KR20130101867A - A method and apparatus outputting an analyzed information for a specific part in the living body - Google Patents

Abstract

PURPOSE: An analysis information outputting method for the specific part of a body and a device thereof are provided to offer accurate analyzing information in medical and cosmetic fields by providing a variety of quantified analysis information for a specific part according to similarity by comparing the image of the specific part with the similarity of multiple comparing target images stored in a database. CONSTITUTION: A device detects inspecting target image for the specific part of a body (100). The device respectively compares the similarity of a variety of comparing target images stored in one of a disease for the specific part related database and a skin related database with an inspecting target image. The device outputs analysis information including the quantified information for the part in which the similarity is recognized (110). [Reference numerals] (100) Detect target images for inspection of the specific part of a body; (102) Perform an image normalization and image quality processing process for the inspection target images; (104) Perform a mutual correction process for the inspection target images and comparison target images respectively; (106) Divide the sizes of the inspection target images and the comparison target images respectively; (108) Compares the similarity among the comparison target images and the inspection target images respectively; (110) Output the quantified analysis results for those images having a certain level of similarity; (AA) Start; (BB) End

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for analyzing information on a specific region in a living body,

The present invention relates to an invention providing various information relating to the presence or absence of disease, the progress or the beauty of a specific part of a living body such as an organ or an organ in a living body of a human or an animal, and more particularly, A technology for extracting images to be compared from a database having various types of images corresponding thereto and providing analysis information of a living body that enables information on diseases for a specific site and various information related to beauty to be output as quantified information .

Various image photographing apparatuses for photographing living bodies such as human or animal organs and skin have been developed. As such a photographing apparatus, a general camera for photographing a skin of a living body may be exemplified. A medical apparatus for photographing an organ of a living body may be an X-ray photographing apparatus, a CT (Computed Tomography) photographing apparatus, an MRI Resonance Imaging), and ultra sound imaging devices.

The biomedical image is used as an important data for determining a plan or a method for diagnosis, treatment, surgery, operation or beauty by identifying a specific part of a living body. Since the reading and diagnosis of such a biomedical image can be performed by a person having specialized knowledge and experience in the relevant field, the need for additional information that can provide an objective judgment basis is required.

However, the information about the biomedical image can be judged only when a medical person or a beauty-related worker carefully examines the biomedical image in the region where the disease occurs, the abnormality of the skin tissue, and the degree of change. Therefore, when a medical person or a beauty worker does not concentrate his attention, it may be difficult to accurately grasp the state of the patient or the beauty user. In addition, since the quantified information about the living body can not be known, the judgment information about the living body may be inconsistent depending on the subject of the medical person or the beauty person.

In order to quantitatively confirm information on a specific part of a living body such as an organ or organ in a living body of a human or an animal, the object of the present invention is to provide a method and a system for analyzing an image of a specific region, The present invention relates to a method and an apparatus for outputting analysis information for a specific region in a living body, which compares the similarities of a plurality of stored comparison images with each other and outputs quantified analysis information according to the similarity.

According to an aspect of the present invention, there is provided a method for outputting analysis information on a specific region of a living body, the method comprising: detecting an image to be inspected for a specific region of a living body; Comparing a plurality of comparison images stored in a database of the disease-related database and the skin-related database with respect to the specific region and the similarity of the image to be examined, respectively; And outputting analysis information including quantified information on a portion where the degree of similarity is recognized.

The specific site is characterized by including organs, living organs and skin of a living body including humans and animals.

The step of comparing the similarity between the comparison images and the inspection object image is characterized by using a Content-Based Image Retrieval (CBIR) method.

The analysis information output method for a specific area in the living body may further include performing image standardization and image processing for improving the image quality of the inspection target image after the inspection target image is detected.

The analysis information output method for a specific region in the living body may further include adjusting image mutual alignment for the inspection target image and the comparison target images after detecting the inspection target image, .

The analysis information output method for a specific region in the living body may further include dividing the inspection target image and the comparison target images into a predetermined size and dividing the divided inspection target partial images and the corresponding divided comparison target portions And comparing the similarities for each of the images.

Wherein each of the divided partial images to be compared and the divided images of the partial images to be inspected are compared with each other by using a distributed processing technique using a plurality of computers, And distributes the similarity comparison to the partial images.

The step of outputting the analysis information may include displaying at least one disease-related information and skin-related information on any one of the organ, the organ, and the skin including human and animal as the specific region on the inspection object image .

