Image encoding apparatus

• US 4,951,140 A
• Filed: 02/22/1989
• Issued: 08/21/1990
• Est. Priority Date: 02/22/1988
• Status: Expired due to Term

• Current Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Original Assignee: Kabushiki Kaisha Toshiba (Toshiba Corporation)
• Inventors: Ueno, Hideyuki, Dachiku, Kenshi
• Primary Examiner(s): Britton, Howard W.
• Application Number:
  US07/313,941
• Time in Patent Office: 545 Days
• Field of Search: 358/105, 358/133, 358/135, 358/136
• US Class Current: 348/413.1
• CPC Class Codes: H04N 19/107 between spatial and temporal predictive coding, e.g. picture refresh
• CPC Class Codes: H04N 19/115 Selection of the code volume for a coding unit prior to 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 19/132 Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
• CPC Class Codes: H04N 19/146 Data rate or code amount at the encoder output
• CPC Class Codes: H04N 19/152 by measuring the fullness of the transmission buffer
• CPC Class Codes: H04N 19/17 the unit being an image region, e.g. an object
• CPC Class Codes: H04N 19/20 using video object coding
• CPC Class Codes: H04N 19/23 with coding of regions that are present throughout a whole video segment, e.g. sprites, background or mosaic
• CPC Class Codes: H04N 19/30 using hierarchical techniques, e.g. scalability H04N19/63 takes precedence
• CPC Class Codes: H04N 19/503 involving temporal prediction adaptive coding with adaptive selection between spatial and temporal predictive coding H04N19/107; adaptive coding with adaptive selection among a plurality of temporal predictive coding modes H04N19/109
• CPC Class Codes: H04N 19/507 using conditional replenishment
• CPC Class Codes: H04N 19/587 involving temporal sub-sampling or interpolation, e.g. decimation or subsequent interpolation of pictures in a video sequence
• CPC Class Codes: H04N 19/593 involving spatial prediction techniques
• CPC Class Codes: H04N 19/61 in combination with predictive coding

Abstract

An image encoding apparatus has a memory for storing image data corresponding to an object for every frame, and a face region detecting circuit for detecting a difference between input frame image data and preceding frame image data, and extracting image data of a face region of the object on the basis of the difference data. More bits are allocated to the extracted face region than those to the remaining image region.

Claims

1. An image encoding apparatus for encoding a plurality of image data corresponding to a movable object having a specific portion and sequentially produced for every frame, comprising:

storing means for storing the image data for every frame;

specific image extracting means for detecting a difference between at least two of the image data to obtain movement data representing a movement of the movable object and including specific movement data corresponding to the specific portion, and extracting the specific movement data from the movement data corresponding to the difference; and

bit allocating means for allocating more bits to the image data corresponding to the extracted specific movement data than to the remaining image data.

2. An apparatus according to claim 1, wherein said specific image extracting means comprises subtracting means for calculating a difference between the image data read out from said storing means in units of frames, and outputting differential data corresponding to the movement data and including a plurality of pixels;

means for comparing the differential data with a predetermined threshold value;

means for detecting from the differential data the pixels each having a level exceeding a predetermined threshold value;

means for determining the specific portion from information corresponding to the number of the pixels detected by said pixel detecting means and outputting the image data representing the specific portion to said bit allocating means.

3. An apparatus according to claim 1, wherein said specific image extracting means comprises subtracting means for calculating a difference between the image data read out from said storing means in units of frames, and outputting differential data;

histogram forming means for forming at least X- and Y-axis histograms from the differential data; and

coordinate data output means for determining coordinates corresponding to the specific portion from the X- and Y-axis histograms and outputting the coordinate data to said bit allocating means.

4. An apparatus according to claim 1, wherein said specific image extracting means comprises means for limiting a processing range of 1-frame image data read out from said storing means to a predetermined region by the coordinate data, labeling means for labeling the limited region, and means for extracting the specific image data in accordance with data output from said labeling means.

5. An apparatus according to claim 1, wherein said bit allocating means comprises means for dividing the image data read out from said storing means into a plurality of blocks and outputting a plurality of data blocks;

means for comparing the data block and data block of the preceding image data, detecting movement of the specific portion from both the data blocks, and outputting movement data;

selecting means for selecting data block corresponding to the movement data from the preceding image data;

means for calculating a difference between the data block selected by said selecting means and the data block obtained by said dividing means and outputting block differential data; and

determining/ encoding means for determining a movement region on the basis of the block differential data and the movement data and encoding the data block corresponding to the movement region by a predetermined number of bits.

6. An apparatus according to claim 5, wherein said determining/encoding means encodes the data block by different numbers of bits, to form a plurality of encoded data blocks, and which further comprises means for selecting one of the encoded data blocks output from said determining/encoding means.

7. An apparatus according to claim 1, wherein said specific image extracting means comprises subtracting means for calculating a difference between the image data read out from said storing means in units of frames, and outputting difference data;

means for comparing the differential data with a predetermined threshold value;

histogram forming means for forming at least X- and Y-axis histograms, by means of a comparison result obtained by said comparing means; and

coordinate data output means for determining coordinates corresponding to the specific portion from the X- and Y-axis histograms and outputting the coordinate data to said bit allocating means.

8. An apparatus according to claim 7, wherein the image data is formed by a predetermined block including a number of pixels, said histogram forming means comprises counting means for counting pixels corresponding to the differential data and exceeding a threshold value and means for forming the histograms by sequentially storing the count values of said counting means.

9. An apparatus according to claim 7, wherein said coordinate data output means comprises mean value means for calculating means for comparing the Y- and X-axis histograms, and means for comparing the Y- and X-axis histograms on the basis of the means values and detecting changes in histograms below the mean values.

