A method, device, computer equipment and storage medium for identifying place type

Abstract

The present application provides a method, device, computer equipment and storage medium for identifying the type of place, which are applied to the field of the field of identifying the type of place. The method includes obtaining the location information and working condition information of multiple construction equipment, constructing at least one undirected graph based on the location information of multiple construction equipment, clustering the vertices in at least one undirected graph, and obtaining corresponding vertex clusters, wherein any vertex cluster corresponds to a gathering place, and the place type of the gathering place corresponding to the vertex cluster is obtained according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained type recognition model. This solution can automatically determine the type of the gathering place, for example, determine whether the gathering place is a parking lot type, or a factory construction type, a subway construction type, or a bridge construction type, etc., and no longer requires manual statistics, which not only saves manpower and material resources, but also can effectively improve statistical efficiency.

Description

Technical Field

[0001] The present application relates to the field of place type identification, and in particular to a place type identification method, apparatus, computer equipment and storage medium. Background Art

[0002] In actual applications, there are many places where a large number of equipment are gathered, such as parking lots, factory construction sites, subway construction sites, and bridge construction sites. In order to efficiently manage various types of places and provide targeted services, it is necessary to identify the type of place.

[0003] The existing technology generally uses manual methods to confirm the type of place. For example, for a place where a large number of cranes are gathered, it is necessary to determine whether the place is a parking lot or a construction site. If the place is a construction site, it is necessary to further confirm whether it is the aforementioned factory construction site, subway construction site, or bridge construction site. These confirmation tasks are all completed manually. Obviously, this manual identification method not only consumes a lot of manpower and material resources, but also has relatively low identification efficiency.

[0004] Therefore, how to efficiently identify the type of place while saving manpower and material resources is a technical problem that needs to be solved urgently by those skilled in the art. Summary of the invention

[0005] In view of this, the embodiments of the present application are dedicated to providing a method, apparatus, computer device and storage medium for identifying place types, so as to solve the problem in the prior art that manual identification of place types not only consumes a lot of manpower and material resources, but also has relatively low identification efficiency.

[0006] On one hand, the present application provides a method for identifying a place type, comprising:

[0007] Obtain location information and working condition information of multiple construction equipment;

[0008] Based on the location information of the plurality of construction equipment, construct at least one undirected graph;

[0009] Clustering the vertices in at least one of the undirected graphs to obtain corresponding vertex clusters; wherein any of the vertex clusters corresponds to a gathering place;

[0010] The place type of the gathering place corresponding to the vertex cluster is obtained according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained type recognition model.

[0011] Optionally, constructing at least one undirected graph based on the location information of the plurality of construction equipments includes:

[0012] Acquire the distance between any two of the construction equipment based on the position information of the plurality of construction equipment;

[0013] Taking the construction equipment as vertices, edges are generated between any two vertices corresponding to the construction equipment whose distance is less than a preset distance threshold, to obtain at least one undirected graph.

[0014] Optionally, clustering vertices in at least one of the undirected graphs to obtain corresponding vertex clusters includes:

[0015] Taking a target vertex in any undirected graph whose number of connected edges is greater than or equal to a preset number threshold as a starting point, obtaining other vertices other than the target vertex according to a breadth-first search algorithm;

[0016] The vertices whose number of connected edges is greater than or equal to a preset number threshold are taken as a set to obtain a corresponding vertex cluster.

[0017] Optionally, before acquiring the type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model, the method further includes:

[0018] The working condition information of the construction equipment corresponding to each of the vertices in the vertex cluster is normalized.

[0019] Optionally, the venue type identification method further includes:

[0020] According to the training samples and the LSTM neural network, the time window data features are obtained; the training samples include sample operating condition information;

[0021] Obtaining output results of the training samples according to the time window data features and the CNN neural network;

[0022] Based on the error between the output result of each training sample and the corresponding label, the error corresponding to each training sample is obtained; the label includes the sample location type corresponding to the training sample;

[0023] Taking the error corresponding to each training sample being within a preset range as a training goal, the parameters of the LSTM neural network and / or the CNN neural network are adjusted to obtain the type recognition model.

[0024] Optionally, after clustering the vertices in at least one of the undirected graphs to obtain corresponding vertex clusters, the method further includes:

[0025] The construction period of the gathering place corresponding to the vertex cluster is obtained according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained period recognition model.

