CN101513366B - PS-OCT-based three dimension visualized imaging method - Google Patents

Abstract

A PS-OCT-based three-dimensional visualization oral caries detection device and detection method. Including: the polarized light beam is irradiated on the measured object through the self-focusing lens, the reflected light carrying the measured object information is returned from the original optical path, and enters the two arms of the interferometer together with the reflected light from the exit end face of the self-focusing lens, and passes through the Faraday rotator After that, interference occurs at the return coupler; adjust the fiber stretcher to realize scanning in the longitudinal Z direction, and at the same time scan the measured object in the horizontal X direction through the displacement platform to obtain two-dimensional information of a certain cross section; adjust the displacement platform so that It scans along the Y direction perpendicular to the cross section to obtain multiple two-dimensional images representing different parts of the sample; the software analyzes and processes the data to obtain the structure and birefringence information of the measured object, and reconstructs the The three-dimensional image of the measured object displays the results. The invention can realize high-precision and radiation-free detection of teeth, and facilitates the diagnosis of dental caries.

Description

Three-dimensional visualization imaging method based on PS-OCT

【Technical field】：

The invention belongs to the technical field of detection of oral caries diseases, and relates to PS-OCT (polarization-sensitive optical coherence tomography) technology and a method and device for three-dimensional reconstruction of oral caries structures. the

【Background technique】:

OCT (Optical Coherence Tomography) technology has developed rapidly in recent years. It is another new biological tissue imaging technology following ultrasound, X-ray computed tomography (XCT), and magnetic resonance imaging (MRI). It integrates confocal microscopy technology and ultra-sensitive detection technology, and combines automatic control and computer image processing technology to realize non-destructive testing of living organisms and obtain high-resolution cross-sectional images of the internal microstructure of biological tissues. the

Caries is a common and frequently-occurring disease of the oral cavity worldwide, and it is the main cause of oral pain and tooth loss. According to the latest statistics from the World Health Organization, dental caries has become the third largest human disease after cancer and cardiovascular and cerebrovascular diseases. In recent years, although the oral health status of the whole people in my country has been gradually improved, the prevalence of dental caries among people of all ages is still high, which poses a great threat to people's life and health, so the control and intervention of dental caries has become an urgent task . Bacteria in the mouth convert food into acids, which demineralize the surface and interior of the tooth, forming pores or decay, resulting in dental caries. the

The main methods of current oral cavity caries diagnosis are doctors' manual diagnosis and X-ray imaging. Since early dental caries is caused by demineralization inside the teeth, there are no obvious signs on the surface of the teeth. Doctors' manual diagnosis cannot diagnose early dental caries changes. However, although X-ray imaging technology can find the lesions inside the teeth, its resolution is limited. At the same time, it has the disadvantages of radiation pollution and inaccurate positioning, which seriously limits the accuracy of early caries diagnosis, and also causes harm to patients and doctors. the

【Invention content】：

The purpose of the present invention is to solve the shortcomings of the above oral caries diagnosis method, and provide a three-dimensional visual oral caries detection device and detection method based on PS-OCT. the

The PS-OCT technology provided by the present invention can obtain the microstructure and polarization characteristics inside the tooth, and at the same time, the three-dimensional structure image of the tooth can be obtained through the three-dimensional visualization technology, so as to diagnose dental caries. the

The three-dimensional visual oral caries detection device based on PS-OCT provided by the present invention includes:

Light source: used to irradiate the measured object to generate a measurement light signal with the measured object information;

Polarization state generator: used to adjust the light emitted by the light source to a specific polarization state according to the needs (adjust according to the polarization properties of the optical fiber, device and sample to make the output signal clear);

Optical fiber circulator: Port 1 is used to receive light from the light source and output from port 2;

Self-focusing lens: connected to port 2 of the fiber optic circulator, and its exit end face has a certain reflectivity, which is used to irradiate the light output from port 2 of the fiber optic circulator on the measured object and directly reflect a part of the light back to the fiber optic circulator The 2-port, and 3-port outputs enter the 1-port of the fiber coupler. At the same time, the reflected light carrying the measured sample information is received and returned to port 2 of the fiber optic circulator and output from port 3 to enter port 1 of the fiber coupler;

Displacement platform: composed of a two-dimensional stepping motor and a sample platform, it is used to fix the measured object, and at the same time, it can drive the measured object to perform a horizontal two-dimensional displacement movement to realize two-dimensional movement of the horizontal X-axis and Y-axis of the measured sample. dimension scan;

