KR20220090840A - Capsule endoscopy and processing method for image - Google Patents

Abstract

The present invention provides a capsule endoscope and an image processing method capable of extracting only an effective area necessary for lesion diagnosis from an image acquired inside the human body and providing it to an external system, a body forming an external shape, a lens mounted on the body, and An image sensor for acquiring an external image through the lens inside the body and an image signal provided from the image sensor are divided into an effective area and an unnecessary area required for lesion diagnosis, so that data on the effective area is provided to an external system It includes an image processing unit.

Description

CAPSULE ENDOSCOPY AND PROCESSING METHOD FOR IMAGE

The present invention relates to a capsule endoscope and an image processing method, and more particularly, to a capsule endoscope capable of imaging a lesion inside a human body and an image processing method.

In general, the digestive system includes the esophagus, stomach, small intestine and large intestine. Various examination methods, including ultrasound examination and tomography examination, are used to identify lesions of the digestive organs. However, the endoscopy method, in which a diagnostic expert visually inspects the inside of the digestive tract and makes a diagnosis, is preferred because it shows a high diagnosis rate. Among them, capsule endoscopy is preferred because it is easy to test, has less objection and pain, and can confirm lesions of the small intestine that are generally difficult to access.

The prior art for the capsule endoscope has already been disclosed by "Korea Patent Publication No. 2020-0101111 (Capsule Endoscope, 2020.08.27.)". In the disclosed invention, a capsule endoscope is inserted into the human body to take an image inside the human body, and the photographed information is transmitted to an external system so that a lesion examination is performed.

Such a capsule endoscope provides image information from the inside of the human body to an external system in real time, or provides stored information to an external system after being discharged from the human body. However, as the amount of data acquired and transmitted by the capsule endoscope increases, data processing efficiency is lowered, so there is a problem in that the inspection time and efficiency are lowered.

Republic of Korea Patent Publication No. 2020-0101111 (Capsule Endoscope, 2020.08.27.)

An object of the present invention is to provide a capsule endoscope and an image processing method capable of extracting only an effective area necessary for lesion diagnosis from an image acquired inside the human body and providing it to an external system.

The capsule endoscope according to the present invention includes a body forming an external shape, a lens mounted on the body, an image sensor for acquiring an external image through the lens inside the body, and an image signal provided from the image sensor to be effective for lesion diagnosis and an image processing unit that divides the region into a region and an unnecessary region so that data on the effective region is provided to an external system.

The image processing unit acquires a frame having a matrix of n*n based on the image signal provided from the image sensor, and after acquiring the effective region from a plurality of rows, the effective region from the top row to the bottom row It is possible to sequentially provide pixel information for each pixel.

After acquiring the number of pixels corresponding to the effective area in the uppermost row of the frame, the image processing unit sets an effective area by increasing the number of pixels from the middle row to the middle row in the height direction of the frame. The effective area can be set by reducing the number of pixels up to the bottom row of the frame.

When acquiring the number of pixels corresponding to the effective region among the uppermost row of the frame, the image processing unit may acquire the number of pixels based on conventionally learned information.

When the image processing unit sequentially provides the pixel information to the external system, the image processing unit may also provide recognition information indicating a row in which the pixels are arranged.

The image processing unit may provide the recognition information before the pixel information.

The identification information may include a sync identification code and a line code indicating a row in which the pixels are arranged.

The synchronization recognition code may include a pattern that is difficult to appear in the digital code.

The synchronization identification code may indicate the start of the transmission of the pixel information in one row.

After acquiring the number of pixels corresponding to the effective area among the uppermost rows of the frame, the image processing unit sets the effective area by increasing the number of pixels in the left and right directions from the height direction of the frame to the set row, and the number of pixels is less than the set row The effective area can be set by decreasing the number of pixels in the left and right directions.

The number of pixels increased or decreased in the left and right directions when the image processing unit sets the effective area may be different from each other.

An effective diameter of the lens may be smaller than a diagonal length of an incident surface of the image sensor.

