A distance-gated lidar image reconstruction method based on an improved centroid method

Abstract

The application provides a distance-gated laser radar image reconstruction method based on an improved centroid method, and solves the technical problem that the existing laser radar echo signal strength is affected by the difference in object surface reflectivity, resulting in difficulty in distinguishing the detection target signal from the background noise, and the distortion of the reconstructed image distance information. It comprises obtaining a plurality of radar slice images, extracting the pixel intensity of a certain pixel point in the plurality of radar slice images and drawing a pixel intensity value trend graph; performing background judgment on the pixel intensity value trend graph, if it is background, then clearing the centroid distance of a certain pixel point; otherwise, it is a target, calculating the centroid distance of a certain pixel point, and iterating until the centroid distance matrix of all pixel points is obtained, and the distance image of the reconstructed target object is obtained through pseudo-color processing. The application can be widely applied in the field of laser radar imaging technology.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of laser radar imaging, and more particularly relates to a range-gated laser radar image reconstruction method based on an improved centroid method. BACKGROUND

[0002] Laser radar is a technology that uses pulsed laser as a transmitting source, receives the reflected echo signal intensity of the laser beam on the target object, and obtains the distance between the target object and the laser to realize the detection of the shape of the target object. The characteristics of short wavelength, narrow pulse width, smaller divergence angle, and higher pulse energy of laser are used to realize the detection of target objects at a farther distance with higher resolution.

[0003] The range-gated technology realizes the reception of echo signals of a specific distance by precisely controlling the transmission time of pulsed laser and the opening time of the gated detector, and is an important method for suppressing the backscattering effect of laser. However, when detecting a large target object, the sampling area needs to cover the entire target object. In order to not increase the opening time of the gating door, the time width of the range-gated door is generally much larger than the delay step length of the multiple continuous step slice images of the detector, resulting in a time overlap between each slice image and the previous and next several images, which increases the background noise of the target object.

[0004] The existing Chinese invention patent with the application number CN201910801487.6 proposes a range-gated laser radar imaging method based on adjacent frame difference, which obtains the effective distances corresponding to the front and rear edges of the ICCD gating door by performing difference operation on the adjacent two intensity images, and realizes fast imaging. However, this patent only considers single target scanning imaging in the imaging process, and does not consider the influence of the difference in reflectivity of the object surface on the echo signal intensity of the laser radar in a complex environment background, which makes it difficult to distinguish the detected target signal from the background noise, and the distance information of the reconstructed image is distorted. SUMMARY

[0005] The purpose of the embodiments of the present application is to provide a range-gated laser radar image reconstruction method based on an improved centroid method to solve the technical problem that the echo signal intensity of the laser radar is affected by the difference in reflectivity of the object surface, which makes it difficult to distinguish the detected target signal from the background noise, and the distance information of the reconstructed image is distorted.

[0006] To achieve the above-mentioned purpose, the embodiments of the present application provide a range-gated laser radar image reconstruction method based on an improved centroid method, comprising the following steps:

[0007] Obtain a plurality of radar slice images, extract the pixel intensity of a certain pixel point in the plurality of radar slice images, and draw a pixel intensity value trend graph;

[0008] If the pixel intensity value trend chart is background, the centroid distance of a pixel point is cleared; otherwise, the centroid distance of a pixel point is calculated, and iteration is performed until the centroid distance matrix of all pixel points is obtained, and the distance image of the reconstructed target object is obtained through pseudo-color processing.

[0009] Preferably, the process of obtaining the centroid distance comprises: extracting the centroid in the pixel intensity value trend chart to obtain the frame number of the centroid corresponding to the radar slice image, multiplying the resolution distance of adjacent two frames and adding the starting distance to obtain the centroid distance.

[0010] The starting distance refers to the distance of the starting node of the radar slice image.

[0011] Preferably, the process of obtaining the centroid comprises:

[0012] The maximum and minimum values of the pixel intensity in the pixel intensity value trend chart are obtained, and the difference between the maximum and minimum values of the pixel intensity is equally divided to obtain a plurality of pixel intensity value gradients.

[0013] The abscissa corresponding to the first and last time when the pixel intensity value is greater than the pixel intensity value gradient is obtained, and the median value is taken to obtain the distance frame approximation value.

[0014] Iteration is performed in sequence until all distance frame approximation values are obtained, and then the average is taken to obtain the centroid.

[0015] Preferably, the process of background judgment comprises:

[0016] The maximum and minimum values of the pixel intensity in the pixel intensity value trend chart are obtained, the range is calculated, and a threshold value is set.

[0017] If the range is less than the threshold value, a pixel point is background, and the centroid distance is set to zero; otherwise, a pixel point is a target, and the centroid distance of all pixel points is calculated.

