CN112184707A - Method and system for judging earth and stone load of muck vehicle based on point cloud data - Google Patents
Method and system for judging earth and stone load of muck vehicle based on point cloud data Download PDFInfo
- Publication number
- CN112184707A CN112184707A CN202011191481.0A CN202011191481A CN112184707A CN 112184707 A CN112184707 A CN 112184707A CN 202011191481 A CN202011191481 A CN 202011191481A CN 112184707 A CN112184707 A CN 112184707A
- Authority
- CN
- China
- Prior art keywords
- muck
- vehicle
- loading
- neural network
- point cloud
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 239000004575 stone Substances 0.000 title claims abstract description 25
- 238000003062 neural network model Methods 0.000 claims abstract description 38
- 239000002893 slag Substances 0.000 claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 7
- 238000005520 cutting process Methods 0.000 claims abstract description 6
- 238000004140 cleaning Methods 0.000 claims description 18
- 238000012549 training Methods 0.000 claims description 18
- 238000010276 construction Methods 0.000 claims description 16
- 238000012360 testing method Methods 0.000 claims description 12
- 239000011435 rock Substances 0.000 claims description 8
- 238000013527 convolutional neural network Methods 0.000 claims description 7
- 238000005303 weighing Methods 0.000 claims description 7
- 238000005507 spraying Methods 0.000 claims description 5
- SAZUGELZHZOXHB-UHFFFAOYSA-N acecarbromal Chemical compound CCC(Br)(CC)C(=O)NC(=O)NC(C)=O SAZUGELZHZOXHB-UHFFFAOYSA-N 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 230000003749 cleanliness Effects 0.000 claims description 3
- 238000005286 illumination Methods 0.000 claims description 2
- 238000001514 detection method Methods 0.000 abstract description 4
- 238000004364 calculation method Methods 0.000 description 10
- 238000005259 measurement Methods 0.000 description 8
- 210000002569 neuron Anatomy 0.000 description 6
- 238000010586 diagram Methods 0.000 description 5
- 239000002689 soil Substances 0.000 description 4
- 239000007921 spray Substances 0.000 description 4
- 239000000126 substance Substances 0.000 description 4
- 238000001914 filtration Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 238000012805 post-processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000007599 discharging Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000010304 firing Methods 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000003908 quality control method Methods 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
- G06T17/20—Finite element generation, e.g. wire-frame surface description, tesselation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10028—Range image; Depth image; 3D point clouds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Biophysics (AREA)
- Mathematical Physics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Geometry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a method for judging earth and stone load of a muck vehicle based on point cloud data, which comprises the following steps: collecting a whole-vehicle image of the slag vehicle, inputting the image into a first neural network model, and identifying the type of the slag vehicle to obtain a standard loading capacity of the corresponding type of the slag vehicle; collecting images of the muck carried by the muck vehicle, inputting the images into a second neural network model, and identifying types of the muck carried by the muck vehicle to obtain the density of the muck; scanning the muck loaded by the muck truck to obtain point cloud data of the muck loaded by the muck truck; denoising and cutting the point cloud data of the muck to obtain a three-dimensional image of the muck; processing the information result of the three-dimensional image recognition to obtain a loading volume; judging whether the loading condition of the muck truck is qualified or not; and calculating to obtain the load of the muck truck and judging whether the muck truck is overweight. The method and the device automatically judge the type of the muck truck, calculate the volume of the earthwork and automatically judge the loading condition of the muck truck, and have the advantages of high detection efficiency, time and labor saving, good accuracy, convenience and practicability.
Description
Technical Field
The invention belongs to the technical field of earth and stone square statistics of engineering projects, and particularly relates to a method and a system for judging earth and stone square loading of a muck truck based on point cloud data.