Optical fiber coupler: used to divide the light input from port 1 into two beams to enter port 3 and port 4 respectively, and enter the two arms of the interferometer, while receiving the light returned from the two arms of the interferometer and interfering at the coupler, And emit the interference signal from port 2;

Adjustable fiber delayer: receiving the light beam output from the port 3 of the fiber coupler, used to adjust the optical path in the optical path;

Compensation optical fiber: connected behind the adjustable optical fiber delayer, used to match the length of the optical fiber of the two arms of the interferometer;

Fiber stretcher: Receive the beam output from the 4 ports of the fiber coupler, used to quickly and accurately adjust the optical path in the optical path, and realize the scanning of the longitudinal Z direction of the measured object;

Two Faraday rotation mirrors: respectively installed behind the compensation fiber and the adjustable fiber delayer, used to reflect the two beams of light emitted by the fiber coupler back to the coupler to cause interference, and at the same time eliminate the optical fiber through the Faraday rotation effect The impact of polarization noise on system performance;

Polarization beam splitter: connected to the 2 ports of the fiber coupler, used to split the interference signal light into two beams of linearly polarized light whose polarization states are perpendicular to each other;

Two photodetectors: respectively used to receive the linearly polarized light in the horizontal and vertical directions output by the polarizing beam splitter, and convert the light intensity signal into a voltage signal and input it to the acquisition card;

Data acquisition card: used to collect the voltage signal converted by two photodetectors, send it to the computer for processing, and control the movement of the displacement platform through its I/O port;

Computer: Provide control signals for the fiber stretcher, polarization generator, and displacement platform, and at the same time process the output signal of the data acquisition card to obtain a two-dimensional tomographic image, use software for three-dimensional reconstruction, and output the signal . the

A three-dimensional visualization oral caries detection method based on PS-OCT, the method comprises in turn:

1. Place the tooth sample of the tested object on the displacement platform;

2. Turn on the light source and adjust the polarization state generator so that the light emitted by the light source can be adjusted to a specific polarization state according to the needs (according to the polarization properties of the optical fiber, device and the sample to be tested, so that the output signal obtained is clear);

3. Let the light beam emitted by the light source pass through the self-focusing lens to generate a reflected light and a transmitted light, so that the transmitted light is irradiated on the measured object;

4. The reflected light carrying the information of the measured object is returned from the original optical path and combined with the reflected light from the exit end face of the self-focusing lens. After entering port 1 of the fiber coupler, it enters the two arms of the interferometer respectively and passes through two Faraday rotators. Interference occurs at the rear return coupler;

The interference signals obtained in steps 5 and 4 are divided into two optical signals of horizontal polarization and vertical polarization after passing through the polarization beam splitter, and enter two photodetectors respectively;

6. Adjust the optical fiber stretcher to scan the measured object in the longitudinal Z direction, and at the same time scan the measured object horizontally in the X direction through the displacement platform to obtain two-dimensional information corresponding to a certain cross-section of the measured object;

7. Adjust the displacement platform so that it travels along the Y direction perpendicular to the cross-section in step 6, repeat steps 2 to 6, and collect two-dimensional information of different cross-sections of the measured object;

8. Send the light intensity signals obtained by the two photodetectors to the computer from the acquisition card, analyze and process the data, and obtain the structure and birefringence information of the measured object;

Ninth, reconstruct the three-dimensional image of the tooth sample of the measured object, and display the result, the method includes in turn:

9.1. Since the two-dimensional grayscale image corresponds to the cross-sectional information of a certain part of the sample, the displacement platform scans the sample in the Y direction perpendicular to the cross-section, and multiple two-dimensional images representing different parts of the sample can be obtained. These multiple two-dimensional images are the original image data required for three-dimensional reconstruction. the

Section 9.2. Preprocessing the image. First, the median filter is used to process the image to remove the salt and pepper noise in the background. In order to make the gray value of the target more prominent, histogram equalization is used to adjust the contrast of the image. Then set the threshold to filter out most of the background noise. Then use the sobel operator to extract the contour of the image, so that the extracted contour contains a lot of unnecessary internal information of the tissue, which needs to be removed. Connecting the contour lines through expansion can be repeated, but in order to ensure the same size, it must be corroded the corresponding number of times. Finally, the eight-connected domain is used for processing to extract the outline of the tissue. Turn the background outside the outline to black, and keep the gray value of the original image inside. the

