A method for rendering acceleration of volume data facing blank space hopping

Abstract

The application provides a kind of blank space jump-oriented body data rendering acceleration method, it is related to the field of body data rendering, comprising: judging whether the body data block to which the current light sampling point belongs exists non-empty voxel, if it is blank body data block, then enter Chebyshev jump stage, based on Chebyshev distance map blank body data block is jumped, if it is not blank body data block then enter voxel level jump stage.Voxel level jump stage, judge whether voxel is non-empty, if it is non-empty, then execute light projection sampling process, if it is empty, then enter the blank voxel based on unsigned distance map jump stage and jump.The adaptive sampling step length adjustment is carried out based on gradient amplitude, body data value and viewpoint distance.When light is emitted from the cube, iteration is stopped.The technical scheme of the application overcomes the problems that the existing technology cannot realize efficient skipping of empty voxels and the light projection algorithm has low calculation efficiency.

Description

Technical Field

[0001] This invention relates to the field of volumetric data rendering, and more specifically to a method for accelerating volumetric data rendering that involves jumping through blank spaces. Background Technology

[0002] Volume data is a numerical data structure used to describe objects or phenomena in three-dimensional space, and is widely used in fields such as medical imaging, scientific computing, and engineering visualization. Volume rendering technology generates two-dimensional images directly from three-dimensional data, allowing for a visual representation of their internal structure and properties. Current mainstream volume rendering methods include texture slicing, snowballing algorithms, and ray casting algorithms, among which ray casting has become a research hotspot due to its high-quality imaging effects. Ray casting methods determine pixel values ​​and generate realistic images by calculating the color contribution at the intersection of light rays emitted from the viewpoint and scene objects. The core process of the algorithm is as follows: light rays are emitted from the viewing point and intersect with the volume data; sampling and color synthesis are performed along the ray propagation path to generate the volume rendering result on the two-dimensional screen. This algorithm has a significant computational redundancy problem: it uses a uniform sampling strategy for empty and non-empty voxels. That is, during the rendering process, the color values ​​and opacities of all voxels need to be mixed and calculated, while the actual imaging contribution only comes from the effective voxel region. This means that a large amount of computational resources are used for empty voxel sampling and interpolation.

[0003] Therefore, there is a need for a volume data rendering acceleration method that enables efficient skipping of empty voxels and improves the computational efficiency of ray casting algorithms by jumping over empty spaces. Summary of the Invention

[0004] The main objective of this invention is to provide a method for accelerating volumetric data rendering that allows jumping in empty space, in order to solve the problems of inefficient jumping of empty voxels and low computational efficiency of ray casting algorithms in the prior art.

[0005] To achieve the above objectives, the present invention provides a method for accelerating volume data rendering for jumping in blank space, specifically including the following steps:

[0006] S1: Divide the volume data into blocks of fixed size. During the ray traversal of the volume data, determine whether the volume data block to which the current ray sampling point belongs has a non-empty voxel. If it is an empty volume data block, enter the Chebyshev jump stage and jump the empty volume data block based on the Chebyshev distance map. If it is not an empty volume data block, enter the voxel-level jump stage.

[0007] S2, in the voxel-level jump phase, determine whether the voxel is non-empty. If it is non-empty, execute the ray casting sampling process; if it is empty, enter the unsigned distance map-based jump phase to jump to empty voxels. S3, in the ray casting sampling process, perform adaptive sampling step size adjustment based on gradient magnitude, volume data value, and viewpoint distance.

[0008] S4, calculate the distance m from the incident ray to the outgoing ray. When sampling volume data each time, calculate the distance n that the ray has traveled in the cube. If n>=m, it means that the ray has exited the cube, and the iteration stops.

[0009] Furthermore, step S1 specifically includes the following steps:

[0010] S1.1, assuming the volume data dimension is The occupied graph dimension is The block size is Then for the normalized sampling point coordinates The coordinates of the volume data in the occupancy graph for:

[0011] .

[0012] S1.2, assuming the normalization step size between two adjacent sampling points is... ,but Step size in the occupancy map for:

[0013] .

[0014] S1.3, verify whether the current block is empty based on the occupancy graph; if it is empty, enter the Chebyshev jump phase, which includes: skipping the remaining part of the current block and skipping empty data blocks based on Chebyshev distance.

