Displacement meter and method for manufacturing an article

Abstract

To provide a displacement meter that can more accurately calculate a displacement amount even when a moving speed of an object to be measured increases.SOLUTION: A displacement meter measures displacement of an object to be measured, and has: an illumination system that illuminates the object to be measured; a photoelectric conversion element array that detects reflected light from the object to be measured; and a calculation unit that calculates an amount of displacement of the object to be measured by using a cross-correlation function of a plurality of images detected by the photoelectric conversion element array at different timings. The calculation unit limits a detection range for performing processing of detecting a position indicating a peak of the cross-correlation function to a part of the entire range of the cross-correlation function, detects the position indicating the peak in the limited detection range, and calculates the displacement amount, on the basis of, the detected position indicating the peak.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present invention relates to a displacement meter and the like.

Background Art

[0002] Patent Document 1 discloses a speckle interferometer that irradiates the surface of a moving object with a laser beam, detects the generated speckle pattern, and measures the displacement amount of the object.

[0003] This speckle interferometer is configured to acquire the speckle pattern of the observation surface irradiated with the laser beam by an image sensor, and calculates the cross-correlation function between the reference data acquired at the reference position and the current data acquired thereafter. Then, the peak position is searched from the obtained cross-correlation function, and the movement amount of the object is calculated based on the peak position. Further, when it is determined that the degree of correlation of the cross-correlation function has decreased, it is disclosed that the current data can be used as new reference data, so that the length can be measured even when the movement distance of the object is large.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] However, when the moving speed of the object is high, the change in the speckle pattern also becomes fast, and in addition to the true peak that should be originally detected in the obtained cross-correlation function, a false peak with a high degree of correlation may occur. When the false peak has a higher degree of correlation than the true peak, the displacement amount cannot be calculated correctly.

[0006] Therefore, an object of the present invention is to provide a displacement meter that can more accurately calculate the displacement amount even when the moving speed of the measurement object is high.

Means for Solving the Problem

[0007] A displacement meter as one aspect of the present invention for solving the above problems is a displacement meter for measuring the displacement of a measurement object, and includes an illumination system for illuminating the measurement object, a photoelectric conversion element array for detecting reflected light from the measurement object, and a calculation unit for calculating the displacement amount of the measurement object using the cross-correlation function of a plurality of images detected by the photoelectric conversion element array at different timings. The calculation unit limits a detection range for performing a process of detecting a position indicating the peak of the cross-correlation function to a part of the entire range of the cross-correlation function, detects the position indicating the peak in the limited detection range, and calculates the displacement amount based on the detected position indicating the peak. Shi , The calculation unit sets a detection range so as to include the position indicating the peak of the cross-correlation function used for calculating the previous displacement amount, based on the position indicating the peak of the cross-correlation function for the entire range of the cross-correlation function. It is characterized by this.

Advantages of the Invention

[0008] According to the present invention, even when the moving speed of the measurement object becomes fast, it is possible to provide a displacement meter capable of calculating the displacement amount more accurately.

Brief Description of the Drawings

[0009]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Mode for Carrying Out the Invention

[0010] Hereinafter, preferred embodiments of the present invention will be described in detail based on the accompanying drawings.

[0011] <First Embodiment> FIG. 1 is a diagram showing a schematic configuration of a non-contact length measuring instrument (displacement meter) according to the first embodiment. As shown in FIG. 1, the displacement meter according to this embodiment is arranged opposite to the displacement meter 100 and measures the displacement of a measurement object moving in the direction of the arrow in the figure in a non-contact manner.

[0012] The light beam emitted from the light source 1 is condensed onto the measurement object by the illumination optical system (illumination system) 2 to illuminate the measurement object. The light source 1 is controlled by the control unit. The light source 1 is a laser diode, an LED, a halogen lamp, or the like. When a laser diode as a coherent light source is selected, the obtained image is composed of speckles. When an LED or a halogen lamp as an incoherent light source is selected, an image reflecting the pattern on the surface of the measurement object is obtained.

[0013] The illumination optical system 2 is composed of a single lens or a lens group. When using a laser diode, it is desirable to select based on the criterion of performing aberration correction so that it can be condensed with a plane wave. Also, when an LED or a halogen lamp is selected, it is only necessary to be able to illuminate the light receiving area, and aberrations and the like are not particularly problematic. It is possible to appropriately select the type of the light source 1 according to the size of the area to be illuminated.

[0014] A part of the reflected light diffusely reflected by the object to be measured is condensed onto the sensor 4 through the light-receiving optical system of the condensing member 3. The condensing member 3 is composed of a single lens or a lens group and can be appropriately selected according to the resolution of the light-receiving optical system.

