Open Access
Add to BibSonomyAbstract
This paper presents a quick monitor method to measure micro height variations directly. Here we apply optical moiré technology with a program designed by ourselves to Traditional Chinese Medical Pulse diagnosis. We analyze the moiré pattern which records the information of pulse, and then examine the conditions of pulse, the conditions of pulse including the location of pulse (by image processing the moiré pattern), the rhythm of pulse (via the frequency), the shape of pulse (via moiré pattern) and the strength of pulse (via amplitude intensity). Therefore, we can quantify the conditions of pulse by the system.
© 2015 Optical Society of America
1. Introduction
A fast and non-destructive optical measurement method is always one of the most efficient and executable ways of measurement no matter it is a long distance ranging or a short distance scale measurement on the surface of an object. Given the rapid advancement of speed and calculating capacity of computers, many fast and automatized detection systems can be developed by combining optical measurement systems and digital image processing technique.
In recent times, contactless optical measurement has been widely researched and applied to short distance scale measurement on the surface of an object because of its advantages. No damage would be made on the surface of objects and no mechanics feedback would affect the probe to lower the precision because the probe will not have direct contact with the surface. Furthermore, with the digital image processing technique, it is fast, automatized and easy-to-operate; also, it can capture thorough and consecutive data to see the changes.
There are some low-cost systems using optical surface measurement techniques. The first system is to use laser ranging to scan the surface point by point. Because the light spots are at different height level of the surface, the relative position can be calculated by triangulation. The resolution of the measurement is usually restricted by geometry calculation and CCD pixels and it is mostly overcome by optical centroid; only that full-field measurement is difficult to be conducted due to the mechanism limit to scan the whole surface [1–5]. The second system is projected fringe technique, which directly projects stripe fringes on the object and calculates the terrain features by the distortion of the stripe; thus, it is easier and faster to calculate as well as obtaining full-field data. However, it is limited by the density of the stripes: when the density is high, black-and-white contrast is lost and difficult to read; when the density is too low, detail variations cannot be captured [6,7].
The resolution is enhanced by phase-shifting method, yet it is not useful in precise measurement for the error value is higher according to the experiment. The third system is the method presented in this paper—shadow moiré, which simply using one light source, grating and CCD filming devices to capture the terrain and the height variations of each spot swiftly and timely. It has the advantages of prompt data and full-field measurement, and the grating cycle can be adjusted as needed. It’s a pity a good measurement method like this is limited by complicated image processing and phase-shifting procedure on the application of moiré.
The innovative image process in this research is easier than previous studies. First, define a small location (a point) on the surface of object; by measuring the changes of the fringe in this location, we can have the conditions of pulse and speculate the result of the whole surface by summarizing the result of each point. Thus, in this paper, traditional moiré processing and calculating methods are replaced by one integrated program to automatize and simplify the procedures. Moiré measurement is upgraded to a simple, fast and automatized detection system, which can monitor surface on its own for a long period of time [8–12].
Currently, without any analysis instrument, the conditions of the pulse refer to the sensation of pulse felt by TCM Doctors’ fingers. The examination of pulse conditions means to distinguish the features of pules according to the location of pulse, the rhythm of pulse, the shape of pulse and the strength of pulse. Since the “pulse” can be only described in words in TCM so far but not be quantified, it cannot be analyzed by scientific methods. The purpose of this paper is to launch an initial quantitative method to quantify TCM data to support the study of TCM.
So far, the commercial TCM pulse detectors can only measure the rhythm of pulse with accuracy. The pulsing equipment mostly uses piezoelectric measuring methods, which have the following disadvantages: Firstly, when the sensor touches the pulse via its head, the coexistence of skin tension plus the underlying soft tissue and the change of arterial axial tension as well as vascular radial pulse force have concurrent effects on the head. In the measurement of the vessel’s radial power stroke, one cannot can eliminate such tensile impacts, thus, the pulse characteristics associated with vascular tension and the axial power stroke of the pulse are indistinguishable; secondly, when the sensor presses on the measured pulse via its head, the sensor causes a deformation due to the resistance of the skin and soft tissue (i.e., pulsing pressure) and the head that is pressed and touched, resulting in more measurement errors [13,14].
