Cascade fuzzy matching shaping system and method for quasi-near-field focusing light beam
Technical Field
The invention relates to the technical field of beam shaping, in particular to a cascade fuzzy matching shaping system and a shaping method of a quasi-near-field focusing beam.
Background
In many cases, the laser power density is required to be sufficient, and the irradiation spot has a certain area, so that the light beam is mostly irradiated on the surface of the substance in a cone light form after being converged by a lens. The irradiation surface which is in the conical light and does not reach the focal plane has regular light spot profile and uniform light intensity distribution, marks the quality of an actual laser irradiation field, and is an important condition parameter for the action of laser and substances. However, after the laser beams are converged by the lens, the light field is rearranged according to a spatial angular spectrum after being focused, the light field carries partial far-field characteristics before reaching the focal plane, and the modulation in the light field is rapidly enhanced along with the increase of the propagation distance, so that the light field is degraded and appears as the split of a medium-low frequency strong region of a light spot and the increase of high-frequency modulation. This is undesirable in industrial applications and scientific research, and the occurrence of any uncontrolled intensity distribution or intensity singularities in the laser-irradiated surface can change the actual effect or mechanism of action of the laser on the material irradiation. Therefore, the effective technical method is adopted to inhibit the splitting and singular point problem of the cone light irradiation surface and improve the intensity distribution, thereby having important practical significance.
The existing laser beam shaping technology is concentrated into two types of unfocused light beam near-field shaping technology and focal spot far-field shaping technology of a focusing focal plane, and the quasi-near-field light beam shaping of the invention is not reported yet. Taking a common Main Oscillation Power Amplification (MOPA) laser system with an oscillation level + amplification level configuration as an example, in the aspect of light beam quality control of the existing laser near field, the control is mainly realized by using a non-uniform film optical element, a binary optical element and a programmable liquid crystal Spatial Light Modulator (SLMs) to reduce the flux modulation degree and reduce the small-scale self-focusing effect in the aperture of a light spot; in far field control, i.e. in focal spot control, wavefront distortion is reduced by using Deformable Mirrors (DMs) and Spatial Filters (SFs) to obtain far field parameters closer to the diffraction limit in the focal region, resulting in better far field distribution of the laser beam. However, the characteristics of the beam cross-section optical field on the focusing path are between the near field and the far field, emphasizing that the corresponding relation between the image planes is destroyed by the manufacturing, and the traditional beam space shaping method cannot effectively improve the near field distribution. Each section of spatial frequency component brought by diffraction, wavefront distortion and local modulation in the front-stage light beam forms complex light field distribution on the section of the rear-stage focused light beam, and the contrast and the modulation degree of the light field are intensified along with the appearance of a local peak.
Disclosure of Invention
The invention provides a cascade fuzzy matching shaping system and a shaping method of a quasi-near-field focusing light beam for solving the problems in the prior art, and the invention provides the following technical scheme:
a quasi-near-field focusing beam cascading fuzzy matching shaping system comprises an SLM1, a rear-stage fundamental frequency light amplifying device 2, a frequency doubling system 3, a focusing lens 4, a sampling mirror 5, a CCD6 and a target 7;
the fundamental frequency light enters an incidence end of a rear-stage fundamental frequency light amplifying device 2 through an exit end of the SLM1, enters an incidence end of a frequency doubling system 3 through the exit end of the rear-stage fundamental frequency light amplifying device 2, enters an incidence end of a focusing lens 4 through the exit end of the frequency doubling system, the exit end of the focusing lens 4 converges the fundamental frequency light on the surface of a material of a target 7, a sampling mirror 5 is arranged between the focusing lens 4 and the target 7, the SLM1, the rear-stage fundamental frequency light amplifying device 2, the frequency doubling system 3, the focusing lens 4, the sampling mirror 5 and the target 7 form a main optical path, the CCD6 is positioned on one side of the main optical path, the CCD6 and the sampling mirror 5 form a sampling optical path, the main optical path laser is subjected to beam splitting sampling through the sampling mirror 5, and the CCD6 collects a focused light field image at the target 7.