[0015] The sensor 4 is composed of an array of photoelectric conversion elements such as a CCD element, a CMOS element, or a line sensor. When a two-dimensional area sensor is selected as the sensor 4, it is possible to detect two-dimensional displacement, and when a line sensor is selected, it is possible to detect one-dimensional displacement.

[0016] The magnification of the light-receiving optical system can be determined from the installation distance between the object to be measured and the condensing member 3, the focal length of the single lens or lens group used in the condensing member 3, and the position of the sensor 4. From the magnification of the light-receiving optical system and the size of the pixels constituting the sensor 4, the conversion length, that is, the resolution, of the pixels serving as the measurement reference is determined.

[0017] The light beam imaged on the sensor 4 is photoelectrically converted and then output to the data processing unit 5 (calculation unit). The data processing unit 5 is composed of a processor such as a microcomputer, an FPGA, or a DSP, and processes the image output from the sensor 4 to calculate the measured length (displacement amount) of the object to be measured.

[0018] The measured length calculated by the data processing unit 5 is output to the input / output unit 6. The input / output unit 6 transmits the calculated measured length as data to the outside of the displacement meter 100. As the configuration of the input / output unit 6, in the case of a wired LAN, an Ethernet-PHY or a connector is applicable. In addition, other serial communications such as USB (Universal Serial Bus) or RS-232C can be configured as long as they are communication means with the outside. Also, it may be wireless rather than limited to wired. The output of the measured length value may be in the specification of outputting as a pulse signal or an analog signal.

[0019] By configuring the input / output unit 6 to receive an instruction from the outside as a command, it is also possible to start or stop the length measurement or notify the state of the device according to the received command.

[0020] Next, the data processing unit 5 will be described with reference to FIG. 2. FIG. 2 is a diagram showing the configuration of the data processing unit 5. The image data of the analog signal sent from the sensor 4 is AD-converted into a digital signal by the image data conversion unit 10. When the sensor 4 outputs a digital signal, the image data conversion unit 10 is omitted.

[0021] The image data output from the image data conversion unit 10 performs preprocessing necessary for calculating the cross-correlation function, such as offset processing, in the image data processing unit 11. Spatial filter processing such as edge enhancement may also be performed.

[0022] The reference image data storage unit 12 stores reference image data that serves as a reference for calculating the length measurement. As a principle for calculating the length measurement, it is necessary to store the reference image data in order to calculate the cross-correlation function from two pieces of image data acquired at different times.

[0023] The correlation function calculation unit 13 calculates the cross-correlation function between the image data output from the image data processing unit 11 and the reference image data output from the reference image data storage unit 12.

[0024] The peak position detection unit 14 searches for the pixel position where the correlation value becomes a peak (extreme value) from the calculated cross-correlation function. The cross-correlation function represented in this embodiment shows the correlation value when two pieces of image data acquired at different times are shifted one pixel at a time.

[0025] The length measurement value calculation unit 15 calculates, as the length measurement (displacement amount), the amount by which the measurement object has been displaced at the time when the two pieces of image data were acquired, from the pixel position indicating the peak detected by the peak position detection unit 14. As described above, the displacement amount of the measurement object with respect to one pixel is determined by the conversion length (resolution) determined by the position of the sensor 4, the focal length of the condensing member 3, etc. Therefore, by multiplying the conversion length by the peak pixel position, the displacement amount of the measurement object can be calculated. The calculated length measurement is output to the input / output unit 6.

[0026] The peak search range setting unit 16 calculates a range (detection range) for searching for a peak position with respect to the cross-correlation function calculated at the next timing from the peak position detected by the peak position detection unit 14. The peak search range is a range that limits the detection range in which the peak position detection unit 14 performs the process of detecting the position indicating the peak of the cross-correlation function to a part of the entire range of the cross-correlation function. The peak search range is set to be changeable. The peak position detection unit 14 detects the position indicating the peak within the limited range, and based on the position indicating the detected peak, the displacement amount calculation unit 15 calculates the displacement amount.

[0027] The detailed displacement amount calculation flow of the data processing unit 6 will be described with reference to the flowchart of FIG. 3 and the diagram showing the correlation function of FIG. 4. First, when receiving an external displacement measurement start command, the non-contact displacement meter 100 turns on the light source 1 and acquires image data sequentially obtained at the sampling rate set by the sensor 4 (S101).

[0028] In S102, it is confirmed whether the reference image data has been acquired. At the timing of first acquiring the image data, there is no reference image data for calculating the cross-correlation function. If there is no reference image data, the acquired image data is stored in the reference image data storage unit 12 (S104). Then, the displacement measurement amount is set to 0 (zero) (S119).