This system makes the most of the strength of shadow moiré measurement and combines the theory of TCM, using the surface height variations to locate the precise position of the TCM pulse diagnosis—guan pulse. And then it is applied to monitor the guan pulse for a period of time to draw the waveform of pulse variations, so as to obtain pulse movement data and calculate the amplitude and frequency value, which can indicate the strength of pulse and the rhythm of pulse respectively. Besides, by image processing the result of moiré pattern, we can get the location of pulse and the shape of pulse [15–18].
2. Method
This research applies Principle of Shadow Moiré to measure the terrain height and the height variations at any location on the surface of the object being measured. In previous studies for calculating moiré, the first step is to locate the fringe by calculating its shape in every single frame; then, compare the results in two frames to find out the change. If we want to know the change of a single point, we have to analyze the whole image in every frame, which is very complicated. Unlike current studies analyzing Moiré by whole image, this paper directly can analyzes a particular point by measuring the change of the fringe to speculate the change of amplitude. Therefore, our method is simple and fast. Here we explain the research method:
2.1 Measurement principle of shadow moiré
Shadow moiré is one of the moiré measurement methods, which is mainly used to measure the height and the amount of deformation. The principle is as shown in Fig. 1. A reference grating was placed on the object. After the light source was applied on the grating, the shadow of the grating would appear on the object. The shadow is called the model grating, or the shadow grating. Meanwhile, a filming device was set up to observe the scattered light and capture the image mapped by reference grating and model grating. This image is the interferometry moiré pattern of the two gratings [20].
In Fig. 1, the light went through the reference grating at 5 spots. Scattered light from spot A, C and E was captured by the camera, revealing the bright fringes; while lights from B and D was block by the reference grating, forming the dark part of the moiré pattern as it cannot be retrieved by CCD [20].
Figure 2 is the partial enlarged figure around Spot E. In the figure, α was the incidence angle of light source to the grating, and β was the observation angle of the camera to the grating. From the geometry relation, it shows:
After transposition, the distance between E and the grating was found (Z):
Z: the distance between the surface to be measured and the gratingP: the periodic of the grating or the period of the Ronchi gratings expressed in line-pair (a pair of black line and blank line);N: the ordinal number of the moiré fringes, N = 1, 2, 3,…α: the incidence angle of light source;β: the observation angle of CCD.2.2 Drawing terrain feature figure and measuring height variations
Equation (2), i.e. Shadow moiré measurement method was used here. When the distance (L) between light source and the grating and the distance (L) between CCD and the grating were both a lot farther than the distance (Z) between object surface and the grating; therefore, α and β could be taken as 2 constants, and K as 1 constant. They were only related to moiré Measurement System. Z (x,y) indicated the any coordinate height on moiré fringes. It formed a linear relation with the ordinal number of the moiré fringes N, i.e. the change value of each fringe represented a fixed height variations value. Via this constant relation of moiré fringes and height, not only static topographic map could be analyzed, but also the dynamic amount of moiré fringes movement, so as to calculate the height variations of the surface to be measured [20].
As shown in Fig. 3, on profile line y = A, when x went from left to right, each fringe number (N) added indicated the surface terrain height went up a fixed height difference Z = K. Also, height and direction had to be changed at the top (bottom) point, so that the altitude terrain figure of profile line y = A could be drawn.
To further use this figure, the analysis spot at the second left fringe on profile line y = A was considered as x = B. When the distances between the fringes and the surface (Z) gradually increase, it meant the height of surface is lowered gradually. Moiré contour line fringes would move to the higher spot. The fringe that was originally at Spot x = B would gradually move and pass that spot, and the number of passing is the fringe ordinal number N. So, whenever an analysis point passed through one fringe cycle, it meant the height of the spot displaced a fixed height difference Z = K. K represented the smallest resolution of moiré measurement height variations. And by analyzing the total periodic number of the fringes passed (N, the total times of the fringe changing cycle), the total height variations at x = B could be calculated Z = KN [20].
2.3 The automatized calculation program
Programmed automatized height variations measurement is the most important feature of this moiré Monitor System. By using the image processing and mathematics graphic designing programs which were developed by myself, the automatized measurement of the height of any spot of the surface can be conducted following the measurement principle of shadow moiré. And the calculation steps of the program is as listed below:
- (a) The program picked a white profile line from the moiré fringe figure as shown in Fig. 4 below and transform the fringe distribution to gray intensity distribution. And it defined the maxima and minima of the gray color scale, and then determine the coordinate position of the bright and dark moiré fringes.
