when the numbers of bits of the encoded error data are k, m, and n (k≧

m≧

n) in said error data synthesizing means producing the fourth component from said encoded error data;

wherein a topmost bit to an (m+1) bit of said quantized error data to which k number of bits has been allocated are stored in a first area of said fourth component from said topmost bit to a (2m+n+1) bit of said fourth component;

an m-th bit to an (n+1) bit of said quantized error data to which k number of bits has been allocated and a topmost bit to an (n+1) bit of the quantized error data to which m number of bits has been allocated are alternately stored in a second area from a (2m+n) bit to a (3n+1) bit of said fourth component; and

an n-th bit to a bottommost bit of said quantized error data to which k number of bits has been allocated, said n-th bit to said bottommost bit of the quantized error data to which m number of bits has been allocated, and all of the bits of the quantized error data to which n number of bits has been allocated are stored cyclically in a third area from a 3n-th bit to a bottommost bit of said fourth component.

10. An image signal processing device according to claim 1, further comprising:

decoding means for reading out said compressed image data from said recording medium and decoding from said compressed image data said quantized orthogonal transformation coefficients and said fourth component;

second inverse quantization means for inversely quantizing said quantized orthogonal transformation coefficients to reproduce said orthogonal transformation coefficients;

error data separating means for separating error data of the components comprising said compressed image data from said fourth component;

adding means for adding said error data to said orthogonal transformation coefficients; and

inverse orthogonal transformation means for performing an inverse orthogonal transform on said orthogonal transformation coefficients outputted from said adding means so as to reproduce image data.

11. An image signal processing device according to claim 1, further comprising:

decoding means for reading out the compressed image data from the recording medium and decoding quantized orthogonal transformation coefficients and an error component from the compressed image data;

second inverse quantization means for inversely quantizing said quantized orthogonal transformation coefficients to reproduce said orthogonal transformation coefficients;

error data separating means for separating error data of components comprising the compressed image data from said error component;

adding means for adding said error data to said orthogonal transformation coefficients; and

inverse orthogonal transformation means for performing an inverse orthogonal transform on the orthogonal transformation coefficients outputted from said adding means so as to reproduce image data.

12. An image signal processing device comprising:

orthogonal transformation means for applying orthogonal transformation to original image data comprising at least one image component to obtain orthogonal transformation coefficients for each image component, said original image data comprising a first, a second and a third component;

quantization means for quantizing said orthogonal transformation coefficients, comprising quantization tables including quantization coefficients to obtain the quantized orthogonal transformation coefficients;

first inverse quantizing and inverse orthogonal transformation means for inverse quantization and inverse orthogonal transformation of said quantized orthogonal transformation coefficients to obtain restored inverse orthogonal transformation coefficients;

error data producing means for obtaining error data of a difference between said restored inverse orthogonal transformation coefficients and said original image data for each component;

encoding means for encoding said error data and said quantized orthogonal transformation coefficients to form compressed image data;

error data synthesizing means for arranging said error data into unit data of a fixed length for each same spatial frequency of components to synthesize a fourth component, and said encoding means encodes said fourth component and said quantized orthogonal transformation coefficients to form compressed image data; and

recording means for recording said compressed image data in a recording medium.

13. An image signal processing device according to claim 12, further comprising:

decoding means for reading out said compressed image data from said recording medium and decoding from said compressed image data said quantized orthogonal transformation coefficients and fourth component;

second inverse quantization means for inversely quantizing said quantized orthogonal transformation coefficients and reproducing said orthogonal transformation coefficients;

error data separating means for separating said error data of the components comprising said compressed image data from said fourth component;

inverse orthogonal transformation means for performing an inverse orthogonal transform on said orthogonal transformation coefficients to reproduce image data; and

adding means for adding said error data to said reproduced image data.

14. An image signal processing device according to claim 12, further comprising:

decoding means for reading out the compressed image data from the recording medium and decoding quantized orthogonal transformation coefficients and an error component from the compressed image data;

second inverse quantization means for inversely quantizing said quantized orthogonal transformation coefficients to reproduce orthogonal transformation coefficients;

error data separating means for separating error data of components comprising the compressed image data from said error component;

adding means for adding said error data to said orthogonal transformation coefficients; and

inverse orthogonal transformation means for performing an inverse orthogonal transform on said orthogonal transformation coefficients outputted from said adding means to reproduce image data.