On the other hand, the image processing method of the capsule endoscope according to the present invention includes the steps of: an image sensor disposed inside the capsule endoscope acquires an external image through a lens; Separating a necessary valid region and an unnecessary region, and providing data on the valid region to an external system.

Capsule endoscope and image processing method according to the present invention extracts an effective area in a preset pattern from photographed image information, and transmits a code for constructing an effective area when transmitting pixel information on the effective area to increase data throughput It has the effect of obtaining a high-quality image while reducing it.

The technical effects of the present invention as described above are not limited to the effects mentioned above, and other technical effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a perspective view schematically showing a capsule endoscope according to the present embodiment;
2 is a conceptual diagram schematically showing the internal structure of the capsule endoscope according to the present embodiment;
3 is an image obtained before applying the image processing method of the capsule endoscope according to the present embodiment;
4 is a flowchart illustrating an image processing method of a capsule endoscope according to the present embodiment;
5 is a conceptual diagram illustrating extraction of an effective area of the capsule endoscope according to the present embodiment;
6 is a conceptual diagram showing recognition information of the capsule endoscope according to the present embodiment;
7 is an image to which the image processing method of the capsule endoscope according to the present embodiment is applied;
8 is an image to which an image processing method of a capsule endoscope according to another embodiment is applied.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the present embodiment is not limited to the embodiment disclosed below, but may be implemented in various forms, and only this embodiment allows the disclosure of the present invention to be complete and the scope of the invention to those of ordinary skill in the art. It is provided for complete information. The shapes of elements in the drawings may be exaggerated for more clear explanation, and elements indicated by the same reference numerals in the drawings mean the same elements.

1 is a perspective view schematically showing a capsule endoscope according to this embodiment, and FIG. 2 is a conceptual diagram schematically showing an internal structure of the capsule endoscope according to the present embodiment.

1 and 2 , the capsule endoscope 100 according to the present invention includes a body 110 .

The body 110 forms the outer shape of the capsule endoscope 100 , and a lens 120 for photographing may be provided at the front end of the body 110 . And inside the body 110, the image through the lens 120 An image sensor 130 for acquiring information is provided. The image sensor 130 is formed of a complementary metal oxide semiconductor (CMOS).

In addition, the body 110 is provided with an image processing unit 140 for processing an image signal obtained from the image sensor 130 and a communication unit 150 for transmitting an image signal to an external system 200 capable of diagnosing a lesion. . Here, the image processing unit 140 processes the image acquired from the image sensor 130 in a preset manner. In addition, the communication unit 150 provides the processed data to the external system 200 so that the lesion examination can be performed.

On the other hand, the lens 120 provided in the capsule endoscope 100 may be provided in a circular shape, and the image sensor 130 may be provided with a rectangular incident surface for obtaining information. Accordingly, the effective diameter of the lens 120 occupies a larger area than the image sensor 130 . Therefore, when an image is taken through the capsule endoscope 100, the effective area A1 and the unnecessary area A2 necessary for lesion diagnosis are acquired together.

3 is an image obtained before applying the image processing method of the capsule endoscope according to the present embodiment.

As shown in FIG. 3 , looking at the image before image processing is applied in the capsule endoscope 100 according to the present embodiment, the effective diameter of the lens 120 is smaller than the diagonal length of the image sensor 130 . It can be seen that the outer part of the area does not acquire light and appears black. Therefore, when an image is transmitted from the capsule endoscope 100 to the external system 200, the effective area A1 and the unnecessary area A2 that do not affect the diagnosis of the lesion are transmitted together, thereby reducing communication efficiency.

Accordingly, the capsule endoscope 100 according to the present embodiment removes an unnecessary area from the image processing unit 140 and acquires only the effective area A1 necessary for lesion diagnosis, so that information on the effective area A1 is transmitted to the external system 200 ) to be provided.

Hereinafter, an image processing method for extracting and providing the effective area A1 necessary for lesion diagnosis will be described in detail. However, detailed descriptions of the above-described components are omitted and the same reference numerals are given.