[0026] Optionally, if the construction equipment is a crane, the aggregation type includes at least a parking lot and a construction site, and the construction site includes at least a factory construction site, a subway construction site, and a bridge construction site;

[0027] The working condition information at least includes the arrangement mode, working time, lifting weight, lifting height, lifting amplitude, and rotation angle of the crane.

[0028] Another aspect of the present application provides a device for identifying a place type, comprising:

[0029] An acquisition module, used to acquire location information and working condition information of multiple construction equipment;

[0030] A construction module, used for constructing at least one undirected graph based on the location information of the plurality of construction equipments;

[0031] A clustering module, used for clustering vertices in at least one of the undirected graphs to obtain corresponding vertex clusters; wherein any of the vertex clusters corresponds to a gathering place;

[0032] The type confirmation module is used to obtain the place type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained type recognition model.

[0033] Optionally, the venue type identification device further includes: a model pre-training module;

[0034] The model pre-training module includes:

[0035] A first result acquisition unit is used to acquire time window data features according to a training sample and an LSTM neural network; the training sample includes sample operating condition information;

[0036] A second result acquisition unit, used to acquire the output result of the training sample according to the time window data features and the CNN neural network;

[0037] An error determination unit, configured to obtain an error corresponding to each training sample based on an error between an output result of each training sample and a corresponding label; the label includes a sample location type corresponding to the training sample;

[0038] A parameter adjustment unit is used to adjust the parameters of the LSTM neural network and / or the CNN neural network with the error corresponding to each training sample being within a preset range as a training target, so as to obtain the type recognition model.

[0039] On the other hand, the present application provides a computer device, including a memory, a processor, and a computer program stored in the memory and executed by the processor, and when the processor executes the computer program, the steps of the place type identification method as described in any one of the above items are implemented.

[0040] On the other hand, the present application provides a storage medium, including: a computer program is stored on the storage medium, and when the computer program is executed by a processor, each step of the place type identification method as described in any one of the above items is implemented.

[0041] According to the place type identification method, device, computer equipment and storage medium provided by the present application, the method includes obtaining the location information and working condition information of multiple construction equipment, constructing at least one undirected graph based on the location information of multiple construction equipment, clustering the vertices in at least one undirected graph, and obtaining corresponding vertex clusters, wherein any vertex cluster corresponds to a gathering place, and according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model, the place type of the gathering place corresponding to the vertex cluster is obtained. The scheme of the present application can automatically determine the type of the gathering place, for example, determine whether the gathering place is a parking lot type, or a factory construction type, a subway construction type, or a bridge construction type, etc., and no manual statistics are required, which not only saves manpower and material resources, but also can effectively improve statistical efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] Figure 1 The figure is a schematic flow chart of a method for identifying a place type according to an embodiment of the present application.

[0043] Figure 2 The figure shows an undirected graph provided by an embodiment of the present application.

[0044] Figure 3 The figure is a schematic diagram of the construction process of an undirected graph provided in one embodiment of the present application.

[0045] Figure 4 Shown is a before-and-after comparison diagram of the construction of an undirected graph provided by an embodiment of the present application.

[0046] Figure 5 Shown is a before and after comparison diagram of an undirected graph provided by an embodiment of the present application.

[0047] Figure 6 The figure is a schematic diagram of the training process of the type recognition model provided in one embodiment of the present application.

[0048] Figure 7 Shown is a schematic diagram of the structure of a place type identification device provided in one embodiment of the present application.

[0049] Figure 8Shown is a schematic diagram of the structure of a computer device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0050] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0051] Application Overview

[0052] In actual applications, there are many places where a large number of equipment are gathered, such as parking lots, factory construction sites, subway construction sites, and bridge construction sites. In order to efficiently manage various types of places and provide targeted services, it is necessary to identify the type of place. The existing technology generally uses manual methods to confirm the type of place, which not only consumes a lot of manpower and material resources, but also has relatively low recognition efficiency.

[0053] Exemplary Methods

[0054] Figure 1 FIG. 2 is a flow chart of a method for identifying a location type according to an embodiment of the present application. Figure 1 As shown, the following steps are included:

[0055] S101. Acquire location information and working condition information of multiple construction equipment.

[0056] The above-mentioned construction equipment refers to cranes, excavators, etc., which may be gathered in places such as parking lots or construction sites in actual applications.

[0057] The location information of the construction equipment refers to information that can indicate the actual geographical location of the construction equipment, for example, the latitude and longitude information of the construction equipment.

