Histogram/variance mechanism for detecting presence of an edge within block of image data

Claim 2

For use with a signal processing mechanism for compressing signals representative of an image for transmission over a bandwidth-limited communication channel, in which blocks of transform coefficients, that have been obtained by subjecting said signals to a spatial frequency block transformation process, are quantized and the quantized transform coefficients encoded for transmission over said channel, a method of detecting the presence of an edge within an image block for which a corresponding block of image transform coefficients is obtained, comprising the steps of: (a) generating a histogram of said image block; (b) determining prescribed variance characteristics of said image block in accordance with transform coefficients associated with said image block; and (c) identifying whether or not an edge is present within an image block in dependence upon said histogram and said prescribed image variance characteristics.

Claim 3

A method according to claim 1, wherein step (c) comprises identifying whether or not an edge is present within an image block in dependence upon the maximum value of said histogram and said prescribed variance characteristics.

Claim 4

A method according to claim 2, wherein said prescribed variance characteristics include the total variance and the ratio of the low frequency variance to the total variance.

Claim 5

A method according to claim 2, wherein step (b) comprises determining the total variance and low frequency variance of said block in accordance with transform coefficients associated with said image block.

Claim 6

A method according to claim 4, wherein step (c) includes the step of (c1) in response to the maximum value of said histogram exceeding a first threshold value, generating a first output signal indicating the presence of a uniform area and proceeding to quantize transform coefficients associated with said image block.

Claim 7

A method according to claim 5, wherein step (b) includes the step of, in response to the maximum value of said histogram not exceeding said first threshold value, determining the total variance and low frequency variance of said block in accordance with the transform coefficients associated with said image block

Claim 8

A method according to claim 6, wherein step (c) further includes the step of (c2) in response to the total variance being less than a second threshold value, generating a second output signal indicating the absence of an edge, and proceeding to quantize the transform coefficients associated with said image block in accordance with a prescribed adaptive quantization process, but in response to the total variance not being less than said second threshold value, comparing the ratio of said low frequency variance to said total variance with a third prescribed threshold value and, upon said ratio exceeding said third prescribed threshold value, generating said first output signal and proceeding to quantize the transform coefficients associated with said image block.

Claim 9

A method according to claim 7, wherein step (c) further includes the step of (c3) in response said ratio not exceeding said third prescribed threshold value, comparing the maximum value of said histogram to a fourth threshold value and, upon the maximum histogram value being less than said fourth threshold value, quantizing the transform coefficients associated with said image block in accordance with said prescribed adaptive quantization process.

Claim 10

A method according to claim 8, wherein step (c) further includes the step of (c4) in response to the maximum value compared in step (c3) being greater than said fourth prescribed threshold, comparing said ratio to a fifth threshold value, and generating said first output signal and proceeding to quantize the transform coefficients associated with said image block upon said ratio exceeding said fifth threshold, but otherwise generating said second output signal and proceeding to quantize the transform coefficients associated with said image block in accordance with said prescribed adaptive quantization process.

Claim 11

A method according to claim 2, wherein said transform coefficients comprise discrete cosine transform coefficients.

Claim 12

For use with a signal processing mechanism for compressing signals representative of an image for transmission over a bandwidth-limited communication channel, in which blocks of transform coefficients, that have been obtained by subjecting said signals to a spatial frequency block transformation process, are quantized and the quantized transform coefficients encoded for transmission over said channel, a method of detecting the presence of an edge within an image block for which a corresponding block of image transform coefficients is obtained, comprising the steps of: (a) generating a histogram of said image block; (b) in response to the maximum value of said histogram exceeding a first threshold value, generating a first output signal indicating the presence of an edge and proceeding to quantize the transform coefficients associated with said image block; (c) in response to the maximum value of said histogram not exceeding said first threshold value, determining the total variance and low frequency variance of said block in accordance with the transform coefficients associated with said image block; (d) in response to the total variance being less than a second threshold value, generating a second output signal indicating the absence of an edge, and proceeding to quantize the transform coefficients associated with said image block in accordance with a prescribed adaptive quantization process; (e) in response to the total variance not being less than said second threshold value, comparing the ratio of said low frequency variance to said total variance with a third prescribed threshold value and, upon said ratio exceeding said third prescribed threshold value, generating said first output and proceeding to quantize the transform coefficients associated said image block; (f) in response said ratio not exceeding said third prescribed threshold value, comparing the maximum value of said histogram to a fourth threshold value and, upon the maximum histogram value being less than said fourth threshold value, generating the transform coefficients associated with said image block in accordance with said second prescribed quantization process; (g) in response to the maximum value compared in step (f) being greater than said fourth prescribed threshold comparing said ratio to a fifth threshold value, and generating said first output signal and proceeding to quantize the transform coefficients associated with said image block said ratio exceeding said fifth threshold, but otherwise generating said second output signal and proceeding to quantize the transform coefficients associated with said image block in accordance with said adaptive quantization process.

