CN114241475A - Self-adaptive automobile type identification method and system based on three-dimensional laser scanning - Google Patents
Self-adaptive automobile type identification method and system based on three-dimensional laser scanning Download PDFInfo
- Publication number
- CN114241475A CN114241475A CN202111389143.2A CN202111389143A CN114241475A CN 114241475 A CN114241475 A CN 114241475A CN 202111389143 A CN202111389143 A CN 202111389143A CN 114241475 A CN114241475 A CN 114241475A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- point cloud
- area
- cloud data
- type
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 62
- 229910000831 Steel Inorganic materials 0.000 claims abstract description 80
- 239000010959 steel Substances 0.000 claims abstract description 80
- 239000002994 raw material Substances 0.000 claims abstract description 20
- 238000000638 solvent extraction Methods 0.000 claims abstract description 13
- 239000002699 waste material Substances 0.000 claims abstract description 10
- 238000000605 extraction Methods 0.000 claims description 32
- 238000007599 discharging Methods 0.000 claims description 28
- 230000003044 adaptive effect Effects 0.000 claims description 19
- 210000000481 breast Anatomy 0.000 claims description 16
- 238000005520 cutting process Methods 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 12
- 230000004888 barrier function Effects 0.000 claims description 8
- 239000000047 product Substances 0.000 abstract description 20
- 239000013067 intermediate product Substances 0.000 abstract description 15
- 238000009851 ferrous metallurgy Methods 0.000 abstract description 7
- 230000003137 locomotive effect Effects 0.000 abstract description 7
- 239000000463 material Substances 0.000 description 10
- 230000011218 segmentation Effects 0.000 description 7
- 238000001914 filtration Methods 0.000 description 6
- XEEYBQQBJWHFJM-UHFFFAOYSA-N Iron Chemical compound [Fe] XEEYBQQBJWHFJM-UHFFFAOYSA-N 0.000 description 4
- 230000011731 head segmentation Effects 0.000 description 4
- 210000000746 body region Anatomy 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 229910052742 iron Inorganic materials 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000003723 Smelting Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/2431—Multiple classes
Landscapes
- Engineering & Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a self-adaptive automobile type identification method and a system based on three-dimensional laser scanning, wherein the method comprises the following steps of firstly, carrying out three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area; then, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; then, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data; and finally, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. The invention can be applied to adaptively identify various vehicle types in the ferrous metallurgy industry, and particularly can be used for identifying a sideboard type vehicle and a flat type vehicle for transporting steel raw materials, steel waste materials, steel intermediate products and/or steel final products. The invention can adaptively acquire the coordinates and the height of the locomotive area and the effective cargo carrying area and the height of the carriage for different vehicle models.
Description
Technical Field
The invention relates to the technical field of machine vision perception, in particular to a method and a system for recognizing a self-adaptive automobile type based on three-dimensional laser scanning.
Background
In the ferrous metallurgy industry, the total transportation amount of smelting raw materials, finished steel products, scrap steel materials and the like is very large, and trucks of different types are massively applied to each link of production according to transportation requirements.
At present, the loading and unloading of vehicles mainly adopt a manual mode, and the efficiency is low and the safety accident rate is high. With the development of information technology, the demand for factory unmanned and low-human is greatly increased. In order to meet the requirements of intelligent factories, automatic identification needs to be carried out on vehicles during loading and unloading, particularly identification on a locomotive protection area and effective loading and unloading positions. In many current intelligent recognition systems, in order to reduce the complexity of a recognition algorithm for loading and unloading vehicles, a mode of a specific scene and a specific vehicle is often adopted, but the mode has insufficient flexibility and cannot meet the requirement that multiple vehicle types meeting transportation conditions can be applied in the same application scene, so that the vehicle recognition algorithm is required to be capable of adapting to the vehicle types.
Disclosure of Invention
In view of the above drawbacks of the prior art, an object of the present invention is to provide an adaptive automobile type recognition method and system based on three-dimensional laser scanning, which are used to solve the problem that in the prior art, multiple vehicles cannot be recognized in the same application scenario.
In order to achieve the above objects and other related objects, the present invention provides a method for recognizing a vehicle type of an adaptive vehicle based on three-dimensional laser scanning, the method comprising the steps of:
carrying out three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area; the target area comprises a loading and unloading area for transporting steel raw materials, transporting steel waste materials, transporting steel intermediate finished products and/or transporting steel final finished products;
clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of a vehicle to be identified in the target area;
acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data;
and determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified.
Optionally, the process of obtaining the vehicle head area of the vehicle to be identified based on the vehicle point cloud data includes:
acquiring point cloud data of a preset height range from the vehicle point cloud data along a direction vertical to a ground plane, and recording the point cloud data as height point cloud;
respectively intercepting point cloud data with a first preset length from two ends of the height point cloud along a direction parallel to a ground plane;
respectively carrying out plane recognition and extraction on the point cloud data at two intercepted ends, and taking one side with a plane as the vehicle head side of the vehicle to be recognized;
determining the parking direction of the vehicle to be identified according to the vehicle head side, and taking a plane corresponding to the vehicle head side as a vehicle head top plane;
and acquiring the geometric boundary of the vehicle head top plane, determining the vehicle head area coordinate information of the vehicle to be identified, and determining the vehicle head area of the vehicle to be identified according to the vehicle head area coordinate information.
Optionally, the process of acquiring the body region of the vehicle to be identified based on the vehicle point cloud data includes:
cutting vehicle head point cloud data from the vehicle point cloud data according to the vehicle head area coordinate information, and taking the remaining vehicle point cloud data as vehicle body point cloud data of the vehicle to be identified;
projecting the vehicle body point cloud data onto a ground plane, extracting two long line sections in the projected point cloud, and taking four end points of the two long line sections as four corner points of a vehicle body rectangular area of the vehicle to be identified;
and intercepting corresponding areas from the vehicle body point cloud data according to the four corner points, and taking the intercepted areas as the vehicle body areas of the vehicles to be identified.
Optionally, the process of determining the vehicle type of the vehicle to be identified according to the head area and the body area of the vehicle to be identified includes:
intercepting four point cloud bands of the boundary of the vehicle body area;
extracting line segments exceeding a second preset length from each point cloud band, and calculating the centroid coordinates of each line segment;
acquiring coordinates of the lowest point cloud in the vehicle body area;
calculating the height difference between the centroid coordinate of each line segment and the lowest point cloud, and judging whether the height difference exceeds a preset threshold value;
if the height difference between the barycentric coordinates of two line segments and the lowest point cloud exceeds the preset threshold value, the vehicle type of the vehicle to be identified is in a baffle type, and the two line segments are used as the line segment where the vehicle baffle is located;
and if the height difference between the centroid coordinates of all the line segments and the lowest point cloud does not exceed the preset threshold, the vehicle type of the vehicle to be identified is a flat plate type, and the corresponding line segment is used as the line segment where the vehicle bottom plate is located.