A cascading fuzzy matching shaping method of a quasi-near-field focusing light beam is realized based on a cascading fuzzy matching shaping system of the quasi-near-field focusing light beam, and comprises the following steps:
the method comprises the following steps: loading a positioning diaphragm with characteristic positioning holes on the SLM1, and determining the spatial conversion relation between the CCD6 measurement light spot and the SLM1 by analyzing the positions of the positioning holes;
step two: loading a standard square gray diaphragm with softened edges on the SLM1 to obtain an output light field of initial radiation light field distribution and a corresponding output fundamental frequency light field, and obtaining a compensation diaphragm according to the output light field and the fundamental frequency light field;
step three: loading a shaping diaphragm of the fundamental frequency light to the SLM1 to obtain a first shaped output light field, performing filtering fuzzy preprocessing on a focused light field image acquired by the CCD6 to obtain a processed light field distribution, combining the processed light field distribution with the shaping diaphragm to obtain a modified transfer function, and performing inversion operation on the modified transfer function and a target light field to obtain a new shaping diaphragm;
step four: and loading a shaping diaphragm for focusing the light beam on the SLM1 to obtain a shaped output light field, comparing the shaped output light field with a target light field, and finally obtaining the output light field distribution meeting the set index after passing through the new diaphragm.
Preferably, the spatial translation relationship of the measurement spot with SLM1 includes axial expansion and rotation of the measurement spot; the spatial deformation information of the light spots is obtained through the change of the center positions of the characteristic positioning holes in the front and back light spots, the rotation of the light spots is obtained through the change of the included angle between the positioning holes, and the axial expansion is calculated through the change of the spacing between the positioning holes.
Preferably, the second step is specifically:
the first step is as follows: the SLM1 is loaded with a standard square gray scale stop S with an edge softening factor of 20%1(x, y), CCD6 obtains the initial radiation light field distribution output light field IO1(x, y) and correspondingly outputting a fundamental frequency light field I1ω(x,y);
The second step is that: output fundamental optical field I1ω(x, y) in combination with a standard square gray stop S with an edge softening factor of 20%1(x, y) obtaining a corresponding transfer function H of fundamental frequency optical amplification1ω(x, y) H is represented by the following formula1ω(x,y):
H1ω(x,y)=I1ω(x,y)/S1(x,y) 1
Wherein, I1ω(x, y) is the output fundamental optical field, S1(x, y) Standard Square Gray stops with an edge softening factor of 20%, H1ωTransfer function H of (x, y) fundamental frequency optical amplification1ω(x,y);
The third step: transfer function H to which fundamental frequency light is amplified1ω(x, y) is compared with the corresponding fundamental frequency light target light field to obtain a shaping diaphragm S of the fundamental frequency light2(x, y) S is represented by the following formula2(x,y):
S2(x,y)=I1ωf(x,y)/H1ω(x,y) 2
Wherein, I1ωf(x, y) is the target fundamental optical field, S2(x, y) a shaping diaphragm for the fundamental light, H1ω(x, y) transfer function of fundamental frequency optical amplification.
Preferably, the third step is specifically:
the first step is as follows: shaping diaphragm S for loading fundamental frequency light on SLM12(x, y) obtaining the output light field I after the first shapingO2(x,y);
The second step is that: the focused light field image I acquired by the CCD6O2(x, y) carrying out filter fuzzy preprocessing, and recording the distribution of the processed light field asIO2 *(x,y);
The third step: the processed light field distribution is combined with a shaping diaphragm to obtain a corrected system transfer function H*(x, y) H is represented by the following formula*(x,y):
H*(x,y)=IO2 *(x,y)/S2(x,y) 3
Wherein H*(x, y) is the modified system transfer function, IO2 *(x, y) is the processed light field distribution, S2(x, y) is a shaping diaphragm of fundamental frequency light;
the third step: h is to be*(x, y) and target light field inversion operation to obtain new shaping diaphragm S3(x, y) S is represented by the following formula3(x,y):
S3(x,y)=IOf(x,y)/H*(x,y) 4
Wherein S is3(x, y) is H*(x, y) and target light field inversion operation to obtain new shaping diaphragm, IOf(x, y) is the target light field distribution, H*(x, y) is the modified system transfer function.
