KR102062372B1 - Apparatus for optical coherence tomography and method for image processing thereof for improving image quality based on region segmentation - Google Patents

Abstract

The present invention relates to an optical coherence tomography apparatus, which divides an A-scan signal into a partial signal, performs correction on the divided partial signal, and combines the corrected partial signals to reconstruct the A-scan signal. It includes an image processing unit for generating. According to the present invention, the influence of the speckle noise can be effectively reduced.

Description

Quality Improvement Optical Coherence Tomography Apparatus and Its Image Processing Method Based on Region Segmentation Technique

The present invention relates to an optical coherence tomography apparatus and an image processing method thereof, and more particularly, to an optical coherence tomography apparatus and an image processing method of improving image quality by performing image processing based on a region segmentation technique. .

Optical Coherence Tomography (OCT) is a technique for creating micrometer-resolution three-dimensional images in optical dispersion media using light, in a non-invasive way, such as optic nerve papilla, retina and cornea. It is to grasp and analyze the internal structure of the body, such as the eyeball. Of these, Time Domain-Optical Coherence Tomography (TD-OCT) has the disadvantage of low noise and low resolution, but Fourier Domain-Optical Coherence Tomography (FD-OCT) Spectral Domain-Optical Coherence Tomography (SD-OCT), which is one of them, is widely used because of its superior speed and resolution.

However, when the image is generated by the optical coherence tomography apparatus, the speckle noise is generated in the image signal due to the influence of the interfering light. Make analysis difficult. Therefore, recently, image quality improvement methods for removing or reducing speckle noise have been developed. Among them, image processing techniques and statistical processing techniques for analyzing two-dimensional image features have been applied.

Korean Laid-Open Patent Publication 10-2014-0021487

SUMMARY OF THE INVENTION An object of the present invention is to provide an optical coherence tomography apparatus and an image processing method capable of improving image quality without using a two-dimensional image processing technique or a statistical processing technique for reducing or eliminating speckle noise. will be.

The optical coherence tomography apparatus according to an embodiment of the present invention for solving the above problems, the light source unit for irradiating the light, and splits the light into two optical paths, and recombines the branched light to cause an interference phenomenon An interferometer, an image receiver for detecting an interference signal from the interferometer and generating an A-scan signal, and dividing the A-scan signal into partial signals grouped by similar signal characteristics, and performing correction on the divided partial signals And an image processor for generating the reconstructed A-scan signal by combining the corrected partial signals.

The image processor may filter the A-scan signal in a frequency domain and divide the partial signal using a frequency section between vertices of the filtered signal as an area.

The image processor may perform correction by multiplying the divided partial signals by using the intensity of the vertex as a weight.

The image processor may perform the correction by differentially weighting the divided partial signals.

The image processor may generate a B-scan signal by interpolating the reconstructed A-scan signal.

The reconstructed A-scan signal includes adjacent first and second reconstructed A-scan signals, and the image processor corresponds to two adjacent depth data and the two depth data of the first reconstructed A-scan signal. Interpolation may be performed using a center of gravity of two adjacent depth data of the second reconstructed A-scan signal.

According to another aspect of the present invention, there is provided an image processing method of an optical coherence tomography apparatus, the method comprising: generating an A-scan signal by detecting an interference signal from an interferometer, and partial signal grouping the A-scan signal into similar signal characteristics Dividing into; performing correction on the divided partial signal; and combining the corrected partial signal to generate a reconstructed A-scan signal.

The dividing step may include filtering the A-scan signal in a frequency domain and dividing the partial signal by using a frequency section between vertices of the filtered signal as an area.

The correcting step may include performing multiplication by multiplying the divided partial signals by using the intensity of the vertex as a weight.

The correcting step may include performing a correction by differentially weighting the divided partial signals.

Interpolating using the reconstructed A-scan signal, and generating a B-scan signal using the interpolated signal.

