Methods for segmenting objects in images
View Patent ↗A method for segmenting a digital image into a plurality of target objects, comprising, generating a plurality of probability maps of the image, wherein each probability map is derived from a different segmentation classifier; generating a combined probability map based on the plurality of probability maps; mapping a plurality of image points based on one or more local object maxima; applying one or more object constraints based at least in part on the mapped points to identify local object information; applying one or more regional thresholds to the combined probability map, given the local object information and a background mask, to segment the image into regions; creating a segmented image at least in part by merging the segmented regions with corresponding local object maxima; and at least temporarily storing or displaying the segmented image on a digital device.
1. A method for segmenting a digital image into a plurality of target objects, comprising,
generating a plurality of probability maps of the image, wherein each probability map is derived from a different segmentation classifier;
generating a combined probability map based on the plurality of probability maps;
mapping a plurality of image points based on one or more local object maxima;
applying one or more object constraints based at least in part on the mapped points to identify local object information;
applying one or more regional thresholds to the combined probability map, given the local object information and a background mask, to segment the image into regions;
creating a segmented image at least in part by merging the segmented regions with corresponding local object maxima; and
at least temporarily storing or displaying the segmented image on a digital device.
2. The method of claim 1 , wherein at least one of the object constraints is a morphological constraint.
3. The method of claim 2 , wherein the morphological constraint is based on a cell nucleus.
4. The method of claim 1 , wherein the target objects are biological cells.
5. The method of claim 1 , wherein segmentation classifiers are selected from a group consisting of size, shape, intensity, texture, wavelets and fractals.
6. The method of claim 1 , wherein the combined probability map is based on a weighted average of the plurality of probability maps.
7. The method of claim 6 , wherein the weighted average is defined empirically.
8. The method of claim 6 , wherein the weighted average is predefined.
9. The method of claim 1 , wherein at least one of the segmentation classifiers is based on object curvature.
10. The method of claim 9 , wherein at least one of the segmentation classifiers is a set of digital filters derived at least in part from a Gaussian function and an harmonic function.
11. The method of claim 9 , wherein one or more of the segmentation classifiers is based on gradient, intensity, wavelets or fractals.