[0029] If there is reference image data, the cross-correlation function is calculated using the acquired image data and the reference image data (S103). FIG. 4(a) shows the cross-correlation function calculated first. The vertical axis represents the correlation value, and the horizontal axis represents the pixel position. A process of searching for a peak is performed from this cross-correlation function. However, at the timing of first acquiring the cross-correlation function, the peak search range has not yet been set (S105).

[0030] If it is determined in S105 that the peak search range is not set, the peak position detection unit 14 detects the pixel position where the correlation value becomes the maximum peak value from the entire range of the cross-correlation function (S107). In the cross-correlation function of FIG. 4(a), the peak P1a is the peak (extreme value) with the maximum value, and its pixel position is set as X1a.

[0031] In the cross-correlation function in this embodiment, the center on the horizontal axis is defined as Xc. When a peak of the cross-correlation function is detected at the center Xc, it means that the image data of the measurement target and the reference image data are not displaced. Also, the moving direction of the measurement target can be determined by which side of the center Xc the peak is detected. If the case where it is on the right side of the center Xc like the peak P1a is regarded as the movement in the plus direction, when a peak is detected on the left side of the center Xc, it indicates that the measurement target is moving in the minus direction.

[0032] In FIG. 4(a), the pixel position of the peak P1a is X1a, and the length measurement value calculation unit 15 calculates the length measurement value by multiplying the number of pixels from the center Xc to X1a by the conversion length (S108). In this case, since X1a is on the right side of the center Xc, it is output as a plus length measurement value.

[0033] The calculated length measurement value is output as length measurement data to the outside from the input / output unit 6 (S109). When pulse output is performed, pulses are generated according to the length measurement amount, and pulse phase adjustment and the like are also performed and output according to the moving direction of the measurement target.

[0034] Next, in the peak search range setting unit 16, a range for searching for the peak position in the next calculated cross-correlation function is calculated (S110). Based on the peak position X1a detected in FIG. 4(a), a predetermined position on the left side of the peak position X1a is calculated as the start position XL1b for the next peak search, and a predetermined position on the right side of the peak position X1a is calculated as the end position XH1b for the peak search. That is, a range including the peak position X1a is calculated as the peak search range. After the calculation, the positions XL1b and XH1b are set in the peak position detection unit 14.

[0035] After calculating and setting the peak search range, the latest image data obtained for the next calculation of the cross-correlation function is updated as the reference image data (S111).

[0036] Next, it is determined whether the length measurement has ended (S112). If a command to end the length measurement has been received from the outside, the acquisition of image data by the sensor 4 and the lighting of the light source 1 are terminated, and the length measurement is ended. If the length measurement is to be continued, the next image data acquisition is performed (S101).

[0037] The cross-correlation function calculated from the newly acquired image data and the updated reference image data is shown in Fig. 4(b). When the measurement object is moving at the same speed, since the sampling rate for acquiring the image data does not change, the peak position of the cross-correlation function occurs at almost the same position.

[0038] Since the peak search range was set in the previous S110, it is determined in S105 that the peak search range is set, and the process proceeds to S106. The peak position detection unit 14 starts searching for the peak from the start position XL1b of the set peak search range. By searching in the range from XL1b to XH1b of the peak search range, the pixel position X1b of the peak P1b can be searched as the peak position. The peak position X1b in Fig. 4(b) is almost the same position as the peak position X1a in Fig. 4(a).

[0039] The length measurement amount is calculated from the detected peak position X1b, and the search ranges XL1c and XH1c for the next peak search are calculated. After the calculation, the positions XL1c and XH1c are set in the peak position detection unit 14. Here too, in the same manner as above, a range including the peak position X1b is set as the peak search range.

[0040] Next, the cross-correlation function calculated after updating the image data is shown in Fig. 4(c). The peak search range is set from XL1c to XH1c, and the peak position is searched within this range. As a result, the pixel position X1c of the peak P1c can be detected as the peak position.

[0041] Thereafter, the same length measurement value calculation flow is repeated until an instruction to end the length measurement is given. Figure 4 shows the change in the cross-correlation function when the moving speed of the measurement object is relatively slow. Due to the slow moving speed, the peak position appears at a position close to the center Xc. Also, since the image data acquired by sensor 4 is also due to the slow moving speed of the measurement object, the state of the surface of the measurement object can be clearly captured. Therefore, since the peak position of the cross-correlation function also stably occurs at the same position, the length measurement amount can also be stably measured.