Optionally, if the vehicle to be identified is a balustrade vehicle, the method further includes:
acquiring a line segment of the tail of the vehicle, a line segment of the vehicle bottom plate and a line segment of the vehicle breast board of the vehicle;
comparing the line segment of the vehicle tail with the line segment of the vehicle bottom plate and the line segment of the vehicle breast board;
if the line segment of the tail of the vehicle and the line segment of the bottom plate of the vehicle are located on the same straight line, the barrier type vehicle is marked as a self-discharging barrier type vehicle;
and if the line segment of the tail of the vehicle and the line segment of the vehicle railing panel are positioned on the same straight line, recording the railing panel type vehicle as a non-self-unloading railing panel type vehicle.
Optionally, if the vehicle to be identified is a balustrade vehicle, the method further includes:
acquiring point cloud data corresponding to the vehicle body area, and intercepting point cloud data with a third preset length at a position close to the vehicle head;
performing plane recognition and extraction on the intercepted point cloud data according to a preset fitting threshold value;
if the plane exists, the existing plane is taken as the top plane of the front plate of the vehicle body, and the railing panel type vehicle is marked as a railing panel type vehicle with the front top plane;
and if no plane exists, marking the railing panel type vehicle as a front-roof-free plane railing panel type vehicle.
Optionally, the method further comprises:
cutting off the area where the front top plate plane of the vehicle body is located from the vehicle body area of the front top plane breast board type vehicle, taking the remaining area as a vehicle loading area, and determining the height of a vehicle container according to the lowest height value and the highest height value of the vehicle loading area;
or the body area of the front-roof-free flat-bed type vehicle is used as a vehicle loading area, and the height of a vehicle container is determined according to the lowest height value and the highest height value of the vehicle loading area.
Optionally, if the vehicle to be identified is a tablet vehicle, the method further includes:
extracting all planes in a body area of the flat plate type vehicle;
classifying the flat plate type vehicle into a single flat plate type vehicle or a multi-flat plate type vehicle according to the extracted plane data; if only one plane is extracted, the flat plate type vehicle is marked as a single flat plate type vehicle; and if a plurality of planes are extracted, the flat plate type vehicle is marked as a multi-flat plate type vehicle.
Optionally, if the vehicle to be identified is a tablet vehicle, the method further includes:
extracting all planes in a body area of the flat plate type vehicle;
the effective areas of all planes are used as vehicle loading areas, and the height value of each plane is calculated according to the vehicle loading areas.
The invention also provides a self-adaptive automobile type recognition system based on three-dimensional laser scanning, which comprises:
the data acquisition module is used for carrying out three-dimensional laser scanning on a target area to acquire three-dimensional point cloud data of the target area; the target area comprises a loading and unloading area for transporting steel raw materials, transporting steel waste materials, transporting steel intermediate finished products and/or transporting steel final finished products;
the vehicle point cloud module is used for clustering and partitioning the three-dimensional point cloud data to acquire vehicle point cloud data of vehicles to be identified in the target area;
the vehicle area module is used for acquiring a vehicle head area and a vehicle body area of the vehicle to be identified according to the vehicle point cloud data;
and the vehicle type identification module is used for determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified.
As described above, the present invention provides a method and a system for recognizing a vehicle type of an adaptive vehicle based on three-dimensional laser scanning, which has the following advantages: firstly, carrying out three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area; then, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; then, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data; and finally, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. Wherein the target areas include a handling area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. The invention can be applied to adaptively identify various vehicle types in the ferrous metallurgy industry, and particularly can be used for identifying a sideboard type vehicle and a flat type vehicle for transporting steel raw materials, steel waste materials, steel intermediate products and/or steel final products. The invention can adaptively acquire the coordinates and the height of the locomotive area and the effective cargo carrying area and the height of the carriage for different vehicle models.
Drawings
Fig. 1 is a schematic flowchart of an adaptive automobile type identification method based on three-dimensional laser scanning according to an embodiment;
FIG. 2 is a schematic illustration of a vehicle classification provided in one embodiment;
fig. 3 is a schematic flow chart of an adaptive automobile type identification method based on three-dimensional laser scanning according to another embodiment;
fig. 4 is a schematic hardware structure diagram of an adaptive automobile type identification system based on three-dimensional laser scanning according to an embodiment.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the present embodiment are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Referring to fig. 1, the present embodiment provides a method for recognizing a vehicle type of an adaptive vehicle based on three-dimensional laser scanning, including the following steps:
s100, carrying out three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area. By way of example, the manner of acquiring three-dimensional point cloud data in the present embodiment includes, but is not limited to: and 3D point cloud data acquired for the specified area based on the 3D scanner correctly installed in the scene. In this embodiment, the target areas include a loading and unloading area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. For example, the target area may refer to a parking space area defined in a garage area for loading and unloading a vehicle and an allowance thereof, and the collected point cloud data in the area includes vehicle point cloud data to be identified.
S200, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; as an example, before performing cluster segmentation on the three-dimensional point cloud data, the embodiment further includes denoising the three-dimensional point cloud data of the target region, and filtering out outliers. The method for denoising the three-dimensional point cloud data in the embodiment can adopt the existing method, for example, the point cloud data larger than the preset distance value is filtered as noise, wherein the preset distance value can be set according to the actual situation, and the embodiment does not limit specific numerical values.
S300, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data. According to the embodiment, a vehicle head physical area, a mass center coordinate and vehicle head orientation information can be obtained through vehicle head plane extraction algorithm and boundary extraction algorithm processing according to effective vehicle point clouds; and according to the effective point cloud data and the vehicle head characteristic information, vehicle body point cloud data are obtained through vehicle head segmentation algorithm and vehicle body clustering extraction algorithm processing. Specifically, as an example, the process of acquiring the vehicle head area of the vehicle to be identified based on the vehicle point cloud data includes: acquiring point cloud data of a preset height range from the vehicle point cloud data along a direction vertical to a ground plane, and recording the point cloud data as height point cloud; respectively intercepting point cloud data with a first preset length from two ends of the height point cloud along a direction parallel to a ground plane; respectively carrying out plane recognition and extraction on the point cloud data at two intercepted ends, and taking one side with a plane as the head side of the vehicle to be recognized, or taking one side with the plane and a lower plane fitting threshold value as the head side of the vehicle to be recognized; determining the parking direction of the vehicle to be identified according to the vehicle head side, and taking a plane corresponding to the vehicle head side as a vehicle head top plane; and acquiring the geometric boundary of the vehicle head top plane, determining the vehicle head area coordinate information of the vehicle to be identified, and determining the vehicle head area of the vehicle to be identified according to the vehicle head area coordinate information. As an example, the first preset length in this embodiment may be set according to an actual situation, and details are not described here. As another example, the process of acquiring the body region of the vehicle to be identified based on the vehicle point cloud data includes: cutting vehicle head point cloud data from the vehicle point cloud data according to the vehicle head area coordinate information, and taking the remaining vehicle point cloud data as vehicle body point cloud data of the vehicle to be identified; projecting the vehicle body point cloud data onto a ground plane, extracting two long line sections in the projected point cloud, and taking four end points of the two long line sections as four corner points of a vehicle body rectangular area of the vehicle to be identified; and intercepting corresponding areas from the vehicle body point cloud data according to the four corner points, and taking the intercepted areas as the vehicle body areas of the vehicles to be identified.