Background
The civil engineering project ubiquitous earth and stone side's transportation, the earth and stone side number of need counting when the dregs car carrying earth and stone side leaves the building site, at present most building site still takes traditional artifical counting mode to count, but traditional counting mode easily receives to be limited to factors such as managers' learning and quality of equipment operating personnel influence and causes all kinds of ticket evasion acts as a rule, lacks phenomenons such as normalized construction and counting error. The phenomenon of insufficient carrying or overload exists in the transportation process of the slag car, and the safety and the environmental protection of the city are affected to a certain extent by the illegal getting-on of the slag car. The utility model discloses a utility model "a hand-held device for engineering transportation earthwork measurement" that publication number is CN208795315U discloses an earthwork metering device, can not judge the loading condition of dregs car automatically. The invention patent with publication number CN109002421A discloses a method and a system for statistics of earth transportation by muck trucks, a server and a storage medium, which directly use images of muck trucks to obtain earth volume, and the accuracy is problematic. The prior art lacks an accurate and efficient automatic statistical method and system for the number of earth and stone squares of the muck truck.
Disclosure of Invention
The invention aims to solve the problems and provides a method and a system for judging earth and stone load of a muck truck based on point cloud data.
The technical scheme adopted by the invention is a method for judging the earth and stone load of a muck vehicle based on point cloud data, which comprises the following steps:
step 1: collecting a whole-vehicle image of the slag vehicle, inputting the image into a first neural network model, and identifying the type of the slag vehicle by using the first neural network model to obtain a standard loading capacity of the corresponding type of the slag vehicle;
step 2: collecting images of the muck carried by the muck car, inputting the images into a second neural network model, and identifying the types of the muck carried by the muck car by using the second neural network model to obtain the density of the muck;
and step 3: scanning the muck loaded by the muck truck by using a three-dimensional laser scanner to obtain point cloud data of the muck loaded by the muck truck;
and 4, step 4: denoising and cutting the point cloud data of the muck to obtain a three-dimensional image of the muck;
and 5: processing the information result of the three-dimensional image recognition to obtain a loading volume;
step 6: comparing the loading volume of the slag car with the standard loading volume obtained in the step (1) and judging whether the loading condition of the slag car is qualified or not;
and 7: and (3) calculating the load of the muck truck according to the loading volume of the muck truck and the density of the muck obtained in the step (2), and judging whether the muck truck is overweight.
Further, step 1 further comprises identifying a cleanliness level of the slag car from the full car image of the slag car using the first neural network model.
Preferably, the first neural network model adopts a Faster R-CNN neural network, and comprises a convolutional neural network, an RPN network and a full connection layer which are connected in sequence.
Preferably, the second neural network model adopts a BP neural network, the input layer comprises 10 neurons, the hidden layer comprises 11 neurons, and the output layer is 1 neuron.
Alternatively, step 1 may also be to collect images of the muck truck, identify the license plate number of the muck truck, and obtain the type of the muck truck and the standard loading capacity of the corresponding type according to the correspondence between the pre-stored license plate number and model.
The method for judging the earth and rock load of the muck vehicle based on the point cloud data further comprises the step of pre-training a first neural network model, and specifically comprises the following steps:
1) setting reference surfaces for the muck trucks of different vehicle types according to the maximum loading limit height of baffles at two sides of the muck trucks to measure the standard loading capacity of the muck trucks;
2) shooting all-vehicle images of the muck vehicles of different vehicle types with different cleaning degrees, combining the all-vehicle images of the muck vehicles with the cleaning degrees of the muck vehicles and standard carrying capacity data of corresponding types of the muck vehicles to form a muck vehicle data set, and dividing the muck vehicle data set into a training set and a testing set;
3) and training and testing the first neural network model by utilizing the muck loading data set.
The method for judging the earth-rock space loading of the muck vehicle based on the point cloud data further comprises the following steps of pre-training a second neural network model, and specifically comprises the following steps:
1) collecting a certain volume of muck of each construction site, and classifying to obtain classified muck samples;
2) weighing each type of muck sample, calculating the density of the muck sample according to the muck weighing and the muck volume, collecting an image of the muck sample to form a muck data set, and dividing the muck data set into a training set and a testing set;
3) and training and testing the second neural network model by utilizing the muck data set.
The muck vehicle loading identification and judgment system utilizing the muck vehicle earth and stone loading judgment method based on point cloud data comprises a computer, and a GPS receiver, a camera, a laser scanner, a camera and a display screen which are respectively connected with the computer.