- (b) The program chose a fixed analysis point from the profile line as shown that the white cross on 4(a) coordinate with the dotted line on 4(b). It also defined all the fringe distribution in 10 seconds and recorded the times when the fringes passed the analysis point, so as to convert the height displacement.
- (c) When the number of the cycle that the analysis point pass through the fringes was less than or more than one cycle, digital interpolation method was used. With the rate of △ x(pixels), which is the excess movement of fringes, in a complete cycle X(pixels), the cycle variation value was calculated △ N = (N + △ x / X) to the second place after the decimal point.
Fig. 4 (a) The distribution of the moiré pattern light intensity (b) the distribution of the gray intensity on the profile line (c) the distribution of the extreme values of the gray scale coordinating the fringe locations.
3. Moiré monitor system
3.1 System structure
The system architecture is shown in Fig. 5. Using a set of linear stripes periodic distribution Ronchi rulings grating as a reference grating, and then fix it at an appropriate distance from the top of the right wrist pulse position near the skin surface to be measured, at the top right with a halogen lamp as the stable light source to illuminate the grating and the object, digital cameras shoot in the upper left shadow moiré pattern on the surface of the object.
The specifications of the light source in this system are: Color temperature: 3200K, output power: 150W, the manufacturer is TAIWAN FIBER OPTICS, and the model number is LSH150F.
The specifications of the camera (DV) in this system are: The image resolution is 720* 480 pixels and the shooting speed is 30 (frames/second), the manufacturer is SONY, and the model number is DCR-TRV40.
In this paper, the angle of the light system is set as α = 40゜, shooting angle is β = 30゜, fringe cycle reference grating pitch P = 600 (μm) (the grating is parallel to the optical table surface), on behalf of the shadow moiré characteristic Eq. (2) to obtain the basic measurement system resolution K = 233.4 (μm).
The system developed in this study can observe and distinguish the cun pulse, guan pulse, and chi pulse at the same time. Here is the method: we first find the radius bone position, and then we use it as a reference in search of pulses. This originates from the description listed in Traditional Chinese Medicine, “The part slightly below the styloid process of radius bone is the guan pulse; the part anterior the guan pulse is the cun pulse, and the part posterior the guan pulse is the chi pulse [17].” Hence, we use the coordinates of the image to locate the positions of cun, guan and chi pulses. For each measurement, we use traditional Chinese physician’s finger to reconfirm the position of cun, guan and chi pulses. Once we locate the rough positions of these pulses, we distinguish the correct positions by finding three points with the largest amplitude change in this rough area. Then, after locating the positions of pulse, this system will record the pulse information in 10 sec and calculate the waveform and the frequency. In our paper, we illustrate this method by analyzing guan pulse, as shown in Figs. 6 and 7.
3.2 Program calculation process
After the moiré pattern video is stored as an * .avi file, the execution of program begins. After six automatized processing steps, as shown in the flowchart in Fig. 8, the final output was highly contour plot and the surface to be measured with the time height variations situations, completing 10 seconds the inner surface of the skin pulse monitoring.
The following is a flow chart of the six process steps and details
- Step 1: Import videos and process.
- Step 2: Correction: zero measurement error due to the object or sloshing around the system itself generated.
- Step 3: select the location to be detected, as our analysis points.
- Step 4: calculate a cross-sectional analysis point’s moiré contour line, and draw terrain graphics.
- Step 5: calculate the change of the fringe cycle in the analysis points during the detection period, and convert it into height.
- Step 6: produce test results, select the loop re-analyze the other points, to complete the three-dimensional detection.
3.3 The accuracy of detection
To demonstrate the system measurement results and real results on the degree of compliance, before measuring the pulse of the first detection accuracy of the system.
3.3.1 Height detection accuracy
Every small change in height adjustment using highly fine-tuning meter rise as the true value of 10um, by height 0 ~250um each moiré system by measuring changes in the amount of 100 times its height and record, in which fine-tune the instrument to the height of each of its interference rangefinders (the manufacturer is Agilent, and the model number is 5529A.) correction height, making fine-tuning instrument error <0.3um, the main source of error due to the random vibration of the external environment, as shown in Fig. 9.
