Space-based optical imaging simulation method for complex targets based on multi-level distributed beam tracing

Abstract

The present invention relates to a space-based optical imaging simulation method for complex targets based on multi-level distributed beam tracing. By integrating the light rays with the same path in the ray tracing method, the multiple reflection energy of the light beam between the surface elements of the target is calculated, effectively reducing the calculation cost while ensuring the accuracy. Based on the principle of ray tracing, the present invention proposes a space-based optical imaging simulation method for complex targets based on multi-level distributed beam tracing. In this method, after generating a large number of reflected light rays by using the distributed ray tracing method each time, the light rays with the same path are integrated and calculated as a whole, effectively solving the problem of the explosive growth of the number of light rays in the distributed ray tracing method. Experiments show that compared with the photon mapping method, the calculation efficiency of this method is better than that of the photon mapping method, and the calculated target optical image has good accuracy.

Description

Technical Field

[0001] The present invention relates to ray tracing calculation technology, and specifically to a complex target space-based optical imaging simulation method based on multi-level distributed beam tracing. Background Art

[0002] The status of target informatization perception and the composite information characteristics of different environments where the target is located have become increasingly prominent in digital information confrontation. The optical characteristics of the target are an important part of digital information, and their research and application in technologies such as target recognition, camouflage, and tracking are very important. However, factors such as the target structure and size are complex and variable, the actual measurement samples are very limited, and the cost of conducting experimental measurements to obtain image data is very high. Therefore, more and more researchers use simulation calculations to obtain the method of target optical imaging characteristics. By constructing a scene geometric model, analyzing the influencing factors of optical imaging characteristics, and establishing a calculation model for target and background optical imaging characteristics. However, when performing optical image simulation calculations on complex targets, the calculation accuracy and calculation efficiency of their optical characteristics often cannot be improved simultaneously, and achieving high-precision simulation of complex target images requires a large amount of time.

[0003] Currently, in order to perform high-precision optical characteristic calculations on complex targets, ray tracing methods based on Monte Carlo are usually adopted. By generating a large number of rays emitted by the light source to simulate real light, the path of the ray propagating and reflecting in the scene is traced until the ray enters the sensor. The photon mapping algorithm (PMA) is a widely used hybrid ray tracing method. It uses photons to replace the rays in the traditional ray tracing method. Photons can be reflected or absorbed by the object surface. When a photon is absorbed, it can be stored at the corresponding position in the photon map for calculating the radiant energy reflected by the object to the sensor. When describing the reflection phenomenon of light on the object surface, ray tracing methods are usually divided into two categories: distributed ray tracing and path tracing. The characteristic of the distributed ray tracing method is that when a ray is reflected on the target surface, multiple reflected rays are generated from the reflection point and continue to be traced. This method has high calculation accuracy, but the number of rays will increase geometrically as the number of reflections increases. The characteristic of the path tracing method is that each ray generates at most one reflected ray and continues to trace this ray until it intersects with the light source; many such paths will be traced for each pixel point of the two-dimensional image, and the distribution probability of the reflection path is calculated according to the surface material. The defect of this method is that the convergence speed is slow. In order to improve the calculation efficiency while ensuring the accuracy, it is very necessary to carry out research on fast and accurate calculation methods for complex target optical characteristics. Summary of the Invention

[0004] In order to overcome the problem of low computational efficiency of multiple reflections of optical radiation in traditional target optical imaging simulation applications, the present invention proposes a multi-level distributed beam tracing method, which can effectively reduce computational costs while ensuring accuracy.