Preferably, the muck vehicle loading identification and judgment system further comprises a controller and a spraying and cleaning device, wherein the control end of the spraying and cleaning device is connected with the controller, and the controller is connected with the computer through a data line.
Preferably, the muck truck loading identification and judgment system further comprises an illumination sensor connected with the controller.
Compared with the prior art, the invention has the beneficial effects that:
1) according to the image of the muck car, the type of the muck car is automatically judged, the point cloud data of the earthwork loaded by the muck car obtained by three-dimensional laser scanning is utilized, the volume of the earthwork is calculated, and the standard loading capacity corresponding to the type of the muck car is compared, so that the loading condition of the muck car is automatically judged, and the muck car loading detection method is high in detection efficiency, time-saving, labor-saving, good in accuracy, convenient and practical;
2) according to the method, the muck is automatically identified and classified according to the image of the muck to obtain the density of the muck, the load of the muck truck is obtained by combining the volume of earth and stone, whether the overweight phenomenon exists or not is automatically judged, manual work is replaced, and the weighing efficiency of the muck truck is improved;
3) according to the image of the muck truck, the cleaning degree of the muck truck and whether the muck truck is subjected to spray cleaning or not are automatically judged, manual work is replaced, the muck truck delivery efficiency is improved, the cleaning degree of the surface of the muck truck is guaranteed, and the management level of the muck truck is improved.
Drawings
The invention is further illustrated by the following figures and examples.
Fig. 1 is a schematic structural view of a muck vehicle loading identification and determination system according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a first neural network model according to an embodiment of the present invention.
FIG. 3 is a diagram of a second neural network model according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of three-dimensional laser scanning of a muck truck according to an embodiment of the invention.
Fig. 5 is a schematic diagram of three-dimensional laser scanning.
Fig. 6 is a schematic diagram of measuring the cubic volume of the earth and the stone by three-dimensional laser scanning according to the embodiment of the invention.
Detailed Description
Example one
In the embodiment, a parking measurement area is divided at an exit of a civil engineering project building site, a muck truck enters the parking measurement area after loading muck, laser scanning is performed on loaded earth and rock after collecting images of the whole muck truck and images of the muck, and automatic judgment of loading conditions is performed, as shown in fig. 4.
The method for judging the earth and rock load of the muck car based on the point cloud data comprises the following steps of sequentially executing,
step 1: collecting a whole-vehicle image of the muck vehicle by using a camera, inputting the image into a first neural network model, and identifying the type and the cleanliness of the muck vehicle by using the first neural network model to obtain a standard loading capacity of the corresponding type of the muck vehicle;
step 2: collecting images of the muck carried by the muck car by using a high-definition camera, inputting the images into a second neural network model, and identifying the types of the muck carried by the muck car by using the second neural network model to obtain the density of the muck;
and step 3: scanning the muck loaded by the muck truck by using a three-dimensional laser scanner to obtain point cloud data of the muck loaded by the muck truck;
and 4, step 4: denoising and cutting the point cloud data of the muck to obtain a three-dimensional image of the muck;
and 5: processing the information result of the three-dimensional image recognition to obtain a loading volume;
step 6: comparing the loading volume of the slag car with the standard loading volume obtained in the step (1) and judging whether the loading condition of the slag car is qualified or not;
and 7: calculating the load of the muck truck according to the loading volume of the muck truck and the density of the muck obtained in the step 2, and judging whether the muck truck is overweight;
and 8: and if the judged result of the cleaning degree of the muck truck is that the dust is more, starting the spray cleaning equipment to perform spray cleaning with the outer side of the muck truck carriage and the tires, and if not, ending the spray cleaning.