[0042] Next, the case where the moving speed of the measurement object becomes fast will be described using Figure 5. Figure 5 shows the cross-correlation function, where the vertical axis represents the correlation value and the horizontal axis represents the pixel position. First, after measuring the object multiple times to obtain multiple images, the peak position detection unit 14 detects the pixel position that becomes the peak value with the maximum correlation value from the entire range of the cross-correlation functions of the multiple images. In the cross-correlation function of Figure 5(a), peak P2a is the peak (extreme value) with the maximum value, and its pixel position is denoted as X2a. Compared with the cross-correlation function of Figure 4, the peak position is farther from the center Xc, indicating that the moving speed of the measurement object is fast.

[0043] Next, based on the peak position X2a detected in Figure 5(a), a predetermined position on the left side of the peak position X2a is calculated as the start position XL2b for the next peak search, and a predetermined position on the right side of the peak position X2a is calculated as the end position XH2b of the peak search. That is, a range including the peak position X2a is calculated as the peak search range. After the calculation, the positions XL2b and XH2b are set in the peak position detection unit 14.

[0044] Then, the next measurement is performed, and the peak position detection unit 14 starts searching for the peak from the start position XL2b of the set peak search range. By searching in the range from XL2b to XH2b of the peak search range, the pixel position X2b of peak P2b can be searched as the peak position.

[0045] Perform length measurement calculation from the detected peak position X2b, calculate the search ranges XL2c and XH2c for the next peak search, and after the calculation, set the positions XL2c and XH2c in the peak position detector 14. Here too, similar to the above, set the range including the peak position X2b as the peak search range.

[0046] Next, update the image data and show the calculated cross-correlation function in Fig. 5(c). The peak search range is set from XL2c to XH2c, and search for the peak position within this range. As a result, the pixel position X2c of the peak P2c can be detected as the peak position.

[0047] When the measurement object is moving, since the measurement object moves within the exposure time of the sensor 4, it is affected by blurring to some extent. Therefore, as the moving speed of the measurement object increases, the image data acquired by the sensor 4 is more affected by blurring, and the acquired image data is in a state with a large blur.

[0048] Thus, in order to acquire image data in a situation where the moving speed of the measurement object is high, both the image data and the reference image data have a large blur. Therefore, as in the cross-correlation function shown in Fig. 5(c), a false peak P2d is generated whose correlation value is calculated to be higher than that of the actual peak P2c generated at the actual moving speed as the moving speed increases. This means that in the process of calculating the cross-correlation function, due to the influence of the blur of the image data, the correlation is calculated to be higher in the false peak position X2d than in the true peak position X2c. If the length measurement value is calculated using the false peak position X2d, a length measurement value different from the length measurement value obtained from the actual movement amount of the object will be calculated.

[0049] Here, by considering the relationship between the sampling rate of the sensor 4 and the moving speed of the object, the peak position of the cross-correlation function can be roughly estimated. Therefore, in this embodiment, the search range can be set based on the information regarding the speed of the measurement object. Also, the search range can be set based on the information regarding the displacement amount of the measurement object per sampling of the photoelectric conversion element array. By performing peak search within the thus-set search range from XL2c to XH2c, only the true peak P2c can be detected. As a result, the true peak position X2c is determined, and by calculating the measurement length from X2c, an accurate measurement length can be output. If peak search is performed over the entire range, a false peak P2d will be detected, resulting in an incorrect calculated measurement length.

[0050] Also, in the calculation of the next peak search range, by calculating with reference to the peak P2c detected within the search range, even if a false peak occurs in the next calculated cross-correlation function, the true peak can be detected.

[0051] Next, the details of the calculation of the peak search range will be described. In the process of limiting the peak search range, it is necessary to appropriately determine the search range. If the search range is narrow, there is a possibility that the true peak will not be included in the search range. On the contrary, if the search range is too wide, there is a possibility that false peaks will be included.

[0052] As an example of calculating the peak search range, it can be calculated from the information regarding the speed of the measurement object that the non-contact length gauge allows, for example, the acceleration. If the maximum acceleration that the non-contact length gauge allows is determined, the maximum movement amount that the measurement object moves during the sampling period for acquiring the image data can be calculated. For example, from the relationship between the maximum acceleration and the sampling period, if the movement amount of the measurement object corresponds to 10 pixels, a range of ±10 pixels or more with respect to the detected peak position may be set as the peak search range. If a search range narrower than ±10 pixels is set, the true peak cannot be detected within the search range, resulting in an incorrect output of the measurement length.