According to an aspect of the present invention, there is provided an apparatus for outputting analysis information on a specific region of a living body, the apparatus comprising: an image detector for detecting an image to be inspected for a specific region of a living body; A database storing at least one of a plurality of comparison images related to the disease and a plurality of comparison images related to the skin for the specific region; A similarity comparison control unit for comparing the similarities of the comparison images stored in the database with the inspection target images; And an analysis information output unit for outputting analysis information including quantified information about a portion where the degree of similarity is recognized.

The specific site is characterized by including organs, living organs and skin of a living body including humans and animals.

The similarity comparison control unit uses a content-based image retrieval (CBIR) scheme.

The analysis information output apparatus for a specific region in the living body may further include an image processing unit for performing image standardization and image processing for improving the image quality of the inspection target image after the inspection target image is detected .

And the image processing unit adjusts mutual image alignment for the inspection object image and the comparison object images after the inspection object image is detected.

Wherein the similarity comparison control unit divides the inspection object image and the comparison object images into a predetermined size and compares the divided inspection subject partial images with the corresponding divided partial images to be compared, do.

The similarity degree comparison control unit is characterized by using a distributed processing technique using a plurality of computers to perform control so as to variably compute the similarity degree comparison between the divided inspection target partial images and the divided comparison target partial images .

Wherein the analysis information output unit displays one or more disease-related information and skin-related information for the organ, the organ, and the skin including human and animal as the specific region on the image to be inspected .

According to the present invention, an image of a specific region is compared with a plurality of comparative images stored in a database, and various quantitative analysis information on a specific region according to the degree of similarity is provided. Information can be provided.

Through this, it is possible to acquire statistical information related to disease and beauty by using quantitative information related to diseases or beauty related to the living body, thereby contributing to health promotion of the people, and systematically building health care .

1 is a flowchart illustrating an image analysis information output method for a specific region in a living body according to an embodiment of the present invention.
2 is a reference diagram of an embodiment for explaining a method of outputting image analysis information on a human retina in vivo.
FIG. 3 is a reference view illustrating the similarity comparison between the divided inspection target partial images and the divided comparison target partial images in each of the plurality of computers.
Fig. 4 is an example of a reference for outputting a result of comparison with a comparison object image stored in a database with respect to human facial skin. Fig.
5 is a block diagram of an embodiment of an image analysis information output apparatus for a specific region in a living body according to the present invention.

Hereinafter, an image analysis information output method for a specific region in a living body according to the present invention will be described with reference to the accompanying drawings.

1 is a flowchart illustrating an image analysis information output method for a specific region in a living body according to an embodiment of the present invention.

First, an image to be inspected for a specific region of a living body is detected (operation 100).

Here, the specific region of the living body includes organs, organs and skin of a living body including a human. For example, the organs include heart, liver, lung and the like corresponding to the internal organs of human and animal, and the organs include human and animal facial, eye, nose, ear, teeth and the like, Animal skin tissues, fingerprints, and the like. In order to detect an image of a specific region of a living body, a general camera for photographing a skin of a living body can be generally used as an apparatus for image photographing. A medical device for photographing a living organ or the like, Device, a computed tomography (CT) imaging device, an MRI (Magnetic Resonance Imaging) imaging device, and an ultrasonic imaging device.

After operation 100, image processing for image normalization and image quality enhancement is performed on the detected inspection object image (operation 102). The image normalization operation corresponds to a stereotyping operation for adjusting the size of a photographed image to a certain standard and deleting an unnecessary portion. Further, in order to improve the image quality, an image processing operation for correcting the brightness and saturation of the photographed image is performed. However, the step 102 is not necessarily a required step, and is a process that can be selectively performed as needed.

After operation 102, image correlation is performed on the image to be examined and the plurality of comparison images stored in the disease-related database or the skin-related database for the specific region, respectively (operation 104). The plurality of images to be compared may be a plurality of images related to the disease, a plurality of images related to the skin, image information about various kinds of living bodies including human or animal, The image information is stored in advance in the database, and the comparative image information is continuously updated by the expert group or the like.

Image cross calibration refers to adjusting the alignment of the image, i.e., the image alignment, for the comparison of the image to be examined and the images to be compared, and means performing the same image processing operation between the images. Step 104 is not necessarily a required step, but is a process that can be selectively performed as needed.

