WO2021179438A1 - 存在投影盲区去除绝对相位噪声方法、装置及存储介质 - Google Patents
存在投影盲区去除绝对相位噪声方法、装置及存储介质 Download PDFInfo
- Publication number
- WO2021179438A1 WO2021179438A1 PCT/CN2020/091748 CN2020091748W WO2021179438A1 WO 2021179438 A1 WO2021179438 A1 WO 2021179438A1 CN 2020091748 W CN2020091748 W CN 2020091748W WO 2021179438 A1 WO2021179438 A1 WO 2021179438A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- absolute phase
- value
- data
- noise
- straight line
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 45
- 238000001514 detection method Methods 0.000 claims description 20
- 230000015654 memory Effects 0.000 claims description 16
- 238000010586 diagram Methods 0.000 claims description 15
- 230000011218 segmentation Effects 0.000 claims description 4
- 238000012545 processing Methods 0.000 claims description 2
- 238000004364 calculation method Methods 0.000 abstract description 16
- 238000012937 correction Methods 0.000 abstract description 2
- 230000008569 process Effects 0.000 description 7
- 230000008859 change Effects 0.000 description 5
- 230000010363 phase shift Effects 0.000 description 3
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000001568 sexual effect Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
- G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
Definitions
- the invention belongs to the field of three-dimensional imaging, and in particular relates to a method, a device and a storage medium for removing absolute phase noise in the presence of a projection blind zone.
- the three-dimensional measurement equipment based on structured light fringe projection has the characteristics of high measurement accuracy, good real-time performance and non-contact, and has been widely used in different fields of life and industry.
- the structured light fringe projection is about projecting the designed fringe pattern onto a three-dimensional object. After the projection, the deformed fringe pattern will be obtained, and then the deformed fringe pattern will be calculated by the phase shift algorithm to obtain the wrapped phase, and then the wrapped phase will be restored to obtain the absolute
- the accuracy of phase and absolute phase affects the reconstruction accuracy of three-dimensional objects.
- Various factors such as the error of the measuring equipment, the interference of the external environment, and the limitation of the algorithm, will inevitably produce noise during the acquisition of the absolute phase, and the noise will affect the reconstruction accuracy of the three-dimensional object.
- most of the current absolute phase noise removal methods cannot filter out a large number of noise points, and the calculation complexity is high and the calculation amount is large.
- the present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention proposes a method for removing absolute phase noise with a blind zone of projection.
- the method for removing absolute phase noise with a blind zone of projection can reduce computational complexity, improve the efficiency of noise removal, and improve the reconstruction accuracy of three-dimensional objects.
- a method for removing absolute phase noise from projection blind areas includes the following steps: obtaining absolute phase value data from a fringe pattern; obtaining an absolute phase line according to the absolute phase data, and according to each absolute phase The straight line divides the corresponding effective area; determines the reference absolute phase straight line to be compared with each of the absolute phase straight lines; the absolute phase value of the absolute phase straight line passing through the same abscissa and the reference absolute phase of the reference absolute phase straight line
- the first difference value is obtained by the first difference value, and the first average value is obtained by the first difference value;
- the second absolute phase value is obtained by the absolute phase value data obtained in the corresponding effective area and the reference absolute phase value of the reference absolute phase line on the same abscissa.
- Two difference values and compare the second difference value with the first average value to determine a reference difference value; preset a threshold range, compare the reference difference value with the second difference value, and divide the absolute phase value data Noise points and non-noise points; after removing the noise points, the absolute phase straight line is corrected to obtain corrected absolute phase value data, and three-dimensional object reconstruction is performed.
- a method for removing absolute phase noise with projection blind spots has at least the following beneficial effects: the present invention uses absolute phase straight lines and reference absolute phase straight lines to judge the acquired absolute phase value data through difference calculations, And filter out the noise points and non-noise points of the absolute phase value data, effectively eliminate the noise points of the absolute phase value, the calculation complexity is low, the calculation is simple, and the denoising efficiency is high. At the same time, the absolute phase value of the non-noise points can be corrected. Improving the accuracy of the absolute phase value and avoiding the influence of noise can improve the reconstruction accuracy of three-dimensional objects.
- “obtaining absolute phase value data from a fringe pattern” includes the following steps: acquiring wrapped phase data from the fringe pattern; and restoring the wrapped phase data to the absolute phase data.
- "obtaining the absolute phase line according to the absolute phase data, and dividing the effective area according to each absolute phase line” includes the following steps: dividing and marking the absolute phase value data; The absolute phase straight line; according to each absolute phase straight line, the corresponding effective area is divided.
- dividing and marking absolute phase value data includes the following steps: detecting valid data and invalid data in the absolute phase data; marking valid data and invalid data respectively; dividing the absolute phase according to valid data and invalid data data.
- the valid data is a non-zero absolute phase value
- the invalid data is a zero absolute phase value
- determining the reference absolute phase straight line to be compared with each of the absolute phase straight lines includes the following steps: determining the reference absolute phase straight line according to the number of valid data in the absolute phase data and through straight line detection .
- the straight line detection adopts Hall transform straight line detection.
- "after removing the noise points, and correcting the absolute phase line to obtain the correct absolute phase value, and performing three-dimensional object reconstruction” includes the following steps:
- the absolute phase value data of the non-noise point and the reference absolute phase value of the reference absolute phase line on the same abscissa obtain the third difference value, and obtain the second average value through the third difference value, and set the second average value as the reference Distance value; correct the absolute phase line according to the reference distance value and the reference absolute phase line; obtain the correct absolute phase value through the corrected absolute phase line, and perform three-dimensional object reconstruction.
- a device for removing absolute phase noise in the presence of a projection blind zone comprises: at least one processor and a memory for communicating with the processor; the memory can be stored by An instruction executed by the at least one processor, the instruction being executed by the at least one processor, so that the processor can execute the method for removing absolute phase noise with a projection blind zone as described above.
- a computer-readable storage medium is characterized in that the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to cause a computer to execute the above The existence of the projection blind zone to remove the absolute phase noise method.
- Fig. 1 is a flow chart of the method for removing absolute phase noise in the presence of a projection blind zone according to the present invention.
- Fig. 2 is a flow chart of obtaining absolute phase value data from a fringe pattern according to the present invention.
- Fig. 3 is a schematic diagram of a deformed fringe pattern of the present invention.
- Fig. 4 is a schematic diagram of the wrapping phase without projection blind zone according to the present invention.
- Fig. 5 is a schematic diagram of the absolute phase of the present invention without a projection blind zone.
- Fig. 6 is a flow chart of obtaining the absolute phase line according to the absolute phase data according to the present invention, and dividing the corresponding effective area according to each absolute phase line.
- Fig. 7 is a flowchart of the absolute phase value data segmentation and labeling of the present invention.
- FIG. 8 is a schematic diagram of the absolute phase of the present invention with noise points and projection blind areas.
- FIG. 9 is a schematic diagram of the absolute phase of the present invention with noise points removed and projection blind areas.
- FIG. 10 is a schematic diagram of the absolute phase of a three-dimensional object with noise points and projection blind areas according to the present invention.
- FIG. 11 is a schematic diagram of the absolute phase of the three-dimensional object with the noise points removed and the projection blind zone of the present invention.
- Fig. 12 is a flow chart of correcting the absolute phase straight line equation of the present invention.
