Contour extraction method and apparatus
AssigneeNEC Corporation
InventorTomoharu Kiyuna
Life of this patent
Estimating the term…
Abstract
A contour extraction method and apparatus is disclosed by which a contour can be extracted automatically at a high speed with a high degree of accuracy without the necessity to set a threshold value explicitly. In the contour extraction method and apparatus, from picture image data of a picture image including an image of a body which makes an object of contour extraction, region belonging probabilities with which individual points of the picture image belong to regions are calculated not based on values themselves of the image data but based on attributes of the points of the picture image. Then, the regions to which the individual points of the picture image belong are delimited using the region belonging probabilities, and then a boundary between the regions is extracted as a contour.
Claims20
A contour extraction method for delimiting, from picture image data of a picture image including an image of a body that includes an object of contour extraction, regions to which individual points of the picture image belong based on attributes of the points of the picture image and extracting a boundary between the regions as a contour, the method comprising:
- (a) initializing parameters which define mixed probability distributions of the attributes of the points of the picture image;
- (b) calculating region belonging probabilities with respect to the points of the picture image;
- (c) updating the parameters so that the mixed probability distributions may be increased;
- (d) calculating an evaluation function to be used as a scale for favorableness of estimation from the mixed probability distributions defined by the updated parameters;
- (e) delimiting the regions to which the points of the picture image belong based on the region belonging probabilities;
- (f) extracting a boundary between the delimited regions;
- and
- (g) discriminating, based on the region belonging probabilities, whether or not the points of the picture image are region internal points, which are points within a determined one of the regions, or region external points, which are points outside the determined region.
A contour extraction apparatus for delimiting, from picture image data of a picture image including an image of a body which makes an object of contour extraction, regions to which individual points of the picture image belong based on attributes of the points of the picture image and extracting a boundary between the regions as a contour, comprising:
- first means for initializing parameters which define a mixed probability distributions of the attributes of the points of the picture image;
- second means for calculating expected values of a region belonging probability with which the points of the picture image belong individually to the regions;
- third means for updating the parameters so that the mixed probability distributions may be increased;
- fourth means for calculating an evaluation function to be used as a scale for favorableness of estimation from the mixed probability distributions defined by the updated parameters;
- fifth means for delimiting the regions to which the points of the picture image belong based on the values of the region belonging probabilities;
- sixth means for extracting a boundary between the delimited regions;
- and
- seventh means for discriminating based on the region belonging probabilities whether or not the points of the picture image are region internal points which are points within a determined one of the regions or region external points which are points outside the determined region.
A region contour extraction method, comprising:
- receiving picture image data of a picture image including an image of a body that includes an object of contour extraction from inputting means;
- calculating region belonging probabilities with respect to which individual points of the picture image belong to regions based on attributes of the points of the picture image;
- delimiting the regions to which the individual points of the picture image belong using the region belonging probabilities;
- and
- extracting a boundary between the regions as a contour.
A region contour extraction method, comprising:
- calculating, from picture image data of a picture image inputted from inputting means and including an image of a body which makes an object of contour extraction, expected values of a region belonging probability with which individual points of the picture image belong to regions based on attributes of the points of the picture image;
- updating parameters which define a mixed probability distribution of the attributes of the points of the picture delimiting the regions to which the individual points image;
- calculating an evaluation function based on the mixed probability distribution calculated from the region belonging probabilities and the parameters;
- and
- delimiting the regions based on the region belonging probabilities when the evaluation function satisfies a predetermined condition and extracting a contour based on the delimited regions.
A contour extraction apparatus, comprising:
- a data analysis section for analyzing a characteristic of image data of a picture image inputted from an inputting apparatus to determine initial values to region parameters which characterize regions of the picture image;
- a region belonging probability calculation section for calculating region belonging probabilities with which individual points of the picture image belong to the regions based on values of the points of the picture image;
- a parameter updating section for updating the region parameters so that a value of an evaluation function may be increased or decreased depending upon the evaluation function employed;
- an evaluation function calculation section for calculating the evaluation function from the updated region parameters;
- a region delimiting section for deciding, when the evaluation function satisfies an ending condition determined
- in advance, to which one of the regions each of the points of the picture image belongs based on the estimated region belonging probabilities;
- a region belonging discrimination section for discriminating whether or not a designated one of the points of the picture image which make a boundary between the regions belongs to a designated one of the regions;
- a boundary extraction section for producing an ordered contour line from the extracted region;
- and
- an outputting apparatus for outputting a result of the contour extraction.
A recording medium on which a program to be executed by a computer is recorded, the program comprising:
- a data analysis step of analyzing a characteristic of image data of a picture image inputted from an inputting apparatus to determine initial values to region parameters which characterize regions of the picture image;
- a region belonging probability calculation step of calculating region belonging probabilities with which individual points of the picture image belong to the regions based on values of the points of the picture image;
- a parameter updating step of updating the region parameters so that a value of an evaluation function may be increased or decreased depending upon the evaluation function employed;
- an evaluation function calculation step of calculating the evaluation function from the updated region parameters;
- a region delimiting step off deciding, when the evaluation function satisfies an ending condition determined in advance, to which one of the regions each of the points of the picture image belongs based on the estimated region belonging probabilities;
- a region belonging discrimination step of discriminating whether or not a designated one of the points of the picture image which make a boundary between the regions belongs to a designated one of the regions;
- and
- a boundary extraction step of producing an ordered contour line from the extracted region.
Dependencies are inferred from the claim text.