CN106203335A - Sign board visibility evaluation methodology based on three-dimensional point cloud - Google Patents

Abstract

The invention discloses a kind of sign board visibility evaluation methodology based on three-dimensional point cloud, invention proposes can the definition of visual field and spacing-visible degree, and using spacing-visible degree as sign board in the space observability evaluation criterion of different observation positions.In three-dimensional point cloud, four element method, alpha shape algorithm and retina image-forming principle is utilized to calculate the visual field intensity of sign board at this viewpoint；Whether the point cloud projection in view frustums is calculated for blocking a cloud by ray method；Utilizing the visibility of the sign board of the correlation calculations viewpoint position of visual field intensity and visibility, calculate traffic signs can visual field.The method automatically, directly perceived, accurately, efficiently, install at traffic mark board safeguard, billboard is thrown in, can be given in terms of roadside greening and architectural design and reasonably instruct and suggestion, has the highest actual promotional value.

Description

Evaluation method of signboard visibility based on 3D point cloud

technical field

The invention relates to a method for evaluating the visibility of a signboard based on a three-dimensional point cloud.

Background technique

Signs refer to signs that use words or symbols to mark features and attract the attention of pedestrians. They are very common in our daily life, including prohibition signs, warning signs, tourist signs, speed limit signs, instruction signs, and reflective signs. , Safety signs, etc. Billboards are also one of its extended uses. As a guiding sign, the sign should give people a striking visual impact. However, in practical applications, there are many factors that restrict its eye-catching effect. Tree shading is one of the most influential factors. Traffic signs are blocked, drivers cannot see the instructions in advance, and it is easy to take the wrong route or slow down to find instructions, causing unnecessary congestion and rear-end collision accidents; billboards are blocked, and investors cannot get the expected return on investment ; And for relevant management departments, the pruning of greening trees can only be carried out by subjective judgment, resulting in repeated pruning of some road sections, some road sections are ignored, low efficiency and unsatisfactory results. Therefore urgently need a kind of method that can intuitively and accurately evaluate the eye-catching degree of signboard in practical application, as signboard installation, billboard puts in and whether greening tree is pruned the guiding method and suggestion.

Contents of the invention

The purpose of the present invention is to overcome the shortcomings of the above-mentioned prior art and provide a method for evaluating the spatial visibility of signboards based on three-dimensional point cloud data.

To achieve the above object, the present invention adopts the following technical solutions:

A method for evaluating the visibility of signboards based on 3D point clouds, including the following steps:

S1. Calculate the reference visual field strength of the signboard at the reference viewpoint: take the normal line passing through the center of the signboard and the position d meters away from the center of the signboard as the reference viewpoint, and calculate the distance of the signboard on the retina when observed at the reference viewpoint Imaging area, define the visible field intensity at this time as 1;

S2. Calculate the visual field intensity of the signboard at the viewpoint: calculate the ratio of the imaging area of the signboard on the retina when observed at the viewpoint to the imaging area of the signboard on the retina when observed at the reference viewpoint, and the ratio is the Visible field strength;

S3. Calculating the spatial visibility of the signboard at the viewpoint: after calculating the center of the signboard, the ratio between the area of the unoccluded part of the signboard and the area of the signboard on the plane with the center point of the signboard and the line connecting the viewpoint as the normal vector, the ratio is the spatial visibility at this point;

S4. Calculate the spatial visibility of the signboard at the viewpoint: the spatial visibility of the point is the product of the visual field strength of the point and the spatial visibility of the point;

S5. Calculate the visible field of the signboard: calculate and store the spherical coordinates of all viewpoints whose distance from the center of the signboard is R according to the spherical equation, repeat steps S2-S4, calculate the spatial visibility of each viewpoint, and establish the observation distance when R is Visible field model for signage.

Further, step S1 specifically includes the following steps:

S11. Using the existing signboard extraction algorithm, extract the signboard point cloud in the point cloud scene;

S12. Calculate the center of the signboard point cloud and the normal vector Normal of the signboard plane;

S13, using the four-element method to rotate the signboard point cloud normal vector into the coordinate space where the signboard plane normal vector Normal is the Z axis;

S14, projecting the signboard point cloud to the XOY plane, using the alpha-shape algorithm to calculate the boundary of the projected point cloud;

S15. Calculate the area of the signboard using the polygonal area formula according to the boundary vertices of the projected point cloud;

S16. Calculate the distance between the reference viewpoint and the center of the signboard, and calculate the imaging area of the signboard on the retina according to the principle of retinal imaging.