The first neural network model adopts a Faster R-CNN network, as shown in fig. 2, and includes a convolutional neural network, an RPN network, and a full connection layer, which are connected in sequence, the convolutional neural network extracts features of an input picture, the RPN network recommends a candidate region, and outputs are converted into a fixed size by using a RoI firing layer. The convolutional neural network comprises 13 conv layers, 13 relu layers and 4 Pooling layers. The pre-training is carried out before the use, and the method specifically comprises the following steps:
1) setting reference surfaces for the muck trucks of different vehicle types according to the maximum loading limit height of baffles at two sides of the muck trucks to measure the standard loading capacity of the muck trucks;
2) shooting all-vehicle images of the muck vehicles of different vehicle types with different cleaning degrees, combining the all-vehicle images of the muck vehicles with the cleaning degrees of the muck vehicles and standard carrying capacity data of corresponding types of the muck vehicles to form a muck vehicle data set, and dividing the muck vehicle data set into a training set and a testing set;
3) and training and testing the first neural network model by utilizing the muck loading data set.
The second neural network model adopts a BP network, an input layer comprises 10 neurons, a hidden layer comprises 11 neurons, and an output layer comprises 1 neuron, as shown in FIG. 3.
The second neural network model is pre-trained before being used, and specifically comprises the following steps:
1) collecting a certain volume of muck of each construction site, and classifying to obtain classified muck samples;
2) weighing each type of muck sample, calculating the density of the muck sample according to the muck weighing and the muck volume, collecting an image of the muck sample to form a muck data set, and dividing the muck data set into a training set and a testing set;
3) and training and testing the second neural network model by utilizing the muck data set.
The muck vehicle loading identification and judgment system utilizing the muck vehicle earth and stone loading judgment method based on point cloud data comprises a computer, and a GPS receiver, a CCD camera, a laser scanner, a camera and a display screen which are respectively connected with the computer, wherein a controller is connected with the computer through a data line, and a control end of a spraying and cleaning device is connected with the controller, as shown in figure 1. In the embodiment, the GPS receiver adopts an RTK carrier phase difference technology, and the positioning precision is centimeter level.
In the system operation process, the laser scanner carries out laser scanning to the earth and rockfill that the dregs car loaded, and the texture of dregs is shot along the same direction of laser scanner to the CCD camera, provides accurate positional information through the GPS receiver. In the data processing process, the accurate three-dimensional space rectangular coordinates (X, Y, Z) of each laser foot point can be obtained through the information measured by the GPS and the inclined distance and angle measured by the laser pulse, and the three-dimensional coordinates of the surface elevation of the loading object with a certain width can be obtained through scanning.
The three-dimensional laser scanner is a main device for data acquisition, and is mainly constructed by a high-speed accurate laser range finder which guides emitted laser through a reflecting prism and scans a detected substance at a uniform angular speed. The principle is that a laser pulse signal is emitted by a three-dimensional laser scanner emitter, after diffused reflection is carried out on the surface of an object, the laser pulse signal is reversely transmitted back to a receiver along the same path, the distance L from a target point C to the scanner can be calculated, and an encoder is controlled to synchronously measure a longitudinal scanning angle observation value a and a transverse scanning angle observation value beta of each laser pulse, and the principle is shown in fig. 5. Selecting a custom elevation for the observed substance, and establishing a coordinate system.
The three-dimensional coordinates of the spatial position of each sampling point on the surface of the loaded object are quickly and accurately acquired by arranging the coordinate point data to obtain a point set of an entity, a three-dimensional model of the loaded object is reconstructed in a computer, and various post-processing and application work is carried out on the modeling data by utilizing self-contained software.
The raw data output by the laser scanner are the distance values and the angle values of the scanning points. The loading limit of the muck truck is taken as a horizontal reference layer for three-dimensional laser recognition, a coordinate system is established in the three-dimensional laser recognition range by the horizontal reference layer, as shown in fig. 6, an X axis is in a transverse scanning plane, a Y axis is perpendicular to the X axis in the transverse scanning plane, and a Z axis is perpendicular to the transverse scanning plane, and coordinates P (X, Y, Z) of an earth and stone surface point are obtained through three-dimensional laser scanning.