And S400, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. As an example, the process of determining the vehicle type of the vehicle to be identified according to the head area and the body area of the vehicle to be identified in the embodiment includes: intercepting four point cloud bands of the boundary of the vehicle body area; extracting line segments exceeding a second preset length from each point cloud band, and calculating the centroid coordinates of each line segment; acquiring coordinates of the lowest point cloud in the vehicle body area; calculating the height difference between the centroid coordinate of each line segment and the lowest point cloud, and judging whether the height difference exceeds a preset threshold value; if the height difference between the barycentric coordinates of two line segments and the lowest point cloud exceeds the preset threshold value, the vehicle type of the vehicle to be identified is in a baffle type, and the two line segments are used as the line segment where the vehicle baffle is located; and if the height difference between the centroid coordinates of all the line segments and the lowest point cloud does not exceed the preset threshold, the vehicle type of the vehicle to be identified is a flat plate type, and the corresponding line segment is used as the line segment where the vehicle bottom plate is located. The second preset length in this embodiment may be set according to an actual situation, and is not described herein again. Similarly, the preset threshold value can also be set according to the actual situation.
Therefore, in the embodiment, firstly, the target area is subjected to three-dimensional laser scanning to obtain three-dimensional point cloud data of the target area; then, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; then, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data; and finally, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. Wherein the target areas include a handling area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. The embodiment can be applied to adaptively identify various vehicle types in the ferrous metallurgy industry, and particularly can be used for identifying a sideboard type vehicle and a flat type vehicle for transporting steel raw materials, steel waste materials, steel intermediate products and/or steel final products.
According to the above description, as shown in fig. 2, in an exemplary embodiment, if the vehicle to be identified is a balustrade vehicle, the method further includes: acquiring a line segment of the tail of the vehicle, a line segment of the vehicle bottom plate and a line segment of the vehicle breast board of the vehicle; comparing the line segment of the vehicle tail with the line segment of the vehicle bottom plate and the line segment of the vehicle breast board; if the line segment of the tail of the vehicle and the line segment of the bottom plate of the vehicle are located on the same straight line, the barrier type vehicle is marked as a self-discharging barrier type vehicle; and if the line segment of the tail of the vehicle and the line segment of the vehicle railing panel are positioned on the same straight line, recording the railing panel type vehicle as a non-self-unloading railing panel type vehicle. It can be seen that the present embodiment can be used to identify both balustrade and flatbed vehicles for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products.
According to the above description, in an exemplary embodiment, if the vehicle to be identified is a balustrade vehicle, the method further includes: acquiring point cloud data corresponding to the vehicle body area, and intercepting point cloud data with a third preset length at a position close to the vehicle head; performing plane recognition and extraction on the intercepted point cloud data according to a preset fitting threshold value; if the plane exists, the existing plane is taken as the top plane of the front plate of the vehicle body, and the railing panel type vehicle is marked as a railing panel type vehicle with the front top plane; and if no plane exists, marking the railing panel type vehicle as a front-roof-free plane railing panel type vehicle.
According to the above, in an exemplary embodiment, the method further includes: cutting off the area where the front top plate plane of the vehicle body is located from the vehicle body area of the front top plane breast board type vehicle, taking the remaining area as a vehicle loading area, and determining the height of a vehicle container according to the lowest height value and the highest height value of the vehicle loading area; or the body area of the front-roof-free flat-bed type vehicle is used as a vehicle loading area, and the height of a vehicle container is determined according to the lowest height value and the highest height value of the vehicle loading area. After the height of the container of the vehicle is determined, the method can assist the iron and steel enterprises to arrange subsequent transportation business, and the method does not limit the subsequent use of the container.
In an exemplary embodiment, as shown in fig. 2, if the vehicle to be identified is a tablet vehicle, the method further includes: extracting all planes in a body area of the flat plate type vehicle; classifying the flat plate type vehicle into a single flat plate type vehicle or a multi-flat plate type vehicle according to the extracted plane data; if only one plane is extracted, the flat plate type vehicle is marked as a single flat plate type vehicle; and if a plurality of planes are extracted, the flat plate type vehicle is marked as a multi-flat plate type vehicle. If the vehicle to be identified is a flat plate type vehicle, the method further comprises the following steps: extracting all planes in a body area of the flat plate type vehicle; the effective areas of all planes are used as vehicle loading areas, and the height value of each plane is calculated according to the vehicle loading areas.
In an embodiment, as shown in fig. 2, the present embodiment further provides a vehicle classification method, including the following steps:
according to the three-dimensional laser point cloud data of the target area of the designated parking space area, whether effective vehicles exist in the parking space is judged through vehicle detection algorithm processing;
according to the point cloud of the effective parking space area, effective vehicle data in a target area are extracted through vehicle clustering segmentation extraction algorithm processing;
according to the effective vehicle point cloud, a vehicle head physical area, a centroid coordinate and vehicle head orientation information are obtained through vehicle head plane extraction algorithm and boundary extraction algorithm processing;
according to the effective point cloud data and the vehicle head characteristic information, vehicle body point cloud data are obtained through vehicle head segmentation algorithm and vehicle body clustering extraction algorithm processing;
according to the effective vehicle body point cloud data, the truck is divided into a railing panel type and a flat plate type through vehicle body feature extraction algorithm and railing panel existence judging algorithm processing;
for the railing panel type truck, the railing panel type truck is divided into a self-unloading type and a non-self-unloading type according to the judgment algorithm processing of the existence of the tail railing panel;
for the two kinds of railing panel type trucks, whether the truck body has a front panel top plane or not is judged through vehicle body top plane recognition algorithm processing according to effective railing panel type vehicle body point cloud data, and the trucks are further divided into a self-discharging railing panel with a front top, a self-discharging railing panel without a front top, a non-self-discharging railing panel with a front top and a non-self-discharging railing panel without a front top.
According to the record, for the railing panel type truck, obtaining algorithm processing according to the height of the container, calculating the lowest height of the railing panel of the container, and recording as the height of the carriage; for the truck with the front roof railing type, processing according to a front roof plane segmentation algorithm to obtain effective cargo carrying rectangular area point cloud data of the truck with the front roof railing type; for a sideboard truck, processing according to an effective cargo area boundary identification algorithm to obtain an effective cargo area boundary; for a flat plate type truck, according to effective plate type truck body point cloud data, processing the point cloud data through a vehicle bottom plane extraction and identification algorithm, calculating the number of bottom plane, and dividing the truck into a single flat plate type and a multi-flat plate type; and for the flat plate type truck, judging the boundary of each bottom plane area according to the processing of the bottom plane extraction algorithm.
In a specific embodiment, as shown in fig. 3, the present embodiment further provides a method for adaptive vehicle type recognition based on three-dimensional laser scanning, where the method is based on 3D point cloud data acquired by a 3D scanner correctly installed in a scene for a target area. The target area refers to a parking space area and the allowance thereof defined in the warehouse area and used for loading and unloading the vehicle, and the collected point cloud data in the area comprises the vehicle point cloud data to be identified. The present embodiment includes the following steps.
