Apparatus and method for defocusing images

Claim 2

The apparatus of Claim 1, wherein the data related to the circle of confusion comprises a radius and an area of each circle of confusion, and wherein the circle-of-confusion computing means includes: a radius calculating unit for computing a radius C of each circle of confusion using an expression: C = |1-p(d-f)/d(p-f)| * f/F where p is a distance from a view point to a focal point, d is a Z value that represents a distance from the view point to each point, f is a focal length, and F is an iris value of a lens being used, and wherein the updating means includes: an area calculating unit for computing an area of each circle of confusion using their respective radiuses C.

Claim 3

The apparatus of Claim 2, wherein the circle-of-confusion computing means further includes a storage unit for storing the radiuses C of the circles of confusion in relation with their respective pixels' x, y coordinates.

Claim 4

The apparatus of Claim 3, wherein the checking means includes: a distance computing unit for computing a linear distance from the target pixel to each neighboring pixel; a comparing unit for comparing the linear distance with the radius C relative to each neighboring pixel by referring to their x, y coordinates; and a comparison-result outputting means for selecting a neighboring pixel whose radius C is longer than the linear distance, and subsequently outputting the selected neighboring pixel as the reference pixel.

Claim 5

The apparatus of Claim 4, wherein the updating means includes: a distributing unit for weighting a value of each reference pixel to determine a distribution within their own circles of confusion, the distribution being distributed to the target pixel; a target-pixel weighting unit for weighting the original value of the target pixel with the area of its circle of confusion; and an updating unit for accumulating the distributions and the weighted value of the target pixel to yield a mean value, the original value of the target pixel being updated by the mean value.

Claim 6

The apparatus of Claim 5, wherein the distributing unit includes a reference-pixel weighting unit for multiplying a value of each reference pixel by a reciprocal of an area of their respective circles of confusion.

Claim 7

The apparatus of Claim 5, wherein the checking means further includes: a spatial-location comparing unit for comparing Z values of the target pixel and the reference pixels, the Z value representing a distance from the view point to their corresponding points, and wherein the image defocusing apparatus further comprising: means for disallowing the updating means to update the original value of the target pixel when the reference pixel has a larger Z value than the target pixel.

Claim 8

The apparatus of Claim 5, wherein the checking means further includes: an output unit for comparing the Z values of the target pixel and the neighboring pixels to select a neighboring pixel whose Z value is smaller than the target pixel, and for subsequently outputting the selected neighboring pixel to the distance computing unit, the Z value representing a distance from the view point to their corresponding points.

Claim 9

An image defocusing apparatus for defocusing an image formed on a screen by modeling an object in a boundary box by updating pixel values on the image to approximate a lens's focusing effects, the apparatus comprising: means for computing data related to a circle of confusion for each point in the boundary box, the circle of confusion being a measure how defocused an out-of-focus point is on the screen; a first checking means for checking a first reference pixel whose corresponding point's circle of confusion overlaps a target pixel in a first direction; a first updating means for updating an original value of the target pixel by taking into account a value of the first reference pixel and the data related to the target pixel's circle of confusion; a second checking means for checking a second reference pixel whose corresponding point's circle of confusion overlaps the target pixel in a second direction, the second direction being orthogonal with respect to the first direction; and a second updating means for updating the values updated by the first updating unit by taking into account a value of the second reference pixel and the data related to the target pixel's circle of confusion.

Claim 10

The apparatus of Claim 9, wherein the data related to the circle of confusion comprises a radius and an area of each circle of confusion, and wherein the circle-of-confusion computing means includes: a radius calculating unit for computing a radius C of each circle of confusion using an expression: C = |1-p(d-f)/d(p-f)| * f/F where p is a distance from a view point to a focal point, d is a Z value that represents a distance from the view point to each point, f is a focal length, and F is an iris value of a lens being used, and wherein the updating means includes: an area calculating unit for computing an area of each circle of confusion using their respective radiuses C.

Claim 11

The apparatus of Claim 10, wherein the circle-of-confusion computing means further includes a storage unit for storing the radiuses C of the circles of confusion in relation with their respective pixels' x, y coordinates.

Claim 12

The apparatus of Claim 11, wherein the first checking means includes: a first distance computing unit for counting the number of pixels between the target pixel and each pixel aligned in an array that contains the target pixel in the first direction, the number thus counted being a distance; a first comparing unit for comparing the distance and the radius C relative to each pixel aligned in the target pixel's array in the first direction by referring to their x, y coordinates; a first comparison-result outputting unit for outputting a pixel aligned in the first direction whose radius is longer than the distance to the first updating means as the first reference pixel, and wherein the second checking means includes: a second distance computing unit for counting the number of pixels between the target pixel and each pixel aligned in an array that contains the target pixel in the second direction, the number thus counted being a distance; a second comparing unit for comparing the distance and the radius C relative to each pixel aligned in the target pixel's array in the second direction by referring to their x, y coordinates; and a second comparison-result outputting unit for outputting a pixel aligned in the second direction whose radius is longer than the distance to the second updating means as the second reference pixel.