The reconstructed A-scan signal includes adjacent first and second reconstructed A-scan signals, and the interpolation step corresponds to two adjacent depth data and the two depth data of the first reconstructed A-scan signal. And performing interpolation using the center of gravity of two adjacent depth data of the second reconstructed A-scan signal.

According to another embodiment of the present invention, the light source unit for irradiating the light, the light is split into two optical paths, the interferometer causing the interference phenomenon by combining the split light again, and detects the interference signal from the interferometer An image processing apparatus of an optical coherence tomography apparatus including an image receiver for generating an A-scan signal includes: an area divider for dividing the A-scan signal into partial signals grouped by similar signal characteristics; And a reconstruction unit that combines the corrected partial signals to generate a reconstructed A-scan signal.

The area divider may filter the A-scan signal in the frequency domain and divide the partial signal using a frequency section between the peaks of the filtered signal as an area.

The correction unit may perform correction by multiplying the divided partial signals by using the intensity of the vertex as a weight.

The correction unit may differentially weight the divided partial signals to perform correction.

The interpolation unit may further include an interpolator configured to generate a B-scan signal by interpolating the reconstructed A-scan signal.

The reconstructed A-scan signal includes adjacent first and second reconstructed A-scan signals, and the interpolation unit corresponds to two adjacent depth data and the two depth data of the first reconstructed A-scan signal. Interpolation may be performed using the center of gravity of two adjacent depth data of the second reconstructed A-scan signal.

As described above, according to the optical coherence tomography apparatus and the image processing method thereof, the A-scan signal is reconstructed by segmenting the regions and performing correction on each of the divided regions. The image quality may be improved by performing interpolation on the reconstructed neighboring A-scan signals and generating a B-scan image.

In addition, according to the optical coherence tomography apparatus and the image processing method according to an embodiment of the present invention, the effect of speckle noise is effectively avoided without using a two-dimensional image processing technique or a statistical processing technique for reducing or eliminating the speckle noise itself. Can be reduced.

1 is a block diagram of an optical coherence tomography apparatus according to an embodiment of the present invention.
FIG. 2 is a block diagram of the image processor shown in FIG. 1.
3 is a schematic diagram illustrating reconstruction of an A-scan signal according to an embodiment of the present invention.
4 is a graph illustrating an input A-scan signal.
5 is a graph illustrating a signal for dividing an area of an A-scan signal according to an embodiment of the present invention.
6 is a graph illustrating a signal derived for dividing an area of an A-scan signal according to an embodiment of the present invention.
7 is a graph illustrating an A-scan signal reconstructed according to an embodiment of the present invention.
8 is a schematic diagram illustrating generating a B-scan image according to an embodiment of the present invention.
9 is a schematic diagram illustrating a method of interpolating a reconstructed A-scan signal according to an embodiment of the present invention.
10 is a schematic diagram illustrating a method of interpolating a reconstructed A-scan signal according to an embodiment of the present invention.
11 is a schematic diagram illustrating a method of interpolating a reconstructed A-scan signal according to an embodiment of the present invention.
12 is a flowchart illustrating an image processing method of an optical coherence tomography apparatus according to another exemplary embodiment of the present invention.

The terms or words used in this specification and claims are not to be construed as limiting in their usual or dictionary meanings, and the inventors may appropriately define the concept of terms in order to best explain their invention in the best way possible. It should be interpreted as meaning and concept corresponding to the technical idea of the present invention based on the principle that the present invention.

Therefore, the embodiments shown in the specification and the configuration shown in the drawings are only the most preferred embodiments of the present invention and do not represent all of the technical idea of the present invention, various equivalents that may be substituted for them at the time of the present application It should be understood that there may be variations and examples.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention.

FIG. 1 is a block diagram of an optical coherence tomography apparatus according to an embodiment of the present invention, and FIG. 2 is a block diagram of the image processor shown in FIG.