And step S1, point cloud data are preprocessed. Specifically, (a) denoising point cloud data, and filtering outliers; (b) and (4) obtaining point cloud in the parking space area through intelligent filtering.
And step S2, clustering and partitioning the obtained parking space area point cloud, and extracting the vehicle point cloud.
And step S3, performing vehicle head identification extraction on the vehicle point cloud data, and acquiring a vehicle head area. Specifically, (a) intercepting a height point cloud H1 in a vehicle point cloud; (b) intercepting point cloud data in the length range of L1 before and after the point cloud obtained in the step (a), respectively carrying out low-threshold vehicle head top plane recognition and extraction on the obtained two point cloud data, and extracting an effective vehicle head top plane side as a vehicle head side, thereby determining the vehicle parking direction, wherein the extracted plane is the vehicle head top plane. (c) And calculating the geometric boundary of the roof plane to obtain the coordinate information of the locomotive area.
And step S4, cutting out possible point cloud data of the vehicle head and vehicle body connection part, and extracting a vehicle body rectangular area. Specifically, (a) segmenting a vehicle head according to the coordinates of a vehicle head area and the direction of the vehicle head, and acquiring vehicle body point cloud; (b) projecting the ground plane of the vehicle body point cloud, extracting two long straight lines in the projected point cloud, wherein four end points of the two long straight lines are four corner points of a rectangular region of the vehicle body, and intercepting the inner regions of the four corner points to obtain the rectangular region of the vehicle body.
Step S5, judging the existence of the vehicle body sideboard in the rectangular region of the vehicle body, and classifying the vehicle into a sideboard type and a flat plate type, specifically, (a) intercepting four W-width point cloud bands on the boundary of the rectangular region of the vehicle body; (b) extracting straight lines with a certain length in each cloud band, and solving the mass center coordinates of each straight line; (c) calculating the height of the lowest point P of the point cloud of the vehicle body area; (d) comparing the coordinates of the mass centers of the straight lines obtained in the step (b) with the height difference of the lowest point P obtained in the step (c), if the height difference between more than two linear mass centers and the point P exceeds a threshold value, the straight lines are straight lines where the vehicle fenders are located, the vehicle is in a fence type, if the height difference between all the linear mass centers and the point P is within the range of the threshold value, the straight lines are straight lines where the vehicle bottom plate is located, and the vehicle is in a flat plate type.
According to the above description, the following steps LS1, LS2, LS3, LS4 are performed for a gate wagon.
And step LS1, judging the existence of the tailgate of the gate type truck obtained in the step S5, dividing the gate type truck into a self-discharging type and a non-self-discharging type, specifically judging that the gate type truck is the self-discharging type if the straight line is the straight line of the vehicle bottom plate according to the judgment result of the position of the tail straight line in the step S5, and judging that the gate type truck is the non-self-discharging type if the straight line is the straight line of the vehicle gate plate.
And LS2, intercepting the point cloud of the length range L2 in front of the vehicle body, identifying and extracting the top plane of the front plate of the vehicle body with a low threshold value for the obtained point cloud, judging that the fence plate type of the front top plane exists if the top plane of the front plate exists, and judging that the fence plate type of the front top plane exists if the top plane of the front plate does not exist. According to the judgment of the integrated step LS1, the breast board type truck can be divided into: the self-discharging fence plate with the front top plane, the self-discharging fence plate without the front top plane, the non-self-discharging fence plate with the front top plane and the non-self-discharging fence plate without the front top plane are adopted.
And step LS3, calculating the area of the top plane of the truck with the front top plane obtained by LS2, cutting the area of the top plane from the point cloud of the rectangular area of the truck body, wherein the rest areas are effective cargo carrying areas, and if the front top plane does not exist, all the rectangular areas of the truck body are effective cargo carrying areas.
And step LS4, calculating the height of the breast board type boxcar, specifically, comparing and taking the height value of the boundary with the lowest height and recording the height value as the height of the cargo box.
According to the above description, for a flatbed truck, the following steps PS1 and PS2 are performed.
And step PS1, calculating the number of plates, and dividing the plates into single plates and multiple plates. Specifically, all planes in the rectangular area of the body of the flatbed truck determined in step S5 are extracted, and if only one plane exists, it is determined as a single-plane type, and if a plurality of planes exist, it is determined as a multi-plane type.
In step PS2, the effective areas of all planes obtained in step PS1 are calculated as effective cargo areas and the height values of the planes are calculated.
In summary, the present invention provides a method for recognizing a vehicle type of an adaptive vehicle based on three-dimensional laser scanning, which includes performing three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area; then, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; then, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data; and finally, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. Wherein the target areas include a handling area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. The method can be applied to adaptively identify various vehicle types in the ferrous metallurgy industry, and particularly can be used for identifying a sideboard type vehicle and a flat type vehicle for transporting steel raw materials, steel waste materials, steel intermediate products and/or steel final products. The method can adaptively acquire the coordinates and the height of the locomotive area and the effective cargo area and the height of the carriage for different vehicle models.
As shown in fig. 4, the present invention further provides a system for recognizing a vehicle type of an adaptive vehicle based on three-dimensional laser scanning, wherein the system comprises:
and the data acquisition module M10 is used for performing three-dimensional laser scanning on the target area to acquire three-dimensional point cloud data of the target area. By way of example, the manner of acquiring three-dimensional point cloud data in the present embodiment includes, but is not limited to: and 3D point cloud data acquired for the specified area based on the 3D scanner correctly installed in the scene. In this embodiment, the target areas include a loading and unloading area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. For example, the target area may refer to a parking space area defined in a garage area for loading and unloading a vehicle and an allowance thereof, and the collected point cloud data in the area includes vehicle point cloud data to be identified.
And the vehicle point cloud module M20 is used for carrying out clustering segmentation on the three-dimensional point cloud data to obtain vehicle point cloud data of the vehicle to be identified in the target area. As an example, before performing cluster segmentation on the three-dimensional point cloud data, the embodiment further includes denoising the three-dimensional point cloud data of the target region, and filtering out outliers. The method for denoising the three-dimensional point cloud data in the embodiment can adopt the existing method, for example, the point cloud data larger than the preset distance value is filtered as noise, wherein the preset distance value can be set according to the actual situation, and the embodiment does not limit specific numerical values.