Section 9.3, resampling, a ray is sent from each pixel on the screen according to the set viewing direction, this ray passes through the three-dimensional data field, K equidistant sampling points are selected along this ray, and are determined by the distance The function value of the nearest 8 data points of a sampling point is averaged to obtain the function value of the sampling point. the

Section 9.4. Classify the data. According to the different data values, divide it into several categories and assign different gray value C and opacity value α to each type of data to generate an image with a transparent effect and reflect the three-dimensional data field. internal structure. the

Section 9.5, Image Synthesis. Combining the gray value and opacity value of each sampling point on each ray from front to back, the gray value at the pixel point where the ray is emitted can be obtained. the

For the convenience of observation, the function of rotation and section is added. The three-dimensional image can be rotated around any axis with the mouse, and the cutting function of any surface can be realized by using the mouse, so that any plane in the three-dimensional space can be observed. the

Advantages and positive effects of the present invention:

The present invention proposes a three-dimensional visualization method for dental caries detection based on all-fiber PS-OCT, and a detection device for realizing the method, which can realize high-precision and radiation-free detection of teeth, and its lateral resolution and axial resolution can be achieved. up to 15um. In this method, a polarization generator is added in the light path, and the structure and birefringence information of the sample can be obtained simultaneously, which facilitates the diagnosis of caries. The optical fiber stretcher is used to scan the sample longitudinally, and at the same time, the displacement platform drives the sample to perform a two-dimensional lateral displacement movement to realize the three-dimensional scanning of the tooth sample. The software is used to reconstruct the teeth in three dimensions, which facilitates the intuitive detection of dental caries. the

[Description of drawings]:

Figure 1 is a schematic diagram of the PS-OCT three-dimensional visualization oral caries detection device;

Figure 2 is a block diagram of the overall structure of the PS-OCT three-dimensional visualization oral caries detection device;

Figure 3 is a schematic diagram of the three-dimensional scanning process of the PS-OCT three-dimensional visualization oral caries detection device;

Fig. 4 is the flowchart of the three-dimensional image reconstruction method in the PS-OCT three-dimensional visualization oral caries detection method;

Fig. 5 is a flow chart of the preprocessing part of the three-dimensional image reconstruction method in the PS-OCT three-dimensional visualization oral caries detection method;

Fig. 6 is a schematic diagram of the image synthesis part of the three-dimensional image reconstruction method in the PS-OCT three-dimensional visualization oral caries detection method;

Fig. 7 is a three-dimensional image of teeth obtained by using a PS-OCT three-dimensional visualized dental caries detection device. the

In the figure, 101 light source, 102 polarization state generator, 103 fiber optic circulator, 104 self-focusing lens, 105 displacement platform, 106 and 108 photodetectors, 107 polarization beam splitter, 109 acquisition card, 110 fiber optic coupler, 111 can Adjustable fiber retarder, 112 compensation fiber, 113 Faraday rotator, 114 fiber stretcher, 115 computer. the

【Detailed ways】:

Embodiment 1, detection device

As shown in Figure 1, the present invention provides a three-dimensional visualization oral caries detection device based on PS-OCT including:

——Light source 101: used to irradiate the measured object and generate a measurement optical signal with the measured object information;

——Polarization state generator 102: used to adjust the light emitted by the light source to a specific polarization state as required (adjust according to the polarization properties of the optical fiber, device and sample to be measured, so that the output signal obtained is clear);

——Fiber optic circulator 103: port 1 is used to receive the light emitted by the light source and output it from port 2;

——Self-focusing lens 104: connected to port 2 of the optical fiber circulator, and its exit end face has a certain reflectivity, which is used to irradiate the light output from port 2 of the optical fiber circulator on the measured object and directly reflect a part of the light back Port 2 of the fiber optic circulator outputs from port 3 and enters port 1 of the fiber coupler. At the same time, the reflected light carrying the measured sample information is received and returned to port 2 of the fiber optic circulator and output from port 3 to enter port 1 of the fiber coupler;

——Displacement platform 105: composed of a two-dimensional stepping motor and a sample platform, it is used to fix the object to be measured, and at the same time, it can drive the object to perform a two-dimensional displacement movement laterally, so as to realize the horizontal X-axis and Y-axis of the sample to be measured two-dimensional scanning. The embodiment of the displacement platform adopts the electric control displacement platform commonly used in the market at present;

