CN-121978877-A - Low speckle noise holographic 3D display method based on improved random pixel separation method

Abstract

The invention provides a holographic 3D display method with low speckle noise based on an improved random pixel separation method. The method comprises the steps of considering a 3D object as being composed of 2D objects with different depths, firstly extracting intensity information of the 2D object to obtain an original image, and extracting random pixel indexes of complementary checkerboards to generate a set of random masks. Subsequently, a set of subgraphs is obtained by superimposing a random mask with the initial image, applying a uniform phase to form the object plane complex amplitude, and calculating the holographic plane complex amplitude based on the principle of angular spectrum diffraction. The complex amplitude is then converted into a phase hologram using a bi-phase encoding method. Finally, the phase holograms are loaded on the spatial light modulator in sequence in a time division multiplexing mode, and a holographic 3D reproduction image with low speckle noise is reconstructed.

Inventors

• WANG DI
• Zhang Guanduo
• LI NANNAN
• ZHAO RUIYI
• HUANG QIAN

Assignees

• 北京航空航天大学

Dates

Publication Date

20260505

Application Date

20251126

Claims (4)

1. 1. A holographic 3D display method with low speckle noise based on an improved random pixel separation method is characterized by mainly comprising the following steps of firstly regarding a 3D object as being composed of 2D objects with different depths, extracting intensity information of the 2D object with any depth surface to obtain an initial image RD (X, Y), generating a pair of complementary checkerboards N 1 and N 2 with the same size as RD (X, Y), extracting pixel indexes of the two checkerboards and randomly scrambling the pixel indexes in sequence to obtain random pixel indexes, evenly distributing the random pixel indexes to generate p random masks, secondly, superposing the initial image RD (X, Y) with the p random masks to generate p corresponding sub-images, applying uniform phases to obtain object plane complex amplitudes D i (X, Y) of each sub-image, carrying out diffraction calculation on D i (X, Y) based on an angular spectrum diffraction theory to obtain a holographic plane amplitude U i (X, Y), thirdly, carrying out phase-shifting on the hologram phase complex amplitudes U i and reconstructing the hologram phase images by a holographic phase shifter when the hologram phase position of the three-dimensional complex phase position is different from the two hologram phase position mirrors, and reconstructing the hologram phase position images by a spatial light modulator to be sequentially reconstructed by a phase-shifting hologram phase position of the hologram phase shifter of the hologram phase position of the hologram 2.
2. 2. The method of claim 1, wherein in the first step, in order to avoid the situation that adjacent pixels are not separated, according to the size of the initial image RD (X, Y), a pair of complementary checkerboards N 1 and N 2 are introduced to completely separate the adjacent pixels, and N 1 and N 2 are each a matrix with a size of [ m, N ], where m and N represent the total number of pixels of the initial image in the horizontal and vertical directions, respectively, and the assignment principle of the complementary checkerboards is as follows: Then, pixel indexes corresponding to the checkerboards N 1 and N 2 are respectively generated, the sequence is randomly disordered, and the random pixel indexes are obtained, wherein the scrambling principle is as follows: Wherein N 1 * and N 2 * are random pixel indexes corresponding to N 1 and N 2 , respectively, pi 1 and pi 2 are two independent random arrangements, a group of random masks B i (i=1, 2,..p) with the number p is generated, the size is [ m, N ], the random pixel indexes of N 1 are sequentially and evenly distributed to the front p/2 masks, the random pixel indexes of N 2 are sequentially and evenly distributed to the rear p/2 masks, and all the random masks B i are overlapped to form a full 1 matrix with the size of [ m, N ].
3. 3. The method of claim 1, wherein in the second step, the random mask B i is superimposed on the initial image RD (X, Y) to obtain p corresponding subgraphs, and the uniform phases are applied to obtain the object plane complex amplitude D i (X, Y), respectively, as follows: wherein (X, Y) is the pixel point coordinates, j is the imaginary number, The Hadamard product representing the element-wise multiplication of the matrix propagates the D i (X, Y) diffraction to the hologram plane based on the principle of angular spectrum diffraction, resulting in the hologram plane complex amplitude U i (X, Y): Wherein F and F -1 represent the fourier transform and inverse fourier transform, respectively, F X and F Y represent the spatial frequencies of the light field in the X and Y directions, respectively, λ is the wavelength of the incident light, and z is the angular spectrum diffraction distance.
4. 4. The method for 3D display of low speckle noise holograms based on improved random pixel separation as claimed in claim 1, wherein in step three, phase information in complex amplitudes of the hologram is extracted to obtain phase holograms, and the complex amplitude information is retained to the maximum extent by decomposing complex amplitudes U i (X, Y) of the hologram into two pure phase values, thereby improving the quality of the hologram.

