Method of and device for image enhancement

Claim 3

The method of claim 1 , wherein when the binarization data values have a connectivity, the step of estimating further comprises: determining a similarity of the binarization data values of pixels corresponding to each of the plurality of windows; and estimating the pixel to be an edge pixel when the binarization data values of the pixel corresponding to each of a plurality of windows have a similarity.

Claim 4

The method of claim 2 , wherein the similarity of at least one of the binarization data values is determined by the equation; ∑ i = 0 3 ⁢ ⁢ ∑ j = 0 3 ⁢ ⁢ a ij ⁢ b ij ⁢ c ij ≥ THc wherein a ij , b ij , c ij are binarization data values of each pixel, THc is a predetermined threshold, and when satisfying with the equation, the binarization data values are considered to have a similarity.

Claim 5

The method of claim 1 , further comprising a step of: emphasizing the pixel determined as an edge.

Claim 6

The method of claim 1 , wherein the step of estimating comprises: estimating a pixel not to be an edge pixel when the difference of a pixel brightness value is less than the predetermined threshold Tha.

Claim 7

The method of claim 1 , wherein the step of estimating comprises: estimating a pixel not to be an edge pixel when the binarization data values have no connectivity.

Claim 8

The method of claim 1 , wherein the step of estimating comprises: estimating the predetermined pixel not to be an edge pixel when the binarization data values of the estimated pixel corresponding to a plurality of windows have no similarity.

Claim 9

The method of claim 1 , wherein the step of calculating binarization data values comprises: generating a plurality of 3×3 window masks by applying different samplings to a 5×9 window of the predetermined edge detection region wherein the brightness values are indicated on each pixel; generating binarization data values from the brightness values of each pixel of a plurality of the sampled 3×3 window masks; and grouping the binarization data values to calculate a connected component value N with respect to each window mask.

Claim 10

The method of claim 8 , wherein at least one of the binarization data values is ‘1’ when the brightness values of each pixel is less than the predetermined threshold THb, and at least one of the binarization values is ‘0’ when the brightness values of each pixel is in excess of the predetermined threshold THb.

Claim 11

The method of claim 8 , wherein the binarization data values are estimated to have connectivity when all the connected component values N of each window mask are ‘1’.

Claim 12

The method of claim 1 , wherein the samplings as applied comprise 5×5, 5×7 and 5×9 samplings with respect to a 600 dpi image.

Claim 13

An image enhancement apparatus, comprising: a luminosity variance measurement unit for determining the highest brightness value and the lowest brightness value among a plurality of pixels included in a predetermined window and for determining the predetermined window has a luminosity variance when the difference between the highest and lowest brightness values exceed a predetermined threshold THa; an edge classification unit for assigning binarization data values to each pixel in the predetermined window having the determined luminosity variance and for determining a connectivity of a plurality of windows corresponding to each predetermined window based on the binarization data values of each pixel within a corresponding predetermined window and for estimating an actual edge of a text region in a predetermined edge detection region of an output screen with respect to a document having a text region and an image region based on the connectivity of the plurality of windows; an edge detection unit for finally confirming an edge estimated pixel to be an edge pixel if the number of the pixels estimated as an edge is in excess of a predetermined preset threshold THd, and if not, determining the edge estimated pixel to not be an edge pixel; and an edge emphasis unit for emphasizing the pixel confirmed as an edge.

Claim 14

An image enhancement apparatus, comprising: a luminosity variance measurement unit for determining the highest brightness value and the lowest brightness value among a plurality of pixels included in a predetermined window and for determining the predetermined window has a luminosity variance when the difference between the highest and lowest brightness values exceed a predetermined threshold THa; an edge classification unit for assigning binarization data values to each pixel in the predetermined window having the determined luminosity variance and for determining a connectivity of a plurality of windows corresponding to each predetermined window based on the binarization data values of each pixel within a corresponding predetermined window and for estimating an actual edge of a text region in a predetermined edge detection region of an output screen with respect to a document having a text region and an image region based on the connectivity of the plurality of windows; an edge detection unit for finally confirming an edge estimated pixel to be an edge pixel if the number of the pixels estimated as an edge is in excess of a predetermined preset threshold THd, and if not, determining the edge estimated pixel to not be an edge pixel; and an edge emphasis unit for emphasizing the pixel confirmed as an edge.

Claim 15

The image enhancement apparatus as claimed in claim 12 , wherein the edge classification unit further comprises: a half tone detection unit for calculating at least one of the binarization data values based on each pixel brightness value of the pixels in the plurality of windows that are formed by applying various samplings to the predetermined edge detection region when the brightness values are in excess of a predetermined threshold THb, determining the connectivity of the calculated binarization data values with respect to each of the plurality of windows, determining a similarity of the binarization data values of pixels corresponding to the plurality of windows when the binarization values have a connectivity, and estimating the predetermined edge detection region as an edge when the binarization data values have a similarity.

Claim 16

The apparatus of claim 13 , wherein at least one of the binarization data values includes a ‘1’ if each pixel brightness is less than the predetermined threshold THb, and at least one of the binarization data values includes a ‘0’ if each pixel brightness is in excess of the predetermined threshold THb.