An imaging device based on light ray customized free-form surface illumination and a free-form surface design method

Abstract

The application discloses an imaging device based on light customization free-form surface and a free-form surface design method, and the method comprises the following steps: determining basic optical parameters and a target pattern, and sampling an incident surface and a focal plane; performing rough matching on incident light and outgoing light of a lens to obtain a rough matching light pair; calculating a surface normal vector of a lens surface by using Snell's law according to the rough matching light pair; constructing an initial surface by using a Poisson equation based on the surface normal vector; performing fine matching on the initial surface to obtain a fine matching light pair; and performing global and local optimization by using a layered optimization method based on the fine matching light pair to obtain design surface data of the free-form surface. The imaging device comprises a light source and the free-form surface. The application can be popularized in different fields and applications by introducing a systematic design idea and an optimization strategy, and modularity and standardization of illumination system design are realized.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of customized lighting, and particularly relates to an imaging device for customized lighting free-form surface based on light rays and a free-form surface design method. BACKGROUND

[0002] Customized lighting is a technology of using light rays emitted by a light source to achieve a required lighting effect, such as road lighting, automobile lighting, etc. In actual application, different light distributions may be required to achieve different lighting effects. However, direct application of a light source cannot usually meet the requirements. Efficient adjustment of the spatial energy distribution of the light source is a research field of customized lighting, and is a classic and challenging problem in the field of non-imaging optics.

[0003] Since the curvature radius is the only design freedom provided by a traditional spherical surface, customized lighting usually uses an aspherical surface to redirect a light beam. Aspherical optics can very effectively handle rotationally or linearly symmetric illumination designs, but due to the rotational or linear limitations of the surface geometry, they usually cannot well solve the problem of asymmetric illumination, and cannot meet the growing demand for advanced lighting systems. Free-form surfaces are three-dimensionally controllable optical surfaces, and their free form provides strong freedom, which can be used to avoid the limitations on the surface geometry and create compact and efficient designs with better performance, and more importantly, using a deformed surface can produce new designs that cannot be achieved using spherical or aspherical surfaces.

[0004] Although free-form surfaces can theoretically achieve any customized lighting, due to the very large degree of freedom, the design is difficult, and today's research is based on complex mathematical derivation, but a definite exact solution has not been found, which limits the degree of freedom in actual design, and it is easy to have no solution. In addition, the method is generally based on illumination design and cannot achieve precise light-level lighting. At the same time, today's design is not systematic, and there is no clear design link. SUMMARY

[0005] The application aims to solve the problems of the prior art, and provides the following solutions.

[0006] An imaging device for customized lighting free-form surface based on light rays, comprising a light source, and further comprising a free-form surface lens, the light source and the free-form surface lens are separately arranged, the light rays emitted by the light source pass through the free-form surface lens, and are directly converged into a required customized pattern lighting.

[0007] The application further provides another imaging device for customized lighting free-form surface based on light rays, comprising a light source, and further comprising a free-form surface lens, the light source is directly placed on the free-form surface lens, an incident surface of the free-form surface lens is a collimating surface, and an exit surface is a free surface.

[0008] The present application also provides a free-form surface design method for customized lighting, which is used for designing the free-form surface lens as described above, and comprises the following steps:

[0009] determining basic optical parameters and a target pattern, and sampling an entrance surface and a focal plane, wherein the optical parameters include: a transmittance refractive index, an exit space refractive index, a focal plane position, a lens position, and a sampling number;

[0010] coarsely matching incident light rays and exit light rays of the lens to obtain a coarse matching light ray pair;

[0011] calculating a surface normal vector of the lens surface according to the coarse matching light ray pair by using Snell's law;

[0012] constructing an initial surface by using a Poisson equation based on the surface normal vector;

[0013] precisely matching the initial surface to obtain a precise matching light ray pair;

[0014] adopting a hierarchical optimization method to perform global and local optimization based on the precise matching light ray pair to obtain design surface data of the free-form surface lens.

[0015] Preferably, the sampling method comprises: adopting concentric circle sampling.

[0016]

[0017]

[0018]

[0019] wherein, l represents a number of sampled radial annuli, N0 represents an ideal sampling number, S represents an area of a sampling pattern, rl represents a radius of the lth annulus from the center, r1 represents a radius of the first annulus from the center, p represents a position of the free-form surface in a coordinate system of the sampling pattern, p0 represents a center of the pattern, and r n represents a radius of the nth annulus, c n represents a sampling number of the nth annulus, N + represents a natural number set except 0.