- FIG. 13 is a schematic diagram of the structure of a device for removing absolute phase noise with a projection blind zone according to the present invention.
- the present invention provides a method for removing absolute phase noise in the presence of a projection blind zone, which includes the following steps:
- Step S100 Obtain absolute phase value data from the fringe pattern
- Step S200 Obtain an absolute phase line according to the absolute phase data, and divide a corresponding effective area according to each absolute phase line;
- Step S300 determining a reference absolute phase line to be compared with each absolute phase line respectively;
- Step S400 Obtain a first difference value through the absolute phase value of the absolute phase line on the same abscissa and the reference absolute phase value of the reference absolute phase line, and obtain a first average value through the first difference;
- Step S500 Obtain a second difference value through the absolute phase value data acquired in the corresponding effective area and the reference absolute phase value of the reference absolute phase line on the same abscissa, and compare the second difference value with the first average value, Determine the reference difference;
- Step S600 preset a threshold value range, compare the reference difference value with the second difference value, and divide the noise points and non-noise points of the absolute phase value data;
- step S700 after removing the noise points, the absolute phase straight line is corrected to obtain the corrected absolute phase value data, and the three-dimensional object is reconstructed.
- the invention judges the acquired absolute phase value data through the absolute phase line and the reference absolute phase line through the difference calculation, and filters out the noise points and non-noise points of the absolute phase value data, and effectively eliminates the noise points of the absolute phase value.
- the calculation complexity is low, the calculation is simple, and the noise removal efficiency is high.
- the absolute phase value of the non-noise point is corrected, which can improve the accuracy of the absolute phase value, avoid the influence of noise, and improve the reconstruction accuracy of three-dimensional objects.
- step S100 absolute phase value data is obtained from the fringe pattern. Specifically, it also includes the following steps:
- Step S110 obtaining package phase data from the fringe pattern
- step S120 the package phase data is restored to absolute phase data.
- the fringe pattern is a graph whose amplitude range of longitudinal sinusoidal change is 0-1, and is composed of multiple periods.
- several fringe patterns are used, and several fringe patterns are graphs of sinusoidal changes of two wavelengths with different phase shifts.
- Project the designed fringe pattern on a three-dimensional object to obtain a deformed fringe pattern (due to the height of the three-dimensional object, the fringe pattern will deform after projection).
- the three-dimensional object is a step
- the deformed fringe pattern as shown in FIG. 3 is obtained, and the projection blind area is shown by arrow A in FIG. 3.
- the deformed fringe pattern is calculated by the phase shift algorithm to obtain the wrapped phase; then the dual-wavelength phase unwrapping algorithm is used to unwrap the wrapped phase data and restore it to absolute phase data.
- the absolute phase data can be used to reconstruct three-dimensional objects.
- the process of obtaining the wrapped phase value data is: for example, using 6 fringe patterns with different phase differences under the same wavelength, and then projecting the fringe pattern onto a three-dimensional object, and using the acquired six deformed fringe patterns to perform phase shifting algorithm Solve.
- the sine change rule of the projected fringe pattern is:
- I pi and I ci are pixel values
- I c is an average gray value
- I'c is a modulation value
- ⁇ p (x p , y p ) and ⁇ c (x c , y c ) are phase values. More specifically, I pi is the pixel value of the fringe pattern, and I ci is the pixel value of the deformed fringe pattern.
- the package phase value can be calculated.
- the graph of the wrapping phase can be obtained as shown in Figure 4.
- the ordinate is the wrapping phase value
- the abscissa is the row coordinate.
- Fig. 4 shows a schematic diagram of the wrapping phase without the projection blind zone. The figure shown in Fig. 4 can be obtained through the above calculation, and the figure of the wrapping phase is in a sawtooth shape. When there is a projection blind zone, the graph of the wrapped phase that can be obtained by the above calculation is also sawtooth, but the graph of the wrapped phase is segmented.
- the process of acquiring the absolute phase value data is: using a dual-wavelength phase unwrapping algorithm, that is, using the difference between the wrapped phases of the two wavelengths, and solving the problem by looking up a table.
- T 1 and T 2 are two kinds of wavelengths, and k 1 and k 2 are the K values (that is, the fringe order) of the absolute phase of the corresponding wavelengths.
- ⁇ c1 (x c , y c ) and ⁇ c2 (x c , y c ) are the wrapped phase values of different wavelengths.
- T 1 and T 2 use 23 and 47 as an example.
- a certain point with the same coordinate as the phase value of the two packages is subtracted, and the difference is rounded to the third value in Table 1.
- the absolute phase value data of each column in the image can be obtained.
- the graph of absolute phase can be obtained as shown in Fig. 5.
- the ordinate is the absolute phase value
- the abscissa is the row coordinate.
- the diagram shown in FIG. 5 is a schematic diagram of the absolute phase without the projection blind zone.
- the graph shown in FIG. 5 can be obtained through the above calculation, and the graph of the absolute phase is linear as a whole.
- the absolute phase figure that can be obtained by the above calculation is also linear, but the absolute phase figure has segments, and the absolute phase figure with the projection blind zone is shown in Figs. 8 and 9.
- the "column” here refers to a column of data that is consistent with the direction of the sine change of the fringe pattern. Since the captured stripes change vertically, in this embodiment, the “column” can also be understood It is a column of pixels in the vertical direction of the image. The “row” in the row coordinates corresponds to the “column”. It can be understood that each row in the image, that is, the row coordinates are expressed as changes in the number of rows of the image.
- step S200 an absolute phase line is obtained according to the absolute phase data, and a corresponding effective area is divided according to each absolute phase line. Specifically, it also includes the following steps:
- Step S210 dividing and marking the absolute phase value data
- Step S220 Obtain a segmented absolute phase straight line after segmentation
- step S230 the corresponding effective area is divided according to each absolute phase line.
- step S210 is to divide and mark the absolute phase value data. Specifically, it also includes the following steps:
- Step S211 Detect valid data and invalid data in the absolute phase data
- Step S212 marking valid data and invalid data respectively
- Step S213 dividing the absolute phase data according to the valid data and the invalid data.
- the valid data is a non-zero absolute phase value
- the invalid data is a zero absolute phase value
- the absolute phase data of a certain column of the image is ooxxooxxoo, and then the absolute phase of the column data is divided by detecting invalid data x, and marked as 1100220033, invalid data is marked with 0, and valid data is marked with 1, 2, 3... ( It can be used to represent the segmentation of the line of absolute phase lines).
- the line of absolute phase lines is divided into three segments.
- the emergence of the projection blind zone is due to the placement of the camera, projection, and three-dimensional objects.
- the part that cannot be covered by the fringe pattern is called This is the projection blind zone, and due to the existence of the projection blind zone, the absolute phase data of the same column will be segmented. Therefore, it is necessary to divide and label the data. After the division, the absolute phase value data of the same column appears as a segmented straight line, that is, the absolute phase line.
- the absolute phase value data of the non-noise point will be located on the absolute phase line.
- the absolute phase value data of the noise point will be outside the absolute phase straight line.
- the slope of the absolute phase line of each column is the same, but the intercept is different.