[0058] The above-mentioned working condition information of the construction equipment refers to the various parameter information generated by the construction equipment during operation. For example, if the construction equipment is a crane, the corresponding working condition information may include the arrangement of the crane, working time, lifting weight, lifting height, lifting amplitude, rotation angle, etc.

[0059] At present, a vehicle-mounted intelligent terminal is installed on the construction equipment, and the location information and working condition information of each construction equipment can be collected through the vehicle-mounted intelligent terminal, and the location information and working condition information of the construction equipment can be transmitted back to the server. In the embodiment of the present application, the location information and working condition information of the construction equipment transmitted back by the vehicle-mounted intelligent terminal can be obtained. In addition, the construction equipment can also be monitored by a drone, and the location information of the construction equipment can be located by the drone.

[0060] In the embodiments of the present application, the location information and working condition information of the construction equipment in a certain area can be obtained. The specific area range corresponding to the above-mentioned certain area can be determined according to the actual situation, for example, the location information and working condition information of all the construction equipment in a specific administrative area can be obtained, and the corresponding management area can be delineated according to the actual management scope, and then the location information and working condition information of the construction equipment in the management area can be obtained.

[0061] S102. Construct at least one undirected graph based on the location information of multiple construction equipment.

[0062] An undirected graph includes two parts, vertices and edges. In the embodiment of the present application, construction equipment is used as the vertices of the undirected graph. It should be noted that in this embodiment, not all edges of the undirected graph are generated between the vertices corresponding to any two construction equipment, but it is determined based on the actual distance of the construction equipment. When the actual distance of the construction equipment meets the requirements, for example, when the actual distance between the two construction equipment is less than the preset distance threshold, it means that the distance between the two equipment is close and they may be in the same place, then the edges of the undirected graph can be generated between the vertices corresponding to the two construction equipment.

[0063] Since the actual distance between any two construction equipments does not meet the requirements, not all vertices will have connected edges. In the embodiments of the present application, one or more undirected graphs may be constructed based on the location information of each construction equipment. Figure 2 In the embodiment shown, two undirected graphs are constructed based on the location information of the construction equipment, namely, a first undirected graph A and a second undirected graph B. It should be noted that: Figure 2 The embodiment shown is only for illustrating that one or more undirected graphs may be constructed based on the location information of each construction equipment, and does not limit the vertices of the construction equipment in all cases to be in accordance with Figure 2 Arrange in the manner shown.

[0064] S103: Cluster the vertices in at least one undirected graph to obtain corresponding vertex clusters.

[0065] The main clustering methods currently used include the k-means algorithm and the DBSCAN algorithm. The k-means algorithm is very sensitive to outliers and noise points, and the clustering effect is not ideal. In the case of a large amount of data, DBSCAN requires a lot of calculations, resulting in a long algorithm calculation time. In order to shorten the calculation time and ensure the clustering effect, in the embodiment of the present application, based on the graph algorithm, the vertices of each undirected graph are clustered.

[0066] This embodiment performs clustering based on the degree of vertices, that is, clustering based on the total number of edges attached to the vertices. Each undirected graph can be traversed to traverse the vertices whose degrees meet the requirements and are connected, and cluster the vertices whose degrees meet the requirements and are connected into a vertex cluster. For example, vertices whose degrees are greater than or equal to a preset number and are connected are clustered into a vertex cluster. In this way, multiple vertex clusters can be obtained.

[0067] Specifically, in the vertex clusters obtained by clustering based on the above method, the degree of any vertex in each vertex cluster is greater than or equal to a preset number. Correspondingly, for any construction equipment among all the construction equipment corresponding to each vertex cluster, there are at least a preset number of construction equipment whose distance from it is less than a preset distance threshold. These construction equipment are in a clustered state. Therefore, any vertex cluster determined based on the graph algorithm corresponds to a gathering place of construction equipment.

[0068] The traversal method may adopt depth-first search or breadth-first search, which is not limited in this embodiment.

[0069] S104. Obtain the place type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model.

[0070] In the embodiment of the present application, a type recognition model is pre-trained. The input of the type recognition model is the working condition information of the construction equipment, such as the arrangement of the crane, working hours, lifting weight, lifting height, lifting amplitude, and rotation angle; the output of the type recognition model is the place type corresponding to the gathering place, such as parking lots and construction sites, including factory construction sites, subway construction sites, bridge construction sites, etc.

[0071] In an embodiment of the present application, the working condition information of all construction equipment in the gathering place is taken as input and input into the type recognition model in chronological order to obtain the place type corresponding to each gathering place.