[0005] The present invention proposes a complex target space-based optical imaging simulation method based on multi-level distributed beam tracing. This method integrates the light rays of the same path in the ray tracing method, calculates the multiple reflection energy of the light beams between the facets on the target surface, and effectively reduces the calculation cost while ensuring the accuracy. The specific process is as follows:

[0006] Step 1: Input the target's triangular facet model and observation parameters, re-divide the target's triangular facet model, calculate the visibility relationship of each facet relative to other facets due to mutual occlusion, and determine whether the radiation energy reflected by each facet can reach other facets;

[0007] Step 2: The incident light generates multiple reflected light rays at each facet of the target surface that can reach other visible facets and sensors, and the reflection results of the reflected light rays at this level are calculated using the distributed ray tracing method;

[0008] Step 3: performing path integration on the light rays reflected to other facets in step 2, superimposing multiple reflected light rays incident on the same facet as incident radiation for the next level of calculation of the facet, and eliminating incident radiation below the first threshold;

[0009] Step 4: Repeat steps 2 and 3 to calculate the reflected radiation of the surface element step by step until the incident radiation of all surface elements is lower than the first threshold, and then stop the loop;

[0010] Step 5: Accumulate the multi-level reflected radiation reflected by each face element to the sensor to obtain the total radiation reflected by each face element to the sensor. According to the reflection of atmospheric radiation, background radiation and the self-heating radiation of the target, complete the mapping of the target face element on the image plane to obtain the target optical characteristic imaging.

[0011] Furthermore, in step 2, a distributed reflected light beam is generated according to the bidirectional reflectance distribution function of the facet, and the radiation energy of the distributed reflected light beam is calculated, and the primary incident radiation is radiation from the sun;

[0012] When calculating the reflected radiation from surface element i to surface element j, the four-path method is used for calculation. The center point of the i-th surface element is taken as the point source, and the four paths are used to calculate the average of its radiation to the vertex and center point of the j-th surface element as the reflected radiation energy received by the j-th surface element. The radiation brightness L received by the j-th surface element is obtained. ij,receive for:

[0013]

[0014] Where E i is the irradiance of the light source on the i-th surface element of the ship's surface, R ij,k is the BRDF of the i-th surface element reflecting from the k-th path to the j-th surface element, A j is the area of the j-th surface element, l ij,k is the radiation transmission distance of the k-th path, θ ij is the angle between the normal vectors of the i-th and j-th surface elements.

[0015] Furthermore, when the distance between two surface elements is less than the second threshold, the number of paths is increased. The number of point sources is 4, that is, the three vertices of the reflecting surface element are also used as point sources. The mean value of the radiation from the i-th surface element to the vertices and the center point of the j-th surface element calculated using 16 paths is used as the reflected radiation energy received by the j-th surface element.

[0016] Furthermore, the second threshold is 1.5 times the diameter of the circumcircle of the i-th surface element.

[0017] Furthermore, in step 3, based on the calculation method of the reflected radiation distribution of surface elements, the reflected light ray paths of a certain level are integrated and calculated. Each surface element reflects N - 1 light rays to other N - 1 surface elements and 1 light ray to the sensor;

[0018] Traverse all N surface elements, and the N light rays reflected by all surface elements to the sensor are used as the target multiple reflected radiations of this level for accumulation with the radiation energies of the previous levels;

[0019] Each surface element will receive N - 1 reflected light rays from other surface elements. The sum of their radiation energies is calculated and it is judged whether it exceeds the first threshold. If it is higher than the first threshold, it is used for the calculation of the reflected radiation of the next level. If it is lower than the first threshold, it is eliminated.

[0020] Furthermore, the reflected radiation distribution of the reflected light rays is calculated. The incident radiation luminance from the j-th surface element to the k-th surface element for the next level calculation is:

[0021]

[0022] Where E ij is the irradiance of the i-th surface element on the j-th surface element during the distributed beam tracing process of this level.

[0023] Furthermore, the first threshold is 1% of the solar radiation incident on the target surface.