And the vehicle area module M30 is used for acquiring a head area and a body area of the vehicle to be identified according to the vehicle point cloud data. According to the embodiment, a vehicle head physical area, a mass center coordinate and vehicle head orientation information can be obtained through vehicle head plane extraction algorithm and boundary extraction algorithm processing according to effective vehicle point clouds; and according to the effective point cloud data and the vehicle head characteristic information, vehicle body point cloud data are obtained through vehicle head segmentation algorithm and vehicle body clustering extraction algorithm processing. Specifically, as an example, the process of acquiring the vehicle head area of the vehicle to be identified based on the vehicle point cloud data includes: acquiring point cloud data of a preset height range from the vehicle point cloud data along a direction vertical to a ground plane, and recording the point cloud data as height point cloud; respectively intercepting point cloud data with a first preset length from two ends of the height point cloud along a direction parallel to a ground plane; respectively carrying out plane recognition and extraction on the point cloud data at two intercepted ends, and taking one side with a plane as the head side of the vehicle to be recognized, or taking one side with the plane and a lower plane fitting threshold value as the head side of the vehicle to be recognized; determining the parking direction of the vehicle to be identified according to the vehicle head side, and taking a plane corresponding to the vehicle head side as a vehicle head top plane; and acquiring the geometric boundary of the vehicle head top plane, determining the vehicle head area coordinate information of the vehicle to be identified, and determining the vehicle head area of the vehicle to be identified according to the vehicle head area coordinate information. As an example, the first preset length in this embodiment may be set according to an actual situation, and details are not described here. As another example, the process of acquiring the body region of the vehicle to be identified based on the vehicle point cloud data includes: cutting vehicle head point cloud data from the vehicle point cloud data according to the vehicle head area coordinate information, and taking the remaining vehicle point cloud data as vehicle body point cloud data of the vehicle to be identified; projecting the vehicle body point cloud data onto a ground plane, extracting two long line sections in the projected point cloud, and taking four end points of the two long line sections as four corner points of a vehicle body rectangular area of the vehicle to be identified; and intercepting corresponding areas from the vehicle body point cloud data according to the four corner points, and taking the intercepted areas as the vehicle body areas of the vehicles to be identified.
And the vehicle type identification module M40 is used for determining the vehicle type of the vehicle to be identified according to the head area and the body area of the vehicle to be identified. As an example, the process of determining the vehicle type of the vehicle to be identified according to the head area and the body area of the vehicle to be identified in the embodiment includes: intercepting four point cloud bands of the boundary of the vehicle body area; extracting line segments exceeding a second preset length from each point cloud band, and calculating the centroid coordinates of each line segment; acquiring coordinates of the lowest point cloud in the vehicle body area; calculating the height difference between the centroid coordinate of each line segment and the lowest point cloud, and judging whether the height difference exceeds a preset threshold value; if the height difference between the barycentric coordinates of two line segments and the lowest point cloud exceeds the preset threshold value, the vehicle type of the vehicle to be identified is in a baffle type, and the two line segments are used as the line segment where the vehicle baffle is located; and if the height difference between the centroid coordinates of all the line segments and the lowest point cloud does not exceed the preset threshold, the vehicle type of the vehicle to be identified is a flat plate type, and the corresponding line segment is used as the line segment where the vehicle bottom plate is located. The second preset length in this embodiment may be set according to an actual situation, and is not described herein again. Similarly, the preset threshold value can also be set according to the actual situation.
Therefore, in the embodiment, firstly, the target area is subjected to three-dimensional laser scanning to obtain three-dimensional point cloud data of the target area; then, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; then, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data; and finally, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. Wherein the target areas include a handling area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. The embodiment can be applied to adaptively identify various vehicle types in the ferrous metallurgy industry, and particularly can be used for identifying a sideboard type vehicle and a flat type vehicle for transporting steel raw materials, steel waste materials, steel intermediate products and/or steel final products.
According to the above description, as shown in fig. 2, in an exemplary embodiment, if the vehicle to be identified is a balustrade vehicle, the system further includes: acquiring a line segment of the tail of the vehicle, a line segment of the vehicle bottom plate and a line segment of the vehicle breast board of the vehicle; comparing the line segment of the vehicle tail with the line segment of the vehicle bottom plate and the line segment of the vehicle breast board; if the line segment of the tail of the vehicle and the line segment of the bottom plate of the vehicle are located on the same straight line, the barrier type vehicle is marked as a self-discharging barrier type vehicle; and if the line segment of the tail of the vehicle and the line segment of the vehicle railing panel are positioned on the same straight line, recording the railing panel type vehicle as a non-self-unloading railing panel type vehicle. It can be seen that the present embodiment can be used to identify both balustrade and flatbed vehicles for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products.
According to the above description, in an exemplary embodiment, if the vehicle to be identified is a balustrade vehicle, the system further includes: acquiring point cloud data corresponding to the vehicle body area, and intercepting point cloud data with a third preset length at a position close to the vehicle head; performing plane recognition and extraction on the intercepted point cloud data according to a preset fitting threshold value; if the plane exists, the existing plane is taken as the top plane of the front plate of the vehicle body, and the railing panel type vehicle is marked as a railing panel type vehicle with the front top plane; and if no plane exists, marking the railing panel type vehicle as a front-roof-free plane railing panel type vehicle.
In accordance with the above, in an exemplary embodiment, the system further comprises: cutting off the area where the front top plate plane of the vehicle body is located from the vehicle body area of the front top plane breast board type vehicle, taking the remaining area as a vehicle loading area, and determining the height of a vehicle container according to the lowest height value and the highest height value of the vehicle loading area; or the body area of the front-roof-free flat-bed type vehicle is used as a vehicle loading area, and the height of a vehicle container is determined according to the lowest height value and the highest height value of the vehicle loading area. After the height of the container of the vehicle is determined, the method can assist the iron and steel enterprises to arrange subsequent transportation business, and the method does not limit the subsequent use of the container.
In an exemplary embodiment, as shown in fig. 2, if the vehicle to be identified is a tablet vehicle, the system further includes: extracting all planes in a body area of the flat plate type vehicle; classifying the flat plate type vehicle into a single flat plate type vehicle or a multi-flat plate type vehicle according to the extracted plane data; if only one plane is extracted, the flat plate type vehicle is marked as a single flat plate type vehicle; and if a plurality of planes are extracted, the flat plate type vehicle is marked as a multi-flat plate type vehicle. If the vehicle to be identified is a flat plate type vehicle, the system further comprises: extracting all planes in a body area of the flat plate type vehicle; the effective areas of all planes are used as vehicle loading areas, and the height value of each plane is calculated according to the vehicle loading areas.
In an embodiment, the present embodiment further provides a vehicle classification system, configured to perform the following steps:
according to the three-dimensional laser point cloud data of the target area of the designated parking space area, whether effective vehicles exist in the parking space is judged through vehicle detection algorithm processing;
according to the point cloud of the effective parking space area, effective vehicle data in a target area are extracted through vehicle clustering segmentation extraction algorithm processing;
according to the effective vehicle point cloud, a vehicle head physical area, a centroid coordinate and vehicle head orientation information are obtained through vehicle head plane extraction algorithm and boundary extraction algorithm processing;
according to the effective point cloud data and the vehicle head characteristic information, vehicle body point cloud data are obtained through vehicle head segmentation algorithm and vehicle body clustering extraction algorithm processing;
according to the effective vehicle body point cloud data, the truck is divided into a railing panel type and a flat plate type through vehicle body feature extraction algorithm and railing panel existence judging algorithm processing;
for the railing panel type truck, the railing panel type truck is divided into a self-unloading type and a non-self-unloading type according to the judgment algorithm processing of the existence of the tail railing panel;
for the two kinds of railing panel type trucks, whether the truck body has a front panel top plane or not is judged through vehicle body top plane recognition algorithm processing according to effective railing panel type vehicle body point cloud data, and the trucks are further divided into a self-discharging railing panel with a front top, a self-discharging railing panel without a front top, a non-self-discharging railing panel with a front top and a non-self-discharging railing panel without a front top.