Preferably, the filter blur preprocessing employs gaussian filter blur.
Preferably, the fourth step is specifically: shaping diaphragm S for loading quasi-near-field focusing light beam on SLM13(x, y) obtaining the shaped output light field IO3(x, y) A, B, CO3(x, y) and the target light field IOf(x, y) comparing, if the set index is not satisfied, at IO3And (x, y) continuing to shape the light field on the basis of the (x, y), adjusting a filtering window, and iterating the new shaping diaphragm until the distribution of the output light field passing through the new diaphragm meets a set index.
The invention has the following beneficial effects:
the quasi-near-field focused beam shaping scheme is proposed based on modifying the gray scale image written on SLM1 in the previous stage system. In the former stage laser system, the flux of the laser light is relatively low and the spot size is small. This therefore means that the damage threshold for the system components is lower and the shaping element size is smaller. In consideration of the problems of the industrial processing level and the processing cost of the shaping device, the preceding-stage beam shaping scheme can effectively avoid the problems and save the shaping cost. And the laser shaping system based on SLM1 has active beam shaping ability, and the closed loop negative feedback type system structure formed by the laser shaping system and the measuring system and the calculating system has the characteristics of convenience, easy operation, high contrast, high resolution and the like. Moreover, after the whole system is solidified, the laser can replace the SLM1 by using some passive shaping devices with the transmittance corresponding to the gray scale of the SLM1 diaphragm image under the stable operation, such as an optical mask and the like, so that the shaping cost is saved.
Drawings
Fig. 1 is a structural diagram of a cascade type fuzzy matching shaping system of a quasi-near-field focusing light beam.
Fig. 2 is a schematic diagram of a positioning diaphragm.
Fig. 3 is a diagram of the initial output optical field profile of the system as measured by CCD 6. Fig. 3- (a) 2-dimensional light field profile, fig. 3- (b) 3-dimensional light field profile, fig. 3- (c) centerline grayscale profile, and fig. 3- (d) power spectral density profile.
Fig. 4 is a fundamental frequency compensated system output optical field profile as measured by CCD 6. Fig. 4- (a) 2-dimensional light field profile, fig. 4- (b) 3-dimensional light field profile, fig. 4- (c) centerline gray scale profile, and fig. 4- (d) power spectral density profile.
Fig. 5 is a final compensated system output optical field profile as measured by CCD 6. Fig. 5- (a) 2-dimensional light field profile, fig. 5- (b) 3-dimensional light field profile, fig. 5- (c) centerline grayscale profile, and fig. 5- (d) power spectral density profile.
Fig. 6 is a diagram of beam quality change during beam shaping. FIG. 6- (a) a graph of flux contrast and intensity modulation, and FIG. 6- (b) a PSD distribution graph.
Detailed Description
The present invention will be described in detail with reference to specific examples.
The first embodiment is as follows:
constructing a light path diagram according to the schematic diagram shown in fig. 1, wherein the system comprises an SLM1, a rear-stage fundamental frequency light amplifying device 2, a frequency doubling system 3, a focusing lens 4, a sampling mirror 5, a CCD6 and a target 7;
the fundamental frequency light enters the incidence end of a rear-stage fundamental frequency light amplifying device 2 through the emergence end of the SLM1, enters the incidence end of a frequency doubling system 3 through the emergence end of the rear-stage fundamental frequency light amplifying device 2, enters the incidence end of a focusing lens 4 through the emergence end of the frequency doubling system, the emergence end of the focusing lens 4 converges the fundamental frequency light on the material surface of a target 7, a sampling mirror 5 is arranged between the focusing lens 4 and the target 7, the SLM1, the rear-stage fundamental frequency light amplifying device 2, the frequency doubling system 3, the focusing lens 4, the sampling mirror 5 and the target 7 form a main light path, the CCD6 is positioned on one side of the main light path and forms a sampling light path with the sampling mirror 5, the laser is subjected to beam splitting sampling through the sampling main light path 5, and the CCD6 collects a focused light field image at the target 7; in the experiment, the consistency of the optical paths of the two paths is strictly kept.