4 is a flowchart illustrating an image processing method of the capsule endoscope according to the present embodiment, and FIG. 5 is a conceptual diagram illustrating extraction of an effective area of the capsule endoscope according to the present embodiment. And Fig. 6 is a conceptual diagram showing recognition information of the capsule endoscope according to the present embodiment, and Fig. 7 is an image to which the image processing method of the capsule endoscope according to the present embodiment is applied.

4 to 7 , in the image processing method of the capsule endoscope 100 according to the present embodiment, the image sensor 130 acquires an image of the inside of the human body through the lens 120 ( S100 ). Accordingly, a frame composed of n*n pixels is provided to the image processing unit 140 . Accordingly, the image processing unit 140 extracts only the effective area A1 from the n*n frame and provides it to the external system 200 .

In this case, the frame is acquired in a quadrangular shape so that a matrix composed of a plurality of pixels corresponds to the incident surface of the image sensor 130 . Here, the area adjacent to the corner of the rectangular frame is an unnecessary area A2 that does not include information on lesion diagnosis, as described above. Accordingly, the image processing unit 140 extracts the effective area A1 based on the matrix pattern of pixels forming the frame.

For example, the image processing unit 140 acquires information about a pixel corresponding to the effective area A1 among a plurality of pixels forming the first row ( S200 ). In addition, the effective area A1 is extracted by increasing the number of pixels from the uppermost row to the middle row (hereinafter referred to as an upper area) in the height direction of the frame to obtain information on the corresponding pixel.

Also, from the middle row to the lowest row (hereinafter, referred to as a sub-region) in the height direction of the frame, the effective area A1 is extracted in a manner that reduces the number of pixels to obtain information on the corresponding pixel.

More specifically, it is assumed that the frame recognized by the image processing unit 140 has a horizontal length of 18 and a vertical length of 10 as shown in FIG. 5 .

In this case, the image processing unit 140 sets the effective area by increasing the number of pixels in each row in the upper area from the first row to the fifth row. For example, the image processing unit 140 may know that two pixels located in the middle among pixels constituting the first row correspond to the effective area A1 based on conventionally learned information. Accordingly, in the first row, the image processing unit 140 acquires two pixels located in the middle as an effective area. In the second row, 2+n pixels located in the middle among pixels constituting the second row are recognized as effective areas, and corresponding pixel information is acquired. As described above, the image processing unit 140 sets the effective area A1 in a manner of increasing the number of pixels up to the fifth row, and extracts 18 pixels as the effective area A1 in the fifth row.

In addition, the image processing unit 140 sets the effective area A1 in such a way that pixels are reduced for each row in the lower area from the sixth row to the tenth row. That is, in the sixth row forming the middle of the frame, 18 pixels may correspond to the effective area A1 as in the fifth row. In the 7th row, 18-n pixels, which are fewer than the 6th row, are recognized as the effective area A1, and corresponding pixel information is acquired. In this way, the image processing unit 140 sets the effective area A1 in a manner of reducing the number of pixels up to the tenth row, and extracts two pixels as the effective area A1 in the tenth row.

Meanwhile, when the extraction of the effective area A1 is completed, the image processing unit 140 provides pixel information corresponding to the effective area A1 to the external system 200 through the communication unit 150 ( S300 ). At this time, the image processing unit 140 transmits the position information of the pixel corresponding to the effective area A1 together so that the external system 200 can construct a frame for the effective area A1 based on the received pixel information. .

That is, when transmitting information about a plurality of pixels, the image processing unit 140 also transmits recognition information 400 for a row in which the corresponding pixel is located. As described above, when two pixels extracted from the first row are transferred to the effective area A1, location information for identifying the first row and two pieces of pixel information are transferred together. In addition, when six pixels extracted from the second row are transferred to the effective area, location information for identifying the second row and six pixel information are transferred together. Next, from the third row to the tenth row, location information indicating each row and pixel information extracted to the effective area A1 are transmitted together.