- the absolute phase line of the column is divided into three segments. According to this, the absolute phase line divides the column into three effective areas, and the effective data of the absolute phase falls respectively. In the three effective areas, the effective absolute phase data of the non-noise points are located on the corresponding absolute phase line, and the effective absolute phase data of the noise points are outside the corresponding absolute phase line. The division of the effective area of the remaining columns in the image is the same as above. After the effective data and invalid data are divided and marked, the number of absolute phase lines in the column is determined, and the effective areas corresponding to the absolute phase lines are divided. The absolute phase value data obtained in, falls into the corresponding effective area. It is worth noting that in the present invention, the "effective area" is proposed to make the subsequent calculation of eliminating noise points clearer and more intuitive, and to divide the data.
- step S300 determine the reference absolute phase line to be compared with each absolute phase line respectively.
- the method includes the following steps: determining the reference absolute phase straight line through straight line detection according to the number of valid data in the absolute phase value data.
- the process of determining the reference absolute phase line is as follows: in each column, look for the column with the most valid data in the absolute phase value data, select the column with the most valid absolute phase value data, and extract the data with the greatest data continuity. The probability is higher, the accuracy is higher, and the accuracy of the noise removal point can be improved, thereby improving the accuracy of the absolute phase.
- the longest absolute phase line in the column is detected in a straight line, and the endpoint coordinates (x 1 , y 1 ) and (x 1, y 1) and (x 2 ,y 2 ), find the equation of a straight line through the two end points,
- the reference absolute phase straight line equation y n k ⁇ x n +b can be obtained, where y n is the reference phase value, and x n is the row coordinate corresponding to the reference phase value. Based on this, the reference absolute phase straight line is determined, and the reference absolute phase straight line equation is obtained.
- the straight line detection adopts the Hall transform line detection.
- the Hall transform line detection can detect all points on the line. Therefore, the line with the most absolute phase data can be detected by the Hall transform line detection, and Detect the endpoint coordinates of the two ends of the longest straight line.
- the Hall transform linear detection has the characteristics of fast detection speed and high accuracy. It can quickly determine the linear equation with reference to the absolute phase straight line, improve the efficiency of noise removal, and can improve the accuracy of the absolute phase.
- the line with the most valid data can also be detected by the LSD line detection algorithm, FLD line detection algorithm, EDlines line detection algorithm, and LSM line detection algorithm.
- other straight line detection algorithms can also be used for straight line detection.
- step S400 the first difference is obtained by the absolute phase value of the absolute phase line on the same abscissa and the reference absolute phase value of the reference absolute phase line, and the first average value is obtained by the first difference.
- step S500 the second difference value is obtained through the absolute phase value data obtained in the corresponding effective area and the reference absolute phase value of the reference absolute phase line of the same abscissa, and the second difference value is compared with the first A mean value is compared to determine the reference difference. It specifically includes the following process: in the effective area corresponding to the absolute phase line selected in step S400, there are several effective absolute phase value data, and the obtained absolute phase value data in the effective area is referenced with the same abscissa.
- the second difference is a difference R n, each of R n second difference compared with the first average value d mean, the difference from the second R a first average value of n, to find the closest second difference value d mean R n, and R n with the second difference value as a reference difference.
- the number of absolute phase value data of non-noise points is much larger than the number of absolute phase value data of noise points.
- the calculated first average value d mean can be regarded as the absolute phase of a non-noise point
- step S600 preset a threshold value range, compare the reference difference value with the second difference value, and divide the noise points and non-noise points of the absolute phase value data.
- the second difference value includes R 1 , R 2 , R 3 , R 4 , and R 5 , where the second difference value closest to the first mean value d mean is R 3 , then the second difference value is set to R 3 It is the reference difference.
- the reference difference is compared with the second difference, and the noise points and non-noise points of the absolute phase value data are divided according to the preset threshold range.
- the threshold range is the absolute value of the difference between the reference difference and the second difference is less than 1.
- the absolute phase value data corresponding to the second difference is a non-noise point
- the threshold range is the reference difference and the second difference.
- the absolute phase value data corresponding to the second difference value whose absolute value of the value difference is greater than 1 is a noise point.
- ⁇ 1 that is, the absolute phase value data corresponding to the second difference R 1 is a non-noise point
- > 1 that is, the second difference R 4 corresponds to
- the method of removing the absolute phase noise of the projection blind zone of the present invention is to separately compare the absolute phase line of each segment with the reference absolute phase line to calculate the noise point and the non-noise point of the absolute phase line of each segment, and Correct the absolute phase straight line of each segment.
- Figure 8 is a schematic diagram of a certain column of absolute phase before noise removal. It can be seen that due to the existence of the projection blind zone, the absolute phase line is segmented. The distance is different, and at the same time, there is the absolute phase value data of the noise point on the graph.
- a continuous straight line shown in Fig. 8 is the reference absolute phase straight line.
- 9 is a schematic diagram of the absolute phase after the noise is removed by the method of removing the absolute phase noise by the method for removing the absolute phase noise of the existing projection blind zone in FIG. 8.
- a continuous straight line shown in FIG. 9 is the reference absolute phase straight line. It can be seen from the denoising results of Figs. 8 and 9 that the method for removing absolute phase noise with projection blind areas of the present invention can effectively remove noise points, improve the accuracy of absolute phase value data, and then improve the accuracy of three-dimensional reconstruction.
- Figure 10 is the absolute phase image of the entire three-dimensional object before denoising.
- the gray value of some pixels is The gray values of nearby pixels are different
- Figure 11 is the absolute phase image of the entire three-dimensional object after denoising.
- the gray value of the pixel is the same as the gray value of the nearby pixels, that is, noise
- the absolute phase noise is effectively removed by the method of removing the absolute phase noise that exists in the projection blind zone of the present invention.
- step S700 after removing the noise points, the absolute phase line is corrected to obtain the correct absolute phase value, and the three-dimensional object is reconstructed. Specifically, it includes the following steps:
- Step S710 Obtain a third difference value through the absolute phase value data of the non-noise points in the effective area and the reference absolute phase value of the reference absolute phase line on the same abscissa, and obtain the second average value through the third difference value, and combine The second average value is set as the reference distance value;
- Step S720 correcting the absolute phase straight line according to the reference distance value and the reference absolute phase straight line
- step S730 the correct absolute phase value is obtained through the corrected absolute phase straight line, and the three-dimensional object is reconstructed.
- the reference distance value d c is used to correct the absolute phase straight line.
- the reference distance value d c is the absolute phase of the non-noise section of the column and the corresponding horizontal line.
- the average value of the difference value of the reference phase value of the coordinate, the average value can make the difference value closer to the correct value, and can improve the accuracy of the absolute phase straight line correction.
- the absolute phase obtained by the above process is the absolute phase with noise.
- the method for removing the absolute phase noise from the projection blind zone of the present invention is required to perform denoising. After denoising, the absolute phase is corrected to obtain the correct absolute phase. Phase, and then reconstruct the three-dimensional object through the corrected absolute phase value data, which can effectively improve the accuracy of the three-dimensional object reconstruction.
- an embodiment of the present invention also provides a device for removing absolute phase noise with a projection blind zone.
- the device for removing absolute phase noise with a projection blind zone can be any type of smart terminal, such as a mobile phone, a tablet, a personal computer, and so on.
- the device for removing absolute phase noise in the presence of a projection blind zone includes: one or more processors and a memory for communicating with the processor.
- a processor is taken as an example in FIG. 13.
- the processor and the memory may be connected by a bus or other methods.
- FIG. 13 uses a bus connection as an example.