Description

Low speckle noise holographic 3D display method based on improved random pixel separation method 1. Technical field The invention relates to a holographic 3D display technology, in particular to a holographic 3D display method with low speckle noise based on an improved random pixel separation method. 2. Background art The holographic 3D display technology utilizes the diffraction principle of light to reproduce all wave-front information of an object, is one of ideal 3D display technologies, and has wide application prospects in the fields of virtual reality, biological microscopy, medical imaging and the like. In recent years, with the continuous development of computer technology, the 3D display technology of computer holography has made remarkable progress, but challenges such as high computational complexity of hologram, small display viewing angle, serious reconstruction astigmatic speckle noise and the like are still faced. In general, speckle noise in holographic 3D displays is mainly due to errors of the hologram algorithm and the use of coherent light sources. In order to improve the reconstruction quality, researchers have proposed a pixel separation method, in which the overlapping area of airy specks is reduced by increasing the sampling interval between adjacent object points, so that speckle noise is suppressed, but speckle noise caused by periodic interference of light sources still exists. Some researchers introduce diffraction region range constraints on the basis of the pixel separation method, which reduces speckle noise while maintaining the resolution of the artwork, but which requires the use of high refresh rate spatial light modulators. Some researchers use a bi-phase constraint macro-pixel separation method to realize the rapid calculation of holograms while reducing speckle noise, but the reproduced images have the problems of resolution loss, limited detail recovery and the like. At present, the mainstream pixel separation method does not add any constraint condition when separating pixels, and the situation that two adjacent pixel points are not completely separated can occur, so that the quality of a reconstructed image is affected. Therefore, how to achieve holographic 3D display with low speckle noise remains a challenge. 3. Summary of the invention In order to suppress speckle noise in holographic 3D display, the invention provides a low speckle noise holographic 3D display method based on an improved random pixel separation method. As shown in fig. 1, the method mainly comprises four steps, namely, a first step, for a 3D object, first treating it as being composed of 2D objects of different depths. For a 2D object of any depth surface, its intensity information is extracted to obtain an initial image RD (X, Y), and a pair of complementary checkerboards N 1 and N 2 of the same size as RD (X, Y) are generated. And then extracting pixel indexes of the two checkerboards, randomly scrambling the sequence of the pixel indexes to obtain random pixel indexes, and evenly distributing the random pixel indexes to generate p random masks. And secondly, respectively superposing the initial images RD (X, Y) with p random masks to generate p corresponding subgraphs, and applying uniform phases to obtain object plane complex amplitudes D i (X, Y) of the subgraphs. Based on the angular spectrum diffraction theory, diffraction calculation is carried out on D i (X, Y) to obtain holographic plane complex amplitude U i (X, Y). And thirdly, converting the holographic surface complex amplitude U i (X, Y) of each sub-image into a corresponding phase hologram by using a bi-phase encoding method. And carrying out phase hologram calculation on the 2D objects with different depths through the processing of the steps, and superposing the phase holograms of the planes with different depths to obtain p phase holograms of the 3D object. And step four, sequentially loading p phase holograms on the spatial light modulator by using a time division multiplexing method, and reconstructing a holographic reconstruction image with low speckle noise when reconstruction light irradiates the spatial light modulator. According to the method, the complementary checkerboard and the random pixel index are introduced, so that the limiting condition is added when the pixels are separated, the condition that two adjacent pixel points are not separated in the traditional random pixel separation method is avoided, and speckle noise caused by the adjacent pixels is effectively reduced. In step one, to avoid the situation that adjacent pixels are not separated, a pair of complementary checkerboards N 1 and N 2 are introduced to completely separate adjacent pixels according to the size of the initial image RD (X, Y). N 1 and N 2 are each matrices of size [ m, N ], where m and N represent the total number of pixels of the initial image in the horizontal and vertical directions, respectively. The assignment principle of the compl