[0072] The place type identification method of this embodiment can automatically determine the type of gathering place, for example, determine whether the gathering place is a parking lot type, a factory construction type, a subway construction type, or a bridge construction type, etc., and manual statistics are no longer required, which not only saves manpower and material resources, but also effectively improves statistical efficiency.

[0073] In another embodiment, if Figure 3 As shown, the steps of the above embodiment are clustered according to the vertices in at least one undirected graph to obtain corresponding vertex clusters. The specific implementation process includes:

[0074] S301. Obtain the distance between any two construction equipment based on the position information of a plurality of construction equipment.

[0075] Based on the preset time point and the location information of each construction equipment, the actual distance between any two construction equipment can be calculated. For example, if the longitude and latitude of two construction equipment are known, the distance between the two construction equipment can be determined through simple mathematical calculations. The above preset time point can be determined according to the actual situation, and this embodiment does not limit it.

[0076] S302: Taking the construction equipment as vertices, generate edges between any two vertices corresponding to the construction equipment whose distance is less than a preset distance threshold, and obtain at least one undirected graph.

[0077] In an embodiment of the present application, the construction equipment and the distance between the construction equipment are converted into an undirected graph. Among them, the vertex represents the construction equipment, and the number of vertices is consistent with the number of construction equipment. The edge is the connection between the vertices. When the distance between two construction equipment is less than or equal to the preset distance threshold, an edge will be generated between the vertices corresponding to the two construction equipment. When the distance between the two construction equipment is greater than the preset distance threshold, no edge will be generated between the vertices corresponding to the two construction equipment. In this way, all the construction equipment and the distance relationship between the construction equipment can be displayed through at least one undirected graph.

[0078] Figure 4 In the embodiment shown, Figure 4 a represents the generation of edges between the vertices corresponding to the construction equipment whose distance is less than the preset distance threshold based on the vertices generated by all the construction equipment. Figure 4 b. The third undirected graph C and the fourth undirected graph D. It should be noted that, Figure 4 The embodiment shown is only for illustrating the generation process of the undirected graph, and does not limit the vertices of the construction equipment to be in accordance with the Figure 4 Arrange in the manner shown.

[0079] The above-mentioned preset distance threshold can be set according to actual conditions and is not limited in this embodiment.

[0080] Based on the above introduction, it can be seen that this embodiment represents the construction equipment whose distance meets the requirements in the form of an undirected graph, which is not only easy to operate and has a short processing time, but also can lay a foundation for subsequent clustering operations.

[0081] In another embodiment, the steps of the above embodiment are clustered according to the vertices in at least one undirected graph to obtain corresponding vertex clusters, and the specific implementation process includes:

[0082] Taking the target vertex whose number of connected edges is greater than or equal to a preset threshold in any undirected graph as the starting point, obtain other vertices except the target vertex according to the breadth-first search algorithm; take the vertices whose number of connected edges is greater than or equal to the preset threshold as a set to obtain the corresponding vertex cluster.

[0083] The target vertex refers to a vertex among all vertices of the undirected graph that meets the condition that the number of connected edges is greater than or equal to a preset number threshold.

[0084] In the embodiment of the present application, the vertices whose number of connected edges in each undirected graph is greater than or equal to a preset number threshold are traversed in a breadth-first search manner. Specifically, a vertex whose number of connected edges is greater than or equal to a preset number threshold is randomly selected as the target vertex, and the target vertex is used as the starting point to search other vertices in a breadth-first manner. When the number of connected edges of the vertex searched by any search branch generated during the breadth-first search process is less than the preset number threshold, the search branch will no longer continue to search until all search branches generated during the breadth-first search process no longer continue to search, and the breadth-first search with the target vertex as the starting point ends.

[0085] When all vertices whose number of connected edges in each undirected graph is greater than or equal to a preset number threshold are traversed, the vertices whose number of connected edges is greater than or equal to the preset number threshold are taken as a set to obtain a corresponding vertex cluster.

[0086] The above-mentioned preset quantity threshold can be set according to actual conditions and is not limited in this embodiment.

[0087] Figure 5 In the embodiment shown, the preset number threshold is set to 2. Figure 5 In a, vertex a, vertex b and vertex c are used as starting points for breadth-first search. After the search is completed, we get Figure 5 b shows the first vertex cluster E, the second vertex cluster F and the third vertex cluster G. The first vertex cluster E, the second vertex cluster F and the third vertex cluster G correspond to three gathering places.