Here, the recognition information 400 capable of identifying a row may include a synchronization recognition code 410 indicating the start and end of a row and a line code 420 indicating information about the row. The synchronization recognition code 410 is a code having a pattern that is difficult to appear in a digital code. [11 22 33 44 55] is exemplified in this embodiment, but this is for explaining the present embodiment, and the pattern is not limited thereto. In addition, the line code 420 may be expressed as [00 00] in the case of the first row and [01 01] in the case of the second row as position information on the row, but this is for explaining the present embodiment and the pattern does not limit

As such, when the image processing unit 140 provides pixel information corresponding to the effective area A1, the synchronization recognition code 410 + line code 420 + pixel information 430 corresponding to the effective area of the corresponding column is transmitted to the communication unit. It is provided to the external system 200 through 150 . Accordingly, the external system 200 removes the synchronization recognition code 430 when the corresponding information is provided from the capsule endoscope 100 and arranges the pixels according to the line code 420 to create a frame for the effective area A1. configurable.

More specifically, for example, when the image processing unit 140 provides pixel information corresponding to the effective area A1 extracted from the first row, the synchronization recognition code 410 [11 22 33 44 55] , [00 00], which is a line code 420, and [25 66], which is an example of pixel information 430, are conveyed. Then, when providing pixel information corresponding to the effective area A1 extracted from the second row, [11 22 33 44 55], which is the synchronization recognition code 410, [01 01], which is the line code 420, and the pixel [55 34 22 65 35 58], which is an example of the information 430, is transmitted sequentially up to the tenth row.

And the external system 200 removes [11 22 33 44 55], which is the pre-approved synchronization recognition code 410 when pixel information is provided, and the line code 420 according to [00 00] as the line code 420 The pixel information 430 following . is configured in the first row. And the external system 200 recognizes that the configuration of the first row has been completed according to the synchronization recognition code 410 in the continuously provided information. Thereafter, the external system 200 removes the synchronization recognition code 410 and configures the pixel information 430 following the line code 420 in the second row according to the line code 420 [01 01], , sequentially configure pixels up to the tenth row. Accordingly, the external system 200 may acquire an image of the effective area A1 from which the unnecessary area A2 is excluded (S400).

Meanwhile, in this embodiment, the image processing unit 140 extracts the effective area A1 while increasing or decreasing the number of pixels in the height direction is described. However, the image processing unit 140 may extract the effective area A1 in another method.

8 is an image to which an image processing method of a capsule endoscope according to another embodiment is applied.

As shown in FIG. 8 , in the image processing method of the capsule endoscope according to another embodiment, the image sensor 130 acquires an image of the inside of the human body through the lens 120 . Accordingly, a frame composed of n*n pixels is provided to the image processing unit 140 . Accordingly, the image processing unit 140 extracts only the effective area A1 from the n*n frame and provides it to the external system 200 .

In this case, the image processing unit 140 acquires information on a pixel corresponding to the effective area A1 among a plurality of pixels forming the first row. Here, the image processing unit 140 may know the number and positions of pixels corresponding to the effective area A1 among pixels constituting the first row based on conventionally learned information.

Thereafter, the image processing unit 140 extracts the effective area A1 by increasing the number of pixels in the left and right directions from the height direction to the set height to obtain information on the corresponding pixel. However, in this case, the number of pixels increasing in the left and right directions may be different from each other. In addition, the image processing unit 140 may extract the effective area A1 in a direction in which the pixel is reduced in the left and right direction to a height less than or equal to a set height in the height direction to obtain information on the corresponding pixel. Even in this case, the number of pixels increasing in the left and right directions may be different from each other, and pixel information may be acquired so that the upper region and the lower region are symmetrical to each other.

More specifically, when the number of effective areas A1 is n among the plurality of pixels forming the first row, the increase/decrease height of the effective area A1 is set to h. In addition, the increasing number of left pixels per row may be set to Lp and the number of right pixels may be set to Rp, and the decreasing number of left pixels per row may be set to Lm and the number of right pixels may be set to Rm. At this time, pixel information is acquired by increasing the number of pixels by Lp+Rp up to the specified height, for example, the fifth row constituting the middle row, and decreasing the number of pixels by Lm+Rm from the sixth row to acquire pixel information do. Here, Lp, Rp, Lm, and Rm may be natural numbers and may be different from each other.