[0018] Preferably, the formula of the pixel intensity value gradient is:

[0019] ;

[0020] In the formula, is the pixel intensity value gradient, is the maximum value of the pixel intensity, is the minimum value of the pixel intensity, is the total number of equal division, =1,2,…… is the equal division.

[0021] Preferably, the formula of the centroid distance is as follows:

[0022] ;

[0023] In the formula, is the centroid distance corresponding to the pixel point (x, y), d0 is the initial distance, j is the frame number of the radar slice image corresponding to the centroid, is the resolution distance of adjacent frames of the range-gated laser radar.

[0024] Preferably, the pixel intensity in the pixel intensity value change trend graph is sorted by the bubble method to obtain the maximum value and the minimum value of the pixel intensity.

[0025] Preferably, the centroid in the pixel intensity value change trend graph is extracted by gradually approaching through the dichotomy method.

[0026] Preferably, the range formula is:

[0027] ;

[0028] In the formula, is the extreme value, is the maximum value in the pixel intensity change trend graph, is the minimum value in the pixel intensity change trend graph.

[0029] Preferably, clearing the centroid distance of a certain pixel point means setting the centroid distance to the initial distance.

[0030] The application has the beneficial effects that the application provides a range-gated laser radar image reconstruction method based on an improved centroid method. First, the intensity value change trend graph of a certain pixel point in a plurality of radar slice images is drawn by obtaining the pixel intensity of the pixel point, background judgment is performed according to the intensity value change trend graph, background judgment of a certain pixel point is realized, the intensity change in the time sequence is considered instead of the intensity value at a single moment, misjudgment caused by the reflectivity difference of the object surface is effectively avoided, and the background and the target are accurately distinguished. Then, the centroid distance of a certain pixel point that is the background is cleared, and the interference of the background noise on the subsequent distance image generation is avoided. Subsequently, the centroid distance of the target corresponding pixel point is calculated based on the pixel intensity value change trend graph, and the centroid distance matrix of all pixel points is obtained through iterative processing. The actual centroid distance information is reflected according to the pixel intensity value change trend, does not depend on a single reflectivity threshold judgment, reduces the distance information distortion caused by the reflectivity difference, and improves the accuracy of the reconstructed image. BRIEF DESCRIPTION OF DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the application, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the application, and other drawings can be obtained by those skilled in the art without creative labor.

[0032] Figure 1 A flowchart of the overall process of a distance-gated laser radar image reconstruction method based on an improved centroid method provided by an embodiment of the present application is shown in FIG. 1.

[0033] Figure 2 A pixel intensity value trend diagram of a pixel point (290, 230) provided by an embodiment of the present application is shown in FIG. 2.

[0034] Figure 3 A background judgment flowchart provided by an embodiment of the present application is shown in FIG. 3.

[0035] Figure 4 A flowchart of extracting a centroid of a pixel intensity value trend diagram provided by an embodiment of the present application is shown in FIG. 4.

[0036] Figure 5 An indoor scene diagram provided by an embodiment of the present application is shown in FIG. 5.

[0037] Figure 6 An indoor scene difference method imaging diagram provided by an embodiment of the present application is shown in FIG. 6.

[0038] Figure 7 An indoor scene improved centroid method imaging diagram provided by an embodiment of the present application is shown in FIG. 7.

[0039] Figure 8 An outdoor scene diagram provided by an embodiment of the present application is shown in FIG. 8.

[0040] Figure 9 An outdoor scene difference method imaging diagram provided by an embodiment of the present application is shown in FIG. 9.

[0041] Figure 10 An outdoor scene improved centroid method imaging diagram provided by an embodiment of the present application is shown in FIG. 10. DETAILED DESCRIPTION

[0042] In order to make the technical problems to be solved by the present application, the technical solutions and beneficial effects clearer, the present application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and do not limit the present application.

[0043] Please refer to Figure 1 A distance-gated laser radar image reconstruction method based on an improved centroid method provided by an embodiment of the present application includes the following steps:

[0044] S1: Obtain a plurality of radar slice images, extract intensity values of a certain pixel point in the plurality of radar slice images, and draw a pixel intensity value trend diagram.

[0045] The range-gated lidar starts its range stepping from an initial distance d0, performing range steps with a consistent step value to acquire K consecutive radar slice images at different distances. The size of each radar slice image is [size missing]. Let the number of pixels in the j-th frame of the radar slice be... Pixel intensity is represented as ,in, , , M is the number of pixels in the vertical direction of the radar slice image, N is the number of pixels in the horizontal direction of the radar slice image, and K is the number of frames of the radar slice image acquired continuously.