The light path shaping is to shape the focused light field on the target 7, the sampling mirror 5 has a certain transmittance and reflectivity, the light of the main light path has a small amount of energy after passing through the sampling mirror 5 and is reflected to enter the sampling light path, the lengths of the sampling light path and the main light path are ensured to be consistent, and the focused light field collected by the CCD6 and the focused light field on the target 7 can be ensured to be consistent.
A core shaping device used in the process of the algorithm is a liquid crystal spatial light modulator (SLM1), an SLM1 is used as a transmission device, the transmission distribution of laser light is consistent with a gray diaphragm image S (x, y) loaded on a computer, and a laser system transfer expression embedded in the SLM1 can be recorded as S (x, y) H (x, y) I (x, y), wherein H (x, y) is a laser system transfer function and represents the sum of optical field distribution changes introduced by each component device of the laser system, and I (x, y) is output optical field distribution. The flux contrast and the intensity modulation degree of the light field can be greatly reduced through the algorithm, and the light beam space intensity profile with regular distribution is obtained.
In order to realize the shaping of the quasi-near-field focusing light beam, the invention provides a cascading fuzzy matching shaping algorithm, which is characterized by comprising the following steps:
the method comprises the following steps: loading a positioning diaphragm with characteristic positioning holes on the SLM1, and determining the spatial conversion relation between the CCD6 measurement light spot and the SLM1 by analyzing the positions of the positioning holes;
step two: loading a standard square gray diaphragm with softened edges on the SLM1 to obtain an output light field of initial radiation light field distribution and a corresponding output fundamental frequency light field, and obtaining a compensation diaphragm according to the output light field and the fundamental frequency light field;
step three: loading a shaping diaphragm of the fundamental frequency light to the SLM1 to obtain an output light field after primary shaping, performing filtering fuzzy preprocessing on a focused light field image acquired by the CCD6 to obtain processed light field distribution, combining the processed light field distribution with the shaping diaphragm to obtain a modified transfer function, and performing inversion operation on the modified transfer function and a target light field to obtain a new shaping diaphragm;
step four: and loading a shaping diaphragm for focusing the light beam on the SLM1 to obtain a shaped output light field, comparing the shaped output light field with a target light field, and finally obtaining output light field distribution which passes through a new diaphragm and meets a set index.
When a laser beam is focused on the surface of a target through a lens, a secondary phase factor is superposed on a light field under the action of the lens, the light field passing through the lens is rearranged according to a spatial angular spectrum, the light field has a certain far field characteristic along with the increase of a propagation distance, and a strong region splitting phenomenon occurs in the light field. The traditional shaping system based on the SLM1 is built on near-field light field compensation between image planes, and each pixel between the image planes satisfies a one-to-one correspondence relationship, which means that by changing the transmittance of a certain pixel of a diaphragm on the SLM1, the intensity of the pixel in the light field to be shaped can be correspondingly controlled. However, the focused light field has no correspondence among pixels due to strong modulation, and the light field to be shaped cannot be directly used for inversion operation when the compensation diaphragm is calculated. Therefore, the filtering fuzzy preprocessing is carried out on the light field image to be shaped, and the cascade corresponding relation between the pixels, namely the one-to-many or many-to-many corresponding relation, is reestablished, so that the compensation diaphragm of the quasi near-field focusing light beam is obtained through inversion operation. This filter-blur preprocessing is essentially a filtering action. First, it can reduce the sharpness of the diaphragm image, thereby reducing the modulation effect introduced by excessive gray scale variation of the picture elements adjacent to the diaphragm. Secondly, due to the modulation effect existing in the filtering, each pixel is associated with other pixels in the filtering window, so that the point-surface or surface-surface correspondence between the pixel and the pixel of the diaphragm written on the SLM1 is realized, namely, the control of the pixel surface on the light field to be shaped is realized through a single pixel or pixel surface on the SLM1 gray diaphragm.