[0024] The present invention adopts the above technical solutions. Compared with the prior art, it has the following technical effects:

[0025] The present invention is applicable to the problem of calculating the optical characteristics of complex three-dimensional targets under a space-based platform. On the premise of ensuring good accuracy of the target optical simulation image, the calculation efficiency of the method of the present invention is increased by 9 times compared with the traditional photon mapping method, and the calculation efficiency is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is the overall flow chart of the present invention;

[0027] Figure 2 is a schematic diagram of the triangular facet model of an Arleigh Burke-class destroyer for which the optical characteristics are to be calculated;

[0028] Figure 3 is the temperature field map of an Arleigh Burke-class destroyer under the conditions of noon in summer;

[0029] Figure 4 is a schematic diagram of the four-path method and the sixteen-path method;

[0030] Figure 5 is a schematic diagram of the cumulative distribution of the reflected radiation of the facets;

[0031] Figure 6 is a comparison chart of the imaging simulation results of the optical (visible and infrared bands) characteristics of an Arleigh Burke-class destroyer. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0032] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings:

[0033] The space-based optical imaging simulation method for complex targets based on multi-level distributed beam tracing proposed by the present invention integrates the rays with the same path in the ray tracing method, calculates the reflected radiation energy from the target facet to the facet, and combines the target model and the parameters of the optical remote sensing camera payload platform to realize the simulation of the target optical image; its process is as Figure 1 shown.

[0034] The specific process of the optical characteristic imaging simulation method for complex targets is as follows:

[0035] Step 1: Input the triangular facet model of the target and the observation parameters, re-divide the facets of the triangular facet model of the target, calculate the visibility relationship of each facet relative to other facets due to mutual occlusion, and judge whether the radiation energy reflected by each facet can reach other facets;

[0036] Step 2: Generate multiple reflected rays that can reach other visible facets and the sensor at each facet on the target surface, and use the distributed ray tracing method to calculate the reflection results of the reflected rays at this level;

[0037] Step 3: Integrate the paths of the light rays reflected from the said step 2 to other facets, superimpose multiple reflected light rays incident on the same facet as the incident radiation for the next-level calculation of this facet, and eliminate the incident radiation below the first threshold;

[0038] Step 4: Repeat steps 2 and 3, calculate the reflected radiation of the facets level by level until the incident radiation of all facets is lower than the first threshold, and stop the loop;

[0039] Step 5: Accumulate the multi-level reflected radiation reflected from each facet to the sensor, obtain the total radiation reflected from each facet to the sensor, and complete the mapping of the target facet on the image plane based on the reflected atmospheric radiation, background radiation, and the self-heating radiation of the said target, so as to obtain the imaging of the target optical characteristics.

[0040] Further, in the said step 2, generate a distributed reflected light beam according to the bidirectional reflectance distribution function of the facet, calculate the radiation energy of the distributed reflected light beam, and the primary incident radiation is the radiation from the sun;

[0041] When calculating the reflected radiation from facet i to facet j, use the four-path method for calculation. Take the center point of the i-th facet as a point source, and use four paths to calculate the mean value of the radiation from the i-th facet to the vertex and center point of the j-th facet as the reflected radiation energy received by the j-th facet, and obtain the radiation luminance L received by the j-th facet ij,receive as:

[0042]

[0043] In the formula, E i is the radiation illuminance of the light source on the i-th facet of the ship's surface, R ij,k is the BRDF of the i-th facet reflecting to the j-th facet under the k-th path, A j is the area of the j-th facet, l ij,k is the radiation transmission distance under the k-th path, θ ij is the included angle between the normal vectors of the i-th and j-th facets.

[0044] Further, when the distance between two facets is less than the second threshold, increase the number of paths. The number of point sources is 4, that is, the three vertices of the reflecting facet are also used as point sources, and use 16 paths to calculate the mean value of the radiation from the i-th facet to the vertex and center point of the j-th facet as the reflected radiation energy received by the j-th facet.

[0045] Further, the second threshold is 1.5 times the diameter of the circumcircle of the i-th facet.