[0046] In an optional embodiment, for the first pixel in the radar slice image Iterate through the pixel intensity of each pixel in the K-frame radar slice images at different distances. ,in, Plot a trend graph of pixel intensity value changes. Pixel The pixel intensity value change trend chart is as follows Figure 2 As shown.

[0047] S2: Analyze the background of the pixel intensity value trend map. If it is the background, reset the centroid distance of a certain pixel to zero; otherwise, if it is the target, calculate the centroid distance of a certain pixel.

[0048] Please see Figure 3 As a flowchart for background judgment, this application uses the bubble sort method to obtain the maximum value in the trend graph of the pixel intensity value change of a pixel (x, y) in a radar slice image. and minimum value The range is calculated based on the maximum and minimum values, using the following formula:

[0049] ;

[0050] In the formula, It is an extreme value. The maximum value in the pixel intensity value change trend graph. This represents the minimum value in the pixel intensity value change trend graph. It is worth noting that this application does not limit the method for obtaining the maximum and minimum values; the appropriate method can be chosen based on the actual situation.

[0051] A threshold T is empirically set based on the background intensity outside the radar gating zone; if the range... Then the pixel point Set the corresponding centroid distance as the background. This pixel The corresponding centroid distance is reset to zero; if the range Then determine the pixel point To obtain the centroid of the pixel intensity value change trend map, a bisection method is used to gradually approximate the target. The corresponding centroid frame number j.

[0052] Specifically, the implementation process of the binary search centroid is as follows: Figure 4 As shown, the specific steps are as follows:

[0053] First, obtain the pixels. The maximum pixel intensity in the pixel intensity value trend graph and minimum value ;

[0054] Next, the maximum pixel intensity and minimum value difference Divide into equal parts, and obtain The gradient of the intensity value of each pixel is given by the following formula:

[0055] ;

[0056] In the formula, The pixel intensity gradient, The maximum pixel intensity. The minimum pixel intensity The total number of equal parts, =1,2,... , for the first equal parts.

[0057] Then, take the pixel intensity value of that pixel that first exceeds the pixel intensity value gradient. The corresponding x-coordinate 'a' and the gradient greater than the pixel intensity value. The x-coordinate b at time a is used as the midpoint between x-coordinate a and x-coordinate b, and is taken as a distance frame approximation for that pixel, as shown in the following formula:

[0058] ;

[0059] Finally, by analogy, we can obtain... pixel intensity gradient The number of frames in the radar slice image corresponding to the centroid of a pixel is obtained by averaging the approximate distance frame values. , obtain pixels The centroid of the pixel intensity variation trend graph.

[0060] In an optional embodiment, a certain pixel is (290, 230), and the resolution distance between two adjacent frames and the pixel value are considered. The centroid is obtained by multiplying the number of frames in the radar slice image corresponding to the radar centroid. corresponding centroid distance , wherein, is the resolution distance of two adjacent frames of the laser radar.

[0061] S3: iteratively calculate the centroid distance of all pixel points to obtain a centroid distance matrix, and perform pseudo-color processing to obtain a reconstructed distance image of the target object.

[0062] For each pixel point in the radar slice image , the centroid distance of the pixel point is calculated in turn , and a centroid distance matrix of the target image is formed , the centroid distance matrix is processed by pseudo-color processing, the distance range corresponding to the color bar is the step range of the laser radar, and the distance image of the detected target object is reconstructed.

[0063] Specific embodiment 1: comparison experiment of one indoor scene reconstruction image.

[0064] Please refer to Figure 5 , which is an indoor scene imaging schematic diagram of the algorithm of the present application. A distance-gated laser radar is used to detect a star-shaped target at a distance of 12 m in an indoor scene. The star-shaped target is placed on a stool 12 m in front of the radar emission end. The distance-gated laser radar starts to step and image from a distance of 10 m, the step time is set to 0.25 ns, the resolution distance of two adjacent frames is 0.0375 m, and a total of 186 continuous step slice images are obtained. Figure 5 is a schematic diagram of the target scene, Figure 6 is the result image of image reconstruction of the target using the difference method. As can be seen from the figure, due to the different reflectivities of the white and black surfaces of the star-shaped target, the target is in the same distance plane as a whole, but the image reconstructed by the difference method cannot distinguish the distance information of the targets with different reflectivities. The distance image of the detected indoor scene obtained by using the algorithm of the present application is shown in Figure 7 , as can be seen from the figure, the reconstructed image of the target in the same distance plane has basically consistent color. The objects in different distance planes can accurately obtain the distance information between the measured target and the laser radar emission end based on the color characteristics of the distance image, and effectively avoid the influence of the surface reflection characteristics of the target.