[0088] In the embodiments of the present application, vertex clusters are determined by a graph algorithm, which can both ensure the clustering effect of the algorithm and improve the computational efficiency of the algorithm compared to clustering algorithms such as the k-means algorithm and the DBSCAN algorithm.

[0089] In another embodiment, the steps of the above embodiment further include: before obtaining the type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model:

[0090] The working condition information of the construction equipment corresponding to each vertex in the vertex cluster is normalized.

[0091] Specifically, in the embodiment of the present application, the working condition information of the construction equipment corresponding to all vertices in each vertex cluster is preprocessed according to different categories. That is to say, the working condition information of all construction equipment in each gathering place is preprocessed according to different categories. For example, the mean and variance of the working condition information of each category are calculated and normalized respectively, and the normalized working condition information of all categories is used as the attribute of the corresponding gathering place, and is input into the type recognition model in chronological order to obtain the place type corresponding to each gathering place.

[0092] In another embodiment, if Figure 6 As shown, the training process of the type recognition model in the above embodiment includes:

[0093] S601. Obtain time window data features based on training samples and LSTM neural network.

[0094] In the embodiment of the present application, the type recognition model is obtained by training a preset neural network using sample working condition information of construction equipment in various gathering places as training samples and sample place types corresponding to the training samples as labels. Exemplarily, the neural network preset in the present application is a LSTM+CNN hybrid neural network.

[0095] Specifically, if the construction equipment is a crane, the training samples include working condition information such as arrangement mode, working time, lifting weight, lifting height, amplitude, rotation angle, etc. Before inputting the training samples into the model for training, the training samples need to be preprocessed, for example, the mean and variance of each category of working condition information are calculated and normalized respectively, and the normalized working condition information of all categories is used as the attribute of the gathering place, and is input into the preset neural network model in chronological order for training. The method of obtaining training samples can refer to the method of obtaining the working condition information of construction equipment in the above embodiment, which will not be repeated here.

[0096] The labels include parking lots and construction sites, and the construction sites include at least factory construction sites, subway construction sites, and bridge construction sites. It should be noted that before model training, the staff can obtain the construction equipment gathering places corresponding to the training samples to determine the gathering type of the gathering places as the labels of the training samples. The gathering types of the gathering places can also be obtained through agent feedback as the labels of the training samples. In addition, drones can be used to take photos or videos of the gathering places, and the staff can determine the gathering types of the gathering places based on the photos or videos.

[0097] The training samples are input into the LSTM neural network, and after calculation by the LSTM neural network, a time window data matrix is ​​output as the time window data feature.

[0098] S602: Obtain output results of training samples according to the time window data features and the CNN neural network.

[0099] After calculation by the LSTM neural network, a time window data matrix is ​​output to the CNN neural network. A second round of calculation is performed in the CNN network and the final place type is output as the output result.

[0100] S603: Based on the error between the output result of each training sample and the corresponding label, obtain the error corresponding to each training sample.

[0101] S604: Taking the error corresponding to each training sample as being within a preset range as a training goal, adjust the parameters of the LSTM neural network and / or the CNN neural network to obtain a type recognition model.

[0102] The loss function can be used to determine the error between the output result of each training sample and the corresponding label, and then the parameters of the LSTM neural network and / or the CNN neural network can be adjusted. Depending on the situation, only the parameters of the LSTM neural network can be adjusted, or only the parameters of the CNN neural network can be adjusted, or the parameters of the LSTM neural network and the CNN neural network can be adjusted simultaneously, so that the error corresponding to each training sample is within a preset range as the training goal, and the model is repeatedly trained until a type recognition model with an accuracy rate that meets the requirements is obtained.

[0103] In this embodiment, the use of LSTM+CNN hybrid neural network can increase the reusability of other construction equipment. Using embedded models only requires fine-tuning of the output layer for reuse. Due to the large dimension of intermediate data, this embodiment uses an automatic encoding framework to improve the model training speed and accuracy, and solves the original data in the form of sliding window optimization to find the best accuracy of transfer learning.

[0104] In another embodiment, after executing the steps of the above embodiment to cluster the vertices in at least one undirected graph and obtain corresponding vertex clusters, the following steps may also be performed:

[0105] According to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained cycle recognition model, the construction cycle of the gathering place corresponding to the vertex cluster is obtained.