And importing the cloud coordinate group data of the measured points into a software Riscan Pro of the three-dimensional scanner, cutting the cloud coordinate group data of the points, eliminating noise points, and keeping required information. In the actual data scanning process, as the scanning area is wider, redundant data is inevitably generated, noise points are removed through the filtering function of the Riccan Pro, filtering point cloud data are obtained, and finally the three-dimensional model is obtained. The volume, cross-sectional area, surface area and the like of the model to any elevation surface can be calculated by utilizing a Riccan Pro analysis. The accuracy of the earth volume calculation is related to the point elevation accuracy and the volume calculation method, and the accuracy of the earth volume calculation is higher by improving the approximate value of the fitting curved surface and the elevation of the original data.
Filtering and fusing the whole data in a post-processing mode through the acquired original data of the GPS, the CCD camera and the laser scanning to finally generate three-dimensional point cloud data, and correcting errors and calculation errors in the measurement to generate a three-dimensional image through error analysis of a three-dimensional laser scanning measurement system.
The point cloud data is used as main data acquired by three-dimensional laser scanning, and is mainly a product obtained by fusing position attitude data determined by a GPS (global positioning system) with distance and angle data of laser scanning.
On the basis of the established earthwork model, the volume calculation function of Riscan Pro software is utilized, the calculation method of the load amount adopts a square grid method, a grid is drawn on the basis of measurement points, the elevation of grid points is interpolated to estimate the earthwork, the calculation accuracy of the earthwork is also influenced by the number and the accuracy of the measurement points, the more the number of the points is, the higher the accuracy is, but the calculation speed is low, and the construction period is long. In addition, the scanner has high scanning speed, so that the working efficiency is greatly improved, the detailed texture characteristics of the object are ensured, and the accuracy of earth volume calculation is greatly improved.
Judging soil and stone materials in a loading area according to the rated loading allowable volume of each type of the muck truck equipped in the project database, and recording the excessive volume part as a positive value; the fraction of volume that is not satisfied is subtracted when a negative value is recorded. And finally, judging that the discharging is released or overloaded or the loading requirement is not met according to the calculation result and needing to be reloaded.
And reporting the scanned image data and volume data to the cloud information processing platform after the qualified vehicle is judged to be released.
The method can be used for not only measuring the volume of soil and stone transported by the muck truck by a construction unit, but also managing the muck truck on site by an owner. Through the transportation volume data of uploading in real time, each unit of being convenient for in time makes clear and definite the engineering situation to in better management and control to enterprise's financial expenditure. In actual construction, the phenomena that the accurate volume of a carrying substance cannot be accurately obtained and the number of metering vehicles is roughly adopted as a settlement voucher commonly exist. The earth and stone engineering accounts for a large share of the total construction cost of the engineering, and disputes are easily caused to later settlement of engineering money due to the fact that the volume cannot be accurately estimated. The method of the invention is convenient for constructing intelligent construction sites, and the accurately obtained engineering quantity of the transport materials has great support effect on the construction quality control.
The system has wide development space, is favorable for establishing an intelligent construction site integrated system through a deep learning network, and has larger application space for the development of residue removal of the residue soil truck through color identification and the detection of whether to set up dustproof cloth.
The image recognition of the target object comprises the steps of firstly extracting relevant characteristic information of the target object in a photographed image, and then converting the extracted image signal characteristics into data which can be recognized by a machine according to the technologies of machine learning, digital signal processing and the like, so that the man-machine interaction is achieved. Valuable features are extracted from the soil and stone substances of the construction site and the transport muck vehicle in the field construction, and a classifier is adopted for modeling and identifying according to the features. Because the types and numbers of the muck vehicles in actual construction are not many, the interference factors are single, and the muck vehicle type and the sand and stone type information for loading and transporting the earthwork in the full actual engineering construction are easy to record and establish the database.
Example two
The muck vehicle loading identification and judgment system in the second embodiment is the same as the muck vehicle loading identification and judgment system in the first embodiment, the license plate number of the muck vehicle is automatically identified from the whole vehicle image of the muck vehicle in the second embodiment, the standard loading capacity of the muck vehicle is obtained according to the vehicle type corresponding to the license plate number, and the method is different from the method for judging the loading condition of earth and stone of the muck vehicle in the first embodiment.