- the memory can be used to store non-transitory software programs and non-transitory computer executable programs, such as the program instructions corresponding to the device for removing absolute phase noise with projection blind spots in the embodiment of the present invention .
- the processor executes the non-transient software program and instructions stored in the memory to realize the above-mentioned method for removing the absolute phase noise with the projection blind zone.
- the memory may include a storage program area and a storage data area, where the storage program area can store an operating system and an application program required by at least one function; the storage data area can store relevant data related to the above method for removing absolute phase noise with the projection blind area.
- the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one magnetic disk storage device, a flash memory device, or other non-transitory solid-state storage devices.
- the memory may optionally include a memory remotely provided with respect to the processor, and these remote memories may be connected to the device for removing absolute phase noise from the projection blind area through a network. Examples of the aforementioned networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.
- a computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are executed by one or more processors, for example, executed by one processor,
- the aforementioned one or more processors can be made to execute the aforementioned method for removing absolute phase noise with a projection blind zone.
- the device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules can be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
- computer storage medium includes volatile and non-volatile data implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data).
- Information such as computer-readable instructions, data structures, program modules, or other data.
- Computer storage media include but are not limited to RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disk (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or Any other medium used to store desired information and that can be accessed by a computer.
- communication media usually contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as carrier waves or other transmission mechanisms, and may include any information delivery media. .
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Computer Graphics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
一种存在投影盲区去除绝对相位噪声方法，包括以下步骤：从条纹图中获取绝对相位值数据；根据绝对相位数据获取绝对相位直线，并划分对应的有效区域；确定参考绝对相位直线；通过同一横坐标的绝对相位直线与参考绝对相位直线获取第一差值，并获取第一平均值；通过有效区域内获取到的绝对相位值数据与同一横坐标的参考绝对相位直线获取第二差值，比较第二差值与第一平均值，确定参考差值；预设阈值范围，比较参考差值与第二差值，划分噪声点和非噪声点；剔除噪声点后，修正绝对相位直线，以得到修正后的绝对相位值数据，进行三维物体重构。所述方法计算复杂度低，去噪声效率高，能够提高绝对相位的准确度和三维物体的重构精度。
Description
本发明属于三维成像领域，尤其涉及一种存在投影盲区去除绝对相位噪声方法、装置以及存储介质。
基于结构光条纹投影的三维测量设备具有测量精度高、实时性好以及非接触性等特点，在生活以及工业的不同领域中得到了广泛的应用。结构光条纹投影即将设计好的条纹图投影到三维物体上，投影后，会得到变形的条纹图，再利用相移算法对变形条纹图进行计算以得到包裹相位，再通过包裹相位恢复从而得到绝对相位，绝对相位的精度影响着三维物体的重构精度。由于测量设备的误差、外界环境的干扰、算法的局限性等各种因素都会导致在绝对相位的获取过程中不可避免地产生噪声，而噪声会影响三维物体的重构精度。但目前大多数地绝对相位噪声去除方法无法滤除大量地噪声点，并且计算复杂度高且计算量大。
发明内容
本发明旨在至少解决现有技术中存在的技术问题之一。为此，本发明提出一种存在投影盲区去除绝对相位噪声方法，该存在投影盲区去除绝对相位噪声方法能够降低计算复杂度，并提高去除噪声的效率，提高三维物体的重构精度。
根据本发明的第一方面实施例的一种存在投影盲区去除绝对相位噪声方法，包括以下步骤：从条纹图中获取绝对相位值数据；根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分对应的有效区域；确定分别与每条所述绝对相位直线相比较的参考绝对相位直线；通过同一横坐标的所述绝对相位直线的绝对相位值与所述参考绝对相位直线的参考绝对相位值获取第一差值，并通过所述第一差值获取第一平均值；通过对应的有效区域内获取到的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获得第二差值，并将所述第二差值与所述第一平均值相比较，确定参考差值；预设阈值范围，将参考差值与第二差值相比较，划分绝对相位值数据的噪声点 和非噪声点；剔除所述噪声点后，并修正所述绝对相位直线，以得到修正后的绝对相位值数据，并进行三维物体重构。
根据本发明实施例的一种存在投影盲区去除绝对相位噪声方法，至少具有如下有益效果：本发明通过绝对相位直线以及参考绝对相位直线，通过差值运算对获取到的绝对相位值数据进行判断，并筛选出绝对相位值数据的噪声点和非噪声点，有效剔除绝对相位值的噪声点，计算复杂度低，运算简单，去噪效率高，同时对非噪声点的绝对相位值进行修正，能够提高绝对相位值得准确度，避免噪声影响，能够提高三维物体的重构精度。
