IMAGING APPARATUS, IMAGING CIRCUIT AND IMAGING METHOD

• US 20080291295A1
• Filed: 05/19/2008
• Published: 11/27/2008
• Est. Priority Date: 05/24/2007
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Nakajima, Ken, Kato, Fumiaki
• Granted Patent: US 8,013,910 B2
• Time in Patent Office: Days
• US Class Current: 348/231.99
• CPC Class Codes: H04N 19/124 Quantisation
• CPC Class Codes: H04N 19/136 Incoming video signal characteristics or properties
• CPC Class Codes: H04N 19/17 the unit being an image region, e.g. an object
• CPC Class Codes: H04N 19/42 characterised by implementation details or hardware specially adapted for video compression or decompression, e.g. dedicated software implementation H04N19/635 takes precedence
• CPC Class Codes: H04N 19/90 using coding techniques not provided for in groups H04N19/10-H04N19/85, e.g. fractals

Abstract

An image by using a solid-state imaging sensor, the imaging apparatus including: a compression section configured to compress image data by dividing the image data into blocks each composed of same color component pixels adjacent to each other as a unit of compression; a memory used for temporarily storing compressed image data; a decompression section configured to decompress the compressed image data read out from the memory; and a signal processing section configured to carry out an image-quality correction process on decompressed image data, wherein each of the blocks is split in advance into two quantization-subject areas, block types are distinguished from each other by the position of the inter-area boundary between the two quantization-subject areas, and the compression section has a dynamic-range computation sub-section, a block-type select sub-section, and a quantization processing sub-section are provided.

Claims

1. An imaging apparatus for taking an image by making use of a solid-state imaging sensor, said imaging apparatus comprising:

a compression section configured to compress image data, which is generated by said solid-state imaging sensor as a result of an imaging process, by dividing said image data into blocks each composed of a fixed number of same color component pixels adjacent to each other and taking each of said blocks as a unit of compression;

a memory used for temporarily storing compressed image data produced by said compression section;

a decompression section configured to decompress said compressed image data read out from said memory; and

a signal processing section configured to carry out an image-quality correction process on decompressed image data produced by said decompression section,

wherein each of said blocks is split in advance into two quantization-subject areas each composed of one or more pixels adjacent to each other,

a plurality of block types are defined in advance for each of said blocks,

said block types defined for any individual one of said blocks are distinguished from each other by the position of the inter-area boundary between said two quantization-subject areas in said individual block, and

said compression section has

a dynamic-range computation sub-section configured to compute a dynamic range of pixel data, of pixels in each of said quantization-subject areas for each of said quantization-subject areas in each of said blocks,

a block-type select sub-section configured to compute a sum of said dynamic ranges, which are each computed by said dynamic-range computation sub-section for one of said quantization-subject areas in any individual one of said blocks, for each of said block types defined for said individual block on the assumption that each of said block types has been selected for said individual block and select said block type having a smallest sum of said dynamic ranges among all said block types defined for said individual block as a proper block type for said individual block, and

a quantization processing sub-section configured to quantize a difference, which is obtained as a result of subtracting the smallest pixel data among all pixels in any individual one of said quantization-subject areas from the pixel data, of each of said pixels in said individual quantization-subject area, in accordance with said dynamic range, which is computed by said dynamic-range computation sub-section for said individual quantization-subject area, for every pixel in said quantization-subject areas which are included in any individual one of said blocks as quantization-subject areas having an inter-area boundary conforming to said proper block type selected by said block-type select sub-section for said individual block.

2. The imaging apparatus according to claim 1 wherein said compressed image data produced by said compression section for any individual one of said blocks includes:

pattern identification information used for identifying said proper block type selected by said block-type select sub-section for said individual block;

dynamic-range information on said dynamic ranges each computed by said dynamic-range computation sub-section for one of said quantization-subject areas included in said individual block as quantization-subject areas having an inter-area boundary conforming to said proper block type;

minimum information showing minimum values each found as said smallest pixel data among all pixels in each one of said quantization-subject areas included in said individual block; and

quantized data produced by said quantization processing sub-section for every pixel in said individual block.

3. The imaging apparatus according to claim 2 wherein said quantization processing sub-section quantizes pixel data of pixels in any individual one of said quantization-subject areas included in said individual block as quantization-subject areas having an inter-area boundary conforming to said proper block type selected for said individual block in order to generate quantized pixel data having a quantized-word length set for said proper block type.

4. The imaging apparatus according to claim 2 wherein

said quantization processing sub-section is configured to function as a right-direction bit shifter for shifting pixel data, of every pixel in a direction to the right, toward the least significant bit of said pixel data by as many bits as indicated by said dynamic range computed by said dynamic-range computation sub-section for said quantization-subject area including said pixel on the assumption that the number of quantization steps included in a quantization process carried out by said quantization processing sub-section is a power of 2, and

said dynamic-range information included in said compressed image data produced by said compression section as information on said dynamic ranges each computed by said dynamic-range computation sub-section for one of said two respective quantization-subject areas in said individual block is a shift quantity representing a bit count equal to the number of bits by which said quantization processing sub-section shifts pixel data, of even pixel included in a particular one of said two respective quantization-subject areas in a direction to the right toward the least significant bit of said pixel data and a shift quantity representing a bit count equal to the number of bits by which said quantization processing sub-section shifts pixel data of even pixel included in the other one of said two respective quantization-subject areas in a direction to the eight toward the least significant bit of said pixel data.