The method for judging the earth and rock load of the muck car based on the point cloud data comprises the following steps of sequentially executing,
step 1: collecting a whole-vehicle image of the muck vehicle, identifying the license plate number of the muck vehicle, and obtaining the type of the muck vehicle and the standard loading capacity of the corresponding type according to the corresponding relation of the pre-stored license plate number and vehicle type;
step 2: collecting images of the muck carried by the muck car, inputting the images into a second neural network model, and identifying the types of the muck carried by the muck car by using the second neural network model to obtain the density of the muck;
and step 3: scanning the muck loaded by the muck truck by using a three-dimensional laser scanner to obtain point cloud data of the muck loaded by the muck truck;
and 4, step 4: denoising and cutting the point cloud data of the muck to obtain a three-dimensional image of the muck;
and 5: processing the information result of the three-dimensional image recognition to obtain a loading volume;
step 6: comparing the loading volume of the slag car with the standard loading volume obtained in the step (1) and judging whether the loading condition of the slag car is qualified or not;
and 7: and (3) calculating the load of the muck truck according to the loading volume of the muck truck and the density of the muck obtained in the step (2), and judging whether the muck truck is overweight.
The second neural network model of example two is the same as the second neural network model of example one, and the same training method is used.
The implementation result shows that the method can accurately measure the type and the volume of the carrying material in real time, has higher measurement accuracy, solves the problems of low efficiency and inaccurate result of the traditional counting method, and has important significance for the establishment of an intelligent construction site system.
Claims (9)
1. The method for judging the earth and stone load of the muck car based on the point cloud data is characterized by comprising the following steps,
step 1: collecting a whole-vehicle image of the slag vehicle, inputting the image into a first neural network model, and identifying the type of the slag vehicle by using the first neural network model to obtain a standard loading capacity of the corresponding type of the slag vehicle;
step 2: collecting images of the muck carried by the muck car, inputting the images into a second neural network model, and identifying the types of the muck carried by the muck car by using the second neural network model to obtain the density of the muck;
and step 3: scanning the muck loaded by the muck truck by using a three-dimensional laser scanner to obtain point cloud data of the muck loaded by the muck truck;
and 4, step 4: denoising and cutting the point cloud data of the muck to obtain a three-dimensional image of the muck;
and 5: processing the information result of the three-dimensional image recognition to obtain a loading volume;
step 6: comparing the loading volume of the slag car with the standard loading volume obtained in the step (1) and judging whether the loading condition of the slag car is qualified or not;
and 7: and (3) calculating the load of the muck truck according to the loading volume of the muck truck and the density of the muck obtained in the step (2), and judging whether the muck truck is overweight.
2. The method for judging earth and rock square loading of the muck vehicle based on the point cloud data as claimed in claim 1, wherein the step 1 further comprises identifying the cleanliness of the muck vehicle according to a whole vehicle image of the muck vehicle by using a first neural network model.
3. The method for judging earth and rock load of the muck vehicle based on the point cloud data as claimed in claim 1, wherein the step 1 can also be to collect images of the muck vehicle, identify the license plate number of the muck vehicle, and obtain the type of the muck vehicle and the standard load capacity of the corresponding type according to the corresponding relationship between the pre-stored license plate number and the pre-stored vehicle type.
4. The method for judging earth and stone load of the muck vehicle based on the point cloud data as claimed in claim 1, wherein the first neural network model adopts fast R-CNN neural network.
5. The method for judging earth and stone load of the muck vehicle based on the point cloud data as claimed in claim 2, further comprising pre-training the first neural network model, specifically comprising:
setting reference surfaces for the muck trucks of different vehicle types according to the maximum loading limit height of baffles at two sides of the muck trucks to measure the standard loading capacity of the muck trucks;
shooting all-vehicle images of the muck vehicles of different vehicle types with different cleaning degrees, combining the all-vehicle images of the muck vehicles with the cleaning degrees of the muck vehicles and standard carrying capacity data of corresponding types of the muck vehicles to form a muck vehicle data set, and dividing the muck vehicle data set into a training set and a testing set;
3) and training and testing the first neural network model by utilizing the muck loading data set.
