Image conversion apparatus

Claim 2

An image conversion apparatus according to Claim 1, wherein said resolution converter means is a spatial resolution converter means (28, 40).

Claim 3

An image conversion apparatus according to Claim 1, wherein said resolution converter means is a temporal resolution converter means.

Claim 4

An image conversion apparatus according to Claim 1, wherein said resolution converter means is a quantization level converter means (4; 202, 301).

Claim 5

An image conversion apparatus according to Claim 1, wherein said image decoder (1) comprises: variable length decoder means (11) for decoding said primary compressed image data in variable length; inverse quantizer means (12) for inverse quantizing the decoded image data for producing an inverse quantized image data; inverse DCT processor means (13) for executing an inverse DCT operation of the inverse quantized image data in units and for generating conversion coefficients; adder means (13') for adding said conversion coefficients with a predicted image data, and for producing a decoded image data with a plurality of conversion coefficients; memory means (14) for storing said decoded image data of one unit; and compensation means (15) for generating said predicted image data which is applied to said adder means, whereby said adder means continuously produces decoded image data.

Claim 6

An image conversion apparatus according to Claim 5, wherein said variable length decoder means (11) generates side information including DCT mode information, motion compensation information, and motion vector information, said DCT mode information being applied to said inverse DCT processor means (13), said motion compensation information being applied to said compensation means (15), and said motion vector information being applied to said memory means (14).

Claim 7

An image conversion apparatus according to Claim 1, further comprising a source of said primary compressed image data An.

Claim 8

An image conversion apparatus according to Claim 7, wherein said source is a broadcasted compressed image data An.

Claim 9

An image conversion apparatus according to Claim 7, wherein said source is a memory means.

Claim 10

An image conversion apparatus according to Claim 1, further comprising a second image coder (3; 3A; 3B) having a second resolution converter means (40; 4, 34; 301) for converting the decoded image data Rn, input thereto from the image decoder, to a second secondary compressed image data Bn2 having a third degree of resolution, and outputting the second secondary compressed image data Bn2.

Claim 11

An image conversion apparatus according to Claim 1, further comprising selecting means (61) for manually selecting a desired degree of resolution of said first resolution converter means.

Claim 12

An image conversion apparatus according to Claim 10, further comprising selecting means (61) for manually selecting one of said first and second resolution converter means.

Claim 13

An image conversion apparatus according to Claim 5, wherein said resolution converter means comprises: a plurality of coefficient selectors each for selecting from among the conversion coefficients output by said adder means (13') a predetermined number of conversion coefficients corresponding to a particular image resolution; and a plurality of variable length coders each for variable length coding the data from the plurality of coefficient selectors.

Claim 14

An image recompression method comprising the steps of: inverse quantizing a compressed image data (a), that has been previously quantized by a first quantization parameter (q), by the use of said first quantization parameter (q) to generate a first inverse quantized image data (b); subtracting a compensation signal (z) based on a previous image data from said first inverse quantized image data (b) and for producing a difference data (b'), requantizing said difference data (b') using a second quantization parameter (qN), and for generating a requantized image data (c); reinverse quantizing said requantized image data (c) using said second quantization parameter (qN) to generate a second inverse quantized image data (d); and adding to said second inverse quantized image data (d) said compensation signal (z) and further subtracting said first inverse quantized image data (b) to generate a compensation signal (z') of the next image data.

Claim 15

An image recompression method comprising the steps of: inverse quantizing a compressed image data (a), that has been previously quantized by a first quantization parameter (q), by the use of said first quantization parameter (q) to generate a first inverse quantized image data (b); subtracting a compensation signal (z) based on a previous image data from said first inverse quantized image data (b) and for producing a difference data (b'), requantizing said difference data (b') using a second quantization parameter (qN), and for generating a requantized image data (c); reinverse quantizing said requantized image data (c) using said second quantization parameter (qN) to generate a second inverse quantized image data (d); and subtracting said difference data (b') from said second inverse quantized image data (d) to generate a compensation signal (z') of the next image data.

Claim 16

An image recompression method comprising the steps of: subtracting a compensation coefficient generated according to a previous image data from a first conversion coefficient block contained in a compressed orthogonal transform image data to generate a second conversion coefficient block; extracting a predetermined number of coefficients from said second conversion coefficient block to generate a third conversion coefficient block; and generating a difference coefficient block using said third conversion coefficient block and said first conversion coefficient block, said difference coefficient being use as a compensation coefficient of the next image data.