As shown in FIG. 1, the optical coherence tomography apparatus according to the exemplary embodiment may include a light source unit 110, an interferometer 120, an optical probe 130, an image receiver 140, and an image processor 150. It includes.

The light source unit 110 may include a collimating lens for converting the laser light source and the light emitted from the laser light source into parallel light. The light source unit 110 generates light and transmits the light to the interferometer 120.

The interferometer 120 is a measuring device using an interference phenomenon of light. The interferometer 120 generates an interference fringe using an interference action of two lights, and analyzes the interference fringe to generate an interference signal with respect to the object 10 located on one optical path. . That is, the interferometer 120 divides the light emitted from the same light source into two light paths, places the object 10 on one light path to make a difference in the two light paths, and then combines the light again to generate an interference phenomenon. Cause

The interferometer 120 may include a beam splitter 122 and a reference mirror 124. The light transmitted from the light source unit 110 is separated into the measurement light and the reference light in the beam splitter 122. Of the light split from the beam splitter 122, the measurement light is transmitted to the optical probe 130, the reference light is transmitted to the reference mirror 124, reflected, and then returned to the beam splitter 122.

On the other hand, the measurement light transmitted to the optical probe 130 is irradiated to the object 10 to be photographed inside the tomography image through the optical probe 130, the response light reflected by the irradiated measurement light from the object 10 Is transmitted to the beam splitter 122 of the interferometer 120 through the optical probe 130. The transmitted response light and the reference light reflected by the reference mirror 124 cause interference in the beam splitter 122. The optical probe 130 may be omitted or included in the interferometer 120.

The image receiver 140 generates an A-scan signal by detecting the interference signal from the interferometer 120. When the A-scan signal generated by the image receiver 140 is transmitted to the image processor 150, the image processor 150 processes the A-scan signal in an appropriate manner to express the object 10 in two or three dimensions. Generate a tomographic image of.

The A-scan signal generated by the image receiver 140 has depth information inside the object 10 until the measurement light is incident into the object 10 and reflected. As such, the depth information of the object 10 is included as information corresponding to one pixel, and a signal consisting of a 1-dimensional signal in the axial direction is called an A-scan signal.

When the A-scan signals are arranged side by side in the horizontal or vertical direction, they form a B-scan signal that forms a planar image in the depth direction, and when the B-scan signals are arranged in multiple layers, a three-dimensional stereoscopic image is obtained. .

The image processor 150 generates the B-scan signal by receiving the A-scan signal from the image receiver 140 and performing image processing on the A-scan signal.

Referring to FIG. 2, the image processing unit 150 receives an A-scan signal and divides an area of the A-scan signal, a region divider 151, a correction unit 153 that corrects each divided region, A reconstruction unit 155 that combines the corrected respective region signals to generate a reconstruction A-scan signal, and an interpolation unit 157 that interpolates neighboring reconstruction A-scan signals to each other to form a B-scan signal.

The image processor 150 may exist as an independent device outside the optical coherence tomography apparatus.

3 is a schematic diagram illustrating reconstructing an A-scan signal according to an embodiment of the present invention, and FIGS. 4 to 7 divide an area of the A-scan signal and reconstruct an A-scan signal according to an embodiment of the present invention. Is a graph illustrating the steps of generating.

First, the area dividing unit 151 receives an input A-scan signal A (i) for the i-th 1D depth information in the 3D stereoscopic image and divides the same into an area grouped by similar signal characteristics. Features include, for example, the size of data values, representative values, and average values. The degree of similarity may be set by setting a threshold value, and the segmentation may be increased or decreased according to the reference value. For example, if the reference value is high, the number of regions may be small. On the contrary, if the reference value is low, the number of regions may be large.

Since the A-scan signal can be represented as one-dimensional data in the depth direction, the meaning of the divided region can be regarded as dividing data having similar characteristics. Therefore, if the input A-scan signal is divided into M regions, the input A-scan signal may be divided into M partial signals R (1), R (2), ..., R (M).