[0065] Specific embodiment 2: comparison experiment of one outdoor complex scene reconstruction image.

[0066] Please refer to Figure 8 , which is an outdoor complex scene imaging schematic diagram of the algorithm of the present application. A distance-gated laser radar is used to detect Figure 8 ​The system starts step imaging from 50m, and the step time is set to 0.25ns, the resolution distance of adjacent two frames is 0.0375m, and 1120 continuous step slice images are obtained. Due to the difference in reflection characteristics of the target, there is a significant difference between the front and side echo signals of the iron house. Figure 9 As a result of using the differential method to reconstruct the radar image, the side of the iron house is farther than the front, but the distance image formed is unable to accurately reflect the distance information of the target due to the weak echo signal. The pseudo-color processed detection outdoor scene distance image obtained by using the image reconstruction method of the present application is as shown in the figure Figure 10 As can be seen from the figure, the color of the iron house from the front to the side gradually changes with the distance, which is consistent with the change of the distance information of the target in the actual scene. The detection distance image formed by the algorithm of the present application can clearly obtain the target image and the corresponding distance information, and the difference in echo signal almost has no effect on the result.

[0067] Those skilled in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be realized by electronic hardware or a combination of computer software and electronic hardware. Whether the functions are realized in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to realize the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0068] The above-described embodiments are only used to illustrate the technical solutions of the present application, and not to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacements for some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A range-gated lidar image reconstruction method based on an improved centroid method, characterized in that, The method comprises the following steps: Obtaining a plurality of radar slice images, extracting pixel intensity of a certain pixel point in the plurality of radar slice images, and drawing a pixel intensity value trend graph; Performing background judgment on the pixel intensity value trend graph, if it is background, clearing the centroid distance of the certain pixel point; Otherwise, it is a target, calculating the centroid distance of the certain pixel point, iterating until the centroid distance matrix of all pixel points is obtained, and performing pseudo-color processing to obtain a reconstructed target object distance image; The process of obtaining the centroid distance comprises: extracting the centroid in the pixel intensity value trend graph, obtaining the frame number corresponding to the radar slice image of the centroid, multiplying the resolution distance of adjacent two frames and adding the starting distance to obtain the centroid distance; The starting distance refers to the distance of the starting node of the radar slice image; The process of obtaining the centroid comprises: Obtaining the maximum and minimum values of the pixel intensity in the pixel intensity value trend graph, equally dividing the difference between the maximum and minimum values of the pixel intensity to obtain a plurality of pixel intensity value gradients; Obtaining the abscissa corresponding to the first and last time when the pixel intensity value is greater than the pixel intensity value gradient, and taking the median value to obtain a distance frame approximation value; Iterating in turn until all the distance frame approximation values are obtained, and then taking the average to obtain the centroid; The formula of the centroid distance is as follows: ； In the formula, is the distance between the pixel point (x, y) and the center of mass corresponding to the pixel point (x, y), d0 is the initial distance, j is the frame number of the radar slice image corresponding to the center of mass, is the resolution distance of the adjacent frames of the range-gated laser radar.

2. The distance-gated laser radar image reconstruction method based on the improved centroid method according to claim 1, characterized in that, The process of background judgment comprises: Obtaining the maximum and minimum values of the pixel intensity in the pixel intensity value trend graph, calculating the range and setting a threshold value; If the range is less than the threshold value, the certain pixel point is background, and the centroid distance is set to zero; otherwise, the certain pixel point is a target, and the centroid distance of all pixel points is calculated.

3. The distance-gated laser radar image reconstruction method based on the improved centroid method according to claim 1, characterized in that, The formula of the pixel intensity value gradient is: ； wherein is the pixel intensity value gradient, is the pixel intensity maximum, is the pixel intensity minimum, is the total number of equal parts, = 1, 2,... is the first equal part.

4. The distance-gated laser radar image reconstruction method based on the improved centroid method according to claim 1, characterized in that, The maximum and minimum values of the pixel intensity are obtained by sorting the pixel intensity in the pixel intensity value trend graph through the bubble method.

5. The distance-gated laser radar image reconstruction method based on the improved centroid method according to claim 1, characterized in that, The centroid in the pixel intensity value trend graph is extracted by gradually approaching through the dichotomy method.

6. The distance-gated laser radar image reconstruction method based on the improved centroid method according to claim 2, characterized in that, The range formula is: ； wherein is a maximum value, is a maximum value in the pixel intensity trend graph, is a minimum value in the pixel intensity trend graph.

7. The distance-gated laser radar image reconstruction method based on the improved centroid method according to claim 1, characterized in that, Clearing the centroid distance of the certain pixel point refers to setting the centroid distance to the starting distance.

Images