[0106] Specifically, in the embodiment of the present application, a cycle recognition model is pre-trained. The input of the cycle recognition model is the working condition information of the construction equipment, such as the arrangement of the cranes, working hours, lifting weight, lifting height, lifting amplitude, and rotation angle; the output of the cycle recognition model is the construction period corresponding to the gathering place, such as the early construction period, construction period, and late construction period.

[0107] In an embodiment of the present application, the operating condition information of all construction equipment in the gathering place is taken as input and input into the cycle recognition model in chronological order to obtain the place cycle corresponding to each gathering place.

[0108] It should be noted that the training samples of the cycle recognition model in this embodiment are the working condition information of the construction equipment. Specifically, if the construction equipment is a crane, the training samples include the arrangement mode, working time, lifting weight, lifting height, amplitude, rotation angle and other working condition information. The label is the construction period, including the pre-construction period, construction period and post-construction period. The construction period of the construction equipment corresponding to the training samples obtained by the staff can be used to determine the construction period of the gathering place as the label of the training sample. The construction period of the gathering place can also be obtained through feedback from the agent as the label of the training sample. In addition, drones can be used to take photos or videos of the gathering place, and the staff can determine the construction period of the gathering place based on the photos or videos. The training process of the cycle recognition model is the same as the training process of the type recognition model in the above embodiment. Those skilled in the art can refer to the records of the type recognition model training process in the above embodiment, and this embodiment will not be repeated.

[0109] For example, if in addition to obtaining the type of gathering place, when the type of gathering place is a factory construction site, a subway construction site, a bridge construction site or other construction site, it is also necessary to simultaneously obtain the construction period of the construction site, in addition to training the type recognition model and the period recognition model in the above embodiment, the working condition information of the construction equipment in each gathering place can be used as a training sample, and the type of gathering place and the construction period of the construction site can be used as labels to train the neural network and obtain a recognition model that can recognize the type of gathering and the construction period. The training process of the recognition model is the same as the training process of the type recognition model in the above embodiment, and will not be repeated here.

[0110] In the embodiment of the present application, when the gathering place is a construction site, the construction period of the construction site can also be determined, which further facilitates efficient management of various types of places and provision of targeted services.

[0111] Exemplary Devices

[0112] Corresponding to the above-mentioned venue type identification method, the present application embodiment also discloses a venue type identification device, see Figure 7 As shown, the device comprises:

[0113] An acquisition module 100 is used to acquire location information and working condition information of multiple construction equipment;

[0114] A construction module 110, configured to construct at least one undirected graph based on location information of a plurality of construction equipment;

[0115] A clustering module 120, configured to cluster vertices in at least one undirected graph to obtain corresponding vertex clusters; wherein any vertex cluster corresponds to a gathering place;

[0116] The type confirmation module 130 is used to obtain the place type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model.

[0117] In this embodiment, the acquisition module 100 acquires the location information and working condition information of multiple construction equipment; the construction module 110 constructs at least one undirected graph based on the location information of multiple construction equipment; the clustering module 120 clusters the vertices in at least one undirected graph to obtain corresponding vertex clusters; wherein any vertex cluster corresponds to a gathering place; the type confirmation module 130 obtains the place type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model. The scheme of this embodiment can automatically determine the type of the gathering place, for example, determine whether the gathering place is a parking lot type, a factory construction type, a subway construction type, or a bridge construction type, etc., and no manual statistics are required, which not only saves manpower and material resources, but also can effectively improve statistical efficiency.

[0118] Optionally, in another embodiment of the present application, the building module 110 includes:

[0119] A first acquisition unit, configured to acquire the distance between any two construction equipment based on the position information of the plurality of construction equipment;

[0120] The second acquisition unit is used to generate edges between any two vertices corresponding to the construction equipment whose distance is less than a preset distance threshold, taking the construction equipment as vertices, and acquire at least one undirected graph.

[0121] Optionally, in another embodiment of the present application, the clustering module 120 includes:

[0122] A search unit is used to take a target vertex whose number of connected edges is greater than or equal to a preset number threshold in any undirected graph as a starting point, and obtain other vertices other than the target vertex according to a breadth-first search algorithm;

[0123] The determination unit is used to take the vertices whose number of connected edges is greater than or equal to a preset number threshold as a set and obtain the corresponding vertex cluster.

[0124] Optionally, in another embodiment of the present application, it further includes:

[0125] According to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model, before obtaining the place type of the gathering place corresponding to the vertex cluster, a normalization unit is used to normalize the working condition information of the construction equipment corresponding to each vertex in the vertex cluster.