According to the record, for the railing panel type truck, obtaining algorithm processing according to the height of the container, calculating the lowest height of the railing panel of the container, and recording as the height of the carriage; for the truck with the front roof railing type, processing according to a front roof plane segmentation algorithm to obtain effective cargo carrying rectangular area point cloud data of the truck with the front roof railing type; for a sideboard truck, processing according to an effective cargo area boundary identification algorithm to obtain an effective cargo area boundary; for a flat plate type truck, according to effective plate type truck body point cloud data, processing the point cloud data through a vehicle bottom plane extraction and identification algorithm, calculating the number of bottom plane, and dividing the truck into a single flat plate type and a multi-flat plate type; and for the flat plate type truck, judging the boundary of each bottom plane area according to the processing of the bottom plane extraction algorithm.
In a specific embodiment, as shown in fig. 3, the present embodiment further provides an adaptive vehicle type recognition system based on three-dimensional laser scanning, where the system is based on 3D point cloud data acquired by a 3D scanner correctly installed in a scene from a target area. The target area refers to a parking space area and the allowance thereof defined in the warehouse area and used for loading and unloading the vehicle, and the collected point cloud data in the area comprises the vehicle point cloud data to be identified. The system is used for executing the following steps:
and step S1, point cloud data are preprocessed. Specifically, (a) denoising point cloud data, and filtering outliers; (b) and (4) obtaining point cloud in the parking space area through intelligent filtering.
And step S2, clustering and partitioning the obtained parking space area point cloud, and extracting the vehicle point cloud.
And step S3, performing vehicle head identification extraction on the vehicle point cloud data, and acquiring a vehicle head area. Specifically, (a) intercepting a height point cloud H1 in a vehicle point cloud; (b) intercepting point cloud data in the length range of L1 before and after the point cloud obtained in the step (a), respectively carrying out low-threshold vehicle head top plane recognition and extraction on the obtained two point cloud data, and extracting an effective vehicle head top plane side as a vehicle head side, thereby determining the vehicle parking direction, wherein the extracted plane is the vehicle head top plane. (c) And calculating the geometric boundary of the roof plane to obtain the coordinate information of the locomotive area.
And step S4, cutting out possible point cloud data of the vehicle head and vehicle body connection part, and extracting a vehicle body rectangular area. Specifically, (a) segmenting a vehicle head according to the coordinates of a vehicle head area and the direction of the vehicle head, and acquiring vehicle body point cloud; (b) projecting the ground plane of the vehicle body point cloud, extracting two long straight lines in the projected point cloud, wherein four end points of the two long straight lines are four corner points of a rectangular region of the vehicle body, and intercepting the inner regions of the four corner points to obtain the rectangular region of the vehicle body.
Step S5, judging the existence of the vehicle body sideboard in the rectangular region of the vehicle body, and classifying the vehicle into a sideboard type and a flat plate type, specifically, (a) intercepting four W-width point cloud bands on the boundary of the rectangular region of the vehicle body; (b) extracting straight lines with a certain length in each cloud band, and solving the mass center coordinates of each straight line; (c) calculating the height of the lowest point P of the point cloud of the vehicle body area; (d) comparing the coordinates of the mass centers of the straight lines obtained in the step (b) with the height difference of the lowest point P obtained in the step (c), if the height difference between more than two linear mass centers and the point P exceeds a threshold value, the straight lines are straight lines where the vehicle fenders are located, the vehicle is in a fence type, if the height difference between all the linear mass centers and the point P is within the range of the threshold value, the straight lines are straight lines where the vehicle bottom plate is located, and the vehicle is in a flat plate type.
According to the above description, the following steps LS1, LS2, LS3, LS4 are performed for a gate wagon.
And step LS1, judging the existence of the tailgate of the gate type truck obtained in the step S5, dividing the gate type truck into a self-discharging type and a non-self-discharging type, specifically judging that the gate type truck is the self-discharging type if the straight line is the straight line of the vehicle bottom plate according to the judgment result of the position of the tail straight line in the step S5, and judging that the gate type truck is the non-self-discharging type if the straight line is the straight line of the vehicle gate plate.
And LS2, intercepting the point cloud of the length range L2 in front of the vehicle body, identifying and extracting the top plane of the front plate of the vehicle body with a low threshold value for the obtained point cloud, judging that the fence plate type of the front top plane exists if the top plane of the front plate exists, and judging that the fence plate type of the front top plane exists if the top plane of the front plate does not exist. According to the judgment of the integrated step LS1, the breast board type truck can be divided into: the self-discharging fence plate with the front top plane, the self-discharging fence plate without the front top plane, the non-self-discharging fence plate with the front top plane and the non-self-discharging fence plate without the front top plane are adopted.
And step LS3, calculating the area of the top plane of the truck with the front top plane obtained by LS2, cutting the area of the top plane from the point cloud of the rectangular area of the truck body, wherein the rest areas are effective cargo carrying areas, and if the front top plane does not exist, all the rectangular areas of the truck body are effective cargo carrying areas.
And step LS4, calculating the height of the breast board type boxcar, specifically, comparing and taking the height value of the boundary with the lowest height and recording the height value as the height of the cargo box.
According to the above description, for a flatbed truck, the following steps PS1 and PS2 are performed.
And step PS1, calculating the number of plates, and dividing the plates into single plates and multiple plates. Specifically, all planes in the rectangular area of the body of the flatbed truck determined in step S5 are extracted, and if only one plane exists, it is determined as a single-plane type, and if a plurality of planes exist, it is determined as a multi-plane type.
In step PS2, the effective areas of all planes obtained in step PS1 are calculated as effective cargo areas and the height values of the planes are calculated.
In summary, the invention provides a three-dimensional laser scanning-based adaptive automobile type identification system, which first performs three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area; then, clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of vehicles to be identified in the target area; then, acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data; and finally, determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified. Wherein the target areas include a handling area for transporting steel raw materials, transporting steel scrap materials, transporting steel intermediate products, and/or transporting steel final products. The system can be applied to adaptively identify various vehicle types in the ferrous metallurgy industry, and particularly can be used for identifying a sideboard type vehicle and a flat type vehicle for transporting steel raw materials, steel waste materials, steel intermediate products and/or steel final products. The system can adaptively acquire the coordinates and the height of the locomotive area and the effective cargo carrying area and height of the carriage for different vehicle models. Therefore, the invention effectively overcomes various defects in the prior art and has high industrial utilization value.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
It should be understood that although the terms first, second, third, etc. may be used to describe preset ranges, etc. in embodiments of the present invention, these preset ranges should not be limited to these terms. These terms are only used to distinguish preset ranges from each other. For example, the first preset range may also be referred to as a second preset range, and similarly, the second preset range may also be referred to as the first preset range, without departing from the scope of the embodiments of the present invention. In addition, the terms "upper", "lower", "left", "right", "middle" and "one" used in the present specification are for clarity of description, and are not intended to limit the scope of the present invention, and the relative relationship between the terms and the terms is not to be construed as a scope of the present invention.