5. The imaging apparatus according to claim 4 wherein said decompression section decompresses image data by the steps of:

driving a left-direction bit shifter to carry out an inverse quantization process to shift quantized pixel data of every pixel in any individual one of said quantization-subject areas in a direction to the left toward the most significant bit of said quantized pixel data by as many bits as indicated by a shift, quantity included in said compressed image data produced by said compression section as said shift quantity for said individual quantization-subject area; and

adding a minimum value included in said compressed image data produced by said compression section as said minimum value of pixel data of all pixels in any individual one of said quantization-subject areas to pixel data obtained as a result of said inverse quantization process carried out on every pixel in said individual quantization-subject area.

6. The imaging apparatus according to claim 1 wherein each of said blocks includes pixel data of pixels existing in a rectangular area in said image taken by making use of said solid-state imaging sensor.

7. The imaging apparatus according to claim 6 wherein, for all said block types, said two quantization-subject areas in each of said blocks are separated from each other by a vertical inter-area boundary line.

8. The imaging apparatus according to claim 1 wherein:

said defined block types are a first block type representing a category of said blocks each split into two quantization-subject areas and a second block type representing a category of said blocks each handled entirely as a single quantization-subject area; and

said block-type select sub-section normalizes said sum of said dynamic ranges each computed for one of said quantization-subject areas included in any particular one of said blocks on the assumption that said first block type has been selected as said proper block type for said particular block, normalizes said dynamic range computed for said single quantization-subject area included in said particular block on the assumption that said second block type has been selected as said proper block type for said particular block and compares said normalized sum of said dynamic ranges with said normalized dynamic range in order to select one of said first and second block types as said proper block type having the smaller one of said normalized sum of said dynamic ranges and said normalized dynamic range for said particular block.

9. The imaging apparatus according to claim 8 wherein, instead of comparing said normalized sum of said dynamic ranges for said first block type with said normalized dynamic range for said second block type, said block-type select sub-section compares said normalized sum with a value smaller than said normalized dynamic range.

10. The imaging apparatus according to claim 9 wherein said quantization processing sub-section sets a quantized-word length of pixel data of every pixel in said quantization-subject areas in a quantization process based on said second block type selected by said block-type select sub-section as said proper block type for said block including said quantization-subject areas at a value greater than the quantized-word length of pixel data of every pixel in said quantization-subject areas in a quantization process based on said first block type selected by said block-type select sub-section as said proper block type for said block including said quantization-subject areas.

11. The imaging apparatus according to claim 9 wherein said block-type select sub-section selects one of said first and second block types as said proper block type having said smaller one of said normalized sum of said dynamic ranges and said value smaller than said normalized dynamic range for said particular block as evidenced by a result of comparing said normalized sum of said dynamic ranges with said value smaller than said normalized dynamic range.

12. The imaging apparatus according to claim 11 wherein:

each of said blocks includes pixel data of 12 pixels, that is 6 pixels per line×

2 lines, in a rectangular area in said image taken by making use of a solid-state imaging sensor;

said first block type is defined to cover

a block type representing said blocks each including a first quantization-subject area placed at the beginning of said block as a quantization-subject area having 4 pixels, that is 2 pixels per line×

2 lines, and a second quantization-subject area placed at a location following said first quantization-subject area as a quantization-subject area having 8 pixels, that is 4 pixels per line×

2 lines,

a block type representing said blocks each including a first quantization-subject area placed at the beginning of said block as a quantization-subject area having 8 pixels, that is 4 pixels per line×

2 lines, and a second quantization-subject area placed at a location following said first quantization-subject area as a quantization-subject area having 4 pixels, that is 2 pixels per line×

2 lines, and

a block type representing said blocks each including a first quantization-subject area placed at the beginning of said block as a quantization-subject area having 6 pixels, that is 3 pixels per line×

2 lines, and a second quantization-subject area placed at a location following said first quantization-subject area as a quantization-subject area having 6 pixels, that is 3 pixels per line×

2 lines;

arid

if the input data of each pixel has a length of 10 bits, said quantization processing sub-section sets a quantized-word length of pixel data of even- pixel in said quantization-subject areas in a quantization process based on said first block type selected by said block-type select sub-section as said proper block type for said block including said quantization-subject areas at 3 bits and a quantized-word length of pixel data of every pixel in said quantization-subject areas in a quantization process based on said second block type selected by said block-type select sub-section as said proper block type for said block including said quantization-subject areas at 4 bits.