Further, the step S2 specifically includes:

S21. Calculate the line vector L of the line between the viewpoint and the center of the signboard, and use the four-element method to rotate the point cloud of the signboard into the coordinate space with the line vector L as the Z axis;

S22, projecting the signboard point cloud to the XOY plane, using the alpha-shape algorithm to calculate the boundary of the projected point cloud;

S23. According to the boundary vertices of the projected point cloud, the area of the signboard is calculated using the polygonal area formula;

S24. Calculate the distance between the viewpoint and the center of the signboard, and calculate the imaging area of the signboard on the retina according to the principle of retinal imaging, and the ratio of the imaging area to the reference visual field intensity is the visual field intensity at the viewpoint.

Further, the step S3 specifically includes:

S31. Repeat step S21 to rotate the coordinates of the viewpoint into the coordinate space with the line vector L as the Z axis;

S32. In the transformed coordinate space, according to the sign board boundary vertex coordinates calculated in step S22, connect the connection line between the viewpoint and each vertex, and calculate the angle between the connection line and the connection line between the viewpoint and the center of the sign board, Then calculate and find the largest angle α;

S33. In the original coordinate space, take the line connecting the viewpoint and the center of the signboard as the direction, and use α as the angle of view to form a viewing cone, and judge the connection between the point cloud between the viewpoint and the point cloud in the viewing cone and the connection between the viewpoint and the center of the signboard Whether the angle formed by the line is greater than α, if > α, discard it, and if ≤ α, proceed to the next step;

S34. Calculate the point of intersection between the line connecting the viewpoint and the point cloud in the viewing frustum and the plane where the signboard is located. The point cloud of this intersection point is recorded as Cloud1, and the point cloud Cloud1 is rotated to the coordinate space with the line vector L as the Z axis by using the four-element method. In , the point cloud after coordinate transformation is recorded as Cloud2;

S35. Project the point cloud Cloud2 onto the XOY plane, and judge whether the point of the Cloud2 projection point cloud is inside the boundary of the signboard projection point cloud calculated in step S22 by the ray method. If the point is inside, it is regarded as an intersection point, otherwise Abandon this point, and the obtained intersection point cloud set is recorded as Cloud3;

S36, using the alpha-shape algorithm to calculate the boundary of the point cloud Cloud3, according to the vertices of the boundary of Cloud3, use the polygonal area formula to calculate the area of the intersection point cloud;

S37. Calculate the area ratio of the point cloud boundary of the intersection point to the signboard calculated in step S24, and subtract the ratio from 1 to obtain the spatial visibility at the viewpoint.

After adopting the technical solution, the present invention has the following advantages compared with the background technology:

It can automatically, intuitively, accurately and efficiently evaluate the visibility of signboards, and give reasonable guidance and suggestions on signboard installation and maintenance, billboard placement, road greening and architectural design, especially in urban roads. Use the on-board laser scanning system to scan to determine the degree of greening trees blocking the sign boards, as a reference for the greening tree pruning cycle and pruning amount.

Description of drawings

Figure 1 is a schematic diagram of the construction of the visual field model;

Figure 2 is a schematic diagram of the occluder during observation;

Fig. 3 is a block diagram of the calculation process of the present invention;

Fig. 4 is a block diagram of calculating the visible field intensity of the present invention;

Fig. 5 is a schematic diagram of polygon area calculation;

6 is a schematic diagram of the principle of retinal imaging;

Fig. 7 is a flow chart of spatial visibility calculation in the present invention;

Fig. 8 is the top view result of point cloud computing at the intersection point;

Fig. 9 side view result of intersection point cloud computing;

Fig. 10 is the result figure of occlusion point cloud calculation;

Figure 11 is the calculation result of the boundary of Cloud3 and the boundary of the signboard;

Fig. 12 is a schematic diagram of a spherical viewpoint point cloud stored in an embodiment;

Fig. 13 is the display result of the visible field with or without occlusion in the embodiment.

detailed description

In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.

Example

In order to intuitively express the degree of visibility of the signboard at different distances and different viewing angles, the definition of the visual field is introduced, as shown in Figure 1. The definition of the visible field is: the center of the signboard is the center of the sphere, and R is the radius. Create a three-dimensional sphere, and define the hemisphere corresponding to the front of the signboard as the visible field of the signboard.