After operation 104, the size of the image to be scanned and the images to be compared are divided into a predetermined size (operation 106). For example, the inspection object image and the comparison object images are divided into sizes of 2 * 2, 4 * 4, 8 * 8, and the like. In step 106, it is not necessarily a required step, and it is a process that can be selectively performed as needed when the data size of the inspection target image and the comparison target images is large.

After operation 106, the similarity between the detected inspection object image and the comparison images stored in the database is compared (operation 108). In order to compare the similarity between the comparison target images and the inspection target image, a content-based image retrieval (CBIR) method can be used. Content-based image retrieval method (CBIR) is a technique for expressing images using visual attributes such as color, texture, and shape of an image, and retrieving desired images from a large-capacity image database using an initial query image.

2 is a reference diagram of an embodiment for explaining a method of outputting image analysis information on a human retina in vivo. In the case of the reference numeral 1 in Fig. 2, the image to be inspected of the retina is detected. The disease-related database stores the comparison images A, B, C, D, E, F, and G for the retina, respectively. 2 is a similarity search engine for comparing the images to be examined of the retina with the images A, B, C, D, E, F and G of the retina for comparison, Method (CBIR) is used to compare the similarities between images. As a result of the similarity degree search by the similarity degree search engine, as shown in the reference numeral 3 in FIG. 2, the similarity degree of the retinal image of the images C, F, Information is shown. Here, when a certain criterion indicates the similarity as a percentage, an image having a certain degree of similarity (for example, 70 [%]) or more is extracted as a similar image.

In operation 108, the inspection target image may be compared with the comparison target images without dividing the inspection target image. In operation 106, if the sizes of the inspection target image and the comparison target images are divided into a predetermined size And compares the partial inspection target partial images with the corresponding divided partial images to be compared. That is, the divided partial images to be inspected and the corresponding partial images to be compared are compared using a content-based image retrieval method (CBIR). Accordingly, an image having a certain percentage (for example, 70 [%]) or more similarity to each of the divided partial images to be inspected and the corresponding divided partial images to be compared is extracted as a similar image.

In order to compare each of the divided inspection target partial images and corresponding divided divided comparison partial images, the divided inspection target partial images and the divided comparison target partial images Can be distributed. FIG. 3 is a reference view illustrating the similarity comparison between the divided inspection target partial images and the divided comparison target partial images in each of the plurality of computers. 3, the reference partial images a, b, c and d and the corresponding partial images a-1 corresponding to the size of 2 * , (b-1), (c-1), and (d-1). In the subfield 2, the subdivided inspection subject partial images and the corresponding comparative partial images are distributed in each of the computers. In this distributed work, as in the case of the identification symbol 3, the similarity comparison operation for the partial image to be compared and the comparative partial image corresponding thereto is distributed. Thereafter, as in the case of the identification symbol 4, the result of the distributed processing is collected, and finally, the image integration processing result reflecting the similarity result for each partial image is extracted,

After operation 108, image analysis information including quantified information on an image portion in which similarity is recognized is output (operation 110). The quantified image analysis information is obtained by comparing the image to be inspected and the image to be compared and comparing the degree of similarity with respect to the image portion similar to the comparative image among the images to be inspected by the degree of similarity such as the size, And outputs it. The quantified image analysis information may be displayed through a display device, or may be printed through a printing device.

In particular, a plurality of disease-related information on a human organ or organ, as a specific region of a living body, is displayed on an image to be examined, and the quantification information of the disease-related information is output. For example, as shown in the reference numeral 3 in Fig. 2, a position on an image to be inspected is displayed for image portions C, F, G similar to a comparison image stored in a disease-related database. At this time, it is possible to display information about the disease related to the location, that is, the disease name, the progress of the disease, and the like.

Further, a plurality of pieces of skin-related information about a human skin as a specific region of a living body are displayed on an image to be inspected and the quantification information of the skin-related information is output.

Fig. 4 is an example of a reference for outputting a result of comparison with a comparison object image stored in a database with respect to human facial skin. Fig.

FIG. 4A shows detection of an image to be inspected using an image capturing apparatus for image detection, and FIG. 4B shows information of various kinds of facial skin related images stored in a skin database. It is. Accordingly, by comparing the images to be inspected for the facial skin with the facial skin-related images stored in the database, images having similarity higher than a certain standard are extracted, and based on information on the extracted similar images, Quantitative skin-related information about health status, brightness, color, etc. is output.

Hereinafter, an image analysis information output apparatus for a specific region in a living body according to the present invention will be described.

FIG. 5 is a block diagram of an image analysis information output apparatus for a specific region in a living body according to the present invention. The image analysis unit 200 includes a database 210, an image processing unit 220, (230) and an analysis information output unit (240).