[0126] Optionally, in another embodiment of the present application, it further includes: a model pre-training module;

[0127] The model pre-training module includes:

[0128] A first result acquisition unit is used to acquire time window data features according to training samples and LSTM neural network; the training samples include sample operating condition information;

[0129] A second result acquisition unit is used to acquire the output result of the training sample according to the time window data characteristics and the CNN neural network;

[0130] An error determination unit, configured to obtain an error corresponding to each training sample based on an error between an output result of each training sample and a corresponding label; the label includes a sample location type of the sample working condition information;

[0131] The parameter adjustment unit is used to adjust the parameters of the LSTM neural network and / or the CNN neural network with the error corresponding to each training sample being within a preset range as the training target, so as to obtain a type recognition model.

[0132] Optionally, in another embodiment of the present application, it further includes:

[0133] After clustering the vertices in at least one undirected graph to obtain the corresponding vertex cluster, a cycle confirmation module is used to obtain the construction cycle of the gathering place corresponding to the vertex cluster based on the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained cycle recognition model.

[0134] Optionally, in another embodiment of the present application, if the construction equipment is a crane, the aggregation type includes at least a parking lot and a construction site, and the construction site includes at least a factory construction site, a subway construction site, and a bridge construction site;

[0135] The working condition information includes at least the arrangement of the cranes, working hours, lifting weight, lifting height, lifting amplitude, and rotation angle.

[0136] The venue type identification device provided in this embodiment belongs to the same application concept as the venue type identification method provided in the embodiment of this application, and can execute the venue type identification method provided in any embodiment of this application, and has the corresponding functional modules and beneficial effects of executing the venue type identification method. For technical details not fully described in this embodiment, please refer to the venue type identification method provided in the embodiment of this application, and will not be repeated here.

[0137] Exemplary Computer Devices

[0138] Corresponding to the above-mentioned venue type identification method, the present application embodiment also discloses a computer device, see Figure 8 As shown, the computer device includes: one or more memories 200 , one or more processors 210 , and a computer program stored in the memory 200 and executed by the processor 210 .

[0139] The processor 210 may be a central processing unit (CPU) or other forms of processing units having data processing capabilities and / or instruction execution capabilities, and may control other components in a computer device to perform desired functions.

[0140] The memory 200 may include one or more computer program products, and the computer program products may include various forms of computer-readable storage media, such as volatile memory and / or non-volatile memory. Volatile memory may include, for example, random access memory (RAM) and / or cache memory (cache), etc. Non-volatile memory may include, for example, read-only memory (ROM), hard disk, flash memory, etc. One or more computer program instructions may be stored on the computer-readable storage medium, and the processor 210 may run the program instructions to implement the venue type identification method of each embodiment of the present application above and / or other desired functions. Various contents such as input signals, signal components, noise components, etc. may also be stored in the computer-readable storage medium.

[0141] In one example, the computer device may further include: an input device 220 and an output device 230 , and these components are interconnected via a bus system and / or other forms of connection mechanisms (not shown).

[0142] The input device 220 may include a device for receiving data and information input by a user, such as a keyboard, a mouse, a camera, a scanner, a light pen, a voice input device, a touch screen, a pedometer, or a gravity sensor.

[0143] The output device 230 may be a device that outputs various information to the outside, such as a display, a speaker, a printer, a communication network and a remote output device connected thereto, and the like.

[0144] Of course, to simplify, Figure 8 Only some of the components in the computer device related to the present application are shown, and components such as a bus, an input / output interface, etc. are omitted. In addition, the computer device may further include any other appropriate components according to specific application conditions.

[0145] Exemplary computer program products and computer-readable storage media

[0146] In addition to the above-mentioned methods and devices, an embodiment of the present application may also be a computer program product, which includes computer program instructions, which, when executed by a processor, enable the processor to execute the steps of the place type identification method according to various embodiments of the present application described in the above-mentioned "Exemplary Method" section of this specification.

[0147] The computer program product may be written in any combination of one or more programming languages ​​to write program codes for performing the operations of the embodiments of the present application, including object-oriented programming languages ​​such as Java, C++, etc., and conventional procedural programming languages ​​such as "C" language or similar programming languages. The program code may be executed entirely on the user computing device, partially on the user device, as a separate software package, partially on the user computing device and partially on a remote computing device, or entirely on a remote computing device or server.