As an example, the graph illustrated in FIG. 4 is a graph illustrating conversion of an input A-scan signal into a frequency domain. In order to extract the area of the input A-scan signal, a Gaussian window (shown in FIG. 5), such as Equation 1, is applied to the A-scan signal, as shown in FIG. It appears as a filtering signal of the scan signal.

[Equation 1]

Here,-(N-1) / 2 ≤ n ≤ (N-1) / 2, α is the inverse of the standard deviation, σ is a Gaussian random value, and N is the number of samples for filtering in the A-scan signal.

Extracting peak-to-peak regions by finding peaks in the filtered signal extracts R (1), R (2), ..., R (M) regions. For example, in FIG. 6, since the vertices are located at the frequencies 1200, 1700, 2100, and 2700, the area in this case is divided into five regions of frequency intervals 0 to 1200, 1200 to 1700, 1700 to 2100, 2100 to 2700, and 2700 to 4000. The partial signal corresponding to each region may be R (1), R (2), R (3), R (4), and R (5).

The correction unit 153 performs appropriate correction on each of the partial signals R (1), R (2), ..., R (M) in which the region is divided, and corrected signals P (1), P. (2), ..., P (M)]. As an example, the correction unit 153 differentially weights each region. In particular, by differentially weighting, the signal values in the areas adjacent to each other are significantly different. By doing so, the contrast between the two adjacent regions is further increased to make the boundary of the material inside the object 10 clearer, thereby facilitating the analysis of the object 10.

For example, if the representative value or the average value of the signals of the two regions were 10 and 20, respectively, the image may be more clearly seen by correcting the signals of the respective regions so that the difference is large, for example, 5 and 30. Referring to the graph of FIG. 6, multiplying each partial signal by the weight of the intensity of the vertices of the partial signals R (1), R (2), R (3), R (4), and R (5) is performed as a weight. The same function can be performed.

The reconstruction unit 155 combines the corrected signals P (1), P (2), ..., P (M) into one to generate a reconstructed A-scan signal AR (i). In the example of FIG. 6, the partial signals R (1), R (2), R (3), R (4), R (5) and the input A-scan signal of FIG. 1 are convolved, and the partial signal R (1) Multiplying the intensities of the vertices of R (2), R (3), R (4), and R (5), respectively, yields a signal of the form shown in FIG. It becomes a signal.

8 is a schematic diagram illustrating generating a B-scan image according to an embodiment of the present invention, and FIGS. 9 to 11 are schematic diagrams illustrating a method of interpolating a reconstructed A-scan signal according to an embodiment of the present invention. to be.

Referring to FIG. 8, the A-scan signals AR (1), AR (2), .. AR (k), .., and AR (m) reconstructed by the reconstruction unit 155 are A-line in one row. Scan signals or a row of A-scan signals. The interpolator 157 interpolates two adjacent reconstructed A-scan signals to generate interpolation signals I (1), I (2), ..., I (m-1) and arranges them side by side to form one B- Generate a scan signal.

Referring to FIG. 9, one interpolation signal I (k) is generated by interpolation using adjacent reconstructed A-scan signals AR (k) and AR (k + 1), and adjacent ones of depth data of AR (k) are interpolated. Depth data of the interpolation signal I (k) is generated by obtaining the center of gravity of two adjacent values among two values and depth data of AR (k + 1). That is, I (k) _L is calculated as the center of gravity of four data values: AR (k) _L, AR (k) _L + 1, AR (k + 1) _L, and AR (k + 1) _L + 1. Here, the subscript L is a variable for representing the L-th depth data of the A-scan signal, and AR (k) _L represents the L-th depth data value of the A-scan signal AR (k).