The image detecting unit 200 detects an image to be inspected for a specific region of a living body and outputs the detection result to the image processing unit 220. The specific site includes organs, organs and skin of a living body including a human. The image detecting unit 200 may correspond to a general camera for photographing a skin of a living body to detect an image of a specific part of a living body, and may be a medical device for photographing a living organ or the like, An imaging device, an MRI imaging device, an ultrasound imaging device, and the like.

On the other hand, the database 210 stores a plurality of comparison images related to the disease or a plurality of comparison images related to the skin for the specific region.

The image processing unit 220 performs image normalization for the detected inspection object image and image processing for image quality improvement. The image processing unit 220 adjusts the size of the photographed image according to a certain standard, and performs an image normalization operation to delete unnecessary portions. In order to improve the image quality, the image processing unit 220 performs an image processing operation for correcting the contrast and saturation of the photographed image.

In addition, the image processing unit 220 fetches the compared images to be compared with the detected images to be inspected from the database 210, and performs image mutual correction on the fetched comparison images and the detected images to be inspected . The image processing unit 220 adjusts the alignment of the image, that is, the alignment of the image, with respect to the inspection target image and the comparison target images, and performs the same image normalization and image quality improvement operations on the inspection target image and the comparison target image, respectively do.

The similarity comparison control unit 230 compares the similarity between the comparison images stored in the database 210 and the inspection object image, and outputs the comparison result to the analysis information output unit 240. The similarity comparison control unit 230 includes a Content-Based Image Retrieval (CBIR) engine for comparing the similarity between the comparison images and the inspection object image. The similarity comparison control unit 230 searches the similarity between images to be compared using visual attributes such as color, texture, and shape between the comparison images and the inspection object image.

2, the similarity comparison control unit 230 compares the comparison images A, B, C, D, E, F, and G of the retina stored in the database 210 with the image to be inspected as a content- (CBIR), and the degree of similarity retrieval results, as exemplified in the reference numeral 3 in FIG. 2, an image C, F , And G are extracted.

On the other hand, the similarity comparison control unit 230 divides the inspection target image and the size of the corresponding comparison target images into a predetermined size, and outputs the divided inspection target partial images and the corresponding divided partial images to be compared The similarity degree is compared. The size of the inspection target image and the size of the comparison target image are divided when the data capacity of the image is equal to or larger than a certain size. For example, the similarity comparison control unit 230 may divide the inspection target image and the comparison target image into sizes of 2 * 2, 4 * 4, 8 * 8, and the like, respectively. Reference numeral 1 in Fig. 3 is an example of images to be inspected and images to be compared which are divided into 2 * 2.

The similarity comparison control unit 230 divides the size of the inspection target image and the comparison target images into a predetermined size, and then, for each divided inspection target partial image and corresponding comparison target partial images, color, texture, Using visual attributes such as shapes, the similarities between the compared images are compared.

The similarity comparison control unit 230 compares the divided partial images to be inspected and the partial images to be compared with each other in each of the plurality of computers, It is also possible to control to distribute the comparison to the divided comparison target partial images. 3, the similarity comparison control unit 230 controls the divided inspection target partial images and the corresponding comparison target partial images to be distributed in each of the plurality of computers. In the distributed work, as in the case of the reference numeral 3, the comparative work of the partial image to be inspected and the comparative partial image corresponding to the divided partial image are distributed.

The analysis information output unit 240 outputs quantitative analysis information for an image portion for which similarity is recognized based on the similarity comparison result provided from the similarity comparison control unit 230. [ The quantified analysis information is obtained by comparing the inspection target image and the comparison target image and comparing the degree of similarity of the image portion similar to the comparison target image among the inspection target images, that is, the similarity of the image size, color, brightness, And so on. The analysis information output unit 240 may include a display device (not shown) for displaying the quantified image analysis information, and may include a printing device (not shown) for printing the quantified image analysis information.

In particular, the analysis information output unit 240 displays a plurality of disease-related information on a human organ as a specific region of a living body on an image to be examined and outputs quantification information of the disease-related information. 2, the analysis information output unit 240 may include an image analysis unit 240 for analyzing the image portion C, F, G similar to the comparison image stored in the disease-related database 210 on the image to be examined Displays the location of. At this time, the analysis information output unit 240 may display disease related information about the location, that is, the disease name, the progress of the disease, and the like.