For example, the number of n representing the effective area of the first row may be two. And h may be set to 6. In addition, Lp is set to 2 and Rp is set to 1, so that the number of pixels increasing from the left may be set to +2, and the number of pixels increasing from the right may be set to +1. In addition, Lm is set to 2, Rm is set to 1, so that the number of pixels decreasing from the left may be set to ??2, and the number of pixels decreasing from the right may be set to ??1.

Accordingly, in the first row, pixel information on the two effective areas A1 is acquired, and until the sixth row, the effective area A1 is increased by +2 in the left direction and +1 in the right direction for each row, and pixel information to acquire In addition, from the 7th to 10th rows occupying the middle area, pixel information is acquired by decreasing the effective area A1 by -2 in the left direction and ?? 1 in the right direction for every row.

In this way, the capsule endoscope 100 extracts the effective area A1 from the photographed image information in a preset pattern, and when transmitting pixel information on the effective area A1, a code for configuring the effective area A1 It has the advantage of being able to acquire high-quality images while reducing data throughput by transmitting

One embodiment of the present invention described above and shown in the drawings should not be construed as limiting the technical spirit of the present invention. The protection scope of the present invention is limited only by the matters described in the claims, and those skilled in the art can improve and change the technical idea of the present invention in various forms. Accordingly, such improvements and modifications will fall within the protection scope of the present invention as long as they are obvious to those of ordinary skill in the art.

Claims (13)

a body forming an outline;
a lens mounted on the body;
an image sensor for acquiring an external image through the lens inside the body; and
and an image processing unit that divides the image signal provided from the image sensor into an effective area and an unnecessary area necessary for lesion diagnosis, and provides data on the effective area to an external system.
According to claim 1,
The image processing unit
acquiring a frame having an n*n matrix based on the image signal provided from the image sensor;
The capsule endoscope according to claim 1, wherein the pixel information on the effective area from the uppermost row to the lowermost row is sequentially provided after each of the effective areas is obtained from a plurality of rows.
3. The method of claim 2,
The image processing unit
After acquiring the number of pixels corresponding to the effective area among the uppermost rows of the frame, the effective area is set by increasing the number of pixels from the height direction of the frame to the middle row, and from the middle row to the lowermost row of the frame Capsule endoscope, characterized in that the effective area is set in such a way that the number of pixels is reduced.
4. The method of claim 3,
The image processing unit
Capsule endoscope, characterized in that when acquiring the number of pixels corresponding to the effective area in the uppermost row of the frame, the number of pixels is acquired based on conventionally learned information.
3. The method of claim 2,
The image processing unit
When sequentially providing the pixel information to the external system,
Capsule endoscope, characterized in that it provides recognition information indicating the row in which the pixels are arranged.
6. The method of claim 5,
The image processing unit
Capsule endoscope, characterized in that the recognition information is provided before the pixel information.
6. The method of claim 5,
The recognition information
Capsule endoscope comprising a synchronization recognition code and a line code indicating a row in which the pixels are arranged.
8. The method of claim 7,
The synchronization recognition code is
Capsule endoscope, characterized in that it contains a pattern that is difficult to appear in digital code.
8. The method of claim 7,
The synchronization recognition code is
A capsule endoscope, characterized in that one row indicates the start of the transmission of the pixel information.
3. The method of claim 2,
The image processing unit
After acquiring the number of pixels corresponding to the effective area among the uppermost rows of the frame, the effective area is set in such a way that the number of pixels is increased in the left and right direction from the height direction of the frame to the set row, and from below the set row in the left and right direction Capsule endoscope, characterized in that the effective area is set in a way that reduces the number of pixels.
11. The method of claim 10,
The image processing unit
Capsule endoscope, characterized in that the number of pixels increased or decreased in the left and right directions in setting the effective area are different from each other.
According to claim 1,
The effective diameter of the lens is
Capsule endoscope, characterized in that smaller than the diagonal length of the incident surface of the image sensor.
acquiring an external image through a lens by an image sensor disposed inside the capsule endoscope;
classifying, by an image processing unit, an effective area and an unnecessary area necessary for lesion diagnosis based on the image signal provided from the image sensor; and
An image processing method of a capsule endoscope comprising the step of providing data on the effective area to an external system.

Images