[0015] Furthermore, step S1.3 specifically includes the following steps:

[0016] S1.3.1, in a certain dimension The formula for jumping up to the remaining part of the current block is:

[0017] ;

[0018] in, This indicates that the current sampling point is in the occupancy map. Non-normalized coordinates in each dimension Represents the first x, y, z directions in the three-dimensional coordinate system. Sampling distance in each dimension, This is for rounding down.

[0019] S1.3.2, the number of skipped empty data blocks based on Chebyshev distance is calculated by using the occupancy graph to determine the Chebyshev distance from the current empty block to the nearest non-empty block, generating the corresponding Chebyshev distance graph; Chebyshev distance Represent two vectors Each coordinate component Maximum absolute value of the difference:

[0020] ;

[0021] in, To obtain the maximum value.

[0022] S1.3.3, combining steps S1.3.2 and S1.3.3, the formula for calculating the jump distance during Chebyshev's jump phase is:

[0023] ;

[0024] in, This represents the distance to jump from the current position to the next non-empty data block. Represents a symbolic function; For Heaviside step function, This represents the offset of the current point within the original empty block. To obtain the minimum value, This is for rounding up.

[0025] S1.3.4, Simplifying the jump distance calculation formula in step S1.3.3, we get:

[0026] .

[0027] Furthermore, step S2 specifically includes the following steps:

[0028] S2.1, iterate through all voxels in the volume data and determine whether each voxel is non-empty; set the UDF value of non-empty voxels to 0, and initialize the distance of other empty voxels to positive infinity.

[0029] S2.2, for empty voxels, the process enters the unsigned distance graph skipping stage, sequentially calculating distances along the x, y, and z axes. Each stage is based on the distance information from the previous stage, employing a bidirectional traversal or alternating search strategy. The distance from each voxel to the nearest non-empty voxel is calculated using the Euclidean distance formula. A three-dimensional unsigned distance field is generated by iteratively finding the minimum value. The three-dimensional unsigned distance field stores the Euclidean distance from the current voxel to the nearest non-empty voxel. The calculation formula is:

[0030] ;

[0031] in, Given the known distance of the current voxel to the nearest non-empty voxel, To represent the distance value of the queried offset voxel in the x-axis direction, For the current voxel coordinates, To take the minimum.

[0032] S2.3, Set the minimum jump threshold based on the volume data scale. Each time the UDF distance value is queried, it is compared with the minimum jump threshold. If the distance value is greater than the minimum jump threshold, a jump is performed. The calculation formula is:

[0033] ;

[0034] in, This represents the diagonal length of the volume data. This represents the maximum number of sampling points.

[0035] Furthermore, step S3 specifically includes the following steps:

[0036] S3.1, Normalize the current ray sampling coordinates, and read the volume data value of the current ray sampling coordinates.

[0037] S3.2, copy the volume data value to the three RGB channels and expand it into grayscale color to obtain the color value of the sampling point. Use the color value and the transparency of the sampling point to perform color accumulation and synthesis.

[0038] S3.3, acquire volume data, gradient magnitude, and view distance; determine the importance of sampling points based on their attributes; and dynamically adjust the sampling density. The weighting formulas for volume data values ​​and gradient magnitude are as follows:

[0039] ;

[0040] ;

[0041] in, Indicates the gradient magnitude weight; Indicates the weight of the volume data value; The number of intervals to divide the volume data value; For the gradient magnitude to fall within the interval The probability, For volume data values ​​to fall within the interval The probability of.

[0042] For viewpoint distance, the range of viewpoint distance for sampling points is divided into three segments: far, medium, and near, and each segment is assigned a corresponding weighting coefficient:

[0043] ;

[0044] in, The viewpoint distance weight.

[0045] S3.4, combining the gradient magnitude weight, volume data value weight, and viewpoint distance weight, yields the adaptive sampling step size adjustment formula:

[0046] ;

[0047] in, , , These represent the normalized volume data gradient magnitude, volume data value, and location distance, respectively. The step size before adjustment. This is the adjusted step size.