[0046] Further, in step 3, based on the calculation method of the reflected radiation distribution of the surface element, the reflected light path of a certain level is integrally calculated. Each surface element reflects N - 1 light rays to the other N - 1 surface elements and 1 light ray to the sensor;

[0047] Traverse all N surface elements, and obtain the N light rays reflected by all surface elements to the sensor as the target multiple - reflection radiation of this level, which is used for cumulative addition with the radiation energy of the previous levels;

[0048] Each surface element will receive N - 1 reflected light rays from other surface elements. Sum up its radiation energy and determine whether it exceeds the first threshold. If it is higher than the first threshold, it is used for the reflected - radiation calculation of the next level. If it is lower than the first threshold, it is eliminated.

[0049] Further, perform the calculation of the reflected - radiation distribution for the reflected light rays. The incident radiation luminance from the j - th surface element to the k - th surface element for the next - level calculation is:

[0050]

[0051] In the formula, E ij is the irradiance of the i - th surface element on the j - th surface element during the distributed beam tracing process at this level.

[0052] Further, the first threshold is 1% of the solar radiation incident on the target surface.

[0053] The technical solution of this application is introduced below in combination with a specific case: The complex - target space - based optical imaging simulation method based on multi - level distributed beam tracing of this application includes the following steps:

[0054] Step 1: Input the triangular - surface - element model of the target and the observation parameters. According to the projection of the optical - sensor image plane, re - divide the surface elements of the ship model. The dividing line is the projection of the image - plane grid line on the target surface, so that each re - divided surface element is completely contained in a certain pixel after being projected onto the image plane. Calculate the visibility of each surface element relative to other surface elements, and use this to judge whether the radiation energy reflected by each surface element can reach other surface elements. Generate a visibility identifier after calculation to facilitate subsequent reduction of unnecessary radiation calculations;

[0055] Step 2: Calculate the reflected light beams between the surface elements of the target surface step by step. When a certain surface element on the target surface reflects, a distributed reflected light beam is generated according to its BRDF, that is, the radiant energy reflected from this surface element to the other surface elements. Judge the distance between two surface elements. If it is greater than the second threshold, the four-path method is used for calculation. Take the center point of the i-th surface element as the point source, and use four paths to calculate the average value of the radiation from it to the vertex and the center point of the j-th surface element as the reflected radiation energy received by the j-th surface element. If it is less than the second threshold, then take the three vertices of the reflecting surface element as the point sources and calculate the radiant energy of a total of 16 paths;

[0056] Step 3: Integrate and calculate the reflected light ray paths at a certain level. Based on the surface element reflected radiation distribution calculation method, each surface element will reflect N - 1 light rays to the other N - 1 surface elements and reflect 1 light ray to the sensor (if both between the surface elements and between the surface element and the sensor are visible); traverse all N surface elements, and the N light rays reflected by all surface elements to the sensor will be obtained as the target multiple reflected radiation at this level, which is used for accumulation with the radiation energy of the previous levels; each surface element will receive N - 1 reflected light rays from other surface elements, sum up its radiation energy and judge whether it exceeds the first threshold. If it is higher than the first threshold, it is used for the reflected radiation calculation of the next level. If it is lower than the first threshold, it is eliminated;

[0057] Step 4: Repeat the geometric calculation of the surface element reflected radiation distribution and the cumulative calculation of the surface element reflected radiation distribution in Steps 2 and 3 until the incident radiation of all surface elements is lower than the first threshold, and stop the loop;

[0058] Step 5: Accumulate the multi-level reflected radiation reflected by each surface element to the sensor to obtain the total radiation reflected by each surface element to the sensor, which is the reflection of the target to solar radiation considering the multiple reflection effect. Calculate the reflection of the target to atmospheric radiation and background radiation (single reflection calculation method) and the self-heating radiation of the target. Based on the geometric structure of the target and the relative position relationship between the target and the sensor, calculate the surface elements on the target surface that can be detected by the sensor, and combine the parameters related to the focal plane imaging of the sensor to complete the mapping of the target surface elements on the image plane and realize the imaging of the target optical characteristics.