[0053] In addition to calculating from the acceleration, the moving direction of the measurement object may be determined to determine the search range. As described above, the cross-correlation function in the present embodiment indicates the moving direction of the measurement object depending on whether the peak position is on the right side or the left side of the center.

[0054] When the moving speed of the measurement object is high, it can be assumed that the moving direction of the measurement object does not reverse within a short sampling period. Therefore, the center of the peak search range may be set to be in the same moving direction as the reference peak position. That is, the moving direction of the measurement object can be detected, and the search range can be set so that the center of the search range is on the moving direction side of the measurement object with respect to the center of the cross-correlation function.

[0055] As described above, by limiting the peak search range, it becomes possible to more accurately detect the true peak without detecting false peaks.

[0056] <Second Embodiment> In the first embodiment, the case where the reference image data is updated to the acquired image data every time has been described. When the moving speed of the measurement object is slow, updating the reference image data every time may cause an error that cannot be ignored with respect to the moving amount of the measurement object. Therefore, by fixing the reference image data until the measurement object moves a certain amount, the reference image data for calculating the cross-correlation function becomes the same, and the error generated when calculating the measured length amount can be suppressed.

[0057] When updating the reference image data every time, when the moving speed of the measurement object is the same, the peak position appearing in the cross-correlation function appears at the same position every time. On the other hand, when the reference image data is fixed, even when the measurement object is moving at the same speed, the peak position appearing in the cross-correlation function does not occur at the same position, and the peak position also moves according to the moving speed.

[0058] In addition, in order to update the reference image data when the object to be measured moves by a certain amount, the peak position of the cross-correlation function calculated next after updating the reference image moves according to the threshold for updating the reference image data. That is, when limiting the peak search range, it is necessary to change the calculation conditions of the peak search range when the reference image data is fixed and when the reference image data is updated.

[0059] In the second embodiment, the setting of the peak search range in the case of fixing and updating the above-described reference image data will be described with reference to FIGS. 6 and 7. The configuration of the non-contact length measuring instrument in the second embodiment is the same as the configuration of the first embodiment shown in FIGS. 1 and 2.

[0060] FIG. 6 shows a processing flow in length measurement considering whether or not the reference image data is updated. With respect to the signal processing flowchart in the configuration of the first embodiment in FIG. 3, the processes of S201 to S203 are added instead of S110. The same processes as in FIG. 3 are denoted by the same reference numerals, and the description thereof will be omitted as appropriate.

[0061] The processes from the start of length measurement to the calculation of the length measurement value (S108) and the output of the calculated length measurement value (S109) are the same as those in the first embodiment. In the first embodiment, after outputting the length measurement value, the search range was calculated and set based on the maximum peak position. In S201 of the present embodiment, it is determined whether or not the peak position detected in S106 or S107 exceeds the determination threshold for updating the reference image data.

[0062] FIG. 7(a) shows the cross-correlation function calculated for the first time in the length measurement value calculation flow of the second embodiment. Here, XTL and XTH represent the update thresholds for determining the update of the reference image data. Although the update of the reference image data is determined by the moving distance of the object to be measured, since the length measurement value is calculated by multiplying the peak position of the cross-correlation function by the above-described conversion length, the number of pixels may be used as the update threshold. In the present embodiment, the update threshold is represented by the number of pixels of the cross-correlation function, and the reference image data is updated when the peak position exceeds the update threshold.

[0063] In Fig. 7(a), peak P3a is detected from the entire range, and the peak position is X3a. At this point, the peak position X3a is inside the update threshold values XTL and XTH, and in S201, it is determined that the update threshold is not exceeded. Therefore, the update of the reference image data (S202) is not performed, and the process proceeds to the calculation of the search range in S203.

[0064] Next, the search range is calculated in S203. Based on the detected peak position X3a, XL3b and XH3b of the search range shown in Fig. 7(b) are calculated. Here, since the reference image data has not been updated, the update of the reference image data is not considered in the calculation of the search range. After calculating the search range, the calculated search range is set in the peak position detector 14.

[0065] Next, it is determined whether to continue length measurement in S111, and the next image data acquisition (S101) is performed. Subsequently, the cross-correlation function is calculated. Since the reference image data has not been updated in the determination of S201 described above, the cross-correlation function is calculated using the unupdated reference image data and the latest acquired image data (S103). The calculated cross-correlation function is shown in Fig. 7(b).

[0066] Next, peak search is performed. Since the peak search range is set from XL3b to XH3b, peak P3b, which is the maximum peak within that range, is detected, and the peak position is X3b (S106). When viewed in the entire range, the maximum peak is peak P3d, but since it is outside the search range, it is not detected.