[0048] The present invention has the following beneficial effects:

[0049] This invention utilizes Chebyshev distance graphs (CDM) to quickly jump between large-scale empty blocks, thereby avoiding invalid block traversal;

[0050] This invention implements empty voxel skipping within non-empty blocks based on unsigned distance field (UDF) to avoid redundant sampling;

[0051] This invention combines body data values, gradient magnitude, and viewpoint distance to dynamically adjust the sampling step size, thereby improving sampling efficiency. Attached Figure Description

[0052] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. In the drawings:

[0053] Figure 1 A flowchart of a volume data rendering acceleration method for skipping in blank space according to the present invention is shown. Detailed Implementation

[0054] The technical solution of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0055] like Figure 1 The illustrated method for accelerating volumetric data rendering by jumping through blank spaces includes the following steps:

[0056] S1: Divide the volume data into blocks of fixed size. During the ray traversal of the volume data, determine whether the volume data block to which the current ray sampling point belongs has a non-empty voxel. If it is an empty volume data block, enter the Chebyshev jump stage and jump the empty volume data block based on the Chebyshev distance map. If it is not an empty volume data block, enter the voxel-level jump stage.

[0057] S2, in the voxel-level jump phase, determine whether the voxel is non-empty. If it is non-empty, execute the ray casting sampling process; if it is empty, enter the unsigned distance map-based jump phase to jump to empty voxels. S3, in the ray casting sampling process, perform adaptive sampling step size adjustment based on gradient magnitude, volume data value, and viewpoint distance.

[0058] S4. When calculating the intersection of the ray and the volume data as the incident point and the exit point, calculate the travel distance m of the incident ray to the exit ray. When sampling the volume data each time, calculate the distance n that the ray has traveled in the cube. If n>=m, it means that the ray has exited the cube, and then stop the iteration.

[0059] This invention provides a method for rapidly skipping invalid volume data within the volume data. Building upon existing methods for accelerating volume rendering blank space skipping using Chebyshev distance maps and 3D unsigned distance, this invention proposes a new method based on layered distance fields and importance-driven approaches, thereby further reducing invalid voxel sampling and improving overall rendering efficiency. This method integrates Chebyshev distance maps and unsigned distance fields to implement block-level and voxel-level spatial skipping mechanisms, effectively eliminating redundant calculations for empty voxels. Furthermore, for valid voxel regions, this invention incorporates a dynamic step-size adjustment mechanism based on volume data values, gradient magnitudes, and view distance to achieve adaptive adjustment of sampling density.

[0060] Specifically, the volume rendering empty space jump acceleration method based on Chebyshev Distance Map (CDM) first divides the volume data into blocks of fixed size, and then iterates through each data block to determine whether there are any non-empty voxels. If a block contains at least one non-empty voxel, it is considered a non-empty block; otherwise, it is an empty block. The occupancy map constructed accordingly records 0 or 1 for each block, representing whether the block is empty or occupied, respectively. A Chebyshev distance map is further constructed based on the occupancy map, where each voxel stores the minimum number of empty blocks that must be traversed from the current position along any direction to the nearest non-empty block. Step S1 specifically includes the following steps:

[0061] S1.1, assuming the volume data dimension is The occupied graph dimension is The block size is Then for the normalized sampling point coordinates The coordinates of the volume data in the occupancy graph for:

[0062] .

[0063] S1.2, assuming the normalization step size between two adjacent sampling points is... ,but Step size in the occupancy map for:

[0064] .

[0065] S1.3, verify whether the current block is empty based on the occupancy graph; if it is empty, proceed to the Chebyshev jump phase, which includes: skipping the remaining part of the current block, skipping empty data blocks based on Chebyshev distance, and skipping subsequent empty data blocks based on Chebyshev distance to reach the nearest non-empty data block.

[0066] Specifically, step S1.3 includes the following steps:

[0067] S1.3.1, in a certain dimension The formula for jumping up to the remaining part of the current block is:

[0068] ;

[0069] in, This indicates that the current sampling point is in the occupancy map. Non-normalized coordinates in each dimension Represents the first x, y, z directions in the three-dimensional coordinate system. Sampling distance in each dimension, This is for rounding down.

[0070] S1.3.2, the number of skipped empty data blocks based on Chebyshev distance is calculated by using the occupancy graph to determine the Chebyshev distance from the current empty block to the nearest non-empty block, generating the corresponding Chebyshev distance graph; Chebyshev distance Represent two vectors Each coordinate component Maximum absolute value of the difference:

[0071] ;

[0072] in, To obtain the maximum value.

[0073] S1.3.3, combining steps S1.3.2 and S1.3.3, the formula for calculating the jump distance during Chebyshev's jump phase is:

[0074] ;

[0075] in, This represents the distance to jump from the current position to the next non-empty data block. The sign function is used to determine the sampling direction; This is a Heaviside step function, used for special logic to handle jumps in the negative direction. This represents the offset of the current point within the original empty block. To obtain the minimum value, To round up, this jumping strategy comprehensively considers the minimum effective jump steps among the three dimensions, ensuring the maximum jump distance from the current position, effectively reducing unnecessary sampling and improving rendering efficiency.