[0020] Preferably, the method for obtaining the coarse matching light ray pair comprises:

[0021] obtaining a point set X of the incident light rays and a point set T(X) of the exit light rays;

[0022] The point set X and the point set T(X) are coarsely matched to find a matching mapping that minimizes a cost function in the point set T(X) to obtain the coarse matching ray pair V and V':

[0023] min T ∑ X c0(X, T(X))

[0024] wherein c0(μ,ν) represents a cost function of coarse matching.

[0025] Preferably, the method for calculating the surface normal vector comprises:

[0026] V' = V + ηN

[0027]

[0028] wherein V represents an incident ray of the coarse matching ray pair, V' represents an outgoing ray of the coarse matching ray pair, N represents a surface normal vector, represents a transmission refraction index, represents an outgoing space refraction index, η represents a coefficient of a unit normal vector, and θ represents an incident angle.

[0029] Preferably, the method for constructing the initial surface comprises:

[0030] The surface normal vector is solved by a gradient, and a surface point cloud is solved by a Poisson equation:

[0031]

[0032]

[0033] H = grad(φ) · N

[0034] wherein φ represents the surface point cloud, represents a divergence, grad represents a gradient, n represents a function normal vector, H represents a current solving value of the equation, and Q represents a value of taking a divergence twice;

[0035] The surface point cloud is subjected to a vector displacement cycle to make a height of a surface center point equal to a position p of the free-form surface and to ensure relative heights of the surface points, and the Poisson equation is solved again until an error is within an error limit to obtain the initial surface.

[0036] e(x, y) = ||φ(x, y) - φ0(x, y)||2> e0

[0037] wherein e represents an error function, φ(x, y) represents a surface point cloud of a current cycle, φ0(x, y) represents a surface function of a previous cycle, and e0 represents an error limit.

[0038] Preferably, the method for obtaining the fine matching light ray pair comprises:

[0039] obtaining a point set X' of incident light rays and a point set T(X') of emergent light rays of the initial surface;

[0040] fine matching the point set X' and the point set T(X'), finding a matching mapping in the point set T(X') that minimizes a cost function, to obtain the fine matching light ray pair:

[0041] min T ∑ X′ c(X', T(X'))

[0042] c(u, v) = ∫(F(u, v) + c0(u, v))dγ(u, v)

[0043] wherein c(u, v) represents the cost function of fine matching, F(u, v) represents the fine matching amount, and γ(u, v) represents the integral domain mapping function.

[0044] Compared with the prior art, the present application has the following beneficial effects:

[0045] 1. The present application introduces numerical calculation to design and optimize various parameters in the lighting system, and changes the structure of the cost function to make it more close to the most natural light matching state. This method allows the designer to more flexibly adjust and optimize the distribution and characteristics of the light without relying on traditional continuous mathematical conditions. Through the method provided by the present application, more accurate lighting control can be achieved in complex geometric structures and multi-light source systems, and the optimal solution can be explored under specific constraint conditions, effectively solving the problem of limited design freedom in traditional methods, thereby greatly improving the design flexibility and applicability of customized lighting systems, and providing solid technical support and theoretical basis for realizing more efficient and intelligent lighting solutions.

[0046] 2. The present application can realize single light ray customization by introducing the refractive index matrix, rewriting the fine matching cost function and not relying on the physical model of the numerical method. The specific properties of each light ray can be independently adjusted, and the propagation path of the light ray can be accurately controlled for different light source conditions. This not only effectively reduces unnecessary light loss, but also avoids irregular distribution of light in space, thereby providing more uniform and required lighting effects in different application scenarios. Therefore, the precise control of single light ray not only breaks through the limitations of traditional illumination control, but also provides a new technical path for realizing higher quality and more efficient lighting effects.