Then each point on the sphere is a viewpoint v _i with a distance R from the signboard, and the viewpoint v _i and the signboard form a viewing cone. The spatial visibility of the human eye on the sign board is affected by the distance of the viewpoint v _i and the observation angle: when the observation distance is constant, for different viewpoints, the observation angle is inversely proportional to the imaging area of the sign board on the retina, and the observation angle deviates from the sign The larger the normal direction of the sign, the smaller the retinal imaging area; when the observation angle is constant, the observation distance is inversely proportional to the imaging area of the sign on the retina, and the farther the distance is, the smaller the retinal imaging area is.

In order to express this relationship intuitively, the influence of viewpoint position on spatial visibility is defined as visible field intensity. In order to measure the intensity of the visible field and normalize the intensity of the visible field, the position at the center of the signboard, along the normal direction of the signboard and away from the center of the signboard d meters (d is usually <10, depending on the actual situation) is defined is the reference viewpoint, record the imaging area of the sign board on the retina at this time, define the visual field intensity at this time as 1, then the visual field intensity of the viewpoint v _i is the imaging area of the sign board on the retina at the viewpoint v _i The ratio of the area to the imaged area of the sign on the retina at the reference point of view.

Therefore, each viewpoint v _i on the hemisphere has a unique visible field strength, and the visible field strength attenuates with the increase of the distance from the viewpoint, and the attenuation with the increase of the angle away from the normal direction of the signboard. The visible field intensity at the reference point of view is 1, and the visible field intensity at infinity or at a 90-degree declination to the normal direction of the signboard is 0.

However, as shown in Figure 2, without considering the illumination, the signboard will still be occluded by different objects at different viewpoints in the actual environment. In the figure, the circle and star represent the occlusion of objects such as leaves and street light poles . Therefore, it is inaccurate to evaluate the spatial visibility of sign boards only by using the visual field strength. The position of the viewpoint is different, the area of the observed sign board is different, and the occlusion area formed by the occluder is also different, which is intuitive. To express this relationship accurately, mark the influence of occlusion degree on spatial visibility as spatial visibility, then the spatial visibility of the viewpoint is defined as: on the plane passing through the center of the signboard and taking the line connecting the center point of the signboard and the viewpoint as the normal vector, The ratio of the area of the unoccluded part of the signboard to the area of the signboard. Then, the larger the occlusion area, the smaller the spatial visibility.

Therefore, the spatial visibility of the signboard is the product of the visual field intensity of the viewpoint and the spatial visibility. Calculating the spatial visibility of each viewpoint in the visual field can obtain the spatial visibility of the visual field.

As shown in Figure 3, the specific calculation steps are as follows:

S1. Calculate the reference visual field strength of the signboard at the reference viewpoint, specifically:

S11. Using the existing signboard extraction algorithm, extract the signboard point cloud in the point cloud scene;

S12. Calculate the center of the signboard point cloud and the normal vector Normal of the signboard plane;

S13, using the four-element method to rotate the signboard point cloud normal vector into the coordinate space where the signboard plane normal vector Normal is the Z axis;

S14, the signboard point cloud is projected to the XOY plane (this time the XOY plane is the plane that crosses the signboard point cloud center and the signboard normal), and uses the alpha-shape algorithm to calculate the boundary of the projected point cloud;

S15. Calculate the area of the signboard using the polygonal area formula according to the boundary vertices of the projected point cloud;

S16. Calculate the distance between the reference viewpoint and the center of the signboard, and calculate the imaging area of the signboard on the retina according to the principle of retinal imaging.

S2 calculates the visible field strength of the signboard at the viewpoint, specifically:

S21. Calculate the line vector L of the line between the viewpoint and the center of the signboard, and use the four-element method to rotate the point cloud of the signboard into the coordinate space with the line vector L as the Z axis;

S22, the signboard point cloud is projected to the XOY plane (this time the XOY plane is the plane that passes the signboard point cloud center and the line vector L is perpendicular), and uses the alpha-shape algorithm to calculate the boundary of the projected point cloud;

S23. According to the boundary vertices of the projected point cloud, the area of the signboard is calculated using the polygonal area formula;

S24. Calculate the distance between the viewpoint and the center of the signboard, and calculate the imaging area of the signboard on the retina according to the principle of retinal imaging, and the ratio of the imaging area to the reference visual field intensity is the visual field intensity at the viewpoint.