Claim 17

An image recompression method as claimed in Claim 16, further comprising the step of adding a compensation coefficient generated from the past image data to the third conversion coefficient block, before the generating step.

Claim 18

An image recompression apparatus comprising: variable length decoder means (110) for decoding from a coded image input data, a first quantization parameter (q), a compressed image data (a), that has been previously quantized by a first quantization parameter (q), and other information data; first inverse quantizer means (130) for inverse quantizing said compressed image data (a) by the use of said first quantization parameter (q) to generate a first inverse quantized image data (b), first subtractor means (210) for subtracting a compensation signal (z) based on a previous image data from said first inverse quantized image data (b) and for producing a difference data (b'), quantizer means (140) for requantizing said difference data (b') using a second quantization parameter (qN), and for generating a requantized image data (c); second inverse quantizer means (170) for reinverse quantizing said requantized image data (c) using said second quantization parameter (qN) to generate a second inverse quantized image data (d); adder means (220) for adding to said second inverse quantized image data (d) said compensation signal (z) and for generating a sum data (e); second subtractor means (230) for subtracting said second inverse quantized image data (d) from said sum data (e) to generate a compensation signal (z') of the next image data; memory means (180) for storing said compensation signal (z'); variable length coder means (150) for converting said second quantization parameter (qN) and said requantized image data (c) to variable length code, and producing said variable length code; multiplexer means (160) for multiplexing said variable length code and said other information data to generate a compressed coded image input data.

Claim 19

An image recompression apparatus according to Claim 18, wherein said memory means (180) comprises: an inverse orthogonal transform processor means (182) applying an inverse orthogonal transform to said compensation signal (z') which is in a form of a coefficient difference; coefficient difference memory means (184) for storing said coefficient difference; and orthogonal transform processor means (186) for applying an orthogonal transform to the coefficient difference.

Claim 20

An image recompression apparatus according to Claim 18, further comprising: inverse orthogonal transform processor means (240) provided in association with said first inverse quantizer means (130) to generate said first inverse quantized image data (b) in an inverse orthogonal transformed format; orthogonal transform processor means (250) provided in association with said subtractor (210) for generating said difference data (b') in an orthogonal transformed format; reinverse orthogonal transform processor means (260) provided in association with said second inverse quantizer means (170) for generating said second inverse quantized image data (d) in an inverse orthogonal transformed format.

Claim 21

An image recompression apparatus comprising: variable length decoder means (110) for decoding from a coded image input data, a first conversion coefficient block (a), and other information data; first subtractor means (210) for subtracting a compensation coefficient (z), based on a previous image data, from said first conversion coefficient block (a) to generate a second conversion coefficient block (a'); selector means (140) for selecting a predetermined number of coefficients from said second conversion coefficient block (a'), and forming a third conversion coefficient block (u); adder means (220) for adding to said third conversion coefficient block (u) said compensation coefficient (z) and for generating a sum data (w); second subtractor means (230) for subtracting said first conversion coefficient block (a) said sum data (w) to generate a compensation coefficient (z') of the next image data; memory means (180) for storing said compensation coefficient (z'); variable length coder means (150) for converting said third conversion coefficient block (u) to variable length code, and producing said variable length code; and multiplexer means (160) for multiplexing said variable length code and said other information data to generate a compressed coded image input data.

Claim 22

An image recompression apparatus according to Claim 21, wherein said memory means (180) comprises: an inverse orthogonal transform processor means (182) applying an inverse orthogonal transform to said compensation coefficient (z') which is in a form of a coefficient difference; coefficient difference memory means (184) for storing said coefficient difference; and orthogonal transform processor means (186) for applying an orthogonal transform to the coefficient difference.

Claim 23

An image recompression apparatus according to Claim 18, further comprising switching means (360) for selectively producing either one of said coded image input data and said compressed coded image input data, and storing means (340) for storing the selected data.

Claim 24

An image recompression apparatus according to Claim 21, further comprising switching means (360) for selectively producing either one of said coded image input data and said compressed coded image input data and storing means (340) for storing the selected data.

Claim 25

An image conversion apparatus according to Claim 1, wherein said image decoder (1) generates side information including DCT mode information, motion compensation information, and motion vector information, which are applied to said first image coder (2).