——Fiber optic coupler 110: it is used to divide the light input from port 1 into two beams to enter port 3 and port 4 respectively, and enter the two arms of the interferometer, and receive the light returned from the two arms of the interferometer at the coupler Interference occurs, and the interference signal is emitted from port 2;

——Adjustable optical fiber delayer 111: receives the light beam output by the port 3 of the optical fiber coupler, and is used to adjust the optical path in the optical path;

——Compensating optical fiber 112: connected to the back of the optical fiber stretcher, used to match the length of the optical fiber of the two arms of the interferometer;

——Fiber stretcher 114: Receives the light beam output from the port 4 of the fiber coupler, and is used to quickly and accurately adjust the optical path in the optical path to realize the scanning of the longitudinal Z direction of the measured object;

——Two Faraday rotator mirrors 113: respectively installed behind the compensation fiber and the adjustable fiber delayer, used to reflect the two beams of light emitted by the fiber coupler back to the coupler to cause interference, and at the same time through the Faraday rotation optical effect, Eliminate the influence of polarization noise in the fiber on the system;

——Polarization beam splitter 107: connected to the 2 ports of the fiber coupler, used to split the interference signal light wave into two beams of linearly polarized light whose polarization states are perpendicular to each other;

——Two photodetectors 106 and 108: respectively used to receive the linearly polarized light in the horizontal and vertical directions output by the polarization beam splitter, and convert the light intensity signal into a voltage signal and input it to the acquisition card;

——Data acquisition card 109: used to collect the voltage signal converted by the two photodetectors, send it to the computer for processing, and control the movement of the displacement platform through its I/O port;

——Computer 115: provides control signals for the fiber stretcher, polarization generator, and displacement platform, and is used to process the output signal of the data acquisition card at the same time to obtain a two-dimensional tomographic image, and use software for three-dimensional reconstruction, output the signal. the

Embodiment 2, detection method

As shown in Figure 2, the overall structure of the PS-OCT 3D visualization platform mainly includes two parts: hardware and software. the

The three-dimensional visual oral caries detection method based on PS-OCT includes:

1. Place the tooth sample of the tested object on the displacement platform;

2nd, turn on the light source 101, adjust the polarization generator 102, the light that the light source 101 sends is adjusted to a specific polarization state as required (adjust according to the polarization properties of the optical fiber, the device and the sample to be tested, so that the output signal obtained is clear) ;

3. Let the light beam emitted by the light source pass through the self-focusing lens 104 to generate a reflected light and a transmitted light, so that the transmitted light is irradiated on the measured object;

4. The reflected light carrying the information of the measured object is returned from the original optical path and combined with the reflected light from the exit end face of the self-focusing lens 104. After entering port 1 of the fiber coupler 110, it enters the two arms of the interferometer respectively and passes through two Faradays. Interference occurs at the return coupler after the optical rotation mirror 113;

The interference signal obtained in the 5th and 4th steps is divided into two optical signals of horizontal polarization and vertical polarization by the polarization beam splitter 107, and enters two photodetectors 106 and 108 respectively;

6. As shown in Figure 3, adjust the optical fiber stretcher 114 to scan the measured object in the longitudinal Z direction, and simultaneously scan the measured object in the X direction through the displacement platform 105 to obtain a certain value corresponding to the measured object. Two-dimensional information of a cross section;

The 7th, adjust displacement platform 105 to make it advance along the Y direction perpendicular to the cross-section in step 6, repeat steps 2 to 6, and collect two-dimensional information of different cross-sections of the measured object;

The 8th, the light intensity signal that two photodetectors 106 and 108 acquires is sent into computer by acquisition card, data is analyzed and processed, obtains the structure and birefringence information of measured object;

Ninth, call the software to reconstruct the three-dimensional image of the tooth sample of the measured object, and display the result. the

The process flow of the 3D reconstruction software is shown in Figure 4. First, since the 2D grayscale image corresponds to the cross-sectional information of a certain part of the sample, the displacement platform scans the sample in the Y direction perpendicular to the cross-section, and the representative sample can be obtained. Multiple 2D images of different parts. These multiple two-dimensional images are the original image data required for three-dimensional reconstruction. The software algorithm assumes that the three-dimensional data f(i, j, k) are distributed on the grid points of a uniform grid or a regular grid. Since the background gray value in the original two-dimensional image data is at the same level as the gray value of the dentin inside the tissue, this will have a great impact on the reconstruction results, so the background gray value should be changed to All black, which is 0. In order to achieve this purpose, it is only necessary to extract the outline of the tissue. The process of this image preprocessing is described in detail below. the

