Search

CN-116645411-B - Stripe direction angle quick unpacking method based on reliability sorting

CN116645411BCN 116645411 BCN116645411 BCN 116645411BCN-116645411-B

Abstract

The invention discloses a stripe direction angle quick unpacking method based on reliability sequencing, which comprises the following steps: firstly, calculating gradients by utilizing an interference pattern, then calculating a density pattern, calculating reliability values by the density pattern, sequencing the reliability values, and then relatively unpacking the direction angles of two adjacent pixel points according to the order of the reliability values in sequence, so as to obtain a pointing pattern. The unfolding path adopted by the invention is a discontinuous path, and the problem that the processing speed is low due to the limitation of the continuous path when the traditional algorithm processes large-size images is solved.

Inventors

  • LI JIANXIN
  • LI ZHIYU

Assignees

  • 南京理工大学

Dates

Publication Date
20260505
Application Date
20230420

Claims (7)

  1. 1. The stripe direction angle quick unpacking method based on reliability sorting is characterized by comprising the following steps of: Step 1, calculating fringe gradient values of an interference pattern; step 2, calculating a fringe density map according to the fringe gradient value; Step 3, calculating a vertical reliability value and a horizontal reliability value by using the stripe density map; Step 4, guiding two adjacent pixel points to relatively unpack through the reliability value position; Step 5, repeatedly executing the step 4 until the unpacking is finished, and obtaining a pointing diagram; and 3, calculating a horizontal reliability value and a vertical reliability value by using the fringe density map, wherein the method specifically comprises the following steps: calculating a horizontal reliability value : ; Calculating vertical reliability values : ; In the formula, Representing interference images in The horizontal reliability value at the location, Representing an image at The vertical reliability value at the location, Representing interference images in Fringe density values at the locations; and step 4, guiding two adjacent pixel points to relatively unpack through the reliability value position, and specifically comprising the following steps: all pixels before unpacking are separately grouped, namely, two adjacent pixels A, B with reliability values are searched, and A is assumed to belong to Group B belongs to Group, calculate two-point direction angle difference Judging A, B whether the packets are identical, if so, skipping the reliability value, otherwise, judging If it is greater than zero, then make Otherwise let ; Judging Whether the number of group elements is greater than Number of group elements, if so All direction angles in the group are added And is added to Group, otherwise All direction angles in the group are added And is added to A group; Before the step 5 is executed, the method further comprises the steps of: Descending order of reliability values obtained in the step 3
  2. 2. The method for rapidly unpacking fringe orientation angles based on reliability ordering according to claim 1, wherein the calculating of the fringe gradient values of the interferograms in step 1 is specifically: calculating the horizontal gradient of an interference image : Wherein 、 Respectively, interference images are in 、 Pixel values at the locations; computing vertical gradients of interference images : 。
  3. 3. The method for rapidly unpacking a stripe direction angle based on reliability sequencing according to claim 2, wherein in step 2, a stripe density map is calculated according to a stripe gradient value, specifically: In the formula, Representing interference images in Fringe density values, symbols at locations Representing the convolution operator, Indicating a core size of Is a gaussian filter of (c).
  4. 4. The method for rapidly unpacking stripe direction angles based on reliability sequencing according to claim 3, wherein step 5 is repeated until unpacking is completed, so as to obtain a direction diagram, namely, unpacking all direction angles by sequentially cycling all edge values, and packing unpacking results in a mode of packing The point diagram is the direction diagram.
  5. 5. A reliability-ordering-based streak direction angle fast unpacking system based on the method of any one of claims 1 to 4, characterized in that the system comprises the following steps: the first module is used for calculating the fringe gradient value of the interference pattern; the second module is used for calculating a fringe density map according to the fringe gradient value; a third module for calculating a vertical reliability value and a horizontal reliability value using the fringe density map; A fourth module, configured to guide two adjacent pixels to relatively unpack through the reliability value position; and the fifth module is used for repeatedly operating the fourth module until the unpacking is finished, and obtaining the directed graph.
  6. 6. The reliability ordering based stripe direction angle fast unpacking system of claim 5, further comprising, prior to execution of the fourth module: and a sixth module for arranging the reliability values obtained by the third module in descending order.
  7. 7. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the method of any of claims 1 to 4 when executing the computer program.