On the other hand, when the AR (k) _L value is smaller than the AR (k + 1) _L value, as shown in FIG. 10, the AR (k) _L + 1 value is also smaller than the AR (k + 1) _L + 1 value. There may be a situation in which the AR (k) _L + 1 value becomes larger than the AR (k + 1) _L + 1 value as shown in FIG. 11. Although the situation of FIG. 10 is general, in the process of reconstructing the A-scan signal, data may be changed drastically as shown in the situation of FIG. 11 by performing correction by changing the weight for each region, which is immediately reflected in the B-scan signal. This can degrade image quality. However, as shown in FIG. 11, when the interpolation value is calculated using the center of gravity of the four data values, the interpolation value itself does not change rapidly even if the data changes abruptly. Therefore, as in the embodiment of the present invention, by interpolating adjacent reconstructed A-scan signals and adjacent depth data by using the center of gravity of four data, it is possible to reduce the possibility of being incorrectly corrected in the boundary area and adversely affecting image quality. have.

As described above, according to the optical coherence tomography apparatus according to the exemplary embodiment of the present invention, the A-scan signal is divided into regions and corrections are performed to differentially weight each of the divided regions, thereby performing contrast between two adjacent regions. The higher the, the clearer the boundaries of the material. This method can effectively reduce the effects of speckle noise without using two-dimensional image processing or statistical processing techniques to reduce or eliminate the speckle noise itself.

Next, an image processing method of an optical coherence tomography apparatus according to another exemplary embodiment will be described with reference to FIG. 12. 12 is a flowchart illustrating an image processing method of an optical coherence tomography apparatus according to another exemplary embodiment of the present invention. In the present embodiment, for convenience of description, the image processing unit 150 of the optical coherence tomography apparatus shown in FIG. 1 is intended. All of the above descriptions also apply to the present embodiment. It will be omitted.

First, the image processor 150 receives an A-scan signal from the image receiver 140 (S110).

The image processor 150 generates a partial signal obtained by dividing the received A-scan signal into regions having similar characteristics (S120).

The image processor 150 corrects the divided partial signals (S130).

The image processor 150 reconstructs the A-scan signal by combining the corrected partial signals into one (S140).

The image processor 150 generates an interpolation signal using the reconstructed A-scan signals (S150). Interpolation is performed by calculating the center of gravity of four data between adjacent reconstructed A-scan signals and between adjacent depth data.

The image processor 150 generates one B-scan signal by arranging the generated interpolation signals side by side (S160).

The foregoing detailed description illustrates the present invention. In addition, the foregoing description merely shows and describes preferred embodiments of the present invention, and the present invention can be used in various other combinations, modifications, and environments. That is, changes or modifications may be made within the scope of the concept of the invention disclosed in this specification, the scope equivalent to the disclosed contents, and / or the skill or knowledge in the art. The above-described embodiments are for explaining the best state in carrying out the present invention, the use of other inventions such as the present invention in other state known in the art, and the specific fields of application and uses of the present invention. Various changes are also possible. Thus, the detailed description of the invention is not intended to limit the invention to the disclosed embodiments. Also, the appended claims should be construed to include other embodiments.

110: light source unit,
120: interferometer,
130: optical probe,
140: video receiving unit,
150: the image processing unit,
151: region divider,
153: correction unit,
155: reconstruction unit,
157: interpolator

Claims (18)