The preprocessing process is shown in Figure 5. First of all, because there is a lot of salt and pepper noise in the background, the median filter is used for processing. The median filter provides excellent denoising ability for a certain type of random noise, and is significantly less blurred than a small-sized linear smoothing filter. It is very effective for dealing with impulse noise, because this noise is superimposed with black and white points on the image. In the next step, in order to make the gray value of the target more prominent, histogram equalization is used to adjust the contrast of the image. Then set the threshold to filter out most of the background noise. Then use the sobel operator to extract the contour of the image. The contour extracted in this way contains a lot of unnecessary internal information of the tissue, and the contour line has disconnected places, which need to be further processed. The contour lines can be connected by repeating the dilation process, but in order to ensure the same size, the corresponding number of times must be corroded. Finally, the eight-connected domain is used for processing to remove the internal information, that is, the outline of the tissue can be extracted. The eight-connected domain can find the largest of all contour lines and record the position. After the outline of the organization is obtained, the background outside the outline is changed to black, and the original gray value is retained inside. This eliminates the effect of background noise. the

Then resampling, a ray is sent from each pixel on the screen according to the set viewing direction, this ray passes through the three-dimensional data field, K equidistant sampling points are selected along this ray, and are determined by a certain distance The function value of the nearest 8 data points of a sampling point is averaged to obtain the function value of the sampling point. Before resampling, it is necessary to convert the three-dimensional data field with function values from object coordinates to corresponding image coordinates. the

The next step is to classify the data, the purpose of which is to correctly divide it into several categories according to the different function values and assign different gray value C and opacity value α to each type of data, where α=1 represents the object It is completely opaque, and α=0 means it is completely transparent, in order to correctly represent the different distributions of multiple substances or the different properties of a single substance, generate an image with a transparent effect, and reflect the internal structure of the three-dimensional data field. At present, this algorithm is used to image the crown part, which is composed of two parts: enamel and dentin. After repeated verification, we use a gray value C of 230 and an opacity α of 0.03 to represent the enamel part. The gray value C is 70 The opacity alpha is 0.1 to represent the dentin part to get a clear image. the

The last step of this algorithm is image synthesis, that is, the gray value and opacity value of each sampling point on each ray are synthesized from front to back, and the gray value at the pixel point where the ray is emitted can be obtained. The front-to-back image synthesis algorithm is to combine the gray value and opacity of various sampling points along the ray from front to back to get the final image. As shown in Figure 5, assume that the gray value of the i-th sampling point is C _now and the opacity is α _now , and the gray value of the i-th sampling point is C _in and the opacity is α _in . The gray value after sampling points is C _out and the opacity is α _out , then:

C _out α _out ＝C _in α _in +C _now α _now (1-α _in ), α _out ＝α _in +α _now (1-α _in )

Combined with Figure 6 and Table 1, the image synthesis process is described as follows:

Assume that the grayscale value of the first sampling point is 70 and the opacity is 0.1. Since it is the first sampling point, the emitted grayscale value and opacity are the same as their own values. Next, calculate the second sampling point. The gray value and opacity of the first sampling point are used as the gray value and opacity of the second sampling point, which are 70 and 0.1 respectively. The second sampling point The gray value of itself is 230, and the opacity is 0.03. From the above formula, it can be obtained that the outgoing gray value is 104.016, and the opacity is 0.127, which is used as the incident gray value of the third sampling point and Opacity is calculated. By analogy, the gray value and opacity of each subsequent sampling point can be obtained. the

Table 1

Sampling point number i 1 2 3 4 The gray value C _now of the i-th sampling point 70 230 230 70 The opacity α _now of the i-th sampling point 0.1 0.03 0.03 0.1 Enter the gray value C _in of the i-th sampling point 0 70 104.016 125.711 The opacity α _in of the i-th sampling point 0 0.1 0.127 0.153 The gray value C _out after the i-th sampling point 70 104.016 125.711 105.726 Opacity α _out after the i-th sampling point 0.1 0.127 0.153 0.238

The front-to-back image synthesis is based on the light absorption and scattering model. The main advantages are that in the process of image synthesis from front to back, the opacity α must gradually increase, indicating that the image of the pixel is close to being completely opaque. The following sampling points will not contribute to the image of the pixel, so they can no longer be calculated, which can improve the speed.