As shown in Figure 4 is the calculation flow diagram of steps S1 and S2, that is, the calculation flow diagram of field intensity, as shown in Figure 5 is a schematic diagram of polygon area calculation, and as shown in Figure 6 is a schematic diagram of the principle of retinal imaging . Among them, the four-element method, alpha-shape and polygon area calculation belong to the prior art, and the specific algorithm will not be described in detail here. It should be noted that the value of alpha will affect the determination of the boundary and should be set according to the actual situation. This embodiment The middle is set to 0.1 meters.

S3. Calculating the spatial visibility of the signboard at the viewpoint, specifically including:

S31. Repeat step S21 to rotate the coordinates of the viewpoint into the coordinate space with the line vector L as the Z axis;

S32. In the transformed coordinate space, according to the sign board boundary vertex coordinates calculated in step S22, connect the connection line between the viewpoint and each vertex, and calculate the angle between the connection line and the connection line between the viewpoint and the center of the sign board, Then calculate and find the largest angle α;

S33. In the original coordinate space, take the line connecting the viewpoint and the center of the signboard as the direction, and take α as the angle of view to form a viewing cone. At this time, the point cloud in the viewing cone is the point that may block the signboard. Cloud, judge whether the angle formed by the connection line between the viewpoint and the point cloud in the viewing frustum and the connection line between the viewpoint and the center of the signboard is greater than α, if >α, discard it, and if ≤α, proceed to the next step;

S34. Calculate the point of intersection between the line connecting the viewpoint and the point cloud in the viewing frustum and the plane where the signboard is located. The point cloud of this intersection point is recorded as Cloud1, and the point cloud Cloud1 is rotated to the coordinate space with the line vector L as the Z axis by using the four-element method. In , the point cloud after coordinate transformation is recorded as Cloud2;

S35, point cloud Cloud2 is projected on the XOY plane (this moment XOY plane promptly crosses the plane that signboard point cloud center is perpendicular to line vector L), judges whether the point of Cloud2 projection point cloud is calculated in step S22 by ray method Inside the boundary of the signboard projection point cloud, if the point is inside, it is regarded as an intersection point, otherwise the point is discarded, and the obtained intersection point cloud is recorded as Cloud3;

S36, using the alpha-shape algorithm to calculate the boundary of the point cloud Cloud3, according to the vertices of the boundary of Cloud3, use the polygonal area formula to calculate the area of the intersection point cloud;

S37. Calculate the area ratio of the point cloud boundary of the intersection point to the signboard calculated in step S24, and subtract the ratio from 1 to obtain the spatial visibility at the viewpoint.

As shown in FIG. 7 is a flow chart of step S3 (that is, calculation of spatial visibility). Figures 8 to 10 are, in order, in step S35: the top-view results of the intersection point cloud calculation, the side-view results of the intersection point cloud calculation, and the occlusion point cloud calculation results. Figure 11 shows the calculation results of the boundary of Cloud3 and the boundary of the signboard.

S4. Calculating the spatial visibility of the signboard at the viewpoint: the spatial visibility of the point is the product of the visible field strength of the point and the spatial visibility of the point.

S5. Calculate the visible field of the signboard:

With the center of the signboard as the center of the sphere and the observation distance R as the radius, a sphere model is established, and the positive hemisphere facing the front of the signboard is taken, θ, are the angles in the horizontal and vertical directions respectively, according to the parameter equation of the ball,

Use the four-element coordinate transformation method to rotate the spherical point cloud to the front of the signboard. Store the spherical coordinates of all viewpoints on the positive hemisphere, repeat steps S2-S4, calculate the spatial visibility of each viewpoint, and establish the visual field model of the signboard when the observation distance is R.

Figure 12 is a schematic diagram of the stored spherical viewpoint point cloud. As shown in Figure 13, the reference point of view in the visual field is 10 meters, the radius of the visual field is 13 meters, and the display results are both horizontal and vertical angles of 6 degrees. The left picture is the display result without occlusion, and the right picture is The display result when there is occlusion.