Claims (10)
1. A self-adaptive automobile type recognition method based on three-dimensional laser scanning is characterized by comprising the following steps:
carrying out three-dimensional laser scanning on a target area to obtain three-dimensional point cloud data of the target area; the target area comprises a loading and unloading area for transporting steel raw materials, transporting steel waste materials, transporting steel intermediate finished products and/or transporting steel final finished products;
clustering and partitioning the three-dimensional point cloud data to obtain vehicle point cloud data of a vehicle to be identified in the target area;
acquiring a vehicle head area and a vehicle body area of the vehicle to be identified based on the vehicle point cloud data;
and determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified.
2. The three-dimensional laser scanning-based adaptive automobile type recognition method according to claim 1, wherein the process of acquiring the head area of the vehicle to be recognized based on the vehicle point cloud data comprises:
acquiring point cloud data of a preset height range from the vehicle point cloud data along a direction vertical to a ground plane, and recording the point cloud data as height point cloud;
respectively intercepting point cloud data with a first preset length from two ends of the height point cloud along a direction parallel to a ground plane;
respectively carrying out plane recognition and extraction on the point cloud data at two intercepted ends, and taking one side with a plane as the vehicle head side of the vehicle to be recognized;
determining the parking direction of the vehicle to be identified according to the vehicle head side, and taking a plane corresponding to the vehicle head side as a vehicle head top plane;
and acquiring the geometric boundary of the vehicle head top plane, determining the vehicle head area coordinate information of the vehicle to be identified, and determining the vehicle head area of the vehicle to be identified according to the vehicle head area coordinate information.
3. The method for recognizing the vehicle type of the self-adaptive automobile based on the three-dimensional laser scanning as claimed in claim 2, wherein the process of acquiring the body area of the vehicle to be recognized based on the vehicle point cloud data comprises the following steps:
cutting vehicle head point cloud data from the vehicle point cloud data according to the vehicle head area coordinate information, and taking the remaining vehicle point cloud data as vehicle body point cloud data of the vehicle to be identified;
projecting the vehicle body point cloud data onto a ground plane, extracting two long line sections in the projected point cloud, and taking four end points of the two long line sections as four corner points of a vehicle body rectangular area of the vehicle to be identified;
and intercepting corresponding areas from the vehicle body point cloud data according to the four corner points, and taking the intercepted areas as the vehicle body areas of the vehicles to be identified.
4. The three-dimensional laser scanning-based adaptive automobile type recognition method according to claim 1 or 3, wherein the process of determining the type of the vehicle to be recognized according to the head area and the body area of the vehicle to be recognized comprises the following steps:
intercepting four point cloud bands of the boundary of the vehicle body area;
extracting line segments exceeding a second preset length from each point cloud band, and calculating the centroid coordinates of each line segment;
acquiring coordinates of the lowest point cloud in the vehicle body area;
calculating the height difference between the centroid coordinate of each line segment and the lowest point cloud, and judging whether the height difference exceeds a preset threshold value;
if the height difference between the barycentric coordinates of two line segments and the lowest point cloud exceeds the preset threshold value, the vehicle type of the vehicle to be identified is in a baffle type, and the two line segments are used as the line segment where the vehicle baffle is located;
and if the height difference between the centroid coordinates of all the line segments and the lowest point cloud does not exceed the preset threshold, the vehicle type of the vehicle to be identified is a flat plate type, and the corresponding line segment is used as the line segment where the vehicle bottom plate is located.
5. The adaptive automobile type recognition method based on three-dimensional laser scanning is characterized in that if the vehicle to be recognized is a sideboard type vehicle, the method further comprises the following steps:
acquiring a line segment of the tail of the vehicle, a line segment of the vehicle bottom plate and a line segment of the vehicle breast board of the vehicle;
comparing the line segment of the vehicle tail with the line segment of the vehicle bottom plate and the line segment of the vehicle breast board;
if the line segment of the tail of the vehicle and the line segment of the bottom plate of the vehicle are located on the same straight line, the barrier type vehicle is marked as a self-discharging barrier type vehicle;
and if the line segment of the tail of the vehicle and the line segment of the vehicle railing panel are positioned on the same straight line, recording the railing panel type vehicle as a non-self-unloading railing panel type vehicle.
6. The adaptive automobile type recognition method based on three-dimensional laser scanning is characterized in that if the vehicle to be recognized is a sideboard type vehicle, the method further comprises the following steps:
acquiring point cloud data corresponding to the vehicle body area, and intercepting point cloud data with a third preset length at a position close to the vehicle head;
performing plane recognition and extraction on the intercepted point cloud data according to a preset fitting threshold value;
if the plane exists, the existing plane is taken as the top plane of the front plate of the vehicle body, and the railing panel type vehicle is marked as a railing panel type vehicle with the front top plane;
and if no plane exists, marking the railing panel type vehicle as a front-roof-free plane railing panel type vehicle.
7. The adaptive automobile type recognition method based on three-dimensional laser scanning is characterized in that the method further comprises the following steps:
cutting off the area where the front top plate plane of the vehicle body is located from the vehicle body area of the front top plane breast board type vehicle, taking the remaining area as a vehicle loading area, and determining the height of a vehicle container according to the lowest height value and the highest height value of the vehicle loading area;
or the body area of the front-roof-free flat-bed type vehicle is used as a vehicle loading area, and the height of a vehicle container is determined according to the lowest height value and the highest height value of the vehicle loading area.
8. The adaptive automobile type identification method based on three-dimensional laser scanning is characterized in that if the vehicle to be identified is a flat plate type vehicle, the method further comprises the following steps:
extracting all planes in a body area of the flat plate type vehicle;
classifying the flat plate type vehicle into a single flat plate type vehicle or a multi-flat plate type vehicle according to the extracted plane data; if only one plane is extracted, the flat plate type vehicle is marked as a single flat plate type vehicle; and if a plurality of planes are extracted, the flat plate type vehicle is marked as a multi-flat plate type vehicle.
9. The adaptive automobile type identification method based on three-dimensional laser scanning is characterized in that if the vehicle to be identified is a flat plate type vehicle, the method further comprises the following steps:
extracting all planes in a body area of the flat plate type vehicle;
the effective areas of all planes are used as vehicle loading areas, and the height value of each plane is calculated according to the vehicle loading areas.