根据本发明的一些实施例，“从条纹图中获取绝对相位值数据”包括以下步骤：从条纹图中获取包裹相位数据；将所述包裹相位数据恢复成所述绝对相位数据。
根据本发明的一些实施例，“根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分有效区域”包括以下步骤：分割与标记所述绝对相位值数据；分割后获得分段的所述绝对相位直线；根据每条所述绝对相位直线划分对应的有效区域。
根据本发明的一些实施例，“分割与标记绝对相位值数据”包括以下步骤：检测绝对相位数据中的有效数据和无效数据；分别标记有效数据和无效数据；根据有效数据和无效数据分割绝对相位数据。
根据本发明的一些实施例，所述有效数据为非0的绝对相位值，所述无效数据为0的绝对相位值。
根据本发明的一些实施例，“确定分别与每条所述绝对相位直线相比较的参考绝对相位直线”包括以下步骤：根据绝对相位数据中的有效数据的数量并通过直线检测确定参考绝对相位直线。
根据本发明的一些实施例，所述直线检测采用霍尔变换直线检测。
根据本发明的一些实施例，“剔除所述噪声点后，并修正所述绝对相位直线，以得到正确的绝对相位值，并进行三维物体重构”包括以下步骤：通过所述有效区域内的非噪声点的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获取第三差值，并通过第三差值获取第二平均值，并将第二平均值设定为参考距离值；根据所述参考距离值和参考绝对相位直线修正绝对相位直线；通过修正后的绝对相位直线获取正确的绝对相位值，并进行三维物体重构。
根据本发明的第二方面实施例的一种存在投影盲区去除绝对相位噪声装置，其特征在于，包括：至少一个处理器和用于与所述处理器通信连接的存储器；所述存储器存储可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述处理器能够执行如上所述的存在投影 盲区去除绝对相位噪声方法。
根据本发明的第三方面实施例的一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机可执行指令，所述计算机可执行指令用于使计算机执行如上所述的存在投影盲区去除绝对相位噪声方法。
本发明的附加方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本发明的实践了解到。
本发明的上述和/或附加的方面和优点从结合下面附图对实施例的描述中将变得明显和容易理解，其中：
图1为本发明的存在投影盲区去除绝对相位噪声方法的流程图。
图2为本发明的从条纹图中获取绝对相位值数据的流程图。
图3为本发明的变形条纹图的示意图。
图4为本发明的不存在投影盲区的包裹相位的图形示意图。
图5为本发明的不存在投影盲区的绝对相位的图形示意图。
图6为本发明的根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分对应的有效区域的流程图。
图7为本发明的绝对相位值数据分割与标记的流程图。
图8为本发明的存在噪声点且存在投影盲区的绝对相位示意图。
图9为本发明的去除噪声点且存在投影盲区的绝对相位示意图。
图10为本发明的存在噪声点且存在投影盲区的三维物体绝对相位示意图。
图11为本发明的去除噪声点且存在投影盲区的三维物体绝对相位示意图。
图12为本发明的修正绝对相位直线方程的流程图。
图13为本发明的存在投影盲区去除绝对相位噪声装置的结构示意图。
下面详细描述本发明的实施例，所述实施例的示例在附图中示出，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。下面通过参考附图描述的实施例是示例 性的，仅用于解释本发明，而不能理解为对本发明的限制。
在本发明的描述中，需要理解的是，涉及到方位描述，例如上、下、前、后、外、内等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。
在本发明的描述中，若干的含义是一个或者多个，多个的含义是两个以上，大于、小于、超过等理解为不包括本数，以上、以下、以内等理解为包括本数。如果有描述到第一、第二只是用于区分技术特征为目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量或者隐含指明所指示的技术特征的先后关系。
本发明的描述中，除非另有明确的限定，设置、安装、连接等词语应做广义理解，所属技术领域技术人员可以结合技术方案的具体内容合理确定上述词语在本发明中的具体含义。
为了对本发明的技术特征、目的和效果有更加清晰的理解，先对照附图详细说明本发明的具体实施方式。
请参考图1，本发明提供一种存在投影盲区去除绝对相位噪声方法，包括以下步骤：
步骤S100,从条纹图中获取绝对相位值数据；
步骤S200,根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分对应的有效区域；
步骤S300,确定分别与每条绝对相位直线相比较的参考绝对相位直线；
步骤S400,通过同一横坐标的绝对相位直线的绝对相位值与参考绝对相位直线的参考绝对相位值获取第一差值，并通过第一差值获取第一平均值；
步骤S500,通过对应的有效区域内获取到的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获取第二差值，并将第二差值与第一平均值相比较，确定参考差值；
步骤S600,预设阈值范围，将参考差值与第二差值相比较，划分绝对相位值数据的噪声点和非噪声点；
步骤S700,剔除噪声点后，并修正绝对相位直线，以得到修正后的绝对相位值数据，并进行三维物体重构。
本发明通过绝对相位直线以及参考绝对相位直线，通过差值运算对获取到的绝对相位值数据进行判断，并筛选出绝对相位值数据的噪声点和非噪声点，有效剔除绝对相位值的噪声点，计算复杂 度低，运算简单，去噪效率高，同时对非噪声点的绝对相位值进行修正，能够提高绝对相位值得准确度，避免噪声影响，能够提高三维物体的重构精度。
请参考图2，在本发明的一些实施例中，步骤S100,从条纹图中获取绝对相位值数据。具体地，还包括以下步骤：
步骤S110，从条纹图中获取包裹相位数据；
步骤S120,将包裹相位数据恢复成绝对相位数据。
其中，条纹图为纵向正弦变化的幅值范围为0-1的图，由多个周期构成。在三维物体重构的过程中会采用若干张条纹图，若干张条纹图为两种波长不同相位移的正弦变化的图。将设计好的条纹图投影在三维物体上，得到变形的条纹图(由于三维物体存在高度，条纹图投影后会发生变形)，请参考图3，在本实施例中，三维物体为一个呈阶梯状的阶梯模型，当条纹图投影至阶梯模型时，得到如图3所示的变形条纹图，其中投影盲区如图3中的箭头A所示。通过相移算法对变形的条纹图进行计算，以得到包裹相位；再利用双波长相位展开算法，将包裹相位数据去包裹恢复成绝对相位数据。通过绝对相位数据能够对三维物体进行重构。
具体地，包裹相位值数据的获取过程为：如采用同一波长下6张不同相位差的条纹图，然后将条纹图投影到三维物体上，利用获取到的六张变形条纹图通过相移算法进行求解。
所投影的条纹图正弦变化规律为：
变形条纹图所对应的点的变化规律为：
两公式代入得到包裹相位值为：
其中，I
pi和I
ci为像素值，I
c为平均灰度值，I'
c为调制值，Φ
p(x
p,y
p)和Φ
c(x
c,y
c)为相位值。更具体地，I
pi为条纹图的像素值，而I
ci为变形条纹图的像素值。
依据上述算法能够计算出包裹相位值。在进行上述计算后，可得到包裹相位的图形如图4所示，在图4中，纵坐标为包裹相位值，横坐标为行坐标。在图4中所示的为不存在投影盲区的包裹相位的示意图，通过上述计算能够得到如图4所示图形，包裹相位的图形呈锯齿状。当存在投影盲区的情况下，通过上述计算能够得到的包裹相位的图形同样呈锯齿状，但是包裹相位的图形存在分段。
计算出包裹相位值后需要对包裹相位值去包裹恢复成绝对相位值。
具体地，绝对相位值数据的获取过程为：通过双波长相位展开算法，即利用两种波长的包裹相位的差值，通过查表进行求解。
通过以下公式对绝对相位数据进行求解：
其中，T
1，T
2为两种波长，k
1，k
2即为对应波长的绝对相位的K值(即条纹阶数)。φ
c1(x
c,y
c)和φ
c2(x
c,y
c)为不同波长的包裹相位值。
例如，在本实施例中，T
1，T
2采用23与47为例，依据上述公式，两包裹相位值相同坐标的某点进行相减，其差值取整，并与表1中第三列数据相匹配，找出该点对应波长的K值。
| k 1(x c) | k 2(x c) | k 2(x c)T 2-k 1(x c)T 1 |
| 0 | 0 | 0 |
| 1 | 0 | -23 |