The above is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed in the present invention can easily think of changes or Replacement should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (4)

1. The signboard visibility evaluation method based on three-dimensional point cloud, is characterized in that, comprises the following steps: S1. Calculate the reference visual field strength of the signboard at the reference viewpoint: take the normal line passing through the center of the signboard and the position d meters away from the center of the signboard as the reference viewpoint, and calculate the distance of the signboard on the retina when observed at the reference viewpoint Imaging area, define the visible field intensity at this time as 1; S2. Taking the point whose distance from the center of the signboard as R as the viewpoint, calculate the visible field strength of the signboard at the viewpoint: calculate the imaging area of the signboard on the retina when observing at the viewpoint and the area of the signboard on the retina when observing at the reference viewpoint The ratio of the imaging area, which is the visible field intensity at this point; S3. Calculating the spatial visibility of the signboard at the viewpoint: after calculating the center of the signboard, the ratio between the area of the unoccluded part of the signboard and the area of the signboard on the plane with the center point of the signboard and the line connecting the viewpoint as the normal vector, the ratio is the spatial visibility at this point; S4. Calculate the spatial visibility of the signboard at the viewpoint: the spatial visibility of the point is the product of the visual field strength of the point and the spatial visibility of the point; S5. Calculate the visible field of the signboard: calculate and store the spherical coordinates of all viewpoints whose distance from the center of the signboard is R according to the spherical equation, repeat steps S2-S4, calculate the spatial visibility of each viewpoint, and establish the observation distance when R is Visible field model for signage. 2. the signboard visibility evaluation method based on three-dimensional point cloud according to claim 1, is characterized in that, step S1 specifically comprises the following steps: S11. Using the existing signboard extraction algorithm, extract the signboard point cloud in the point cloud scene; S12. Calculate the center of the signboard point cloud and the normal vector Normal of the signboard plane; S13, using the four-element method to rotate the signboard point cloud normal vector into the coordinate space with the normal vector Normal as the Z axis; S14, projecting the signboard point cloud to the XOY plane, using the alpha-shape algorithm to calculate the boundary of the projected point cloud; S15. Calculate the area of the signboard using the polygonal area formula according to the boundary vertices of the projected point cloud; S16. Calculate the distance between the reference viewpoint and the center of the signboard, and calculate the imaging area of the signboard on the retina according to the principle of retinal imaging. 3. the signboard visibility evaluation method based on three-dimensional point cloud according to claim 1, is characterized in that, described step S2 specifically comprises: S21. Calculate the line vector L of the line between the viewpoint and the center of the signboard, and use the four-element method to rotate the point cloud of the signboard into the coordinate space with the line vector L as the Z axis; S22, projecting the signboard point cloud to the XOY plane, using the alpha-shape algorithm to calculate the boundary of the projected point cloud; S23. According to the boundary vertices of the projected point cloud, the area of the signboard is calculated using the polygonal area formula; S24. Calculate the distance between the viewpoint and the center of the signboard, and calculate the imaging area of the signboard on the retina according to the principle of retinal imaging, and the ratio of the imaging area to the reference visual field intensity is the visual field intensity at the viewpoint. 4. the signboard visibility evaluation method based on three-dimensional point cloud according to claim 1, is characterized in that, described step S3 specifically comprises: S31. Repeat step S21 to rotate the coordinates of the viewpoint into the coordinate space with the line vector L as the Z axis; S32. In the converted coordinate space, according to the sign board border vertex coordinates calculated in step S22, connect the line between the viewpoint and each vertex, and calculate the angle between the line and the line between the view point and the center of the sign board, Then calculate and find the largest angle α; S33. In the original coordinate space, take the line connecting the viewpoint and the center of the signboard as the direction, and use α as the angle of view to form a viewing cone, and judge the connection between the point cloud between the viewpoint and the point cloud in the viewing cone and the connection between the viewpoint and the center of the signboard Whether the angle formed by the line is greater than α, if > α, discard it, and if ≤ α, proceed to the next step; S34. Calculate the point of intersection between the line connecting the viewpoint and the point cloud in the viewing frustum and the plane where the signboard is located. The point cloud of this intersection point is recorded as Cloud1, and the point cloud Cloud1 is rotated to the coordinate space with the line vector L as the Z axis by using the four-element method. In , the point cloud after coordinate transformation is recorded as Cloud2; S35. Project the point cloud Cloud2 onto the XOY plane, and judge whether the point of the Cloud2 projection point cloud is inside the boundary of the signboard projection point cloud calculated in step S22 by the ray method. If the point is inside, it is regarded as an intersection point, otherwise Abandon this point, and the obtained intersection point cloud set is recorded as Cloud3; S36, using the alpha-shape algorithm to calculate the boundary of the point cloud Cloud3, according to the vertices of the boundary of Cloud3, use the polygonal area formula to calculate the area of the intersection point cloud; S37. Calculate the area ratio of the point cloud boundary of the intersection point to the signboard calculated in step S24, and subtract the ratio from 1 to obtain the spatial visibility at the viewpoint.