In addition, the analysis information output unit 240 displays a plurality of skin-related information on human skin as a specific region of a living body on an image to be inspected, and outputs quantification information of such skin-related information. The analysis information output unit 240 outputs quantified skin-related information about brightness, color, etc. on the basis of information about an image for which a degree of similarity higher than a certain standard is recognized.

Meanwhile, the method inventions of the present invention described above can be implemented as computer readable codes / instructions / programs. For example, it may be implemented in a general-purpose digital computer that operates the code / instructions / program using a computer-readable recording medium. The computer-readable recording medium includes a storage medium such as a magnetic storage medium (e.g., a ROM, a floppy disk, a hard disk, a magnetic tape, etc.), an optical reading medium (e.g., a CD-ROM, a DVD, .

The apparatus and method for outputting image analysis information for a specific region of a living body according to the present invention have been described with reference to the embodiments shown in the drawings for the purpose of understanding, but the present invention is not limited thereto. It will be understood that various modifications and equivalent embodiments are possible. Accordingly, the true scope of the present invention should be determined by the appended claims.

200:
210: Database
220: Image processing unit
230:
240: Analysis information output section

Claims (16)

Detecting an image to be inspected for a specific region of a living body;
Comparing the similarity of the plurality of comparison images stored in any one of a disease-related database and a skin-related database with the similarity of the inspection target image with respect to the specific site; And
And outputting analysis information including quantified information on the part of which the similarity is recognized. The method of claim 1,
Wherein the specific region includes organ, organ, and skin of a living body including humans and animals. 2. The method of claim 1, wherein comparing the similarities of the images to be compared with the images to be inspected comprises:
And a content-based image retrieval (CBIR) method is used. The method according to claim 1, wherein the analysis information output method for a specific region in the living body
And after detecting the inspection target image, performing image processing on the inspection target image to improve image standardization and image quality. The method according to claim 1, wherein the analysis information output method for a specific region in the living body
And after detecting the inspection target image, adjusting image mutual alignment with respect to the inspection target image and the comparison target images, respectively. The method according to claim 1, wherein the analysis information output method for a specific region in the living body
The method may further include dividing the inspection target image and the comparison target images into a predetermined size,
And comparing similarity with each of the divided inspection target partial images and the corresponding divided comparison partial images. The method of claim 6, wherein comparing the similarities between the divided comparison target partial images and the divided inspection target partial images, respectively,
Outputting image analysis information for a specific part of the living body, characterized in that the similarity comparison between the divided inspection partial images and the divided comparison partial images is distributed using a distributed processing technique using a plurality of computers. Way. 3. The method of claim 2, wherein the step of outputting the analysis information comprises:
Related information and skin-related information on any one of the organ, the organ, and the skin including the human and the animal as the specific region is displayed on the image to be inspected. The method comprising: An image detector for detecting an image to be inspected for a specific part of a living body;
A database storing at least one of a plurality of comparison images related to the disease and a plurality of comparison images related to the skin for the specific region;
A similarity comparison control unit for comparing the similarities of the comparison images stored in the database with the inspection target images; And
And an analysis information output unit for outputting analysis information including quantified information on the part of which the similarity is recognized. 10. The method of claim 9,
Wherein the specific region includes an organ of a living body including a human and an animal, a living organism, and a skin. 10. The apparatus according to claim 9, wherein the similarity comparison control unit
And a content-based image retrieval (CBIR) method is used. 10. The apparatus according to claim 9, wherein the analysis information output device for a specific region in the living body
Further comprising an image processing unit for performing image processing for image normalization and image quality enhancement on the inspection target image after detecting the inspection target image. 13. The image processing apparatus according to claim 12, wherein the image processing unit
And the image mutual alignment is adjusted for each of the inspection object image and the comparison object images after the inspection object image is detected. 10. The apparatus according to claim 9, wherein the similarity comparison control unit
The inspection target image and the comparison target image are divided into a predetermined size, and the similarity is compared with each of the divided inspection target partial images and the corresponding divided comparison partial images. Analysis information output device. The method of claim 14, wherein the similarity comparison control unit
A distributed processing technique using a plurality of computers is used to perform a distributed processing for comparing the similarity of the divided inspection partial images and the divided comparison partial images to a specific part of the living body. Analysis information output device. The apparatus of claim 10, wherein the analysis information output unit
Related information and skin-related information on any one of the organ, the organ, and the skin including the human and the animal as the specific region is displayed on the image to be inspected. And an analysis information output device.

Images