Claim 13

For use with a signal processing mechanism for compressing signals representative of an image for transmission over a bandwidth-limited communication channel, in which blocks of transform coefficients, that have been obtained by subjecting said signals to a spatial frequency block transformation process, are quantized and the quantized transform coefficients encoded for transmission over said channel, a method of detecting the presence of an edge within an image block for which a corresponding block of image transform coefficients is obtained, comprising the steps of: (a) generating a histogram of said image block; (b) in response to the maximum value of said histogram exceeding a first threshold value, generating a first output signal indicating the presence of an edge and proceeding to quantize the transform coefficients associated with said image block; (c) in response to the maximum value of said histogram not exceeding said first threshold value, determining the total variance and low frequency variance of said block in accordance with the transform coefficients associated with said image block; (d) in response to the total variance being less than a second threshold value, generating a second output signal indicating the absence of an edge, and proceeding to quantize the transform coefficients associated with said image block in accordance with a prescribed adaptive quantization process; (e) in response to the total variance not being less than said second threshold value, comparing the ratio of said low frequency variance to said total variance with a third prescribed threshold value and, upon said ratio exceeding said third prescribed threshold value, generating said first output and proceeding to quantize the transform coefficients associated said image block; (f) in response said ratio not exceeding said third prescribed threshold value, comparing the maximum value of said histogram to a fourth threshold value and, upon the maximum histogram value being less than said fourth threshold value, generating the transform coefficients associated with said image block in accordance with said second prescribed quantization process; (g) in response to the maximum value compared in step (f) being greater than said fourth prescribed threshold comparing said ratio to a fifth threshold value, and generating said first output signal and proceeding to quantize the transform coefficients associated with said image block said ratio exceeding said fifth threshold, but otherwise generating said second output signal and proceeding to quantize the transform coefficients associated with said image block in accordance with said adaptive quantization process.

Claim 14

For use with an imagery signal processing system for compressing signals representative of an image, for transmission over a bandwidth-limited communication channel, in which blocks of transform coefficients, that have been obtained by subjecting said signals to a spatial frequency block transformation process, are quantized and the quantized transform coefficients encoded for transmission over said channel, a signal processing arrangement for detecting the presence of an edge within an image block for which a corresponding block of image transform coefficients is obtained, comprising: first means, responsive to said spatial frequency block transformation process, for generating first signals representative of a histogram of said image block, and generating second signals, representative of prescribed variance characteristics of said image block, in accordance with transform coefficients associated with said image block; and second means for processing said first and second signals from said first means and identifying whether or not an edge is present within an image block in dependence upon said histogram and said prescribed image variance characteristics.

Claim 15

For use with an imagery signal processing system for compressing signals representative of an image, for transmission over a bandwidth-limited communication channel, in which blocks of transform coefficients, that have been obtained by subjecting said signals to a spatial frequency block transformation process, are quantized and the quantized transform coefficients encoded for transmission over said channel, a signal processing arrangement for detecting the presence of an edge within an image block for which a corresponding block of image transform coefficients is obtained, comprising: first means, responsive to said spatial frequency block transformation process, for generating first signals representative of a histogram of said image block, and generating second signals, representative of prescribed variance characteristics of said image block, in accordance with transform coefficients associated with said image block; and second means for processing said first and second signals from said first means and identifying whether or not an edge is present within an image block in dependence upon said histogram and said prescribed image variance characteristics.

Claim 16

An imagery signal processing arrangement according to claim 12, wherein said second means includes means for identifying whether or not an edge is present within an image block in dependence upon the maximum value of said histogram and said prescribed variance characteristics.

Claim 17

An imagery signal processing arrangement according to claim 13, wherein said prescribed variance characteristics include the total variance and the ratio of the low frequency variance to the total variance.

Claim 18

An imagery signal processing arrangement according to claim 13, wherein said first means includes means for determining the total variance and low frequency variance of said block in accordance with transform coefficients associated with said image block.

Claim 19

An imagery signal processing arrangement according to claim 15, wherein said second means includes means, responsive to the maximum value of said histogram exceeding a first threshold value, for generating a first output signal which indicates the presence of a uniform area and in response to which the transform coefficients associated with said image block are quantized.

Claim 20

An imagery signal processing arrangement according to claim 16, wherein said first means includes means, responsive to the maximum value of said histogram not exceeding said first threshold value, for determining the total variance and low frequency variance of said block in accordance with the transform coefficients associated with said image block.

Claim 21

An arrangement according to claim 17, wherein said third means includes means, responsive to the total variance being less than a second threshold value, for generating a second output signal indicating the absence of an edge, in response to which the transform coefficients associated with said image block are quantized in accordance with an adaptive quantization mechanism.

Claim 22

An arrangement according to claim 18, wherein said third means includes means, responsive to the total variance not being less than said second threshold value, for comparing the ratio of said low frequency variance to said total variance with a third prescribed threshold value and, upon said ratio exceeding said third prescribed threshold value, generating said first output signal in response to which the transform coefficients associated with said image block are quantized.

Claim 23

An arrangement according to claim 19, wherein said third means further includes means, responsive to said ratio not exceeding said third prescribed threshold value, for comparing the maximum value of said histogram to a fourth threshold value and, upon the maximum histogram value being less than said fourth threshold value, generating said second output signal in response to which transform coefficients associated with said image block are quantized in accordance with said adaptive quantization mechanism.

Claim 24

An arrangement according to claim 20, wherein said third means further includes means, responsive to the maximum value of said histogram being greater than said fourth prescribed threshold, for comparing said ratio to a fifth threshold value and, upon said ratio exceeding said fifth threshold, generating said first output signal in response to which the transform coefficients associated with said image block are quantized, but otherwise generating said second output signal in response to which the transform coefficients associated with said image block are quantized in accordance with said adaptive quantization mechanism.

Claim 25

An arrangement according to claim 14, wherein said transform coefficients comprise discrete cosine transform coefficients.