[0047] 3.The application introduces systematic design ideas and optimization strategies, and proposes a customized lighting design link of rough matching-initial structure-precise matching-multidimensional adaptive optimization. This universal design link can not only improve the efficiency of customized lighting design, but also be popularized in different fields and applications to realize the modularization and standardization of lighting system design. The application contributes to the systematization and standardization of the customized lighting field. BRIEF DESCRIPTION OF DRAWINGS

[0048] In order to more clearly illustrate the technical solutions of the present application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings described in the following embodiments are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0049] Figure 1 A structure schematic diagram of an imaging device according to the first embodiment of the present application;

[0050] Figure 2 A structure schematic diagram of an imaging device according to the second embodiment of the present application;

[0051] Figure 3 A method flowchart according to the third embodiment of the present application;

[0052] Figure 4 A target pattern schematic diagram according to the third embodiment of the present application;

[0053] Figure 5 A sampling situation schematic diagram according to the third embodiment of the present application;

[0054] Figure 6 A rough matching situation schematic diagram according to the third embodiment of the present application;

[0055] Figure 7 An initial surface schematic diagram according to the third embodiment of the present application;

[0056] Figure 8 A free-form surface design result schematic diagram according to the third embodiment of the present application. DETAILED DESCRIPTION

[0057] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0058] In order to make the above objectives, features and advantages of the present application more obvious and easy to understand, the present application will be further described in detail below with reference to the drawings and specific embodiments.

[0059] Embodiment one

[0060] In this embodiment, an imaging device is provided, comprising a light source and a free-form lens, the light source and the free-form lens are arranged separately, as shown in the figure, the light emitted by the light source is directly converged into the required custom pattern illumination through the free-form lens. This imaging device abandons the fragile, easily contaminated and low light energy efficiency film pattern illumination method, has the advantages of compact structure, high light energy utilization rate, and is not easy to be damaged, etc., greatly improves the product performance and competitiveness of custom pattern illumination. Figure 1

[0061] Embodiment two

[0062] In this embodiment, an imaging device is provided, comprising a light source and a free-form lens, the light source is directly placed on the free-form lens, as shown in the figure, the incident surface of the free-form lens is a collimating surface, and the exit surface is a free surface. This imaging device directly uses an LED extended light source without a separate collimating lens, and at the same time, does not use a film pattern illumination method, has the advantages of more compact structure, small space occupancy rate, high light energy utilization rate, and is not easy to be damaged, etc., greatly improves the product performance and use scenarios of custom pattern illumination, reduces the cost, and increases the competitiveness of the product. Figure 2

[0063] Embodiment three

[0064] In this embodiment, as shown in the figure, a free-form surface design method for light custom illumination includes the following steps: Figure 3

[0065] S1. Determine the basic optical parameters and the target pattern, and sample the incident surface and the focal plane, the optical parameters include: transmission refractive index, exit space refractive index, focal plane position, lens position and sampling number.

[0066] In this embodiment, the basic optical parameters include: lens refractive index Exit space refractive index Focal plane position f=130, free-form surface position p=30, free-form surface aperture D=50, and sampling number is 30000. The target pattern to be customized is shown in the figure. Figure 4

[0067] The sampling method includes: in this embodiment, in order to ensure the circular shape of the cut aperture and ensure the design accuracy, a uniform concentric circle sampling is adopted:

[0068] ​​​​

[0069]

[0070]

[0071] Where l represents the number of radial rings for sampling, N0 represents the ideal number of samples, which is 30000 in this embodiment, S represents the area of ​​the sampling pattern, which is S = 625π in this embodiment, r l This represents the radius at the l-th ring from the center outwards, in this embodiment r l =25, r1 represents the radius of the first ring from the center outwards, p represents the position of the sampling pattern on the freeform surface in the rectangular coordinate system, p0 represents the center of the pattern, r n Let c represent the radius of the nth annulus. n N represents the number of samples in the nth ring. + Represents the set of natural numbers excluding 0; solving the formula yields:

[0072]

[0073] Sampling details as follows Figure 5 As shown.

[0074] S2. Perform coarse matching on the incident and outgoing rays of the lens to obtain coarsely matched ray pairs.

[0075] Methods for obtaining coarsely matched ray pairs include: acquiring the point set X of the incident ray and the point set T(X) of the outgoing ray; performing a coarse match on the point set X and the point set T(X), such as... Figure 6 As shown, the matching mapping that minimizes the cost function is found in the point set T(X), resulting in the coarse matching ray pairs V and V':

[0076] min T ∑ X c0(X, T(X))

[0077] Where c0(μ,ν) represents the cost function of coarse matching.

[0078] S3. Calculate the surface normal vector of the lens surface using Snell's law based on the coarse matching ray pair.