The inventor verifies the effectiveness of the technical method of the invention through practical experiments. The point-surface type or surface-surface type corresponding relation between the target surface light field collected by the CCD6 and the writing diaphragm of the SLM1 is realized by filtering fuzzy preprocessing. In the traditional light field compensation in the image transmission system, due to the one-to-one correspondence relationship of the image elements between the image surfaces, the intensity of the post-stage point image element can be adjusted by controlling the pre-stage point image element. The action of the lens enables the light field to be rearranged according to the spatial angular spectrum, and the light field has certain far field characteristics. Therefore, for the compensation of the focusing light beam with the quasi-near-field characteristic on the non-image surface, the light intensity of the point pixel on the light field is subjected to strong modulation action of the light field of the surrounding pixel group. Through fuzzification pretreatment, the even-slip effect realizes the correspondence from point image elements to surface image elements or from surface image elements to surface image elements, establishes the relation between a rear-stage light field to be shaped and a front-stage light field, and realizes the light beam shaping by controlling the local transmittance of the front-stage light field.
In practical experiments, the flux contrast of an optical field is reduced from 37.2% to 21.4% by shaping an output quasi-near-field focusing light beam, the intensity modulation degree is reduced from 3.33:1 to 1.86:1, and the low-frequency modulation in the optical field can be inhibited to a certain degree in the power spectral density distribution, namely, a local strong area in the optical beam is weakened, the intensity distribution in the optical field is more concentrated, and the splitting phenomenon of the strong area is weakened.
The second embodiment is as follows:
the present invention will be described in detail with reference to the accompanying drawings.
The cascade fuzzy matching shaping algorithm example of the quasi near field focusing light beam is characterized by comprising the following steps:
step 1, constructing a light path diagram according to the schematic diagram shown in fig. 1, converging a fundamental frequency light beam on the surface of a target material sequentially through an SLM1, a post-stage fundamental frequency light amplification and frequency doubling system 3 and a focusing lens 44, then sampling a light field of the target surface through a sampling mirror 5 on a focusing path, and strictly keeping the two light paths consistent in an experiment. The SLM1 is loaded with a positioning diaphragm that is a circular diaphragm with a characteristic positioning hole as shown in FIG. 2. In consideration of defects existing in processing and growing of optical devices and crystals and inevitable human errors in optical path calibration, certain spatial deformation exists between the rear-stage light spot and the front-stage light spot, and the deformation is mainly expressed as axial expansion and rotation of the light spots. The SLM1 is loaded with a positioning diaphragm, so that an output light spot containing a positioning hole can be collected at the later stage of the laser system, and spatial deformation information can be obtained through the change of the circle center position of a characteristic positioning hole in the light spot at the former stage and the later stage. The light spot rotation is obtained through the change of the included angle between the positioning holes, and the axial expansion is calculated through the change of the distance between the positioning holes. In the process of the invention, the spatial conversion relationship between the loading diaphragm and the fundamental frequency light and between the loading diaphragm and the irradiation light field on the SLM1 is mainly required to be established. The step lays a foundation for subsequent mathematical operation.