10. A self-adaptive automobile type recognition system based on three-dimensional laser scanning is characterized in that the system comprises:
the data acquisition module is used for carrying out three-dimensional laser scanning on a target area to acquire three-dimensional point cloud data of the target area; the target area comprises a loading and unloading area for transporting steel raw materials, transporting steel waste materials, transporting steel intermediate finished products and/or transporting steel final finished products;
the vehicle point cloud module is used for clustering and partitioning the three-dimensional point cloud data to acquire vehicle point cloud data of vehicles to be identified in the target area;
the vehicle area module is used for acquiring a vehicle head area and a vehicle body area of the vehicle to be identified according to the vehicle point cloud data;
and the vehicle type identification module is used for determining the vehicle type of the vehicle to be identified according to the vehicle head area and the vehicle body area of the vehicle to be identified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111389143.2A CN114241475B (en) | 2021-11-22 | 2021-11-22 | Self-adaptive automobile model recognition method and system based on three-dimensional laser scanning |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111389143.2A CN114241475B (en) | 2021-11-22 | 2021-11-22 | Self-adaptive automobile model recognition method and system based on three-dimensional laser scanning |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114241475A true CN114241475A (en) | 2022-03-25 |
CN114241475B CN114241475B (en) | 2024-06-25 |
Family
ID=80750381
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111389143.2A Active CN114241475B (en) | 2021-11-22 | 2021-11-22 | Self-adaptive automobile model recognition method and system based on three-dimensional laser scanning |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114241475B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114777648A (en) * | 2022-04-20 | 2022-07-22 | 中冶赛迪重庆信息技术有限公司 | Slab sensing measurement method and system |
CN114898207A (en) * | 2022-04-13 | 2022-08-12 | 中冶赛迪信息技术(重庆)有限公司 | Vehicle self-adaptive identification method and system based on three-dimensional point cloud and image data |
CN117420562A (en) * | 2023-10-21 | 2024-01-19 | 广州市西克传感器有限公司 | Top-open carriage identification measurement system based on three-dimensional data of cradle head |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100121577A1 (en) * | 2008-04-24 | 2010-05-13 | Gm Global Technology Operations, Inc. | Three-dimensional lidar-based clear path detection |
CN104680795A (en) * | 2015-02-28 | 2015-06-03 | 武汉烽火众智数字技术有限责任公司 | Vehicle type recognition method and device based on partial area characteristic |
CN112365718A (en) * | 2020-10-25 | 2021-02-12 | 北京因泰立科技有限公司 | Laser vehicle type recognition method and device |
-
2021
- 2021-11-22 CN CN202111389143.2A patent/CN114241475B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100121577A1 (en) * | 2008-04-24 | 2010-05-13 | Gm Global Technology Operations, Inc. | Three-dimensional lidar-based clear path detection |
CN104680795A (en) * | 2015-02-28 | 2015-06-03 | 武汉烽火众智数字技术有限责任公司 | Vehicle type recognition method and device based on partial area characteristic |
CN112365718A (en) * | 2020-10-25 | 2021-02-12 | 北京因泰立科技有限公司 | Laser vehicle type recognition method and device |
Non-Patent Citations (2)
Title |
---|
YIMING ZENG等: "RT3D: Real-Time 3-D Vehicle Detection in LiDAR Point Cloud for Autonomous Driving", 《IEEE ROBOTICS AND AUTOMATION LETTERS》, vol. 3, no. 4, 31 October 2018 (2018-10-31), pages 3434 - 3440, XP055981985, DOI: 10.1109/LRA.2018.2852843 * |
刘三林: "基于激光测量数据的车型识别研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, no. 1, 15 January 2018 (2018-01-15), pages 138 - 1586 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114898207A (en) * | 2022-04-13 | 2022-08-12 | 中冶赛迪信息技术(重庆)有限公司 | Vehicle self-adaptive identification method and system based on three-dimensional point cloud and image data |
CN114898207B (en) * | 2022-04-13 | 2024-07-02 | 中冶赛迪信息技术(重庆)有限公司 | Vehicle self-adaptive identification method and system based on three-dimensional point cloud and image data |
CN114777648A (en) * | 2022-04-20 | 2022-07-22 | 中冶赛迪重庆信息技术有限公司 | Slab sensing measurement method and system |
CN114777648B (en) * | 2022-04-20 | 2023-09-05 | 中冶赛迪信息技术(重庆)有限公司 | Plate blank sensing measurement method and system |
CN117420562A (en) * | 2023-10-21 | 2024-01-19 | 广州市西克传感器有限公司 | Top-open carriage identification measurement system based on three-dimensional data of cradle head |
Also Published As
Publication number | Publication date |
---|---|
CN114241475B (en) | 2024-06-25 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN114241475B (en) | Self-adaptive automobile model recognition method and system based on three-dimensional laser scanning | |
EP1903535A1 (en) | Method for identifying vehicles in electronic images | |
CN111291603B (en) | Lane line detection method, device, system and storage medium | |
US20140176679A1 (en) | Method for Automatically Classifying Moving Vehicles | |
CN112070838B (en) | Object identification and positioning method and device based on two-dimensional-three-dimensional fusion characteristics | |
DE102015208782A1 (en) | Object detection device, driving support device, object detection method, and object detection program | |
US10286901B2 (en) | Map of the surroundings for driving areas with random altitude profile | |
CN109325389A (en) | Lane detection method, apparatus and vehicle | |
CN102842038A (en) | Environment recognition device and environment recognition method | |
CN113184707A (en) | Method and system for preventing lifting of container truck based on laser vision fusion and deep learning | |
CN115496757B (en) | Hydraulic flap excess material amount detection method and system based on machine vision | |
EP3696135B1 (en) | Forklift and system with forklift for the identification of goods | |
CN114648744A (en) | Method for determining semantic collision-free space | |
Ramanathan et al. | Spatiotemporal vehicle tracking, counting and classification | |
CN114399550B (en) | Three-dimensional laser scanning-based automobile saddle extraction method and system | |
CN109978879B (en) | Box corner in-groove state detection method based on railway wagon loading video monitoring | |
CN115291240B (en) | Detection method and system for perception and identification of retaining wall behind unloading point in mining area scene | |
CN108090895B (en) | Container lockhole contour extraction method based on image processing | |
CN115601435A (en) | Vehicle attitude detection method, device, vehicle and storage medium | |
CN113177557B (en) | Bowling prevention method and system based on machine vision and deep learning | |
CN115457506A (en) | Target detection method, device and storage medium | |
CN115015962A (en) | Collision detection method, system, equipment and storage medium based on shore bridge | |
Cai et al. | Knowledge template based multi-perspective car recognition algorithm | |
CN117710725A (en) | Container door recognition method and system based on self-vehicle perception and vehicle | |
CN112347664B (en) | Modeling method and device for irregular loading space |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 401329 No. 5-6, building 2, No. 66, Nongke Avenue, Baishiyi Town, Jiulongpo District, Chongqing Applicant after: MCC CCID information technology (Chongqing) Co.,Ltd. Address before: 401329 No. 5-6, building 2, No. 66, Nongke Avenue, Baishiyi Town, Jiulongpo District, Chongqing Applicant before: CISDI CHONGQING INFORMATION TECHNOLOGY Co.,Ltd. |
|
GR01 | Patent grant | ||
GR01 | Patent grant |