[0079] Methods for calculating surface normal vectors include:

[0080] V′=V+ηN

[0081]

[0082] Where V represents the incident ray of the coarse-matched ray pair, V' represents the outgoing ray of the coarse-matched ray pair, and N represents the surface normal vector. represents the transmission refractive index, represents the exit space refractive index, η represents the coefficient of the unit normal vector, and θ represents the incident angle.

[0083] S4. Constructing an initial surface based on the surface normal vector by using the Poisson equation.

[0084] The method for constructing the initial surface includes: performing gradient solving on the surface normal vector, and solving the surface point cloud by using the Poisson equation:

[0085]

[0086]

[0087] H = grad (φ) · N

[0088] wherein φ represents the surface point cloud, represents the divergence, grad represents the gradient, n represents the function normal vector, H represents the current solving value of the equation, and Q represents the value of taking the divergence twice; when the surface normal vector N takes (x, y, -1), the horizontal axis component of the point cloud gradient is x, and the vertical axis component of the point cloud gradient is y, that is, φ x = x, and φ y = y. The vector displacement cycle is performed on the surface point cloud, the height of the surface center point is equal to the position p of the free-form surface, the relative height of each surface point is ensured, the Poisson equation solving is performed again, until the error is within the error limit, and the initial surface is obtained, as shown in Figure 7

[0089] e (x, y) = || φ (x, y) - φ0 (x, y) ||2> e0

[0090] wherein e represents the error function, φ (x, y) represents the surface point cloud of the current cycle, φ0 (x, y) represents the surface function of the last cycle, and e0 represents the error limit, which is 100 in the embodiment.

[0091] S5. Precise matching is performed on the initial surface to obtain a precise matching light ray pair.

[0092] The method for obtaining the precise matching light ray pair includes: obtaining a point set X' of incident light rays and a point set T (X') of exit light rays of the initial surface; performing precise matching on the point set X' and the point set T (X'), finding a matching mapping in the point set T (X') that makes the cost function minimum, and obtaining the precise matching light ray pair:

[0093] min T ∑ X′ c (X', T (X'))

[0094] c (u, v) = ∫ (F (u, v) + c0 (u, v)) dγ (u, v)​

[0095] wherein c(μ, v) represents the cost function of the fine matching, F(u, v) represents the fine matching amount, represents the change of the optical path difference of the lens, contains the supplementary operation customized for single light rays, such as reducing the light intensity in a certain area, and γ(u, v) represents the integral domain mapping function.

[0096] S6. Based on the fine matching light ray pairs, a layered optimization method is used for global and local optimization to obtain the design surface data of the free-form surface.

[0097] In the embodiment, the optimization operation is performed according to the fine matching light ray pairs, and the optimization is performed in an inner layer and an outer layer. The number of the inner layer is 10, and the number of the outer layer is 10. The outer layer determines the optimization granularity. Generally, the granularity is from coarse to fine, and is performed from sampling 1000 to sampling 10000. The granularity change is shown in the following formula:

[0098] J n = J0e αt

[0099] wherein J n represents the change of the granularity, J0represents the initial sampling granularity, a represents the growth coefficient, and t represents the cycle number; for the outer layer, a = (lnJ n -lnJ0) / 9, J0= 1000, and J9= 10000; for the inner layer, a = (lnJ n -lnJ0) / 9, J0is determined according to the error level when the sampling number is between 50 and 60, and J9= 0.001. After the optimization granularity is selected, the simulation operation is performed according to the Snell's law to obtain the result, and then the error value is obtained according to the target and the result. Then the inner layer is classified according to the adaptive error value, and the selected block is locally optimized. The error level of the region is selected to decrease with the change of J n . The surface result obtained after the optimization is shown in Figure 8 . The surface result is imported into the Solidworks software. First, the point cloud scanning is used to establish the grid, and then the three-dimensional surface is reconstructed, and the three-dimensional model is established by using the “stretching” operation in the three-dimensional drawing.

[0100] Example Four

[0101] According to the brand logo of a certain brand of automobile, the free-form surface design method of the application is used to design the microstructure on the free-form surface, to manufacture the free-form surface corresponding to the logo pattern, and then combined with the light source to manufacture a projection device. The light source can use an LED light source. The projection device can be installed at the position of the vehicle headlamp or the surrounding, or the door side, to form a welcome light effect, a prompt light effect, or other light effects.