[0076] S1.3.4, To simplify implementation and further improve jumping efficiency, this invention approximates the spatial jumping method based on Chebyshev distance maps. The original method requires precise calculation of the jump step size based on the ray direction and block boundary position to avoid out-of-bounds missampling; this invention directly multiplies the shortest block-level distance stored in the distance map by the number of voxels contained in each block to estimate the maximum number of voxels that can be jumped, and then completes the jump along the ray direction accordingly. The jump distance calculation formula in step S1.3.3 is simplified to obtain:

[0077] .

[0078] The simplification involves directly multiplying the shortest block-level distance stored in the distance graph by the number of volume elements contained in each block to estimate the maximum jump distance.

[0079] This strategy omits complex block boundary judgment logic, significantly reducing computational overhead. Subsequent comparative experiments show that this simplified strategy can still generate image results with good visual quality while maintaining high rendering efficiency.

[0080] Specifically, the volume rendering empty space jump acceleration method based on the three-dimensional unsigned distance field (UDF) achieves accurate voxel-level spatial jumps by querying the distance from the current point to the nearest non-empty voxel. Compared with the jump method based on Chebyshev distance map, this method has higher spatial resolution and jump accuracy. Step S2 specifically includes the following steps:

[0081] S2.1, iterate through all voxels in the volume data and determine whether each voxel is non-empty; set the UDF value of non-empty voxels to 0, and initialize the distance of other empty voxels to positive infinity;

[0082] S2.2, for empty voxels, the process enters the unsigned distance graph skipping stage, sequentially calculating distances along the x, y, and z axes. Each stage is based on the distance information from the previous stage, employing a bidirectional traversal or alternating search strategy. The distance from each voxel to the nearest non-empty voxel is calculated using the Euclidean distance formula. A three-dimensional unsigned distance field is generated by iteratively finding the minimum value. The three-dimensional unsigned distance field stores the Euclidean distance from the current voxel to the nearest non-empty voxel. The calculation formula is:

[0083] ;

[0084] in, Given the known distance of the current voxel to the nearest non-empty voxel, To represent the distance value of the queried offset voxel in the x-axis direction, For the current voxel coordinates, To take the minimum.

[0085] S2.3, because the spatial jump method based on the three-dimensional unsigned distance field directly calculates the skippable sampling distance based on the distance value of the current sampling point, small-step jumps may occur frequently when the distance value is small, which introduces additional computational overhead and reduces rendering efficiency. This invention introduces a minimum jump step size constraint strategy, setting a minimum jump threshold based on the volume data scale. Each time the UDF distance value is queried, it is compared with the minimum jump threshold. If the distance value is greater than the minimum jump threshold, a jump is performed, effectively avoiding inefficient small jumps and improving overall rendering performance. Minimum jump threshold The calculation formula is:

[0086] ;

[0087] in, This represents the diagonal length of the volume data. This represents the maximum number of sampling points.

[0088] Specifically, to further improve the rendering efficiency of non-empty voxel regions, this invention introduces the concept of importance-driven rendering and proposes an adaptive sampling step size adjustment strategy based on gradient magnitude, volumetric data value, and viewpoint distance. This strategy determines the importance of sampling points based on their attributes and dynamically adjusts the sampling density: in regions with longer viewpoint distances, smaller gradient magnitudes, or lower volumetric data values, the voxels in these regions contribute less to the final image and therefore have lower importance. The sampling step size is dynamically increased for these regions to reduce redundant computation and improve overall rendering performance. Step S3 specifically includes the following steps:

[0089] S3.1, Normalize the current ray sampling coordinates, and read the volume data value of the current ray sampling coordinates.

[0090] S3.2, copy the volume data value to the three RGB channels and expand it into grayscale color to obtain the color value of the sampling point. Then, use the color value and the transparency of the sampling point to perform color accumulation and synthesis.

[0091] S3.3, acquire volume data, gradient magnitude, and view distance; determine the importance of sampling points based on their attributes; and dynamically adjust the sampling density. The weighting formulas for volume data values ​​and gradient magnitude are as follows:

[0092] ;

[0093] ;

[0094] in, Indicates the gradient magnitude weight; Indicates the weight of the volume data value; The number of intervals to divide the volume data value; For the gradient magnitude to fall within the interval The probability, For volume data values ​​to fall within the interval The probability of.

