Image processing apparatus and method, and computer-readable memory

Claim 2

An image processing apparatus comprising: first feature extraction means for extracting a first feature from first encoded data of a first image; second feature extraction means for extracting a second feature from second encoded data of a second image; first decoding means for obtaining a first reconstructed image by decoding the first encoded data; second decoding means for obtaining a second reconstructed image by decoding the second encoded data; correction means for correcting one of the first and second reconstructed images based on the first and second features, wherein said correction means performs a correction to reduce a color offset or a contrast difference between the first and second reconstructed images; and synthesis means for synthesizing the first and second reconstructed images.

Claim 3

The apparatus according to claim 1 , wherein the first image is a background image.

Claim 4

The apparatus according to claim 1 , wherein the second image is a principal object image.

Claim 5

The apparatus according to claim 1 , wherein the first and second features are luminance values of images.

Claim 6

The apparatus according to claim 1 , wherein the first and second features are average values of luminance values of images.

Claim 7

The apparatus according to claim 1 , wherein the first and second features are maximum values of luminance values of images.

Claim 8

The apparatus according to claim 1 , wherein the first and second features are chromaticity values of images.

Claim 9

The apparatus according to claim 1 , wherein the first and second features are maximum luminance values of achromatic color of images.

Claim 10

The apparatus according to claim 3 , wherein said correction means corrects the principal object image.

Claim 11

The apparatus according to claim 1 , wherein the first and second encoded data are obtained by encoding image signals in a frequency domain.

Claim 12

The apparatus according to claim 1 , wherein the first and second features are chromaticity values of images.

Claim 13

The apparatus according to claim 1 , wherein the first and second encoded data are obtained by moving-image encoding moving image signals in a frequency domain.

Claim 14

An image processing method comprising: a first feature extraction step of extracting a first feature from first encoded data of a first image; a second feature extraction step of extracting a second feature from second encoded data of a second image; a first decoding step of obtaining a first reconstructed image by decoding the first encoded data; a second decoding step of obtaining a second reconstructed image by decoding the second encoded data; a correction step of correcting one of the first and second reconstructed images based on the first and second features, wherein said correction step includes performing a correction to reduce a color offset or a contrast difference between the first and second reconstructed images; and a synthesis step of synthesizing the first and second reconstructed images.

Claim 15

An image processing method comprising: a first feature extraction step of extracting a first feature from first encoded data of a first image; a second feature extraction step of extracting a second feature from second encoded data of a second image; a first decoding step of obtaining a first reconstructed image by decoding the first encoded data; a second decoding step of obtaining a second reconstructed image by decoding the second encoded data; a correction step of correcting one of the first and second reconstructed images based on the first and second features, wherein said correction step includes performing a correction to reduce a color offset or a contrast difference between the first and second reconstructed images; and a synthesis step of synthesizing the first and second reconstructed images.

Claim 16

The method according to claim 13 , wherein the first image is a background image.

Claim 17

The method according to claim 13 , wherein the second image is a principal object image.

Claim 18

The method according to claim 13 , wherein the first and second features are luminance values of images.

Claim 19

The method according to claim 13 , wherein the first and second features are average values of luminance values of images.

Claim 20

The method according to claim 13 , wherein the first and second features are maximum values of luminance values of images.

Claim 21

The method according to claim 13 , wherein the first and second features are chromaticity values of images.

Claim 22

The method according to claim 13 , wherein the first and second features are maximum luminance values of achromatic color of images.

Claim 23

A method according to claim 15 , wherein said correction step includes correcting the principal object image.

Claim 24

The method according to claim 13 , wherein the first and second encoded data are obtained by encoding image signals in a frequency domain.

Claim 25

The method according to claim 13 , wherein the first and second features are chromaticity values of images.

Claim 26

The method according to claim 13 , wherein the first and second encoded data are obtained by moving-image encoding moving image signals in a frequency domain.