delete A light source unit for irradiating light,
An interferometer for splitting the light into two optical paths and recombining the branched light to cause an interference phenomenon,
An image receiver for detecting an interference signal from the interferometer and generating an A-scan signal; and
The image processor divides the A-scan signal into partial signals grouped with similar signal characteristics, performs correction on the divided partial signals, and combines the corrected partial signals to generate a reconstructed A-scan signal.
Including;
And the image processing unit filters the A-scan signal in a frequency domain and divides the partial signal using a frequency section between vertices of the filtered signal as an area. In claim 2,
And the image processor performs a correction by multiplying the divided partial signals by using the intensity of the vertex as a weight. In claim 2,
And the image processing unit differentially weights the divided partial signal to perform correction. A light source unit for irradiating light,
An interferometer for splitting the light into two optical paths and recombining the branched light to cause an interference phenomenon,
An image receiver for detecting an interference signal from the interferometer and generating an A-scan signal; and
The image processor divides the A-scan signal into partial signals grouped with similar signal characteristics, performs correction on the divided partial signals, and combines the corrected partial signals to generate a reconstructed A-scan signal.
Including;
And the image processor generates a B-scan signal by interpolating the reconstructed A-scan signal. In claim 5,
The reconstructed A-scan signal includes adjacent first and second reconstructed A-scan signals, and the image processor corresponds to two adjacent depth data and the two depth data of the first reconstructed A-scan signal. The optical coherence tomography apparatus performs interpolation using the center of gravity of two adjacent depth data of the second reconstructed A-scan signal. delete As an image processing method of an optical coherence tomography apparatus,
Detecting an interference signal from the interferometer and generating an A-scan signal,
Dividing the A-scan signal into partial signals grouped by similar signal characteristics;
Performing correction on the divided partial signal, and
Combining the corrected partial signals to generate a reconstructed A-scan signal
Including;
The dividing step includes performing filtering on the A-scan signal in a frequency domain and dividing the partial signal by using a frequency section between vertices of the filtered signal as an area. In claim 8,
The correcting step includes multiplying the divided partial signals by using the intensity of the vertex as a weight to perform correction. In claim 8,
The correcting step includes performing a correction by differentially weighting the divided partial signals. As an image processing method of an optical coherence tomography apparatus,
Detecting an interference signal from the interferometer and generating an A-scan signal,
Dividing the A-scan signal into partial signals grouped by similar signal characteristics;
Performing correction on the divided partial signals;
Combining the corrected partial signals to produce a reconstructed A-scan signal;
Interpolating using the reconstructed A-scan signal, and
Generating a B-scan signal using the interpolated signal
Image processing method comprising a. In claim 11,
The reconstructed A-scan signal includes adjacent first and second reconstructed A-scan signals, and the interpolation step corresponds to two adjacent depth data and the two depth data of the first reconstructed A-scan signal. And performing interpolation using a center of gravity of two adjacent depth data of the second reconstructed A-scan signal. delete A light source unit for radiating light, an interferometer for splitting the light into two optical paths, an interferometer causing interference by recombining the split light, and an image receiving unit generating an A-scan signal by detecting an interference signal from the interferometer An image processing apparatus of an optical coherence tomography apparatus, comprising:
An area divider for dividing the A-scan signal into partial signals grouped by similar signal characteristics;
A correction unit for performing correction on the divided partial signal, and
A reconstruction unit for generating the reconstructed A-scan signal by combining the corrected partial signals
Including;
And the area divider filters the A-scan signal in a frequency domain and divides the partial signal using a frequency section between the peaks of the filtered signal as an area. The method of claim 14,
And the correcting unit to multiply the divided partial signals by using the intensity of the vertex as a weight to perform correction. The method of claim 14,
And the correcting unit differentially weights the divided partial signals to perform correction. A light source unit for radiating light, an interferometer for splitting the light into two optical paths, an interferometer causing interference by recombining the split light, and an image receiving unit generating an A-scan signal by detecting an interference signal from the interferometer An image processing apparatus of an optical coherence tomography apparatus, comprising:
An area divider for dividing the A-scan signal into partial signals grouped by similar signal characteristics;
A correction unit for performing correction on the divided partial signals;
A reconstruction unit which combines the corrected partial signals to generate a reconstructed A-scan signal, and
And an interpolator configured to generate a B-scan signal by interpolating the reconstructed A-scan signal. The method of claim 17,
The reconstructed A-scan signal includes adjacent first and second reconstructed A-scan signals, and the interpolation unit corresponds to two adjacent depth data and the two depth data of the first reconstructed A-scan signal. And interpolation using a center of gravity of two adjacent depth data of the second reconstructed A-scan signal.

Images