Loading patent…
AssigneeBoard of Supervisors of Louisiana State University and Agricultural and Mechanical College
InventorsOleg S. Pianykh, John M. Tyler
Estimating the term…
A method of compressing image data from a set of individual images wherein the images have a correlation between images of approximately greater than 0.5. A wavelet transform is applied to a reference image of the set in order to determine a series of reference coefficients wi(1). The same wavelet transform is applied to an adjacent image of said set in order to determine another series of coefficients wi(2). The series of coefficients wi(k−1) and wi(k) (for all k≧2) are then compared in order to determine a coefficient b(k) and a series of coefficients ci(k), such that wi(k) may be determined from the relationship wi(k)=└b(k)wi(k−1)┘+ci(k). Alternatively, b(k) may be set equal to one rather than be determined. The coefficients for the reference image wi(1), b(k) (if b(k) is not equal to one), and ci(k) are then entropy encoded in order to form a compressed set of image data.
A method of compressing image data comprising the steps of:
A method of compressing image data according to claim 1, wherein said coefficient b(k) equals one and said coefficient ci(k) equals the coefficient di(k), wherein di(k) is the difference between wi(k) and wi(k−1).
A method of compressing image data according to claim 6, wherein the value n equals 25% of the total number of wavelets N representing the image.
A method of compressing image data according to claim 6, wherein the value n is determined by comparing the absolute sum of coefficients wi(k) to the absolute sum of differences di(k).
A method of compressing image data according to claim 6, wherein the value n is determined by comparing the entropy of coefficients wi(k) to the entropy of differences di(k).
An apparatus for compressing image data comprising:
An apparatus for compressing image data comprising:
Dependencies are inferred from the claim text.