[0067] Next, the length measurement value is calculated (S108). The length measurement value is calculated from the peak position X3b. Since the reference image has not been updated, the previously calculated length measurement value is also included. Therefore, it is necessary to subtract the previously calculated length measurement value using the peak position X3a from the length measurement value calculated using the peak position X3b. Also, the length measurement value may be calculated using the number of pixels of the difference between the currently detected peak position X3b and the previously detected peak X3a. The length measurement value output in S109 is the length measurement value calculated by the difference.

[0068] Next, in the update determination of the reference image data (S201), as shown in FIG. 7(b), since the detected peak position X3b exceeds the update threshold XTH, the process proceeds to S202, and the reference image data is updated to the latest image data.

[0069] Due to the update of the reference image data, in calculating the search range in S203, it is necessary to take into account the fact that the reference image data has been updated. FIG. 7(c) shows the search range XL3c - XH3c when the reference image data is not taken into account. The search range XL3c - XH3c calculated here is calculated based on the detected peak position X3b. However, since the true peak P3c that should be detected originally does not fall within the search range XL3c - XH3c due to the update of the reference image data, a false peak P3e is detected.

[0070] Therefore, when the reference image data is updated, it is necessary to take into account the amount of movement of the peak position. When the reference image data is updated, the amount of movement of the peak position is determined by the update threshold. Therefore, the search range may be calculated based on the amount of movement from the peak position X3b according to the update threshold. FIG. 7(c’) shows the search range XL3c’ - XH3c’ calculated taking into account the update of the reference image data. When a peak is detected within this search range, the true peak P3c can be detected.

[0071] By the process described above, it is possible to calculate the peak search range taking into account whether the reference image data has been updated, and to more accurately detect the true peak.

[0072] In this embodiment, when the reference image data is updated, a method of determining the reference position taking into account the amount of movement of the peak position determined from the update threshold and calculating the search range from that reference position has been described. However, the search range may be widened by taking into account the amount of movement of the peak position. In that case, since the search range becomes wider, it is necessary to set a range that does not include false peaks.

[0073] <Third Embodiment> In the first and second embodiments, the true peak detection flow by limiting the search range for searching the peak position from the cross-correlation function and the calculation flow for specifically setting the search range were described. When limiting the search range, it is necessary to use the true peak as a reference. If a false peak is used as a reference, the true peak will not fall within the search range continuously, and thus the correct measurement length value cannot be calculated. Therefore, in the third embodiment, a method for selecting the true peak when limiting the search range will be described using the flowchart of FIG. 8 and the examples of the cross-correlation functions of FIGS. 9 and 10.

[0074] In the flowchart of FIG. 8, the process of S301 is added to the flowchart of FIG. 3 in the first embodiment. The processes described in the first and second embodiments will be omitted as appropriate.

[0075] The processes from the start of the length measurement to the calculation of the length measurement value (S108) and the output of the calculated length measurement value (S109) are the same as those in the first embodiment. The cross-correlation function calculated in S103 is shown in FIG. 9(a).

[0076] In S301, a determination is made as to whether or not to limit the peak search range. Here, a determination threshold is set for the correlation degree (correlation value) of the detected peak. In FIG. 9(a), since it is the cross-correlation function calculated first, the maximum peak is searched over the entire range and the peak P4a is detected, but the correlation degree does not exceed the determination threshold Cth. Therefore, in S301, it is determined that the condition for limiting the search range is not satisfied, and the calculation / setting of the peak search range (S110) is not performed, and the process proceeds to the update of the reference image data (S111).

[0077] The correlation degree in the cross-correlation function indicates that the higher the peak level, the higher the correlation. Therefore, when the peak level is low, that is, when it is lower than the determination threshold Cth, there is also a possibility of a false peak, so the limitation of the peak search range is not performed.

[0078] Subsequently, the cross-correlation function calculated from the acquired image data is shown in Fig. 9(b). Here, since the peak search range is not set, the maximum peak is searched from the entire range according to the determination in S105. At this time, since the detected peak P4b exceeds the determination threshold Cth, the process proceeds to calculate the peak search range (S110) based on the determination in S301.

[0079] In the calculation of the peak search range in S110, the search range XL4c - XH4c is calculated based on the peak position X4b and set in the peak position detector 14.

[0080] Subsequently, the cross-correlation function calculated from the acquired image data is shown in Fig. 9(c). Here, since the peak search range XL4c - XH4c is set, the peak P4c within that search range is detected. Since the peak P4d is outside the peak search range, it is not detected.

