Image analysis – Histogram processing – For setting a threshold
Patent
1989-12-01
1992-09-22
Moore, David K.
Image analysis
Histogram processing
For setting a threshold
382 18, 382 22, 358426, 358433, G06K 946, H04N 1415
Patent
active
051504332
ABSTRACT:
An edge detection mechanism for an imagery signal compression and transmission system processes histogram and variance data to identify the presence of a uniform area with a block of transform signals. The presence of an edge within an image block, for which corresponding transform coefficients are obtained, is identified by initially generating a histogram of the image block. If the maximum value of the histogram exceeds a first threshold value, a uniform area is inferred, and adaptive quantization that would otherwise be carried out prior to executing a minimum redundancy encoding scheme is bypassed. If the maximum value of the histogram does not exceed the first threshold value, a set of variance calculations is carried out for use with a further sequence of edge detection decisions. The total variance and low frequency variance of the block are determined in accordance with the transform coefficients associated with the image block. The results of these variance calculations are then processed with further thresholds and the histogram maximum value. If the absence of an edge is inferred, the transform coefficients associated with the image block are quantized using adaptive quantization.
REFERENCES:
patent: 4774569 (1988-09-01), Morton et al.
patent: 4774574 (1988-09-01), Poly et al.
patent: 4780761 (1988-10-01), Daly et al.
patent: 4797945 (1989-01-01), Suzuki et al.
Dugas Edward
Eastman Kodak Company
Moore David K.
Stellrerht Barry
LandOfFree
Histogram/variance mechanism for detecting presence of an edge w does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Histogram/variance mechanism for detecting presence of an edge w, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Histogram/variance mechanism for detecting presence of an edge w will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-1075899