[0102] Example Five

[0103] According to the static plane advertisement design work, the free curved surface design method is used to design the microstructure on the free curved surface, the free curved surface corresponding to the plane advertisement design work is manufactured, and the projection device is manufactured by combining with the light source, wherein the light source can adopt the LED light source. The projection device can be installed in the hall, elevator room and other public areas of the building to form an advertisement effect.

[0104] The above-described embodiments are only used to describe the preferred modes of the present application, and are not used to limit the scope of the present application. Without departing from the design spirit of the present application, various modifications and improvements of the technical solutions of the present application made by those skilled in the art shall fall within the protection scope determined by the claims of the present application.

Claims

1. A method for designing freeform surfaces for customized lighting, characterized in that, Includes the following steps: The basic optical parameters and target pattern are determined, and the incident surface and focal plane are sampled. The optical parameters include: transmission refractive index, exit space refractive index, focal plane position, lens position, and number of samples. The incident and outgoing rays of the lens are coarsely matched to obtain coarsely matched ray pairs; Based on the coarsely matched ray pair, the surface normal vector of the lens surface is calculated using Snell's law; Based on the surface normal vector, an initial surface is constructed using the Poisson equation. The initial surface is precisely matched to obtain precisely matched ray pairs; Based on the precisely matched light pair, a hierarchical optimization method is used to perform global and local optimization to obtain the design surface data of the freeform lens; The sampling method includes: using concentric circle sampling. ； ； ； in, l Indicates the number of radial rings in the sampling. N 0 represents the ideal number of samples. S Indicates the area of ​​the sampled pattern. r l Indicates the number of circles extending outwards from the center. l The radius at the ring, r 1 represents the number of circles extending outwards from the center. 1 The radius of each ring, p This indicates the position of the sampled pattern on a freeform surface in a rectangular coordinate system. p 0 represents the center of the pattern. r n Indicates the first n The radius of the ring, c n Indicates the first n The number of samples per ring, N + It represents the set of natural numbers excluding 0.

2. The freeform surface design method for customized lighting according to claim 1, characterized in that, The method for obtaining the coarsely matched ray pair includes: Obtain the point set of the incident light ray. X and the point set of the emitted rays T ( X ); For the point set X and the set of points T ( X Perform coarse matching on the point set. T ( X Find the matching mapping that minimizes the cost function, and obtain the coarse matching ray pair. V and V’ : ； in, c 0( u , ν ) represents the cost function for coarse matching.

3. The freeform surface design method for customized lighting according to claim 2, characterized in that, The method for calculating the surface normal vector includes: ； ； in, V This represents the incident light of a coarsely matched ray pair. V’ This represents the outgoing rays of a coarsely matched ray pair. N Represents the surface normal vector. Indicates the transmission refractive index. Indicates the refractive index of the exit space. η The coefficient representing the unit normal vector. θ Indicates the angle of incidence.

4. The freeform surface design method for customized lighting according to claim 3, characterized in that, The method for constructing the initial surface includes: The gradient of the surface normal vector is calculated, and the surface point cloud is solved using the Poisson equation. ； ； ； in, This represents the surface point cloud. grad indicates taking the divergence, and grad indicates calculating the gradient. n Represents the normal vector of a function. H This represents the current solution value of the equation. Q This indicates taking the divergence value twice; Perform vector displacement loops on the surface point cloud so that the height of the surface center point is related to the position of the freeform surface. p The Poisson equation is solved again to ensure that the relative heights of each surface point are equal and that the error is within the error limit, thus obtaining the initial surface. ； in, e Represents the error function. This represents the surface point cloud of the current loop. This represents the surface function of the previous loop. e 0 indicates the error limit.

5. The freeform surface design method for customized lighting according to claim 4, characterized in that, The method for obtaining the precisely matched ray pair includes: Obtain the point set of incident rays on the initial surface. X’ and the point set of the emitted rays T ( X’ ); For the point set X’ and the set of points T ( X’ Perform fine matching on the point set. T ( X’ Find the matching mapping that minimizes the cost function to obtain the finely matched ray pair: ； ； in, c ( u , ν ) represents the cost function for fine matching. F ( u , v ) indicates the fine match quantity. γ ( u , v ) represents the mapping function in the integral domain.

Images