Description

Stripe direction angle quick unpacking method based on reliability sorting Technical Field The invention belongs to the technical field of image processing, and particularly relates to a stripe direction angle quick unpacking algorithm based on reliability sorting. Background The fringe direction contains important information of the interference fringe pattern and is an important component of many existing denoising algorithms and single-frame demodulation techniques. The stripe direction angle is generally obtained through the stripe intensity gradient, and the stripe intensity gradient is used for replacing the actual phase gradient to calculate the direction angle, so that the direction angle is wrapped between [ -pi/2, pi/2 ], and the direction graphic package is the process of restoring the direction angle wrapped between [ -pi/2, pi/2 ] into a real pointing diagram. In order to well realize direction angle unpacking operation, scholars at home and abroad have proposed a unpacking method such as regularized phase tracking and a quality diagram guided unpacking method. Both of the above algorithms are based on the most reliable pixel of the previous unpacking, and this class of algorithms is attributed to the continuous path algorithm. The algorithm is limited by a continuous path when processing a large-size image, and the unpacking time is long, so that the real-time requirement of single-frame interferogram processing cannot be met, and therefore, a rapid unpacking algorithm of a discontinuous path is needed. Disclosure of Invention The invention aims to provide a fast unpacking method for a pattern, which has high unpacking speed and high anti-interference capability, aiming at the defects and the shortcomings of the prior art. The technical scheme for achieving the purpose of the invention is that on one hand, a stripe direction angle quick unpacking method based on reliability sequencing is provided, and the method comprises the following steps: Step 1, calculating fringe gradient values of an interference pattern; step 2, calculating a fringe density map according to the fringe gradient value; Step 3, calculating a vertical reliability value and a horizontal reliability value by using the stripe density map; Step 4, guiding two adjacent pixel points to relatively unpack through the reliability value position; and 5, repeatedly executing the step 4 until the unpacking is finished, and obtaining the directed graph. Further, the step 1 of calculating the fringe gradient value of the interference pattern specifically includes: calculating the horizontal gradient of an interference image Wherein f (i, j), f (i-1, j) are pixel values of the interference image at the (i, j), (i-1, j) positions, respectively; computing vertical gradients of interference images Further, in the step 2, a fringe density map is calculated according to the fringe gradient value, specifically: wherein D q (i, j) represents the fringe density value of the interference image at the (i, j) position, the sign Representing the convolution operator, K ρ represents a gaussian filter with a kernel size p. Further, in step 3, the calculating the horizontal reliability value and the vertical reliability value according to the fringe density chart specifically includes: Calculating a horizontal reliability value E h：Eh(i,j)＝Dq(i,j)+Dq (i+1, j); calculating a vertical reliability value E v：Ev(i,j)＝Dq(i,j)+Dq (i, j+1); Where E h (i, j) represents the horizontal reliability value of the interference image at the (i, j) position, E v (i, j) represents the vertical reliability value of the image at the (i, j) position, and D q (i+1, j) represents the fringe density value of the interference image at the (i+1, j) position. Further, in step 4, the guiding the two adjacent pixels to relatively unpack through the reliability value position specifically includes: Searching two adjacent pixel points A, B with reliability values, assuming that A belongs to G A group and B belongs to G B group, calculating a two-point direction angle difference delta β, judging A, B whether the groups are the same, if so, skipping the reliability value, otherwise, judging whether cos (delta β) is larger than zero, if so, letting alpha=0, otherwise, letting alpha=1; Judging whether the number of the elements in the G A group is larger than that of the elements in the G B group, if so, adding alpha multiplied by pi to all direction angles in the G B group and adding the angles to the G A group, otherwise, adding alpha multiplied by pi to all direction angles in the G A group and adding the angles to the G B group. And further, repeatedly executing the step 4 until the unpacking is finished in the step 5 to obtain the pointing diagram, namely, sequentially cycling all edge values to unpack all direction angles, and packing unpacking results between [ -pi, pi ] to obtain the pointing diagram. Further, before the step 5 is executed, the method further includes executing: