Semiautomatic drawing tool for image segmentation
A semiautomatic drawing tool is configured to generate an image segmentation contour that corresponds to a visible edge within a digital image. Individual contour points define the shape and location of the image segmentation contour, and the semiautomatic drawing tool determines the locations of these contour points based on both user input and on the location of a steep luminance gradient that corresponds to the visible edge. The semiautomatic drawing tool determines the location of a particular point of the image segmentation contour based on the current location of an input device pointer, e.g., a cursor, the location of the steepest luminance gradient, and the location of previously selected points of the image segmentation contour.
1. A computer-implemented method for segmenting an image, the method comprising:
calculating, by a processor, a direction vector based on a first point of an image segmentation contour and a second point of the image segmentation contour, wherein the image segmentation contour at least partially separates a first region of the image from a second region of the image; and
constructing, by the processor, a luminance intensity profile for a portion of the image that intersects the direction vector at an intersection point, wherein the first region of the image has a first average luminance associated therewith, the second region has a second average luminance associated therewith, and the first average luminance is different from the second average luminance.
2. The method of claim 1 , further comprising:
selecting a point on the luminance intensity profile; and
accepting the point on the luminance intensity profile as an additional point of the image segmentation contour.
3. The method of claim 2 , wherein selecting the point on the luminance intensity profile comprises selecting a point on the luminance intensity profile that substantially corresponds to a location proximate a steepest luminance gradient of the intensity profile.
4. The method of claim 1 , further comprising determining a location of the intersection point based on user input.
5. The method of claim 4 , wherein the user input comprises a current cursor position in the image.
6. The method of claim 5 , wherein determining the location of the intersection point based on user input comprises performing an orthogonal projection of the current cursor position onto the direction vector.
7. The method of claim 5 , wherein determining the location of the intersection point based on user input comprises performing an orthogonal projection onto the direction vector from a point disposed on a line between an endpoint of the contour and the current cursor position.
8. The method of claim 1 , wherein the first point comprises a current endpoint of the image segmentation contour.
9. The method of claim 1 , wherein the second point comprises a point of the image segmentation contour that is offset a predetermined number of points from an endpoint of the image segmentation contour.
10. The method of claim 9 , further comprising setting a value of the predetermined number of points based on a user input.
11. The method of claim 1 , wherein the first average luminance is visibly different from the second average luminance.
12. The method of claim 11 , wherein the first average luminance comprises an average intensity of a particular color, and the second average luminance intensity comprises the average intensity of the particular color.
13. The method of claim 1 , wherein the luminance intensity profile comprises a smooth curve.
14. The method of claim 1 , wherein the luminance intensity profile comprises a predetermined number of discrete points.
15. The method of claim 14 , wherein the luminance intensity profile comprises an arc on which the discrete points are disposed.
16. The method of claim 15 , wherein a portion of the discrete points are located in the first region and a portion of the discrete points are located in the second region.
17. The method of claim 1 , wherein the portion of the image is disposed in both the first region of the image and the second region of the image.
18. The method of claim 1 , wherein the first region of the image has a first texture property value associated therewith, the second region has a second texture property value associated therewith, and the first texture property is visibly different from the second texture property.
19. A computing device configured to segment an image, the computing device comprising:
a processor;
a memory coupled to the processor;
an input circuitry coupled to the processor and configured to generate a pointing device on a displayed digital image and to input image location information to the processor; and
an output circuitry coupled to the processor and configured to output image data from the processor,
wherein the processor is configured to:
calculate a direction vector based on a first point of an image segmentation contour and a second point of the image segmentation contour, wherein the image segmentation contour at least partially separates a first region of the displayed digital image from a second region of the displayed digital image; and
construct a luminance intensity profile for a portion of the displayed digital image that intersects the direction vector at an intersection point, wherein the first region of the displayed digital image has a first average luminance associated therewith, the second region of the displayed digital image has a second average luminance associated therewith, and the first average luminance is different from the second average luminance.
20. The computing device of claim 19 , wherein the processor is configured to:
select a point on the luminance intensity profile;
accept the point on the luminance intensity profile as an additional point of the image segmentation contour to form an updated image segmentation contour; and
display the updated image segmentation contour via the output circuitry.
21. The computing device of claim 19 , wherein the processor is configured to determine a location of the intersection point based on user input via the input circuitry.
22. The computing device of claim 21 , wherein the user input comprises a current cursor position in the displayed digital image.
23. A non-transitory computer readable storage medium comprising computer-executable instructions, wherein in response to execution of the instructions by a computer, causes the computer to:
calculate a direction vector based on a first point of an image segmentation contour and a second point of the image segmentation contour, wherein the image segmentation contour at least partially separates a first region of an image from a second region of the image; and
construct a luminance intensity profile for a portion of the displayed image that intersects the direction vector at an intersection point, wherein the first region of the image has a first average luminance associated therewith, the second region of the image has a second average luminance associated therewith, and the first average luminance is different from the second average luminance.