6. The method for judging earth-rock space loading of the muck vehicle based on the point cloud data as claimed in claim 2, further comprising pre-training a second neural network model, specifically comprising:
1) collecting a certain volume of muck of each construction site, and classifying to obtain classified muck samples;
2) weighing each type of muck sample, calculating the density of the muck sample according to the muck weighing and the muck volume, collecting an image of the muck sample to form a muck data set, and dividing the muck data set into a training set and a testing set;
3) and training and testing the second neural network model by utilizing the muck data set.
7. The system for identifying and judging muck vehicle loading by using the method for judging muck vehicle earthwork loading based on point cloud data as claimed in claims 1 to 6 is characterized by comprising a computer, and a GPS receiver, a camera, a laser scanner, a camera and a display screen which are respectively connected with the computer.
8. The muck vehicle loading identification and judgment system according to claim 7, further comprising a controller and a spraying and cleaning device, wherein a control end of the spraying and cleaning device is connected with the controller, and the controller is connected with the computer through a data line.
9. The muck vehicle loading identification system according to claim 7, further comprising an illumination sensor connected to the controller.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011191481.0A CN112184707A (en) | 2020-10-30 | 2020-10-30 | Method and system for judging earth and stone load of muck vehicle based on point cloud data |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202011191481.0A CN112184707A (en) | 2020-10-30 | 2020-10-30 | Method and system for judging earth and stone load of muck vehicle based on point cloud data |
Publications (1)
Publication Number | Publication Date |
---|---|
CN112184707A true CN112184707A (en) | 2021-01-05 |
Family
ID=73917679
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202011191481.0A Pending CN112184707A (en) | 2020-10-30 | 2020-10-30 | Method and system for judging earth and stone load of muck vehicle based on point cloud data |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN112184707A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113837614A (en) * | 2021-09-26 | 2021-12-24 | 北京京东振世信息技术有限公司 | Cargo carrying amount monitoring method, cargo carrying amount monitoring system, electronic device and readable medium |
CN114005092A (en) * | 2021-12-29 | 2022-02-01 | 深圳市思拓通信系统有限公司 | Muck truck bearing capacity monitoring method, controller and system |
CN114758184A (en) * | 2022-06-14 | 2022-07-15 | 福建南方路面机械股份有限公司 | Deep learning-based muck classification processing guide method and device and readable medium |
CN115542337A (en) * | 2022-11-28 | 2022-12-30 | 成都维泰油气能源技术有限公司 | Drilling return rock debris monitoring method and device and storage medium |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106767710A (en) * | 2016-12-22 | 2017-05-31 | 上海华测导航技术股份有限公司 | A kind of Earth Volume of Road Engineering measuring method and system |
CN106839977A (en) * | 2016-12-23 | 2017-06-13 | 西安科技大学 | Shield dregs volume method for real-time measurement based on optical grating projection binocular imaging technology |
CN110766039A (en) * | 2019-09-02 | 2020-02-07 | 厦门卫星定位应用股份有限公司 | Muck truck transportation state identification method, medium, equipment and muck truck |
WO2020120702A1 (en) * | 2018-12-12 | 2020-06-18 | Marel Salmon A/S | A method and a device for estimating weight of food objects |
CN111554097A (en) * | 2020-03-31 | 2020-08-18 | 深圳市奥拓电子股份有限公司 | Method, device and system for detecting vehicle through intelligent lamp pole and electronic equipment thereof |
-
2020
- 2020-10-30 CN CN202011191481.0A patent/CN112184707A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106767710A (en) * | 2016-12-22 | 2017-05-31 | 上海华测导航技术股份有限公司 | A kind of Earth Volume of Road Engineering measuring method and system |
CN106839977A (en) * | 2016-12-23 | 2017-06-13 | 西安科技大学 | Shield dregs volume method for real-time measurement based on optical grating projection binocular imaging technology |
WO2020120702A1 (en) * | 2018-12-12 | 2020-06-18 | Marel Salmon A/S | A method and a device for estimating weight of food objects |