13. The imaging apparatus according to claim 1 wherein said dynamic-range computation sub-section computes said dynamic range for any individual one of said quantization-subject areas by finding a difference between the maximum value of pixel data of all pixels in said individual quantization-subject area and the minimum value of pixel data of all pixels in said individual quantization-subject area.

14. The imaging apparatus according to claim 1, said imaging apparatus further comprising;

a pre-compression section provided at a stage preceding said compression section as a section configured to carry out a pre-compression procedure including a gradation changing process in order to change the gradation of said image taken by making use of said solid-state imaging sensor according to human visual performance characteristics as well as a bit-count reduction process in order to reduce the number of bits included in image data obtained as a result of said gradation changing process and supply image data obtained as a result of said bit-count reduction process to said compression section; and

a post-decompression section provided at a stage succeeding said decompression section as a section configured to decompress image data output by said decompression section in accordance with a post-decompression procedure set by reversing said pre-compression procedure followed by said pre-compression section.

15. The imaging apparatus according to claim 1 wherein:

said compression section, said decompression section and said signal processing section are connected to said memory by a common bus;

after said compressed image data generated by said compression section is stored in said memory through said common bus, said compressed image data is supplied to said compression section through said common bus; and

decompressed image data, generated by said decompression section is supplied to said signal processing section without transferring said decompressed image data through said common bus.

16. An imaging circuit for taking an image, said imaging circuit having a structure created on a single chip as a structure comprising:

a solid-state imaging sensor configured to convert incoming light into an electrical imaging analog signal;

an analog-to-digital conversion circuit configured to convert said electrical imaging analog signal generated by said solid-state imaging sensor into a digital image signal; and

a compression circuit configured to compress said digital image signal, which is generated by said analog-to-digital conversion circuit, by dividing said digital image signal into blocks each composed of a fixed number of same color component pixels adjacent to each other and taking each of said blocks as a unit of compression,

wherein each of said blocks is split in advance into two quantization-subject areas each composed of one or more pixels adjacent to each other.

a plurality of block types are defined in advance for each of said blocks,

said block types defined for any individual one of said blocks are distinguished from each other by the position of the inter-area boundary between said two quantization-subject areas in said individual block, and

said compression circuit has

a dynamic-range computation section configured to compute a dynamic range of pixel data of pixels in each of said quantization-subject areas for each of said quantization-subject areas in each of said blocks,

a block-type select section configured to compute a sum of said dynamic ranges, which are each computed by said dynamic-range computation section for one of said quantization-subject areas in any individual one of said blocks, for each of said block types defined for said individual block on the assumption that each of said block types has been selected for said individual block and to select said block type having a smallest sum of said dynamic ranges among all said block types defined for said individual block as a proper block type for said individual block, and

a quantization processing section configured to quantize a difference, which is obtained as a result of subtracting the smallest pixel data among all pixels in any individual one of said quantization-subject areas from the pixel data of each of said pixels in said individual quantization-subject area, in accordance with said dynamic range, which is computed by said dynamic-range computation section for said individual quantization-subject area, for every pixel in said quantization-subject areas which are included in any individual one of said blocks as quantization-subject areas having an inter-area boundary conforming to said proper block type selected by said block-type select section for said individual block.

17. An imaging method for taking an image by making use of a solid-state imaging sensor, said imaging method comprising:

compressing, performed by a compression section, image data, which is generated by said solid-state imaging sensor as a result of an imaging process, by dividing said image data into blocks each composed of a fixed number of same color component pixels adjacent to each other and by taking each of said blocks as a unit of compression and storing compressed image data obtained as a result of said compression process;

decompressing, performed by a decompression section, compressed image data read out from said memory; and

correcting, performed by a signal processing section, image-quality of decompressed image data produced by said decompression section,

wherein each of said blocks is split in advance into two quantization-subject areas each composed of one or more pixels adjacent to each other,

a plurality of block types are defined in advance for each of said blocks,

said block types defined for any individual one of said blocks are distinguished from each other by the position of the inter-area boundary between said two quantization-subject areas in said individual block, and

said compression step is carried out by executing the steps of

computing a dynamic range of pixel data of pixels in each of said quantization-subject areas for each of said quantization-subject areas in each of said blocks,

computing a sum of said dynamic ranges, which are each computed at said dynamic-range computation sub-step for one of said quantization-subject areas in any individual one of said blocks, for each of said block types defined for said individual block on the assumption that each of said block types has been selected for said individual block and selecting said block type having a smallest sum of said dynamic ranges among all said block types defined for said individual block as a proper block type for said individual block, and

quantizing a difference, which is obtained as a result of subtracting the smallest pixel data among all pixels in any individual one of said quantization-subject areas from the pixel data of each of said pixels in said individual quantization-subject area, in accordance with said dynamic range, which is computed at said dynamic-range computation sub-step for said individual quantization-subject area, for every pixel in said quantization-subject areas which are included in any individual one of said blocks as quantization-subject areas having an inter-area boundary conforming to said proper block type selected at said block-type select sub-step for said individual block.