[0059] A specific calculation example of the space-based optical imaging of a complex target with multi-level distributed beam tracing is as follows:

[0060] Figure 2 It is the triangular surface element model and coordinate definition of the Arleigh Burke-class destroyer. Its surface is divided into multiple triangular surface elements. The ship's surface is painted with gray paint. The measured BRDF data of the gray paint in the visible light band is parameter-fitted through the Phong model to obtain the diffuse reflection parameter k d = 0.5, and the specular reflection parameter k s= 0.5; the measured emissivity of the infrared band is about 0.95, so the reflectivity of the infrared band is 0.05; the temperature field of the ship surface is the temperature field at noon in summer obtained by prior calculation, such as Figure 3 As shown; the method of the present invention is now applied to calculate the optical characteristic image of the target under given input parameters (simulation band, solar zenith angle, observation zenith angle and imaging distance, etc.).

[0061] Application step 1: Input the triangular face model and observation parameters of the target, and re-divide the face of the ship model according to the projection of the optical sensor image plane. The dividing line is the projection of the image plane grid line on the target surface, so that each face after re-division is completely contained in a certain pixel after being projected onto the image plane.

[0062] Application step 2: 20,000 photons are emitted to each facet, and the distributed reflected radiation of the target facet is calculated using the four-path (sixteen-path) method, including the radiation reflected from the facet to other facets and the radiation reflected to the sensor. Figure 4 Schematic diagram of the four-path and sixteen-path methods.

[0063] Application step three: Calculate the reflected radiation received by each facet in this level from other facets as the incident radiation for the next level, and remove the incident radiation below the first threshold. Figure 5 Schematic diagram of the cumulative distribution of surface element reflected radiation.

[0064] Application step 4: Repeat steps 2 and 3 until the incident radiation of all face elements is lower than the first threshold, and then stop the loop.

[0065] Application step 5: Accumulate the multi-level reflected radiation reflected from each facet to the sensor to obtain the total radiation reflected from each facet to the sensor; calculate the reflection of the target on the atmospheric radiation, sea surface radiation and the self-heating radiation of the target; the spectral radiation intensity of the kth facet of the target at the camera entrance pupil can be expressed as:

[0066]

[0067] Where E sun (λ), E atom (λ), E sea (λ) are solar irradiance, atmospheric irradiance and sea surface irradiance, respectively. k is the area of ​​the kth face element, n k is the surface element normal vector, o k is the observation vector of the kth surface element, s k is the illumination vector, BRDF(λ,n k ,o k ,s k ) is the BRDF of the kth surface element, is the surface element occlusion factor.

[0068] Figure 6 are the imaging simulation results of the Arleigh Burke-class destroyer under different observation conditions. To verify the error of the proposed method for the optical characteristic imaging simulation of the Arleigh Burke destroyer (compared with the result of the photon mapping method that emits 50,000 photons / surface element), calculations were performed under the combined conditions of different observation angles (0-90°) and illumination angles (0-180°) in the visible and mid-infrared bands respectively. The results show that the optical image of the Arleigh Burke-class destroyer calculated by the proposed method has good accuracy, with a maximum relative error of 9.24% and a maximum relative root mean square error of 6.68%. Compared with the 270 minutes required for the photon mapping method, the proposed method only requires 29 minutes, and the calculation efficiency is improved by about 9 times. The proposed method can better calculate the highlights and shadows on the surface of the destroyer under typical illumination conditions, and the error mainly appears at the highlighted areas on the ship's surface, which is due to the result of using the sixteen-path mean method to approximately represent the radiant energy transfer between surface elements.

[0069] The specific embodiments described above have further elaborated on the purpose, technical solutions, and beneficial effects of the present invention. It should be understood that the above are only specific embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present invention shall be included within the protection scope of the present invention.