Step 2, loading a standard square diaphragm S with an edge softening factor of 20% on the SLM11(x, y), the edge softening factor is defined as the ratio of the difference in width corresponding to the spot flux falling between 90% and 10% of the maximum flux to twice the spot size. An initial irradiation light field distribution I can be obtained at this timeO1(x, y) as shown in FIG. 3. Fig. 3(a) and (b) are 2-dimensional and 3-dimensional shapes of the light spot, respectively, and according to the intensity information of the light spot, the flux contrast of the initial target surface light field reaches 37.2% without the SLM1, and the modulation degree reaches 3.33: 1. Fig. 3(c) shows the x-axis line gray scale distribution at the center of the spot, and a strong region splitting phenomenon can be observed. Fig. 3(d) is a PSD distribution of the light field, from which it can be seen that both the high frequency modulation and the low frequency modulation are relatively high. The fundamental frequency light field I correspondingly output at the moment1ω(x, y) and S1The (x, y) operation can obtain the corresponding transfer function H of fundamental frequency optical amplification1ω(x,y),H1ω(x,y)=I1ω(x,y)/S1(x,y)。
Then, willIdeal fundamental optical field combination I1ωf(x, y) and transfer function H1ω(x, y) obtaining a compensation diaphragm S of the fundamental frequency light2(x,y),S2(x,y)=I1ωf(x,y)/H1ω(x,y)。
Step 3, loading shaping diaphragm S of fundamental frequency light on SLM12(x, y), the irradiated light field is denoted as IO2(x,y)。S2The writing of (x, y) reduces the effect of gain and transmission non-uniformity during fundamental amplification and the beam quality is improved somewhat, as shown in fig. 4. According to the calculation of 2-dimensional and 3-dimensional intensity information of light spots in the images of fig. 4(a) and (b), the flux contrast of the light field of the primarily compensated target surface is basically leveled to 34.3%, and the intensity modulation degree is reduced to 2.88: 1. In FIG. 4(c), the splitting phenomenon of the strong region of the central x-axis gray distribution of the light spot is improved, but the local strong region is too much. The PSD distribution in fig. 4(d) does not vary much. To IO2(x, y) carrying out image preprocessing, adding Gaussian blur or mean value blur, wherein a filtering window selected in an experiment is a 5 multiplied by 5 square window which is adjustable in practical situation, and a processed light field is marked as IO2 *(x, y). It and diaphragm S2(x, y) calculation to obtain the transfer function H of the modified system*(x,y),H*(x,y)=IO2 *(x,y)/S2(x,y)。
Then H is added*(x, y) and the target light field IOf(x, y) inversion operation to obtain new shaping diaphragm S3(x,y),S3(x,y)=IOf(x,y)/H*(x,y)。
Step 4, loading shaping diaphragm S of quasi-near-field focusing light beam on SLM13(x, y) the shaped system output light field is IO3(x, y). If the experimental requirement is not met, the fuzzy window can be properly adjusted, and the final compensation diaphragm S (x, y) is obtained by continuous iterative approximation on the diaphragm until the requirement is met, in the example, the contrast of the light beam flux is required to be reduced to be below 25%, and the ratio of the intensity modulation degree is less than 2. The final compensated target surface light field distribution is shown in fig. 5. Fig. 5(a) and (b) show the 2-and 3-dimensional topography of the spot, respectively, calculated to reduce the flux contrast by 21.4% and the intensity modulation of the light field by 1.86: 1. FIG. 5(c) clear in line gray distributionThe intensity distribution of the light spots is more concentrated and regular. In the PSD distribution of fig. 5(d), the large-scale strong region corresponding to the low-frequency modulation in the optical field is suppressed. In the whole shaping process, the parameter change of the light field is shown in fig. 6, fig. 6(a) shows the change of flux contrast and intensity modulation, the quality of the light beam is improved along with the deepening of the shaping process, fig. 6(b) shows the change of the PSD of the light field in the shaping process, the low-frequency large-scale intensity is effectively inhibited, and a relatively regular light beam space profile is obtained.
The above description is only a preferred embodiment of the cascaded fuzzy matching shaping system and the shaping method for the quasi-near-field focused light beam, and the protection scope of the cascaded fuzzy matching shaping system and the shaping method for the quasi-near-field focused light beam is not limited to the above embodiments, and all technical solutions belonging to the idea belong to the protection scope of the present invention. It should be noted that modifications and variations which do not depart from the gist of the invention will be those skilled in the art to which the invention pertains and which are intended to be within the scope of the invention.