Method and means of least squares designed filters for image segmentation in scanning cytometry

Claim 2

A computer-executed method of separating an object from a background in a pixelated image with a best-designed digital filter, the method comprising: selecting a digital filter for creating contrast in an image, the digital filter including a neighboring operator of second or higher order for processing neighborhoods of pixels in pixel array; setting a contrast range (R,Q) in which R<Q; receiving a first pixel array defining a pixelated image including one or more objects and a background; applying the digital filter to the first pixel array to create a filtered image; receiving a second pixel array defining a reference image, the reference image including at least one object included in the pixelated image and a background, in which pixels included in the at least one object are distinguished from pixels included in the background by a predetermined amount of contrast; determining a merit function value by: setting the merit function value to a specified value if, either a pixel in the background of the pixelated image is less than or equal to R and a corresponding pixel in the reference image is in a background of the reference image, or a pixel in an object of the pixelated image is greater than or equal to Q and a corresponding pixel in the reference image is an object of the reference image; or, calculating a minimum merit function value otherwise; changing the neighborhood operator of the digital filter to a new neighborhood operator in response to a minimum merit function value; and creating a segmented image in which an object of the pixelated image is separated from the background in the pixelated image by applying the digital filter with a new neighborhood operator to the pixelated image.

Claim 3

The method of claim 1, wherein the neighborhood operator is a convolution kernel.

Claim 4

The method of claim 1, wherein the neighborhood operator is a second order filter.

Claim 5

The method of claim 1, wherein the neighborhood operator is a Volterra series.

Claim 6

The method of claim 1, wherein determining includes thresholding the merit function value by: causing the merit function value to equal zero when: the pixel in the background of the pixelated image is less than or equal to R and the corresponding pixel in the reference image has a magnitude equal to a predetermined background value, or the pixel in the object of the pixelated image is greater than or equal to Q and the corresponding pixel in the reference image has a magnitude equal to a predetermined object value; or calculating a positive, non-zero merit function value otherwise.

Claim 7

The method of claim 5 further including deriving a transformed image by filtering a third pixel array with the digital filter, using the new neighborhood operator.

Claim 8

The method of claim 1, wherein the one or more objects are cells, or portions of cells.

Claim 9

The method of claim 1, wherein receiving a second pixel array includes: buffering the second pixel array; and determining further includes: defining an array of weights, each weight corresponding to a location of one of the pixels in the reference image; and multiplying the merit function value by a weight at the corresponding reference image pixel location to create a weighted merit function value.

Claim 10

The method of claim 8, wherein the weights are 1 at or adjacent edges of the one or more objects and 0 elsewhere in the one or more objects.

Claim 11

The method of claim 8, wherein the weights are 1 at or adjacent edges of the one or more objects, 0 inside the one or more objects, and 0 in the background.

Claim 12

The method of claim 8, wherein determining further includes thresholding the merit function value by: causing the merit function value to equal zero when: the pixel in the background of the pixelated image is less than or equal to R and the corresponding pixel in the reference image has a magnitude equal to a predetermined background value, or the pixel in the object of the pixelated image is greater than or equal to Q and the corresponding pixel in the reference image has a magnitude equal to a predetermined object value; or calculating a positive, non-zero merit function value otherwise.

Claim 15

The method of claim 12, wherein the neighborhood operator is a convolution kernel.

Claim 16

The method of claim 12, wherein the neighborhood operator is a second order filter.

Claim 17

The method of claim 12, wherein the neighborhood operator is a Volterra series.

Claim 18

The method of claim 12, wherein determining includes thresholding the merit function value by: causing the merit function value to equal zero when: the pixel in the background of the pixelated image is less than or equal to R and the corresponding pixel in the reference image has a magnitude equal to a predetermined background value, or the pixel in the object of the pixelated image is greater than or equal to Q and the corresponding pixel in the reference image has a magnitude equal to a predetermined object value; or calculating a positive, non-zero merit function value otherwise.

Claim 19

The method of claim 16, wherein the one or more objects are cells, or portions of cells.

Claim 20

The method of claim 12, wherein receiving a second pixel array includes: buffering the second pixel array; and determining further includes: defining an array of weights, each weight corresponding to a location of one of the pixels in the reference image; and multiplying the merit function value by a weight at the corresponding reference image pixel location to create a weighted merit function value.

Claim 21

The method of claim 18, wherein the weights are 1 at or adjacent edges of the one or more objects and 0 elsewhere in the one or more objects.

Claim 22

The method of claim 18, wherein the weights are 1 at or adjacent edges of the one or more objects, 0 inside the one or more objects, and 0 elsewhere in the background.

Claim 23

The method of claim 18 wherein further includes: causing the merit function value to equal zero when: the pixel in the background of the pixelated image is less than or equal to R and the corresponding pixel in the reference image has a magnitude equal to a predetermined background value, or the pixel in the object of the pixelated image is greater than or equal to Q and the corresponding pixel in the reference image has a magnitude equal to a predetermined object value; or calculating a positive, non-zero merit function value otherwise.

Claim 24

The method of claim 12, wherein applying the digital filter to the third pixel array creates or enhances contrast between the one or more objects and the background.

Claim 25

An image segmentation system, comprising: means for acquiring an array of pixels defining a pixelated image including one or more objects and a background; a digital filter for producing a transformed array of pixels representing the transformation of a pixelated image, the digital filter including a neighborhood operator of second or higher order for processing intensities of pixels in a pixel array; means coupled to the means for acquiring and to the digital filter for setting values of the neighborhood operator in response to a first pixel array defining a pixelated image including one or more objects and a background and a second pixel array defining a reference image, the reference image including at least one object included in the pixelated image and a background in which pixels included in the at least one object are distinguished from pixels included in the background by a predetermined amount of contrast; means for determining a merit function value by: the pixel in the background of the pixelated image is less than or equal to R and the corresponding pixel in the reference image has a magnitude equal to a predetermined background value, or the pixel in the object of the pixelated image is greater than or equal to Q and the corresponding pixel in the reference image has a magnitude equal to a predetermined object value; or calculating a positive, non-zero merit function value otherwise; means for changing the neighborhood operator to a new neighborhood operator in response to a minimum merit function value; and means for applying the new neighborhood operator of the digital filter to a third array of pixels defining an image to be segmented, the image to be segmented including one or more objects and a background, such that the new neighborhood operator creates or enhances contrast between the one or more objects in the background.

Claim 26

An image segmentation system, comprising: means for acquiring an array of pixels defining a pixelated image including one or more objects and a background; a digital filter for producing a transformed array of pixels representing the transformation of a pixelated image, the digital filter including a neighborhood operator of second or higher order for processing intensities of pixels in a pixel array; means coupled to the means for acquiring and to the digital filter for setting values of the neighborhood operator in response to a first pixel array defining a pixelated image including one or more objects and a background and a second pixel array defining a reference image, the reference image including at least one object included in the pixelated image and a background in which pixels included in the at least one object are distinguished from pixels included in the background by a predetermined amount of contrast; means for determining a merit function value by: the pixel in the background of the pixelated image is less than or equal to R and the corresponding pixel in the reference image has a magnitude equal to a predetermined background value, or the pixel in the object of the pixelated image is greater than or equal to Q and the corresponding pixel in the reference image has a magnitude equal to a predetermined object value; or calculating a positive, non-zero merit function value otherwise; means for changing the neighborhood operator to a new neighborhood operator in response to a minimum merit function value; and means for applying the new neighborhood operator of the digital filter to a third array of pixels defining an image to be segmented, the image to be segmented including one or more objects and a background, such that the new neighborhood operator creates or enhances contrast between the one or more objects in the background.