Claims

1. Space-based optical imaging simulation method for complex targets based on multi-level distributed beam tracing, characterized in that, The method comprises the following steps: Step 1: Input the target's triangular facet model and observation parameters, re-divide the target's triangular facet model, calculate the visibility relationship of each facet relative to other facets due to mutual occlusion, and determine whether the radiation energy reflected by each facet can reach other facets; Step 2: The incident light generates multiple reflected light rays at each facet of the target surface that can reach other visible facets and sensors, and the reflection results of the reflected light rays at this level are calculated using the distributed ray tracing method; In the step 2, Generate a distributed reflection beam according to the bidirectional reflectance distribution function of the surface element, calculate the radiation energy of the distributed reflection beam, and the primary incident radiation is the radiation from the sun; When calculating the reflected radiation from surface element i to surface element j, the four-path method is used for the calculation. The center point of the i-th surface element is taken as a point source, and four paths are used to calculate the mean value of the radiation from it to the vertex and the center point of the j-th surface element as the reflected radiation energy received by the j-th surface element, and the radiance L received by the j-th surface element is obtained. ij,receive It is: where, E i is the irradiance of the light source on the i-th surface element of the ship's surface, R ij,k is the BRDF of the i-th surface element reflecting from the k-th path to the j-th surface element, A j is the area of the j-th surface element, l ij,k is the radiation transmission distance under the k-th path, θ ij is the angle between the normal vectors of the i-th and j-th surface elements; Step 3: performing path integration on the light rays reflected to other facets in step 2, superimposing multiple reflected light rays incident on the same facet as incident radiation for the next level of calculation of the facet, and eliminating incident radiation below the first threshold; In step 3, Based on the calculation method of the reflected radiation distribution of the surface element, the reflection light path of a certain level is integrated and calculated. Each surface element reflects N-1 rays to other N-1 surface elements and reflects 1 ray to the sensor. Traverse all N face elements, and obtain N rays reflected by all face elements to the sensor as the target multiple reflection radiation of this level, which is used to accumulate with the radiation energy of the previous levels; Each face element will receive N-1 reflected rays from other face elements, sum up their radiation energies and determine whether they exceed the first threshold. If they are higher than the first threshold, they will be used for the next level of reflected radiation calculation. If they are lower than the first threshold, they will be discarded. Step 4: Repeat steps 2 and 3 to calculate the reflected radiation of the surface element step by step until the incident radiation of all surface elements is lower than the first threshold, and then stop the loop; Step 5: Accumulate the multi-level reflected radiation reflected by each face element to the sensor to obtain the total radiation reflected by each face element to the sensor. According to the reflection of atmospheric radiation, background radiation and the self-heating radiation of the target, complete the mapping of the target face element on the image plane to obtain the target optical characteristic imaging.

2. The complex target space-based optical imaging simulation method of multi-level distributed beam tracking according to claim 1 is characterized in that: When the distance between two surface elements is less than the second threshold, the number of paths is increased, and the number of point sources is 4, that is, the three vertices of the reflecting surface element are also used as point sources. 16 paths are used to calculate the average value of the radiation from the i-th surface element to the j-th surface element vertex and center point as the reflected radiation energy received by the j-th surface element.

3. The complex target space-based optical imaging simulation method for multi-level distributed beam tracing according to claim 2, wherein The second threshold is 1.5 times the diameter of the enclosing circle of the i-th surface element.

4. The complex target space-based optical imaging simulation method for multi-stage distributed beam tracing according to claim 1, wherein The reflected radiation distribution of the reflected light is calculated, and the incident radiation brightness of the jth surface element to the kth surface element used for the next level of calculation is: where E ij is the irradiance of the i-th elemental area on the j-th elemental area during the distributed beam tracing at this level.

5. The complex target space-based optical imaging simulation method for multi-level distributed beam tracing according to claim 1, characterized in that The first threshold is 1% of the solar radiation incident on the target surface.

Images