10. An image encoding apparatus for encoding a plurality of image data corresponding to a movable object having a specific portion and sequentially produced for every frame, comprising:

storing means for storing the image data for every frame;

specific image extracting means for outputting differential data corresponding to a difference between at least two of the image data to obtain movement data representing a movement of the movable object and including specific movement data corresponding to the specific portion, and extracting specific movement data from the movement data corresponding to the differential data;

buffer means for matching a generated data amount with an output data amount;

encoding means for encoding the specific movement data and outputting encoded data;

parameter generating means for generating a quantization parameter according to the specific movement data extracted by said extracting means and a remaining amount of said buffer means;

means for quantizing the encoded data output from said encoding means in accordance with the quantization parameter so as to allocate more bits to the specific image data than to the remaining image data, and outputting quantized data; and

output means for converting the quantized data to output data and outputting the output data to said buffer means.

11. An apparatus according to claim 10, wherein said parameter generating means comprises means for outputting quantizing step size data corresponding to the buffer amount, means for multiplying the quantizing step size data with one and a predetermined value α

(α

>

1) and outputting one- and predetermined number-fold quantizing step size data, and means for selecting the one- and predetermined number-fold quantizing step size data in accordance with the differential data and outputting the selected data to said quantizing means.

12. An apparatus according to claim 10, wherein said storing means stores image data corresponding to a human body including a head portion and a background, and said specific image extracting means comprises subtracting means for calculating a difference between the image data read out from said storing means for every frame and the preceding image data, and outputting difference data, means for forming at least X- and Y-axis histograms on the basis of the differential data, and coordinate data output means for determining coordinates corresponding to a vertex and side surfaces of the head portion and outputting coordinate data to said bit allocating means.

13. An apparatus according to claim 10, wherein said output means scan-converts the quantized data from sad quantizing means in accordance with the specific image data and decreasing the number of bits allocated to a region other than the specific image data.

14. An image encoding apparatus comprising:

generating means for generating a plurality of image data corresponding to a movable object having a specific portion and sequentially produced for every frame;

encoding means for encoding the image data and outputting encoded image data;

specific portion detecting means for detecting a difference between at least two of the image data to obtain movement data representing a movement of the movable object and including specific movement data corresponding to the specific portion, and outputting specific region data corresponding to the specific movement data from the image data;

determining means for comparing the specific region data with predetermined data to determine the validity of the specific region data, and outputting determination data; and

means for supplying the specific region data to said encoding means in response to the determination data from said determining means so that more bits are allocated to the image data corresponding to the specific portion than to remaining image data to encode the image data.

15. An apparatus according to claim 14, further comprising storing means for storing preceding specific region data and means for enlarging the specific region data stored in said storing means in accordance with a predetermined rule.

16. An image encoding apparatus comprising:

storing means for storing image data corresponding to an object having a specific portion for every frame;

means for dividing the image data into a plurality of blocks including M ×

N pixels;

predictor output means for predicting a pixel corresponding to each corner of each of said blocks with reference to already encoded pixels which are included in other adjacent blocks, and outputting a predictor; and

quantizing/encoding means for detecting an error between the predictor and an actual value of the pixel, and quantizing and encoding differential data corresponding to the error in accordance with a relationship between the error and a predetermined value,

wherein said predictor output means predicts a pixel other than the corner pixel with reference to the corner pixel and pixels in other blocks adjacent to the corner pixel, and

said quantizing/encoding means detects an error between the predictor and the actual value of the pixel, and quantizes and encodes differential data corresponding to the error in accordance with the relationship between the error and a predetermined value.

17. An apparatus according to claim 16, wherein said predictor output means comprises means for obtaining a predictor on the basis of an already encoded pixel which is adjacent to a pixel to be predicted in view of a space and time.

18. An apparatus according to claim 16, wherein said quantizing/encoding means comprises means for quantizing data indicating a value exceeding a predetermined threshold value or a threshold value corresponding to an output buffer amount at equal intervals or unequal intervals.

19. An image transmission apparatus comprising:

image data output means for generating image data of a movement region of an input image;

difference image output means for outputting a difference image of images of two adjacent frames of the input image;

detecting means for detecting an outline of the movement region in the difference image output from the difference image output means;

first storing means for storing input image data outside the outline of the movement region detected by said detecting means as a background while updating the stored data;

transmitting means for transmitting the background image data stored in said first storing means and the movement region image data;

receiving means for receiving the movement region image data and the background image data transmitted from said transmitting means;

means for reproducing the received movement region image data;

second storing means for storing the received background image data until new background data is received;

synthesizing means for synthesizing the background image data stored in said second storing means and the reproduced movement region image data; and

means for outputting the movement region image data and the background image data synthesized by said synthesizing means;

wherein said image data output means divides the input image into a plurality of blocks, and performs motion compensated coding with reference to only pixels other than a background in each block.

20. An image transmission apparatus comprising:

image data output means for generating image data of a movement region of an input image;

difference image output means for outputting a difference image of images of two adjacent frames of the input image;

detecting means for detecting an outline of the movement region in the difference image output from the difference image output means;

first storing means for storing input image data outside the outline of the movement region detected by said detecting means as a background while updating the stored data;

transmitting means for transmitting the background image data stored in said first storing means and the movement region image data;

receiving means for receiving the movement region image data and the background image data transmitted from said transmitting means;

means for reproducing the received movement region image data;

second storing means for storing the received background image data until new background data is received;

synthesizing means for synthesizing the background image data stored in said second storing means and the reproduced movement region image data; and

means for outputting the movement region image data and the background image data synthesized by said synthesizing means;

wherein said image data output means divides the input image into a plurality of blocks, and performs transform coding while substituting a luminance value of a background in each block with a value determined by a predetermined rule.