| 2 | 0 | -46 |
| 2 | 1 | 1 |
| 3 | 1 | -22 |
| 4 | 1 | -45 |
| … | … | … |
| 34 | 17 | 17 |
表1
通过上述算法能够获得图像中每一列的绝对相位值数据。在进行上述计算后，可得到绝对相位的图形如图5所示，在图5中，纵坐标为绝对相位值，横坐标为行坐标。在图5中所示的为不存在投影盲区的绝对相位的示意图，通过上述计算能够得到如图5所示图形，绝对相位的图形整体呈直线状。当存在投影盲区的情况下，通过上述计算能够得到的绝对相位的图形同样呈直线状，但是绝对相位的图形存在分段，存在投影盲区的绝对相位的图形如图8及图9所示。
在本实施例中，此处的“列”是指与条纹图正弦变化方向一致的一列数据，由于所拍摄的条纹是呈垂直变化的，因此，在本实施例中，“列”也可以理解为图像垂直方向上像素点构成的一列。而行坐标中的“行”则与“列”相对应，可以理解图像中的每一行，即行坐标则表示为图像行数的变化。
请参考图6，在本发明的一些实施例中，步骤S200，根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分对应的有效区域。具体地，还包括以下步骤：
步骤S210,分割与标记绝对相位值数据；
步骤S220,分割后获得分段的绝对相位直线；
步骤S230,根据每条绝对相位直线划分对应的有效区域。
请参考图7，在本发明的一些实施例中，步骤S210，分割与标记绝对相位值数据。具体地，还包括以下步骤：
步骤S211,检测绝对相位数据中的有效数据和无效数据；
步骤S212,分别标记有效数据和无效数据；
步骤S213,根据有效数据和无效数据分割绝对相位数据。
在本发明的一些具体实施例中，有效数据为非0的绝对相位值，而无效数据则为0的绝对相位值。
更具体地，例如，假设o为有效数据，x为无效数据。图像某列的绝对相位数据为ooxxooxxoo，再通过检测无效数据x对该列数据的绝对相位进行分割，并标记成1100220033，无效数据用0标记，有效数据用1、2、3……进行标记(可以用于表示该列绝对相位直线的分段情况)，在本例子中，该列的绝对相位直线分割成三段。
其中，投影盲区的出现是由于相机、投影以及三维物体的摆放位置，会存在条纹图无法覆盖的部分，但这个无法覆盖的部分却被相机所拍摄，因此，将条纹图无法覆盖的部分称之为投影盲区，而由于投影盲区的存在，会导致同一列的绝对相位数据出现分段现象。因此，需要进行数据的分割与标记，分割后，同一列的绝对相位值数据呈现出分段的直线状，即为绝对相位直线，非噪声点的绝对相位值数据会位于绝对相位直线上，而噪声点的绝对相位值数据会位于绝对相位直线外。而每一列的绝对相位直线的斜率均相同，截距不同。
在上述提出的绝对相位值数据的分割与标记的例子中，该列的绝对相位直线分割成三段，据此，绝对相位直线将该列的分成三个有效区域，绝对相位的有效数据分别落入三个有效区域内，而非噪声点的有效的绝对相位数据位于对应的绝对相位直线上，而噪声点的有效的绝对相位数据则在对应的绝对相位直线外。图像中其余列的有效区域的划分同上，均是根据有效数据和无效数据的分割与标记后，确定该列的绝对相位直线的条数后，划分与绝对相位直线对应的有效区域，从条纹图中获取到的绝对相位值数据则是落入对应的有效区域内。值得注意的是，在本发明中，“有效区域“的提出是为了让后续剔除噪声点的计算更加清晰直观，对数据进行的一个划分。
请参考图1、图6及图7，步骤S300,确定分别与每条绝对相位直线相比较的参考绝对相位直线。具体地，包括以下步骤：根据绝对相位值数据中的有效数据的数量并通过直线检测确定参考绝对相位直线。
更具体地，确定参考绝对相位直线的过程如下：在每一列中寻找绝对相位值数据中的有效数据最多的一列，选取有效的绝对相位值数据最多的一列，提取到数据连续性最大的数据的几率较大，准确度较高，能够提高去噪声点的准确度，从而提高绝对相位的精度。在筛选出有效的绝对相位值数据最多的一列后，对该列中最长的绝对相位直线进行直线检测，检测最长的绝对相位直线两侧的端点坐标(x
1,y
1)和(x
2,y
2)，通过两端点求直线方程，
可得到参考绝对相位直线方程y
n＝k×x
n+b，其中y
n为参考相位值，x
n为参考相位值对应的行坐标。据此，确定了参考绝对相位直线，并得到参考绝对相位直线方程。
由于绝对相位直线的斜率相同而截距不同，绝对相位直线方程可设为：y
mn＝k×x
n+b
m，其中y
mn为该段的绝对相位值，x
n为参考相位值对应的行坐标，绝对相位直线的斜率k与参考绝对相位直线的斜率k相同，而截距各不相同，而截距b
m则等于包裹相位值φ和条纹阶数K值之和。
在本发明的一些具体实施例中，直线检测采用霍尔变换直线检测，霍尔变换直线检测能够检测直线上所有的点，因此，能够通过霍尔变换直线检测检测绝对相位数据最多的一列，并检测最长直线的两端的端点坐标。霍尔变换直线检测具有检测速度快以及准确率高的特点，能够迅速确定参考绝对相位直线的直线方程，提高去噪声效率，并且能够提高绝对相位的准确度。除了采用霍尔变换直线检测对有效数据最多的一列进行直线检测外，还可以通过LSD直线检测算法、FLD直线检测算法、EDlines直线检测算法以及LSM直线检测算法对有效数据最多的一列进行直线检测，当然，还可以采用其他的直线检测算法进行直线检测。
请参考图1，步骤S400，通过同一横坐标的绝对相位直线的绝对相位值与参考绝对相位直线的参考绝对相位值获取第一差值，并通过第一差值获取第一平均值。具体包括以下过程：先选取某一列中的某一分段绝对相位直线，该段绝对相位直线对应一个有效区域，通过绝对相位直线方程y
mn＝k×x
n+b
m以及参考绝对相位直线方程y
n＝k×x
n+b计算对应横坐标的第一差值，即第一差值为d
n＝y
mn-y
n(n＝1,2,3……)，并对第一差值求取第一平均值，即第一平均值为即d
mean＝(d
1+d
2+…+d
n)/s，其中s为差值的个数，s的个数也与该有效区域内绝对相位值的有效数据的个数相同。
请继续参考图1，步骤S500,通过对应的有效区域内获取到的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获取第二差值，并将第二差值与第一平均值相比较，确定参考差值。具体包括以下过程：在与步骤S400中选取的绝对相位直线对应的有效区域中，存在若干个有效的绝对相位值数据，将有效区域内的，获取到的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值y
n做差值，该差值为第二差值R
n，将每一个第二差值R
n与第一平均值d
mean相比较，从第二差值R
n中找出最接近第一平均值d
mean的第二差值R
n，并以该第二差值R
n作为参考差值。由于一般来说非噪声点的绝对相位值数据的个数远大于噪声点的绝对相位值数据的个数，因此，所求出的第一平均值d
mean可认为是一个非噪声点的绝对相位值值与参考绝对相位值的差值，再找出最接近第一平均值d
mean的第二差值R
n，作为参考差值，可认为参考差值所对应的绝对相位值数据为非噪声数据。
请继续参考图1，步骤S600,预设阈值范围，将参考差值与第二差值相比较，划分绝对相位值数据的噪声点和非噪声点。假设第二差值包括R
1，R
2，R
3，R
4，R
5，其中，最接近第一平均值d
mean的第二差值为R
3，则将第二差值为R
3设为参考差值。将参考差值与第二差值相比较，并根据预设的阈值范围划分绝对相位值数据的噪声点和非噪声点。在本实施例中，阈值范围为参考差值与第二差值之差的绝对值小于1的第二差值对应的绝对相位值数据为非噪声点，阈值范围为参考差值与第二差 值之差的绝对值大于1的第二差值对应的绝对相位值数据为噪声点。例如，|R
1-R
3|＜1,即第二差值R
1所对应的绝对相位值数据为非噪声点；|R
4-R
3|＞1，即第二差值R
4所对应的绝对相位值数据为噪声点。据此，可以划分出获取到的绝对相位值数据哪些为噪声点，哪些为非噪声点，划分后剔除噪声点的绝对相位值数据。本发明的投影盲区去除绝对相位噪声方法是将每一分段的绝对相位直线单独分别与参考绝对相位直线进行比对计算，划分每一分段的绝对相位直线的噪声点和非噪声点，并对每一分段的绝对相位直线进行修正。
请参考图8及图9，其中，图8为去噪声前的某一列绝对相位示意图，可以看见，由于投影盲区的存在，绝对相位直线是分段的，其中，绝对相位直线的斜率相同而截距不同，同时，图上具有噪声点的绝对相位值数据，图8中所示的一条连续的直线为参考绝对相位直线。而图9则是图8通过本发明的存在投影盲区去除绝对相位噪声方法去除噪声后的绝对相位示意图，同样的，图9中所示的一条连续的直线为参考绝对相位直线。通过图8及图9的去噪结果能够看到，本发明的存在投影盲区去除绝对相位噪声方法能够有效去除噪声点，提高绝对相位值数据的准确度，继而提高三维重构的精度。