In addition, for the convenience of observation, the function of rotation and section is added. The three-dimensional image can be rotated around any axis with the mouse, and the cutting function of any surface can be realized by using the mouse, so that any plane in the three-dimensional space can be observed. The tooth image reconstructed using the software is shown in Figure 7. the

Claims (1)

1. three-dimensional visualization formation method based on PS-OCT, this method comprises successively:

1st, measured object tooth sample is placed on the displacement platform;

2nd, open light source, regulate the polarization state generator, the light that light source is sent according to the polarization property of optical fiber, device and sample is adjusted to a specific polarization state as required, makes the output signal that obtains clear;

3rd, the light beam that makes light source send passes through GRIN Lens, produces a reflected light and a transmitted light, and transmitted light is radiated on the measured object;

4th, the reflected light that carries measured object information is returned by original optical path, synthetic a branch of with the reflected light of GRIN Lens outgoing end face, enter 1 port of fiber coupler, the light that fiber coupler is imported 1 port is divided into two bundles and enters 3 ports and 4 ports respectively, and enter two arms of interferometer, interfere through two Faraday mirror reflection back into optical fibers coupling mechanism places;

5th, the interference signal of the 4th step acquisition is divided into horizontal polarization and vertical polarization two-beam signal behind polarization beam apparatus, enters two photodetectors respectively;

6th, regulate at 4 ports of fiber coupler and the optical fiber stretcher between the Faraday mirror, measured object is carried out the scanning of vertical Z direction, by displacement platform measured object is carried out the transversal scanning of directions X simultaneously, obtain two-dimensional signal corresponding to a certain transversal section of measured object;

7th, regulate displacement platform it is advanced along the Y direction vertical with transversal section in the 6th step, repeated for the 2nd to 6 step, gather the two-dimensional signal of the different transversal section of measured object;

8th, the light intensity signal that two photodetectors are obtained is sent into computing machine by capture card, and data are analyzed and handled, and draws the structure and the birefringence information of measured object;

9th, rebuild the 3-D view of measured object tooth sample, and the result shown that it comprises:

9.1st, because the transversal section information at a certain position of two dimensional gray image counter sample, by displacement platform sample is scanned in the direction Y direction perpendicular to the transversal section, can obtain the two-dimensional images of representative sample different parts, described two-dimensional images is exactly the required raw image data of three-dimensional reconstruction;

9.2nd, image is carried out pre-service, at first selects for use median filter that image is handled, remove the salt-pepper noise in the background, for the gray-scale value that makes target more outstanding, the contrast that adopts histogram equalization to regulate image; Come the ground unrest of the filtering overwhelming majority again by setting threshold; Use the profile of sobel operator extraction image then, handle that by reexpansion outline line is coupled together, and corrode corresponding number of times; Handle with eight connected domains more at last, just can extract the outline line of tissue, the background of outline line outside is become black, the inside keeps the gray-scale value of original image;

9.3rd, resample, each pixel from screen sends a ray according to the direction of observation of setting, this ray passes 3 d data field, select K equidistant sampled point along this ray, and ask on average by the functional value of nearest 8 data points of a certain sampled point of distance, promptly draw the functional value of this sampled point;

9.4th, data are classified,, it is divided into some classes and gives different gray-scale value C and opacity value α for each class data, have the image of transparent effect, the inner structure of reflection 3 d data field with generation according to the data value difference;

9.5th, image is synthetic, and the gray-scale value of each sampled point on every ray and opacity value are synthesized after by forward direction, promptly can obtain sending the gray-scale value at the pixel place of this ray, and wherein synthetic method is according to following formula:

C _outα _out＝C _inα _in+C _nowα _now(1-α _in)，α _out＝α _in+α _now(1-α _in)

Wherein, the gray-scale value of i sampled point is C _Now, opacity is α _Now, the gray-scale value that enters i sampled point is C _In, opacity is α _In, be C through the gray-scale value behind i the sampled point _Out, opacity is α _Out

9.6th, observe for convenience, the function that has increased rotation and dissectd can realize that with mouse 3-D view rotates around arbitrary axis, uses the dissect function of mouse realization to arbitrary face simultaneously, can observe the arbitrary plane in the three dimensions.

Images