Claim 27

A computer-readable memory storing a program for implementing an image processing method, the program comprising: program code of a first feature extraction step of extracting a first feature from first encoded data of a first image; program code of a second feature extraction step of extracting a second feature from second encoded data of a second image; program code of a first decoding step of obtaining a first reconstructed image by decoding the first encoded data; program code of a second decoding step of obtaining a second reconstructed image by decoding the second encoded data; program code of a correction step of correcting one of the first and second reconstructed images based on the first and second features, wherein the correction step includes performing a correction to reduce a color offset or a contrast difference between the first and second reconstructed images; and program code of a synthesis step of synthesizing the first and second reconstructed images.

Claim 28

A computer-readable memory storing a program for implementing an image processing method, the program comprising: program code of a first feature extraction step of extracting a first feature from first encoded data of a first image; program code of a second feature extraction step of extracting a second feature from second encoded data of a second image; program code of a first decoding step of obtaining a first reconstructed image by decoding the first encoded data; program code of a second decoding step of obtaining a second reconstructed image by decoding the second encoded data; program code of a correction step of correcting one of the first and second reconstructed images based on the first and second features, wherein the correction step includes performing a correction to reduce a color offset or a contrast difference between the first and second reconstructed images; and program code of a synthesis step of synthesizing the first and second reconstructed images.

Claim 29

An image processing apparatus for synthesizing a plurality of images, comprising: background feature extraction means for extracting a background feature from encoded data of at least one background image; object feature extraction means for extracting an object feature including statistic information of image information from encoded data of at least one object image; background decoding means for generating a reconstructed background image by decoding the encoded data of the background image; object decoding means for generating a reconstructed object image by decoding the encoded data of the object image; correction means for correcting the reconstructed object image based on the background and object features; and synthesis means for synthesizing the reconstructed background image and the reconstructed object image corrected by said correction means, wherein said correction means performs a correction to reduce a color offset or a contrast difference between the reconstructed background image and the reconstructed object image.

Claim 30

An image processing apparatus for synthesizing a plurality of images, comprising: background feature extraction means for extracting a background feature from encoded data of at least one background image; object feature extraction means for extracting an object feature including statistic information of image information from encoded data of at least one object image; background decoding means for generating a reconstructed background image by decoding the encoded data of the background image; object decoding means for generating a reconstructed object image by decoding the encoded data of the object image; correction means for correcting the reconstructed object image based on the background and object features; and synthesis means for synthesizing the reconstructed background image and the reconstructed object image corrected by said correction means, wherein said correction means performs a correction to reduce a color offset or a contrast difference between the reconstructed background image and the reconstructed object image.

Claim 31

The apparatus according to claim 26 , wherein said object feature extraction means comprises calculation means for calculating a histogram based on the statistic information of the image information, and said correction means determines a correction method for the object image based on the histogram.

Claim 32

The apparatus according to claim 26 , wherein said object feature extraction means extracts DC information of block images included in the encoded data as the statistic information of the image information.

Claim 33

The apparatus according to claim 26 , wherein said object feature extraction means extracts low-frequency information of block images included in the encoded data as the statistic information of the image information.

Claim 34

The apparatus according to claim 29 , wherein one or both of said background decoding means and object decoding means comprise: decoding means for decoding the encoded data to obtain quantized data; dequantization means for calculating frequency domain data from the quantized data; and fast inverse discrete cosine transform means for calculating space domain data from the frequency domain data, wherein said fast inverse discrete cosine transform means comprises output means for outputting an arbitrary number of stages of radix butterfly operation results, and wherein said object feature extraction means extracts the arbitrary number of stages of radix butterfly operation results as the low-frequency information of the image information.

Claim 35

The apparatus according to claim 26 , wherein said correction means comprises time-sequence adaptive means for slowly changing an input/output relationship between input and output signals of said correction means time-sequentially.

Claim 36

The apparatus according to claim 26 , wherein said object feature extraction means extracts maximum and minimum values of pixel values from one of DC information and low-frequency information of block images included in the encoded data as the statistic information of the image data.

Claim 37

The apparatus according to claim 26 , wherein said object feature extraction means extracts a variance and average value of pixel values from one of DC information and low-frequency information of block images included in the encoded data as the statistic information of the image data.