请参考图10及图11，在本实施例中，以阶梯模型为例，图10为去噪前的三维物体整体的绝对相位图像，在存在噪声的情况下，部分像素点的灰度值与其附近的像素点的灰度值不同，而图11为去噪后的三维物体整体的绝对相位图像，在去除噪声后，像素点的灰度值与其附近的像素点的灰度值相同，即噪声通过本发明的存在投影盲区去除绝对相位噪声方法后被有效去除。
请参考图12，在本发明的一些实施例中，步骤S700,剔除噪声点后，并修正绝对相位直线，以得到正确的绝对相位值，并进行三维物体重构。具体地，包括以下步骤：
步骤S710，通过有效区域内的非噪声点的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获取第三差值，并通过第三差值获取第二平均值，并将第二平均值设定为参考距离值；
步骤S720，根据参考距离值和参考绝对相位直线修正绝对相位直线；
步骤S730，通过修正后的绝对相位直线获取正确的绝对相位值，并进行三维物体重构。
更具体地，在剔除掉该有效区域内的噪声点的绝对相位值数据后，计算剩余的非噪声点的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值的第三差值S
n,并计算第三差值的第二平均值，即第二平均值d
c＝(S
1+S
2+…+S
n)/m，其中m为非噪声点的绝对相位值数据的个数。将第二平均值设定d
c为参考距离值，通过参考距离值和参考绝对相位直线修正绝对相位直线，计算修 正后的绝对相位直线的截距b
m'，b
m'＝b+d
c，其中b为参考绝对相位直线的截距，d
c为参考距离值；则修正后的绝对相位直线方程为y
mn'＝k×x
n+b
m'，最后，通过修正后的绝对相位直线方程重新计算正确的绝对相位值数据。
由于之前存在噪声的情况下，导致绝对相位值具有一定的误差，因此，采取参考距离值d
c对绝对相位直线进行修正，参考距离值d
c为该列该段非噪声的绝对相位与对应横坐标的参考相位值的差值的平均值，该平均值能够使得该差值更接近正确值，能够提高绝对相位直线修正的精度。
上述过程得到的绝对相位是存在噪声的绝对相位，在三维物体重构之前需要通过本发明的存在投影盲区去除绝对相位噪声方法进行去噪，去噪后对绝对相位进行修正，以得到正确的绝对相位，再通过修正后的绝对相位值数据进行三维物体重构，能够有效提高三维物体重构的精确度。
请参考图13，本发明实施例还提供了一种存在投影盲区去除绝对相位噪声装置，该存在投影盲区去除绝对相位噪声装置可以是任意类型的智能终端，如手机、平板电脑、个人计算机等。
进一步地，存在投影盲区去除绝对相位噪声装置包括：一个或多个处理器和用于与处理器通信连接的存储器。其中图13中以一个处理器为例。处理器和存储器可以通过总线或其他方式连接，图13以通过总线连接为例。
存储器作为一种非暂态计算机可读存储介质，可用于存储非暂态软件程序以及非暂态性计算机可执行程序，如本发明实施例中的存在投影盲区去除绝对相位噪声装置对应的程序指令。处理器通过运行存储在存储器中的非暂态软件程序以及指令，从而实现上述的存在投影盲区去除绝对相位噪声方法。
存储器可以包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需要的应用程序；存储数据区可存储上述存在投影盲区去除绝对相位噪声方法的相关数据等。此外，存储器可以包括高速随机存取存储器，还可以包括非暂态存储器，例如至少一个磁盘存储器件、闪存器件、或其他非暂态固态存储器件。在一些实施例中，存储器可选包括相对于处理器远程设置的存储器，这些远程存储器可以通过网络连接至该存在投影盲区去除绝对相位噪声装置。上述网络的实例包括但不限于互联网、企业内部网、局域网、移动通信网及其组合。
本发明的第三方面，提供了计算机可读存储介质，该计算机可读存储介质存储有计算机可执行指令，该计算机可执行指令被一个或多个处理器执行，例如，被一个处理器执行，可使得上述一个或多个处理器执行上述存在投影盲区去除绝对相位噪声方法。
以上所描述的装置实施例仅仅是示意性的，其中作为分离部件说明的单元可以是或者也可以不是物理上分开的，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。
本领域普通技术人员可以理解，上文中所公开方法中的全部或某些步骤、系统可以被实施为软件、固件、硬件及其适当的组合。某些物理组件或所有物理组件可以被实施为由处理器，如中央处理器、数字信号处理器或微处理器执行的软件，或者被实施为硬件，或者被实施为集成电路，如专用集成电路。这样的软件可以分布在计算机可读介质上，计算机可读介质可以包括计算机存储介质(或非暂时性介质)和通信介质(或暂时性介质)。如本领域普通技术人员公知的，术语计算机存储介质包括在用于存储信息(诸如计算机可读指令、数据结构、程序模块或其他数据)的任何方法或技术中实施的易失性和非易失性、可移除和不可移除介质。计算机存储介质包括但不限于RAM、ROM、EEPROM、闪存或其他存储器技术、CD-ROM、数字多功能盘(DVD)或其他光盘存储、磁盒、磁带、磁盘存储或其他磁存储装置、或者可以用于存储期望的信息并且可以被计算机访问的任何其他的介质。此外，本领域普通技术人员公知的是，通信介质通常包含计算机可读指令、数据结构、程序模块或者诸如载波或其他传输机制之类的调制数据信号中的其他数据，并且可包括任何信息递送介质。
上面结合附图对本发明实施例作了详细说明，但是本发明不限于上述实施例，在所述技术领域普通技术人员所具备的知识范围内，还可以在不脱离本发明宗旨的前提下作出各种变化。
Claims (10)
- 一种存在投影盲区去除绝对相位噪声方法，其特征在于，包括以下步骤：从条纹图中获取绝对相位值数据；根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分对应的有效区域；确定分别与每条所述绝对相位直线相比较的参考绝对相位直线；通过同一横坐标的所述绝对相位直线的绝对相位值与所述参考绝对相位直线的参考绝对相位值获取第一差值，并通过所述第一差值获取第一平均值；通过对应的有效区域内获取到的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获取第二差值，并将所述第二差值与所述第一平均值相比较，确定参考差值；预设阈值范围，将参考差值与第二差值相比较，划分绝对相位值数据的噪声点和非噪声点；剔除所述噪声点后，并修正所述绝对相位直线，以得到修正后的绝对相位值数据，并进行三维物体重构。
- 如权利要求1所述的存在投影盲区去除绝对相位噪声方法，其特征在于，“从条纹图中获取绝对相位值数据”包括以下步骤：从条纹图中获取包裹相位数据；将所述包裹相位数据恢复成所述绝对相位数据。
- 如权利要求1所述的存在投影盲区去除绝对相位噪声方法，其特征在于，“根据绝对相位数据获取绝对相位直线，并根据每条绝对相位直线划分有效区域”包括以下步骤：分割与标记所述绝对相位值数据；分割后获得分段的所述绝对相位直线；根据每条所述绝对相位直线划分对应的有效区域。
- 如权利要求3所述的存在投影盲区去除绝对相位噪声方法，其特征在于，“分割与标记绝对相位值数据”包括以下步骤：检测绝对相位数据中的有效数据和无效数据；分别标记有效数据和无效数据；根据有效数据和无效数据分割绝对相位数据。
- 如权利要求4所述的存在投影盲区去除绝对相位噪声方法，其特征在于，所述有效数据为非 0的绝对相位值，所述无效数据为0的绝对相位值。
- 如权利要求1所述的存在投影盲区去除绝对相位噪声方法，其特征在于，“确定分别与每条所述绝对相位直线相比较的参考绝对相位直线”包括以下步骤：根据绝对相位值数据中的有效数据的数量并通过直线检测确定参考绝对相位直线。
- 如权利要求6所述的存在投影盲区去除绝对相位噪声方法，其特征在于，所述直线检测采用霍尔变换直线检测。
- 如权利要求1所述的存在投影盲区去除绝对相位噪声方法，其特征在于，“剔除所述噪声点后，并修正所述绝对相位直线，以得到正确的绝对相位值，并进行三维物体重构”包括以下步骤：通过所述有效区域内的非噪声点的绝对相位值数据与同一横坐标的参考绝对相位直线的参考绝对相位值获取第三差值，并通过第三差值获取第二平均值，并将第二平均值设定为参考距离值；根据所述参考距离值和参考绝对相位直线修正绝对相位直线；通过修正后的绝对相位直线获取正确的绝对相位值，并进行三维物体重构。
- 一种存在投影盲区去除绝对相位噪声装置，其特征在于，包括：至少一个处理器和用于与所述处理器通信连接的存储器；所述存储器存储可被所述至少一个处理器执行的指令，所述指令被所述至少一个处理器执行，以使所述处理器能够执行如权利要求1-8任一项所述的存在投影盲区去除绝对相位噪声方法。
- 一种计算机可读存储介质，其特征在于，所述计算机可读存储介质存储有计算机可执行指令，所述计算机可执行指令用于使计算机执行如权利要求1-8任一项所述的存在投影盲区去除绝对相位噪声方法。