Measuring height error of the system and the real height of the order of 100 times the system measurement results were averaged, the maximum amount of measurement error Emax<1.9 (um), shown in Fig. 10.
3.3.2 The frequency detection accuracy
Every increase in the use of signal generator 0.1Hz as a true value by 0.7 ~2.0Hz, given a fixed frequency speaker frequency, each measured by the amount of moiré system 10 times and recorded. Where the signal generator with digital oscilloscope calibration frequency, frequency control frequency error <0.01Hz, the error depends on the accuracy of the oscilloscope, shown in Fig. 11.
Measurement system and the real frequency error, to the system measurement results averaged 10 times, the maximum amount of measurement error Emax<0.0166 (Hz), Fig. 12.
4. Pulse measurement results
Use of moiré topography system to monitor changes in the characteristics of the entire surface, the program first started using TCM for a way to find out the approximate location of pulse point to the location to do a three-dimensional surface height variations analysis, to find the location of the largest skin surface height variations, is defined as the precise location of TCM pulse point (guan pulse), and then monitoring the position of the skin surface and recording the change real time of the vibration waveform, amplitude and frequency to calculate the value at that point pulse beat.
4.1 Detecting the height variations around the right-hand guan pulse
Plane displacement calculation program execution, the variations of the height detection the right-hand surface of the skin around the point of the pulse, the calculation region shown in Fig. 13; the results plotted as a 3D height variations distribution graph, Fig. 14.The maximum amount of height change happens in t = 2.67s and its position is in (x,y) = (219,221), which is the location of guan pulse.
4.2 Pulse waveform of pulse movements monitor
On our guan pulse position coordinate points, monitor the height variations for 10 seconds, and depict the curve of the height variations values and the time, the curve is a pulse waveform of right-hand surface of the skin of the pulse point of the vibration changes with time, as shown in Fig. 15. To obtain the pulse waveform, then fast Fourier transform, the time domain into the frequency domain, can be obtained by vibration of the frequency of the pulse of frequency shows 1.58 (Hz) = 94.8beats per minute, as shown in Fig. 16.
4.3The discussion of experiments results
From Figs. 13–16, we can quantify the pulse by these experiment data: in Fig. 13, we can measure the shape of pulse by moiré pattern; we find out the location of pulse by finding the maximum amount of the height change in Fig. 14; then we use the relation between time and amplitude of the waveform at the location of pulse to illustrate the strength of pulse, as shown in Fig. 15; finally, in Fig. 16, we calculate the vibration of the frequency at the location of pulse to define the rhythm of pulse. According to these data and the description in TCM theory, the pulse result we have in the experiment probably is the “Full Pulse” which is defined as follow: “Full pulse is marked by wide size and full content, beating like roaring waves and sensibility under light pressure and surges as well as sudden flowing and ebbing [17].” Unlike other pulse measurement instruments which can only provide the information about the rhythm of pulse, our system describes all conditions of pulse including the shape, the location, the strength, and the rhythm of pulse, which can give TCM doctors a whole scope of patients’ status rather than only by doctors’ fingers. Therefore, the contribution of this paper is providing a comprehensive pulse analysis method to TCM study or diagnosis.
5. Conclusion
This research presents a moiré measurement system, which can calculate the measurement process automatically, yet still maintain great measurement resolution and accuracy. And the advantage of automatization is fast height measurement and long cycle height variations monitor. By using the characteristic of moiré, monitoring is expended from a spot to the whole surface, and obtains the 3D graphics of the height variations real time. In addition, a series of detection are made to the measurement of the system, so as to achieve the measurement accuracy within 10um and matches each location on moiré pattern. It suggests the system can be applied to the pulse palpation monitoring on wrist skin surface.
Furthermore, following the TCM pulse diagnosis method and procedure, this moiré measurement system is elevated to an optical pulse palpation system. It can measure the conditions of pulse very quickly and clearly. Together with the ancient TCM pulse palpation theory, this system can be a scientific pulse diagnosis equipment for us to have an initial idea about our health condition.
Acknowledgments
This work was supported by the National Central University Research Fund of 2015 (NO. 104G903-5).