[0095] For viewpoint distance, the range of viewpoint distance for sampling points is divided into three segments: far, medium, and near, and each segment is assigned a corresponding weighting coefficient:

[0096] ;

[0097] in, The viewpoint distance weight.

[0098] S3.4, combining gradient magnitude weights, volumetric data value weights, and viewpoint distance weights, yields an adaptive sampling step size adjustment formula to dynamically adjust the sampling step size, thereby improving rendering efficiency while maintaining image quality:

[0099] ;

[0100] in, , , These represent the normalized volume data gradient magnitude, volume data value, and location distance, respectively. The step size before adjustment. This is the adjusted step size.

[0101] This strategy can effectively control sampling density without significantly affecting image quality.

[0102] To verify the effectiveness of the method of this invention, simulation experiments were conducted on five volumetric datasets of different sizes and sparsity levels on an open scientific visualization dataset platform, and the results of different methods were compared. The datasets are shown in Table 1:

[0103] Table 1 Datasets and Key Metrics

[0104]

[0105] While keeping the rendering window size the same, the rendering speeds of five volume rendering methods were compared: the classic ray casting method, the space jump method based on the occupancy map, the volume rendering space jump acceleration method based on the Chebyshev distance map, the volume rendering space jump acceleration method based on the unsigned distance field, and the volume rendering space jump acceleration method based on the layered distance field and importance driven proposed in this invention. The comparison results are shown in Table 2.

[0106] Table 2 Comparison of rendering acceleration effects of the method on different datasets

[0107]

[0108] The comparative experimental results are shown in Table 2. Regarding acceleration performance, the proposed method achieves an average frame rate improvement of 1.4-4.4 times on various datasets, significantly higher than traditional ray casting algorithms (1.0x) and comparative methods such as OCCM (1.1x-3.1x), CDM (1.2-4.1x), and UDF (1.2-2.7x). This advantage is most pronounced on the highly sparsity aneurism dataset (1.01%) (4.38x speedup), fully validating the effectiveness of the method provided in accelerating empty voxel jumps.

[0109] The method of this invention can significantly improve the rendering efficiency of volumetric data, thereby accelerating the rendering process. While accelerating rendering, it is essential to ensure that image quality is not significantly affected. This invention further employs the Structural Similarity Index (SSIM) to quantitatively analyze the rendering results. The experiment uses the rendering results of the traditional ray casting algorithm as a reference standard. By calculating the SSIM values ​​of the images rendered by each acceleration method and the reference image, the balance between performance improvement and quality preservation of different methods is objectively evaluated. The results are shown in Table 3.

[0110] Table 3. Results of Pulse Signal Screening and Parameter Estimation

[0111]

[0112] Regarding image quality preservation, quantitative evaluation using the Structural Similarity Index (SSIM) shows that the method provided by this invention can maintain a high fidelity of over 0.95 (range 0.9587-0.9996) in all test scenarios, indicating that it can effectively improve rendering speed while ensuring rendering quality.

[0113] Therefore, the volume rendering blank space jump acceleration method based on hierarchical distance field and importance driven proposed in this invention can effectively improve the rendering speed of volume data while ensuring high rendering quality.