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010173834.8A CN111445400B (zh) | 2020-03-13 | 2020-03-13 | 存在投影盲区去除绝对相位噪声方法、装置及存储介质 |
| CN202010173834.8 | 2020-03-13 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2021179438A1 true WO2021179438A1 (zh) | 2021-09-16 |
Family
ID=71652307
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/CN2020/091748 WO2021179438A1 (zh) | 2020-03-13 | 2020-05-22 | 存在投影盲区去除绝对相位噪声方法、装置及存储介质 |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111445400B (zh) |
| WO (1) | WO2021179438A1 (zh) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080204698A1 (en) * | 2005-02-23 | 2008-08-28 | Leica Geosystems Ag | Phase Noise Compensation For Interferometric Absolute Rangefinders |
| CN101726264A (zh) * | 2009-12-30 | 2010-06-09 | 深圳先进技术研究院 | 一种针对投射条纹图像的残差滤波方法 |
| CN102156963A (zh) * | 2011-01-20 | 2011-08-17 | 中山大学 | 一种混合噪声图像去噪方法 |
| CN110230997A (zh) * | 2019-06-04 | 2019-09-13 | 江南大学 | 一种基于改进单调法的阴影区相位噪声校正方法 |
| CN110567398A (zh) * | 2019-09-02 | 2019-12-13 | 武汉光发科技有限公司 | 双目立体视觉三维测量方法及系统、服务器及存储介质 |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103020913B (zh) * | 2012-12-18 | 2015-06-24 | 武汉大学 | 基于分段校正的遥感影像条带噪声去除方法 |
-
2020
- 2020-03-13 CN CN202010173834.8A patent/CN111445400B/zh active Active
- 2020-05-22 WO PCT/CN2020/091748 patent/WO2021179438A1/zh active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080204698A1 (en) * | 2005-02-23 | 2008-08-28 | Leica Geosystems Ag | Phase Noise Compensation For Interferometric Absolute Rangefinders |
| CN101726264A (zh) * | 2009-12-30 | 2010-06-09 | 深圳先进技术研究院 | 一种针对投射条纹图像的残差滤波方法 |
| CN102156963A (zh) * | 2011-01-20 | 2011-08-17 | 中山大学 | 一种混合噪声图像去噪方法 |
| CN110230997A (zh) * | 2019-06-04 | 2019-09-13 | 江南大学 | 一种基于改进单调法的阴影区相位噪声校正方法 |
| CN110567398A (zh) * | 2019-09-02 | 2019-12-13 | 武汉光发科技有限公司 | 双目立体视觉三维测量方法及系统、服务器及存储介质 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111445400A (zh) | 2020-07-24 |
| CN111445400B (zh) | 2023-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10325151B1 (en) | Method of extracting image of port wharf through multispectral interpretation | |
| JP2018155690A (ja) | 表面欠陥検査方法及び表面欠陥検査装置 | |
| US20230086961A1 (en) | Parallax image processing method, apparatus, computer device and storage medium | |
| CN111161169B (zh) | 基于霍夫变换的绝对相位噪声去除方法、装置和存储介质 | |
| CN103400366A (zh) | 基于条纹结构光的动态场景深度获取方法 | |
| CN110349205B (zh) | 一种物体体积的测量方法及装置 | |
| KR20210032367A (ko) | 스테레오-시간적 이미지 시퀀스들로부터 향상된 3-d 데이터 재구성을 위한 방법들 및 장치 | |
| CN113902652B (zh) | 散斑图像校正方法、深度计算方法、装置、介质及设备 | |
| CN107367245B (zh) | 光学三维轮廓测量中的无效点探测与剔除方法 | |
| WO2021179438A1 (zh) | 存在投影盲区去除绝对相位噪声方法、装置及存储介质 | |
| CN110378853B (zh) | 深度图处理方法和装置 | |
| CN111028169A (zh) | 图像校正方法、装置、终端设备和存储介质 | |
| CN116152119B (zh) | 用于条纹结构光的相位去噪方法、装置、设备及介质 | |
| CN111723753B (zh) | 卫星遥感图像的去条带方法、装置和电子设备 | |
| Lei et al. | A novel algorithm based on histogram processing of reliability for two-dimensional phase unwrapping | |
| CN112070700B (zh) | 一种去除深度图像中突起干扰噪声的方法与装置 | |
| CN111340728B (zh) | 基于3d点云分割的点云去噪方法、装置和存储介质 | |
| US20230384085A1 (en) | Phase unwrapping method based on multi-view constraints of light field and related components | |
| CN112967205A (zh) | 基于格雷码滤波器的异常点纠正方法、存储介质和系统 | |
| CN113074634A (zh) | 一种快速相位匹配方法、存储介质和三维测量系统 | |
| CN109902695B (zh) | 一种面向像对直线特征匹配的线特征矫正与提纯方法 | |
| CN113744200B (zh) | 一种摄像头脏污检测方法、装置及设备 | |
| CN113358064B (zh) | 一种光学动态三维测量的相位解包裹方法和装置 | |
| CN114820376A (zh) | 条带噪声的融合校正方法、装置、电子设备及存储介质 | |
| Jiang et al. | An improved canny operator in the application of medical image segmentation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20923948 Country of ref document: EP Kind code of ref document: A1 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| 122 | Ep: pct application non-entry in european phase |
Ref document number: 20923948 Country of ref document: EP Kind code of ref document: A1 |