References and links
1. R. G. Dorsch, G. Häusler, and J. M. Herrmann, “Laser triangulation: fundamental uncertainty in distance measurement,” Appl. Opt. 33(7), 1306–1314 (1994). [CrossRef] [PubMed]
2. M. C. Amann, T. Bosch, M. Lescure, R. Myllylä, and M. Rioux, “Laser ranging: a critical review of usual techniques for distance measurement,” Opt. Eng. 40(1), 10–19 (2001). [CrossRef]
3. G. Manneberg, S. Hertegård, and J. Liljencrantz, “Measurement of humanvocal fold vibrations with laser triangulation,” Opt. Eng. 40(9), 2041–2044 (2001). [CrossRef]
4. J. H. Wu, R. S. Chang, and J. A. Jiang, “A novel pulse measurement system by using laser triangulation and a CMOS image sensor,” Sensors (Basel Switzerland) 7(12), 3366–3385 (2007). [CrossRef]
5. R. S. Chang, J. H. Chiu, F. P. Chen, J. C. Chen, and J. L. Yang, “A Parkinson’s Disease Measurement System Using Laser Lines and a CMOS Image Sensor,” Sensors (Basel) 11(12), 1461–1475 (2011). [CrossRef] [PubMed]
6. J. C. Perrin and A. Thomas, “Electronic processing of moiré fringes: application to moiré topography and comparison with photogrammetry,” Appl. Opt. 18(4), 563–574 (1979). [PubMed]
7. V. Srinivasan, H. C. Liu, and M. Halioua, “Automated phase-measuring profilometry of 3-D diffuse objects,” Appl. Opt. 23(18), 3105–3108 (1984). [CrossRef] [PubMed]
8. D. Post, B. Han, and P. G. Ifju, “Moiré methods for engineering and science - moiré interferometry and shadow moiré,” Top. Appl. Phys. 77, 151–196 (2000). [CrossRef]
9. B. Han, D. Post, and P. Ifju, “Moiré interferometry for engineering mechanics: current practices and future developments,” J. Strain Anal. Eng. Des. 36(1), 101–117 (2001). [CrossRef]
10. A. S. Voloshin, C. P. Burger, and R. E. Rowlands, “Composites analysis, by fractional moiré fringe system,” J. Comp. Mat. 19(6), 513–524 (1985). [CrossRef]
11. J. E. A. Liao and A. S. Voloshin, “Enhancement of the shadow-moiré method through digital image processing,” Exp. Mech. 33(1), 59–63 (1993). [CrossRef]
12. H. Du, H. Zhao, B. Li, J. Zhao, and S. Cao, “Phase-shifting shadow moiré based on iterative self-tuning algorithm,” Appl. Opt. 50(36), 6708–6712 (2011). [CrossRef] [PubMed]
13. E. King, D. Cobbin, S. Walsh, and D. Ryan, “The reliable measurement of radial pulse characteristics,” Acupunct. Med. 20(4), 150–159 (2002). [CrossRef] [PubMed]
14. G. Wei, Y. Q. Lu, and J. Z. Oyang, “Chaos and fractal theories for speech signal processing,” Tien Tzu Hsueh Pao 24, 34–39 (1996).
15. J. I. Davies and A. D. Struthers, “Pulse wave analysis and pulse wave velocity: a critical review of their strengths and weaknesses,” J. Hypertens. 21(3), 463–472 (2003). [CrossRef] [PubMed]
16. D. Normile, “Asian medicine. the new face of traditional Chinese medicine,” Science 299(5604), 188–190 (2003). [CrossRef] [PubMed]
17. Z. G. Wang and J. Ren, “Current status and future direction of Chinese herbal medicine,” Trends Pharmacol. Sci. 23(8), 347–348 (2002). [CrossRef] [PubMed]
18. X. Gong and N. J. Sucher, “Stroke therapy in traditional Chinese medicine (TCM): prospects for drug discovery and development,” Phytomedicine 9(5), 478–484 (2002). [CrossRef] [PubMed]
19. L. F. Wang, Diagnostics of Traditional Chinese Medicine (Shanghai University of Traditional Chinese Medicine, 2002).
20. D. Post, B. Han, and P. Ifju, High Sensitivity Moiré: Experimental Analysis for Mechanics and Materials (Springer, 2008).