[0114] Of course, the above description is not intended to limit the present invention, and the present invention is not limited to the examples given above. Any changes, modifications, additions or substitutions made by those skilled in the art within the scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A method for rendering acceleration of volume data oriented to white space hopping, characterized in that, Specifically comprising the following steps: S1, the body data is blocked according to the fixed size, in the process of ray traversal body data, judge whether the current ray sampling point belongs to the body data block of non empty voxel, if it is blank body data block, then enter chebyshev jump stage, based on chebyshev distance map jump blank body data block, if it is not blank body data block, then enter voxel level jump stage; S2, in the voxel level jump stage, judge whether the voxel is non empty, if it is non empty, then execute ray projection sampling process, if it is empty, then enter the jump stage based on unsigned distance map jump blank voxel; S3, in the ray projection sampling process, based on gradient amplitude, body data value and viewpoint distance, the adaptive sampling step is adjusted; S4, the distance m of the incident light to the outgoing light is calculated, when the body data is sampled each time, the distance n of the ray traversed in the cube is calculated, if n >= m, then it is explained that the ray shoots out the cube, then stop iteration; Step S2 specifically comprises the following steps: S2.1, traverse all voxels in the body data, judge whether each voxel is non empty; The UDF value of non empty voxel is set to 0, and the distance of other empty voxels is initialized to positive infinity; S2.2, for the empty voxels, then enter the unsigned distance map based jump stage; in turn along the x, y, z axis direction distance propagation calculation, each stage is based on the distance information of the previous stage, using bidirectional traversal or alternating search strategy, through the Euclidean distance formula to calculate the distance value of each voxel to the nearest non-empty voxel, through the iteration of the minimum value operation to generate a three-dimensional unsigned distance field; the three-dimensional unsigned distance field stores the Euclidean distance from the current voxel to the nearest non-empty voxel ; the calculation formula is: ； wherein, is the distance from the current voxel to the nearest non-empty voxel known for the current voxel, is a distance value representing the distance of the queried offset voxel in the x-axis direction, is the current voxel coordinate, is to take the minimum; S2.3, set a minimum hop threshold based on the scale of the volume data Each time the UDF distance value is queried, the UDF distance value is compared with the minimum hop threshold, and when the distance value is greater than the minimum hop threshold, a hop is performed, the minimum hop threshold The calculation formula is: ； wherein, represents the diagonal length of the volume data, is the maximum number of sampling points; Step S3 specifically comprises the following steps: S3.1, normalize the current ray sampling coordinate, read the body data value of the current ray sampling coordinate; S3.2, copy the body data value to the RGB three channels to expand to gray color, get the color value size of the sampling point, and use the color value size and the transparency of the sampling point to synthesize the color accumulation; S3.3, get the body data, gradient amplitude and view distance, determine the importance according to the attribute of the sampling point, and dynamically adjust the sampling density, wherein the weight calculation formula of the body data value and the gradient amplitude is: ； ； in, Indicates the gradient magnitude weight; Indicates the weight of the volume data value; The number of intervals to divide the volume data value; For the gradient magnitude to fall within the interval The probability, For volume data values ​​to fall within the interval The probability of; For the viewpoint distance, the view distance range of the sampling point is divided into three sections of far, middle and near, and each interval is assigned a corresponding weight coefficient: ； wherein, is the view distance weight; S3.4, combine the gradient amplitude weight, the body data value weight and the viewpoint distance weight, and get the adaptive sampling step adjustment formula: ； wherein, , , are normalized gradient magnitude, volume data value and position distance respectively, is the step length before adjustment, is the step length after adjustment.

2. The method of claim 1, wherein, Step S1 specifically comprises the following steps: S1.1, assuming the volume data dimension is , the occupancy map dimension is , the block size is , then for a normalized sample point coordinate , the coordinate of the volume data in the occupancy map is : ； S1.2, assuming a normalized step size of then In the occupancy map, the step size is: ； S1.3, verify whether the current block is empty based on the occupancy map; If it is empty, then enter the chebyshev jump stage, which includes: skip the remaining part of the current block, skip the blank body data block based on chebyshev distance.

3. The method of claim 2, wherein, Step S1.3 specifically comprises the following steps: S1.3.1, in some dimension The formula for jumping the remaining part of the current block is: ； wherein, represents the non-normalized coordinate of the current sampling point in the occupancy map in the i-th dimension, represents the non-normalized coordinate of the current sampling point in the occupancy map in the i-th dimension, represents the sampling distance in the i-th dimension direction of the three-dimensional coordinate system XYZ, represents the sampling distance in the i-th dimension direction of the three-dimensional coordinate system XYZ, is the floor function. S1.3.2, the number of blank data block is calculated by the Chebyshev distance from the current blank block to the nearest non-blank block, and the corresponding Chebyshev distance map is generated; the Chebyshev distance denotes the maximum value of the absolute values of the differences between the respective coordinate components of the two vectors the maximum value of the absolute values of the differences​ ； wherein is taken to be the maximum value; S1.3.3, combine step S1.3.2 and step S1.3.3, the jump distance calculation formula of chebyshev jump stage is: ； wherein, denotes the distance to the next non-empty body data block reached by jumping at the current position, denotes the sign function; is the Heaviside step function, denotes the offset of the current point in the original empty block, is the minimum, is the ceiling. S1.3.4, simplify the jump distance calculation formula of step S1.3.3, get: 。

Images