Method and device for dividing area of image of particle in urine
View Patent ↗An objective is to provide a method and an apparatus for accurately extracting a region of an object particle from a urine particle image obtained by taking an image of urine particles in a urine specimen having varying properties. First, a first object region is extracted using one or more of an R image, a G image, and a B image of a urine particle image taken by an image input optical system configured to input particle images. Then, a density distribution and a size of the first object region of one or more of the R image, the G image, and the B image are calculated. Based on these feature parameters, the first object region is classified into a predetermined number of groups. A second object region is extracted from a local region including the first object region, by using one or more of the R image, the G image, and the B image, depending on each of the groups. This configuration allows stable region segmentation for each particle image even for a urine specimen in which urine particles having different sizes and tones coexist.
1. A method for region segmentation of urine particle images, comprising the steps of:
extracting a first object region larger than an object particle included therein, from a urine particle image taken by an image input optical system, by using at least one of an image with red components of the urine particle image (an R image), an image with green components of the urine particle image (a G image), and an image with blue components of the urine particle image (a B image);
calculating a size of the first object region based on at least one of an area or a perimeter of the first object region and a density distribution of the first object region, in the first object region, of each of the at least one of the R image, the G image, and the B image, and classifying the first object region into one group of a predetermined number of groups, based on the size of the first object region and the density distribution of the first object region; and
extracting a second object region, representing the object particle, from the first object region, by using group-specific region segmentation, which includes use of the at least one of the R image, the G image, and the B image.
2. A method for region segmentation of urine particle images, comprising:
first step of creating a density histogram of each of a red (R) density of a urine particle image taken by an image input optical system, a green (G) density of the urine particle image taken by the image input optical system, and a blue (B) density of the urine particle image taken by the image input optical system, by using a red (R) image, a green (G) image, and a blue (B) image of the urine particle image taken by the image input optical system, and thereby obtaining at least one parameter indicating a shape of the density histogram;
a second step of extracting a first object region larger than an object particle included therein, from the urine particle image, by using the at least one parameter, and at least one of the R image, the G image, and the B image;
a third step of calculating a size of the first object region based on at least one of an area or a perimeter of the first object region and a density distribution of the first object region, in the first object region, of each of the at least one of the R image, the G image, and the B image, and of classifying the first object region into a one group of a predetermined number of groups, based on the size of the first object region, and the density distribution of the first object region; and
a fourth step of extracting a second object region, representing the object particle, from the first object region, by using group-specific region segmentation, including use of the at least one parameter, and the at least one of the R image, the G image, and the B image.
3. The method for region segmentation of urine particle images according to claim 2 , characterized in that the second step includes the steps of:
calculating a density threshold of each of at least one of the R image, the G image, and the B image, based on the at least one parameter obtained in the first step; and
extracting an object region, by applying the density threshold to each of the images.
4. The method for region segmentation of urine particle images according to claim 2 , characterized in that the second step includes the steps of:
obtaining values, by performing filtering processing on the at least one of the R image, the G image, and the B image; and
extracting an object region, by performing threshold processing on the values.
5. The method for region segmentation of urine particle images according to claim 2 ,
wherein in the second step, a region is extracted as the first object region, by performing a logical OR operation on the following values:
a region segmentation result extracted through region segmentation, by calculating a density threshold of each of the at least one of the R image, the G image, and the B image, from the at least one parameter obtained in the first step, and by applying the threshold to each of the images, and
a region segmentation result extracted through region segmentation by performing threshold processing on values obtained, by performing filtering processing on the at least one of the R image, the G image, and the B image.
6. The method for region segmentation of urine particle images according to claim 2 , characterized in that the fourth step includes the steps of:
selecting at least one of the R image, the G image, and the B image, for each of the groups specified in the third step;
calculating a threshold for each of the selected color images by using the at least one parameter obtained in the first step; and
performing region segmentation for each of the groups by using the threshold.
7. The method for region segmentation of urine particle images according to claim 2 , characterized in that the fourth step includes the steps of:
selecting at least one of the R image, the G image, and the B image, for each of the groups specified in the third step;
calculating a threshold for each of the selected color images by using the at least one parameter obtained in the first step;
obtaining a first region segmentation result by performing region segmentation for each of the groups by using the threshold;
performing filtering processing on the one or more of the R image, the G image, and the B image, for each of the groups specified in the third step;
obtaining a second region segmentation result by performing threshold processing on values obtained through the filtering processing thus selected; and
extracting the second object region, by performing a logical OR operation by using the first region segmentation result and the second region segmentation result.
8. The method for region segmentation of urine particle image according to claim 2 ,
wherein for a group specified in the third step as being classified into a predetermined group, the region segmentation result obtained in the second step is outputted in the fourth step.
9. An apparatus for region segmentation of a urine particle image, comprising:
a means for extracting a first object region larger than an object particle included therein, from a urine particle image taken by an image input optical system, by using at least one of a red (R) image of the urine particle image, a green (G) image of the urine particle image, and a blue (B) image of the urine particle image;
a means for calculating a size of the first object region based on at least one of an area or a perimeter of the first object region and a density distribution of the first object region, in the first object region, of each of the at least one of the R image, the G image, and the B image, and for classifying the first object region into one group of a predetermined number of groups, based on the size of the first object region, and the density distribution of the first object region; and
a means for extracting a second object region, representing the object particle, from the first object region, by using group-specific region segmentation, including use of the at least one of the R image, the G image, and the B image.