Claim 13

The apparatus of Claim 12, wherein the first updating means includes: a first distributing unit for weighting a value of the first reference pixel to determine a distribution within their own circles of confusion, the distribution being distributed to the target pixel, and wherein the second updating means includes: a second distributing unit for weighting a value of the second reference pixel to determine a distribution within their own circles of confusion, the distribution being distributed to the target pixel.

Claim 14

The apparatus of Claim 13, wherein the first distributing unit includes a first reference-pixel weighting unit for multiplying an original value of each reference pixel by a reciprocal of an area of their respective circles of confusion, and the second distributing unit includes a second reference-pixel weighting unit for multiplying an original value of each reference pixel by a reciprocal of an area of their respective circles of confusion.

Claim 15

The apparatus of Claim 14, further comprising: means for withholding the value of the target pixel updated by the first updating means to transfer them to the second updating means.

Claim 16

The apparatus of Claim 13, wherein the first checking means includes: a first spatial-location comparing unit for comparing Z values of the target pixel and the first reference pixel, the Z value representing a distance from the view point to their corresponding points, and wherein the second checking unit includes: a second spatial-location comparing means for comparing the Z values of the target pixel and the second reference pixel, and wherein the first updating means include; a first disallowing unit for disallowing the first updating means to update the original value of the target pixel when the target pixel has a smaller Z value, and wherein the second updating means include: a second disallowing unit for disallowing the second updating means to update the original value of the target pixel when the target pixel has a smaller Z value.

Claim 17

The apparatus of claim 13, wherein the first checking means further includes: a first output unit for comparing the Z values of the target pixel and the neighboring pixels to select a neighboring pixel whose Z value is smaller than the target pixel, and for subsequently outputting the selected neighboring pixel to the first distance computing unit, the Z value representing a distance from the view point to their corresponding points, and wherein the second checking means further includes: a second output unit for comparing the Z values of the target pixel and the neighboring pixels to select a neighboring pixel whose Z value is smaller than the target pixel, and for subsequently outputting the selected neighboring pixel to the second distance computing unit.

Claim 19

The method of Claim 18, wherein the first step comprises the substep of: (a) computing a radius C of each circle of confusion using an expression: C = |1-p(d-f)/d(p-f)| * f/F where p is a distance from a view point to a focal point, d is a Z value that represents a distance from the view point to each point, f is a focal length, and F is an iris value of a lens being used.

Claim 20

The method of Claim 19, wherein the second step comprises the sub-steps of: (a) computing a distance from the target pixel to each neighboring pixel; (b) comparing the distance with a radius C of the overlapping circle of confusion relative to each neighboring pixel; (c) selecting a neighboring pixel whose radius C is longer than the distance as the reference pixel; (d) selecting another target pixel; and (e) proceeding the third step when all the pixels on the image have been selected as the target pixel and been through the substeps (a) through (c) of the second step.

Claim 21

The method of Claim 20, wherein the third step comprises the substeps of: (a) weighting a value of each reference pixel to determine a distribution to the target pixel; (b) computing an area of each overlapping circle of confusion and the target pixel's circle of confusion to weight the original value of the target pixel with the area of its circle of confusion; (c) accumulating the distributions and the weighted value of the target pixel to yield a mean value, and updating the original value of the target pixel with the mean value; (d) selecting another target pixel; and (e) terminating to update the target pixel's value when all the pixels have been selected as the target pixel and been through the substeps (a) through (c) of the third step.

Claim 22

The method of Claim 21 further comprising the steps of: (4) comparing Z values of the target pixels and neighboring pixels, the Z value representing a distance from a view point to their corresponding points, and (5) disallowing to update the original value of the target pixel when the neighboring pixel has a larger Z value than the target pixel.

Claim 23

The method of Claim 21, wherein the second step further comprises the substep of: (f) comparing Z values of the target pixels and neighboring pixels to select a neighboring pixel whose Z value is smaller than the target pixel prior to the substep (a) of the second step, the Z value representing a distance from a view point to their corresponding points.

Claim 25

The method of Claim 24, wherein the second and fourth steps respectively comprise the substep of: (a) comparing Z values of the target pixels and neighboring pixels, the Z value representing a distance from a view point to their corresponding points, and wherein the third and fifth steps respectively comprise the substep of: (a) disallowing to update the original value of the target pixel when the neighboring pixel has a larger Z value than the target pixel.

Claim 26

The method of Claim 25, wherein the second and fourth steps further and respectively comprise the substep of: (b) comparing the Z values of the target pixels and neighboring pixels to select a neighboring pixel whose Z value is smaller than the target pixel prior to the substep (a) of the second step and the substep (a) of the fourth step, respectively.