[0081] As described above, whether the peak level of the cross-correlation function exceeds the determination threshold is used as the determination condition for limiting the peak search range. The case where the determination condition is whether the peak positions occur at the same position will be described with reference to Fig. 10.

[0082] Fig. 10(a) shows the cross-correlation function calculated first. Since it is the first peak position search, the maximum peak is detected from the entire range, and the peak P5a is detected (S108). The peak position at this time is X5a. Here, since the detection of the peak position is the first time, the process of setting the peak search range (S110) is not performed.

[0083] The cross-correlation function calculated from the subsequently acquired image data is shown in Fig. 10(b). Here, since the peak search range is not set, the maximum peak is detected from the entire range in the same way as the first peak position search (S107). The maximum peak detected from the entire range is P5b, and the peak position is X5b.

[0084] In S301, it is determined whether the peak position X5a detected for the first time and the subsequently detected peak position X5b have the same number of pixels. If the peak positions X5a and X5b are at the same position, it is determined that the detected peak position is the true peak, and the process proceeds to the calculation and setting of the peak search range (S110). Then, based on the peak position X5b, the peak search range XL5c - XH5c is calculated and set in the peak position detector 14.

[0085] In FIG. 10(c) showing the subsequently calculated cross-correlation function, in order to perform maximum peak detection from the peak search range XL5c - XH5c, the peak P5c is detected. Here, since the peak P5d is outside the peak search range, it is not detected.

[0086] As described above, the search range can be set when the positions indicating the plurality of peaks of the sequentially obtained cross-correlation functions are the same. When the peak positions continue at the same position, there is a high possibility that it is the true peak, and it can be used as a determination condition for whether to set the peak search range.

[0087] In this embodiment, it is determined whether the first peak position X5a and the subsequently detected peak position X5b are at the same position. However, for the peak positions to be exactly at the same position, the measurement object needs to move at exactly the same speed. When the measurement object starts to move, the acceleration changes, so peaks do not occur at the same position. Also, since the peak positions are represented discretely, even if the measurement object is moving at the same speed, depending on the calculation results, it is conceivable that the peak position detected for the first time may shift forward or backward. Therefore, when determining whether the peak positions are at the same position, a range for determining the same position may be set.

[0088] Also, without making the condition that they occur continuously at the same position, the number of times may be integrated for each peak position within a certain period, and the integrated number of times may be used as a condition for setting the peak search range.

[0089] <Fourth Embodiment> In the third embodiment, the process of setting the peak search range based on the true peak position was described by providing conditions for limiting the peak search range. This aims to avoid the situation where the true peak position cannot be detected when the peak search range is set at a false peak position.

[0090] Ideally, the peak search range could continue to be set at the true peak position. However, if the true peak position occurs outside the peak search range due to some reason, it becomes difficult to detect the true peak position when the peak search range is limited.

[0091] Therefore, in the fourth embodiment, the processing when a true peak occurs outside the peak search range under the condition of limiting the peak search range will be described using the flowchart of FIG. 11. For the same processing as in the first to third embodiments, all reference numerals are changed in FIG. 11.

[0092] In S401, image data is acquired, and in S402, a cross-correlation function is calculated. Here, it is assumed that the reference image data is already set.

[0093] Next, in S403, it is determined whether the peak search range is set. Since the peak search range is not set at the first calculation of the cross-correlation function, the maximum peak is detected from the entire range (S404). Then, in S410, a length measurement value is calculated from the peak position, and the length measurement value is output (S413).

[0094] The peak search range is set based on the detected peak position (S415), and the reference image is updated in S417. The processing flow for limiting the peak search range up to this point is the same as the content described in the first embodiment.

[0095] Next, a cross-correlation function is calculated from the acquired image data. Since a peak search range is set in the determination of S403, the process proceeds to S405 to detect the maximum peak from the peak search range. The peak detected here is defined as the first peak. Subsequently, in S406, the maximum peak is detected from a range wider than the set search range, for example, the entire range. The peak detected over the entire range is defined as the second peak.

[0096] After the detection of the first peak and the second peak is completed, it is determined whether the first peak detected within the peak search range and the second peak detected over the entire range are at the same position (S407). If the first peak and the second peak are at the same position, a length measurement value is calculated based on the first peak detected within the peak search range (S411).

[0097] If the first peak and the second peak are not at the same position, different counts are incremented (S408), and it is determined in S409 whether the counted number exceeds a specified number. If the specified number is not exceeded, the process proceeds to S411, and a length measurement value is calculated based on the first peak detected within the peak search range.

[0098] If the number of times the first peak and the second peak are not at the same position exceeds the specified number, it is determined that the true peak has occurred outside the peak search range, and a length measurement value is calculated based on the second peak detected from the entire range (S412). As the specified number, for example, the count value can be at least 2 or more.