Claim 38

The apparatus according to claim 26 , wherein said correction means converts the object image by a linear function.

Claim 39

The apparatus according to claim 26 , wherein said correction means converts the object image by an interval spline function.

Claim 40

The apparatus according to claim 26 , wherein said correction means comprises: detection means for detecting a presence/absence of a significant color offset from the object feature extracted by said object feature extraction means; and color correction means for correcting the color offset based on a detection result of the detection means.

Claim 41

The apparatus according to claim 36 , wherein the detection means performs a detection based on the statistic information included in the extracted object feature if a condition in which an absolute value of a difference between an average value and a variance is not more than a given threshold value is satisfied between each respective color signal, and further detects the presence/absence of the significant color offset in a specific region of a histogram based on the statistic information when the condition is satisfied.

Claim 42

The apparatus according to claim 36 , wherein the color correction means linearly corrects color signals to make the color signals have equal maximum values.

Claim 43

The apparatus according to claim 36 , wherein the color correction means does not correct a blue signal.

Claim 44

The apparatus according to claim 26 , wherein said correction means comprises: detection means for detecting a significant contrast difference between the object feature extracted by said object feature extraction means and the background feature extracted by said background feature extraction means; and contrast correction means for correcting a contrast based on a detection result of the detection means.

Claim 45

The apparatus according to claim 40 , wherein the detection means extracts maximum and minimum pixel values obtained from the object and background features, and the contrast correction means performs a correction to decrease an absolute value of a difference between the maximum pixel values and an absolute value of a difference between the minimum pixel values in the object and background images, which have different maximum or minimum pixel values.

Claim 46

The apparatus according to claim 40 , wherein the detection means extracts maximum and minimum pixel values obtained from the object and background features, and the contrast correction means performs a correction to decrease an absolute value of a difference between variances in the object and background images, which have substantially equal maximum or minimum pixel values.

Claim 47

An image processing method for synthesizing a plurality of images, comprising: a background feature extraction step of extracting a background feature from encoded data of at least one background image; an object feature extraction step of extracting an object feature including statistic information of image information from encoded data of at least one object image; a background decoding step of generating a reconstructed background image by decoding the encoded data of the background image; an object decoding step of generating a reconstructed object image by decoding the encoded data of the object image; a correction step of correcting the reconstructed object image based on the background and object features; and a synthesis step of synthesizing the reconstructed background image and the reconstructed object image corrected in said correction step, wherein said correction step includes performing a correction to reduce a color offset or a contrast difference between the reconstructed background image and the reconstructed object image.

Claim 48

An image processing method for synthesizing a plurality of images, comprising: a background feature extraction step of extracting a background feature from encoded data of at least one background image; an object feature extraction step of extracting an object feature including statistic information of image information from encoded data of at least one object image; a background decoding step of generating a reconstructed background image by decoding the encoded data of the background image; an object decoding step of generating a reconstructed object image by decoding the encoded data of the object image; a correction step of correcting the reconstructed object image based on the background and object features; and a synthesis step of synthesizing the reconstructed background image and the reconstructed object image corrected in said correction step, wherein said correction step includes performing a correction to reduce a color offset or a contrast difference between the reconstructed background image and the reconstructed object image.

Claim 49

The method according to claim 43 , wherein the object feature extraction step comprises a calculation step of calculating a histogram based on the statistic information of the image information, and said correction step includes determining a correction method for the object image based on the histogram.

Claim 50

The method according to claim 43 , wherein the object feature extraction step includes extracting DC information of block images included in the encoded data as the statistic information of the image information.

Claim 51

The method according to claim 43 , wherein the object feature extraction step includes extracting low-frequency information of block images included in the encoded data as the statistic information of the image information.

Claim 52

The method according to claim 46 , wherein one or both of the background decoding step and object decoding step comprise: a decoding step of decoding the encoded data to obtain quantized data; a dequantization step of calculating frequency domain data from the quantized data; and a fast inverse discrete cosine transform step of calculating space domain data from the frequency domain data, wherein the fast inverse discrete cosine transform step comprises an output step of outputting an arbitrary number of stages of radix butterfly operation results, and said object feature extraction step includes extracting the arbitrary number of stages of radix butterfly operation results as the low-frequency information of the image information.

