Method and apparatus for extracting and measuring object of interest from an image

1. A method for extracting an object of interest from an image, comprising:

acquiring an initial image containing an object of interest;

in accordance with one or more given primary features of the object of interest, extracting one or more feature regions which match the one or more primary features to obtain a feature extraction image; and

respectively computing an energy function of each of the feature regions or combinations of the feature regions, and taking the feature region whose energy function is the extreme value as the object of interest;

wherein the energy function is defined as the sum of the intensities of the points in the feature regions of the initial image, and/or the sum of intensities of the points in the feature regions of the feature extraction image, and/or a joint probability of the intensity in the feature regions of the initial image.

2. The method of claim 1 , wherein extracting feature regions comprises:

in accordance with the given primary features of the object of interest, selecting a template matching the primary features; and

processing the initial image with the template and extracting one or more feature regions which match the primary features to obtain the feature extraction image.

3. The method of claim 1 , wherein at least one of the primary features of the object of interest is intensity.

4. The method of claim 1 , further comprising, after extracting the one or more feature regions, removing the one or more feature regions which fail to meet a given screening condition in the feature extraction image.

5. The method of claim 1 , further comprising, after extracting the one or more feature regions, extracting candidate objects from the feature regions extracted from the extraction feature image; and

calculating the energy functions of the candidate objects, and taking the candidate objects whose energy functions are extreme values as the object of interest.

6. The method of claim 5 , wherein extracting candidate objects comprises thinning the feature regions and taking one or more thinned feature regions as the candidate objects.

7. The method of claim 6 , further comprising, after thinning the feature regions, fitting the thinned feature regions to obtain at least one fitted feature region in accordance with a given secondary feature of the object of interest, and taking the fitted feature regions as the candidate objects.

8. The method of claim 7 , wherein the secondary feature comprises a morphological feature.

9. The method of claim 8 , further comprising, after fitting the thinned feature regions, setting an examination condition in accordance with a given third feature, examining the fitted feature regions with the examination condition, and discarding the fitted feature regions which fail to meet the examination condition.

10. The method of claim 9 , wherein the third feature comprises at least one parameter statistics of the object of interest.

11. The method of claim 1 , further comprising, after extracting the feature regions, weighting the feature extraction image to the initial image, wherein, in the energy function, the initial image is the initial image which has been weighted.

12. The method of claim 1 , further comprising optimizing the object of interest.

13. An apparatus for extracting an object of interest from an image comprising a computer including a non-transitory computer-readable medium storing program modules executable by the computer, the modules including:

an image acquisition module configured for acquiring an initial image containing the object of interest;

a feature extraction module configured for extracting, in accordance with one or more given primary features of the object of interest, one or more feature regions which match the one or more primary features to obtain a feature extraction image;

an object-of-interest determination module configured for respectively computing an energy function of each of the feature regions or combinations of the feature regions, and taking the feature region whose energy function is the extreme value as the object of interest;

a candidate object determination module configured for extracting candidate objects from the feature regions; and

a thinning sub-module in communication with feature extraction module or screening module and configured for thinning the feature regions and taking the thinned feature region as the candidate objects.

14. The apparatus of claim 13 , wherein the feature extraction module comprises:

a template selection sub-module configured for selecting a template matching the one or more primary features; and

an image processing sub-module configured for processing the initial image with the template to obtain the feature extraction image.

15. The apparatus of claim 13 , further comprising:

a screening module in communication with the feature extraction module and configured for removing the feature regions which fail to meet a given screening condition.

16. The apparatus of claim 13 , further comprising:

a fitting sub-module in communication with the thinning sub-module and configured for fitting the thinned feature regions to obtain at least one fitted feature region in accordance with a given secondary feature of object of interest, and taking the fitted feature regions as the candidate objects.

17. The apparatus of claim 13 , further comprising:

an examination sub-module in communication with fitting sub-module and configured for setting an examination condition in accordance with a given third feature, examining the fitted feature regions with the examination condition, and discarding the fitted feature regions which fail to meet the examination condition.

18. The apparatus of claim 13 , further comprising:

a weighting module configured for weighting the feature extraction image to the initial image.

19. The apparatus of claim 13 , further comprising:

an optimizing module configured for optimizing the object of interest.

20. A method for extracting an object of interest from an image, comprising:

acquiring an initial image containing an object of interest;

in accordance with one or more given primary features of the object of interest, extracting one or more feature regions which match the one or more primary features to obtain a feature extraction image;

respectively computing an energy function of each of the feature regions or combinations of the feature regions, and taking the feature region whose energy function is the extreme value as the object of interest;

after extracting the one or more feature regions, extracting candidate objects from the feature regions extracted from the extraction feature image, wherein extracting candidate objects comprises thinning the feature regions and taking one or more thinned feature regions as the candidate objects; and

calculating the energy functions of the candidate objects, and taking the candidate objects whose energy functions are extreme values as the object of interest.