[0099] Thereafter, the calculated length measurement value is output (S414), and the set peak search range is set to the entire range (S416). This is because it is determined that the true peak position has moved outside the set peak search range, and by expanding the peak search range, the true peak can be detected again.

[0100] By the process described above, when a true peak occurs outside the peak search range due to some influence, it becomes possible to detect the true peak again by expanding the peak search range.

[0101] Although the preferred embodiments of the present invention have been described, the present invention is not limited to these embodiments, and various modifications and changes are possible within the scope of the gist. It is also possible to combine the signal processing methods in the above-described embodiments. Further, the calculation of the correlation degree of the cross-correlation function itself may be limited to a part of the entire range of the cross-correlation function, and the peak position may be detected from the cross-correlation function calculated by limiting it to a part of the range. The limited range can be determined based on the information described in the above embodiments such as the previous peak position. Further, a UI for making the search range of the peak position changeable may be provided. For example, on a display connected to a displacement meter, the search range of the peak position can be changed by a user inputting a numerical value or the like to an input interface for setting the search range of the peak position.

[0102] <Fifth Embodiment> Next, a method for manufacturing an article (such as a metal plate, a press-worked product, paper, fibers, etc.) using the above-described displacement meter will be described. The article is manufactured in a process of conveying an object (target object) by a conveying device, a process of measuring the conveyance amount of the conveyed object as displacement using the above-described displacement meter, and a process of performing processing such as cutting and pressing on the conveyed object at the timing when a desired conveyance amount is detected. Alternatively, it is also possible to detect the deviation from a predetermined value of the conveyance amount and stop the processing.

[0103] According to this article manufacturing method, since the measurement error by the displacement meter is reduced, it is possible to manufacture an article of higher quality than before.

Claims

1. A displacement meter that measures the displacement of a measurement object, an illumination system for illuminating the measurement object; a photoelectric conversion element array for detecting reflected light from the measurement object; a calculation unit that calculates a displacement amount of the measurement object using a cross-correlation function of a plurality of images detected at different times by the photoelectric conversion element array, the calculation unit limits a detection range in which a process of detecting a position indicating a peak of the cross-correlation function is performed to a portion of an entire range of the cross-correlation function, detects a position indicating the peak in the limited detection range, and calculates the amount of displacement based on the detected position indicating the peak; the calculation unit sets a detection range based on a position indicating a peak of the cross-correlation function used in a previous calculation of the displacement amount with respect to an entire range of the cross-correlation function, so as to include the position indicating the peak.

2. The displacement meter according to claim 1 , wherein the calculation unit changes the detection range.

3. 3. The displacement meter according to claim 1, wherein the calculation section sets the detection range based on information regarding a speed of the object to be measured.

4. 4. The displacement meter according to claim 1, wherein the calculation section sets the detection range based on information relating to an amount of displacement of the measurement object per sampling of the photoelectric conversion element array.

5. 5. The displacement meter according to claim 1, wherein the calculation unit detects a moving direction of the object to be measured, and sets the detection range so that a center of the detection range is on the moving direction side of the object to be measured with respect to a center of the cross-correlation function.

6. The calculation unit determines a reference image that serves as a reference from among the images detected by the photoelectric conversion element array, and calculates a cross-correlation function between the reference image and an image subsequently detected by the photoelectric conversion element array; 6. The displacement meter according to claim 1, wherein the detection range is determined based on a condition for updating the reference image.

7. 7. The displacement meter according to claim 1, wherein the calculation unit determines whether a peak value of the cross-correlation function exceeds a threshold value, and sets the detection range when the peak value of the cross-correlation function exceeds the threshold value.

8. 8. The displacement meter according to claim 1, wherein the calculation section sets the detection range when the positions showing a plurality of peaks of the cross-correlation function obtained sequentially are the same.

9. 9. The displacement meter according to claim 1, wherein the detection range is changed when a position indicating a peak of the cross-correlation function exceeds a threshold value.

10. The calculation unit detects a peak of the cross-correlation function in the limited detection range and in a range wider than the limited detection range, comparing a first position indicating a first peak detected in the limited detection range with a second position indicating a second peak detected in a range wider than the limited detection range; Counting the number of times the first position and the second position differ; 10. The displacement meter according to claim 1, wherein, when the count value reaches at least two times, the detection range is set to a range wider than the limited detection range.

11. A step of measuring a displacement amount of an object using the displacement meter according to any one of claims 1 to 10; and performing processing on the object based on the measured amount of displacement.

Images