CN110766039A (en) * | 2019-09-02 | 2020-02-07 | 厦门卫星定位应用股份有限公司 | Muck truck transportation state identification method, medium, equipment and muck truck |
CN111554097A (en) * | 2020-03-31 | 2020-08-18 | 深圳市奥拓电子股份有限公司 | Method, device and system for detecting vehicle through intelligent lamp pole and electronic equipment thereof |
Non-Patent Citations (1)
Title |
---|
徐文怀,孔凡德,汝斌: "《信息化学习能力开发导论》", 31 May 2020 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113837614A (en) * | 2021-09-26 | 2021-12-24 | 北京京东振世信息技术有限公司 | Cargo carrying amount monitoring method, cargo carrying amount monitoring system, electronic device and readable medium |
CN114005092A (en) * | 2021-12-29 | 2022-02-01 | 深圳市思拓通信系统有限公司 | Muck truck bearing capacity monitoring method, controller and system |
CN114005092B (en) * | 2021-12-29 | 2022-04-26 | 深圳市思拓通信系统有限公司 | Muck truck bearing capacity monitoring method, controller and system |
CN114758184A (en) * | 2022-06-14 | 2022-07-15 | 福建南方路面机械股份有限公司 | Deep learning-based muck classification processing guide method and device and readable medium |
CN114758184B (en) * | 2022-06-14 | 2022-09-06 | 福建南方路面机械股份有限公司 | Deep learning-based muck classification processing guide method and device and readable medium |
CN115542337A (en) * | 2022-11-28 | 2022-12-30 | 成都维泰油气能源技术有限公司 | Drilling return rock debris monitoring method and device and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112184707A (en) | Method and system for judging earth and stone load of muck vehicle based on point cloud data | |
Wang et al. | Computational methods of acquisition and processing of 3D point cloud data for construction applications | |
CN107314741A (en) | Measurement of cargo measuring method | |
CN113252700A (en) | Structural crack detection method, equipment and system | |
CN108288059A (en) | A kind of building waste monitoring method based on high-definition remote sensing technology | |
CN106044663A (en) | Visual technology-based stone mine forklift with weight measuring function and weight measuring method of stone mine forklift | |
CN112309108A (en) | Truck-mounted overrun detection system and method | |
CN113869629A (en) | Laser point cloud-based power transmission line safety risk analysis, judgment and evaluation method | |
Chu et al. | A review on pavement distress and structural defects detection and quantification technologies using imaging approaches | |
WO2019080380A1 (en) | Wheelmark transverse distribution measuring system and method | |
CN110926426B (en) | Real-time detection device and detection method for residual muck volume of subway shield segment | |
Wang et al. | Automated pavement distress survey: a review and a new direction | |
KR20100093709A (en) | Measuring method for mass of loaded soil on dump truck and calculating method of soil volume conversion factor | |
CN111462073B (en) | Quality inspection method and device for point cloud density of airborne laser radar | |
CN112070756A (en) | Three-dimensional road surface disease measuring method based on unmanned aerial vehicle oblique photography | |
Medeiros Jr et al. | 3D pavement macrotexture parameters from close range photogrammetry | |
CN103528531A (en) | Intelligent Internet of Things image detection system for small vehicle parameters | |
Shatnawi et al. | Road pavement rut detection using mobile and static terrestrial laser scanning | |
CN114295069A (en) | Side slope deformation monitoring method and system for unmanned aerial vehicle carrying three-dimensional laser scanner | |
CN112232435A (en) | Method for calculating volume of earth and rock cubic of muck vehicle based on convolutional neural network | |
Guan et al. | Multi-scale asphalt pavement deformation detection and measurement based on machine learning of full field-of-view digital surface data | |
Feng et al. | Investigation of 3D terrestrial laser scanning techniques for potential application to rock mechanics | |
CN113538668A (en) | Geological model construction method, geological model construction system, storage medium and electronic equipment | |
Wang et al. | Identification of rocky ledge on steep, high slopes based on UAV photogrammetry | |
CN113124753A (en) | Volume measurement method and device |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210105 |