Claim 53

The method according to claim 43 , wherein said correction step comprises a time-sequence adaptive step of slowly changing an input/output relationship between input and output signals in said correction step time-sequentially.

Claim 54

The method according to claim 43 , wherein said object feature extraction step includes extracting maximum and minimum values of pixel values from one of DC information and low-frequency information of block images included in the encoded data as the statistic information of the image data.

Claim 55

The method according to claim 43 , wherein said object feature extraction step includes extracting a variance and average value of pixel values from one of DC information and low-frequency information of block images included in the encoded data as the statistic information of the image data.

Claim 56

The method according to claim 43 , wherein said correction step includes converting the object image by a linear function.

Claim 57

The method according to claim 43 , wherein said correction step includes converting the object image by an interval spline function.

Claim 58

The method according to claim 43 , wherein said correction step comprises: a detection step of detecting a presence/absence of a significant color offset from the object feature extracted in said object feature extraction step; and a color correction step of correcting the color offset based on a detection result in the detection step.

Claim 59

The method according to claim 53 , wherein the detection step includes detecting, based on the statistic information included in the extracted object feature if a condition in which an absolute value of a difference between an average value and a variance is not more than a given threshold value is satisfied between respective color signals, and further includes detecting the presence/absence of the significant color offset in a specific region of a histogram based on the statistic information when the condition is satisfied.

Claim 60

The method according to claim 53 , wherein the color correction step includes linearly correcting color signals to make the color signals have equal maximum values.

Claim 61

The method according to claim 53 , wherein a blue signal is not corrected in the color correction step.

Claim 62

The method according to claim 53 , wherein said correction step comprises: a detection step of detecting a significant contrast difference between the object feature extracted in said object feature extraction step and the background feature extracted in said background feature extraction step; and a contrast correction step of correcting a contrast based on a detection result in the detection step.

Claim 63

The method according to claim 57 , wherein the detection step includes extracting maximum and minimum pixel values obtained from the object and background features, and the contrast correction step includes performing a correction to decrease an absolute value of a difference between the maximum pixel values and an absolute value of a difference between the minimum pixel values in the object and background images, which have different maximum or minimum pixel values.

Claim 64

The method according to claim 57 , wherein the detection step includes extracting maximum and minimum pixel values obtained from the object and background features, and the contrast correction step includes performing a correction to decrease an absolute value of a difference between variances in the object and background images, which have substantially equal maximum or minimum pixel values.

Claim 65

A computer-readable memory storing a program for implementing an image processing method for synthesizing a plurality of images, the program comprising: program code of a background feature extraction step of extracting a background feature from encoded data of at least one background image; program code of an object feature extraction step of extracting an object feature including statistic information of image information from encoded data of at least one object image; program code of a background decoding step of generating a reconstructed background image by decoding the encoded data of the background image; program code of an object decoding step of generating a reconstructed object image by decoding the encoded data of the object image; program code of a correction step of correcting the reconstructed object image based on the background and object features; and program code of a synthesis step of synthesizing the reconstructed background image and the reconstructed object image corrected in the correction step, wherein the correction means includes performing a correction to reduce a color offset or a contrast difference between the reconstructed background image and the reconstructed object image.

Claim 66

A computer-readable memory storing a program for implementing an image processing method for synthesizing a plurality of images, the program comprising: program code of a background feature extraction step of extracting a background feature from encoded data of at least one background image; program code of an object feature extraction step of extracting an object feature including statistic information of image information from encoded data of at least one object image; program code of a background decoding step of generating a reconstructed background image by decoding the encoded data of the background image; program code of an object decoding step of generating a reconstructed object image by decoding the encoded data of the object image; program code of a correction step of correcting the reconstructed object image based on the background and object features; and program code of a synthesis step of synthesizing the reconstructed background image and the reconstructed object image corrected in the correction step, wherein the correction means includes performing a correction to reduce a color offset or a contrast difference between the reconstructed background image and the reconstructed object image.