[0148] In addition, an embodiment of the present application may also be a computer-readable storage medium on which computer program instructions are stored. When the computer program instructions are executed by a processor, the processor executes the steps of the place type identification method according to various embodiments of the present application described in the above "Exemplary Method" section of this specification.

[0149] Computer readable storage media can adopt any combination of one or more readable media. The readable medium can be a readable signal medium or a readable storage medium. The readable storage medium can include, for example, but is not limited to, a system, device or device of electricity, magnetism, light, electromagnetic, infrared, or semiconductor, or any combination of the above. More specific examples (non-exhaustive list) of readable storage media include: an electrical connection with one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0150] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0151] The block diagrams of the devices, apparatuses, equipment, and systems involved in this application are only illustrative examples and are not intended to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagram. As will be appreciated by those skilled in the art, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open words, referring to "including but not limited to", and can be used interchangeably with them. The words "or" and "and" used here refer to the words "and / or" and can be used interchangeably with them, unless the context clearly indicates otherwise. The words "such as" used here refer to the phrase "such as but not limited to", and can be used interchangeably with them.

[0152] It should also be noted that in the apparatus, device and method of the present application, each component or each step can be decomposed and / or recombined. Such decomposition and / or recombination should be regarded as equivalent solutions of the present application.

[0153] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0154] It should be understood that the qualifiers "first", "second", "third", "fourth", "fifth" and "sixth" used in the description of the embodiments of the present application are only used to more clearly explain the technical solutions and cannot be used to limit the scope of protection of the present application.

[0155] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. A method for identifying a place type, characterized in that: include: Acquire location information and working condition information of multiple construction equipment, where the location information of the construction equipment is the latitude and longitude information of the construction equipment; Acquire the distance between any two of the construction equipment based on the position information of the plurality of construction equipment; Taking the construction equipment as vertices, generating edges between any two vertices corresponding to the construction equipment whose distance is less than a preset distance threshold, and obtaining at least one undirected graph; Taking a target vertex in any undirected graph whose number of connected edges is greater than or equal to a preset number threshold as a starting point, obtaining other vertices other than the target vertex according to a breadth-first search algorithm; The vertices whose number of connected edges is greater than or equal to a preset number threshold are taken as a set to obtain corresponding vertex clusters; wherein any vertex cluster corresponds to a gathering place; According to the working condition information of the construction equipment corresponding to each of the vertices in the vertex cluster and a pre-trained type recognition model, obtaining the place type of the gathering place corresponding to the vertex cluster; If the construction equipment is a crane, the types of the gathering places include at least parking lots and construction sites, and the construction sites include at least factory construction sites, subway construction sites, and bridge construction sites; The working condition information at least includes the arrangement mode, working time, lifting weight, lifting height, lifting amplitude, and rotation angle of the crane.

2. The method according to claim 1, characterized in that Before obtaining the type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and the pre-trained type recognition model, the method further includes: The working condition information of the construction equipment corresponding to each of the vertices in the vertex cluster is normalized.

3. The method according to claim 1, characterized in that Also includes: According to the training samples and LSTM neural network, the time window data features are obtained; The training samples include sample operating condition information; Obtaining output results of the training samples according to the time window data features and the CNN neural network; Based on the error between the output result of each training sample and the corresponding label, the error corresponding to each training sample is obtained; the label includes the sample location type corresponding to the training sample; Taking the error corresponding to each training sample being within a preset range as a training goal, the parameters of the LSTM neural network and / or the CNN neural network are adjusted to obtain the type recognition model.

4. The method according to claim 1, characterized in that After clustering the vertices in at least one of the undirected graphs to obtain corresponding vertex clusters, the method further includes: The construction period of the gathering place corresponding to the vertex cluster is obtained according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained period recognition model.

5. A place type identification device, characterized in that: include: An acquisition module is used to acquire location information and working condition information of multiple construction equipment, where the location information of the construction equipment is the latitude and longitude information of the construction equipment; A construction module, configured to obtain the distance between any two of the construction equipment based on the position information of the plurality of construction equipment; Taking the construction equipment as vertices, generating edges between any two vertices corresponding to the construction equipment whose distance is less than a preset distance threshold, and obtaining at least one undirected graph; A clustering module is used to take a target vertex whose number of connected edges is greater than or equal to a preset number threshold in any undirected graph as a starting point, and obtain other vertices other than the target vertex according to a breadth-first search algorithm; The vertices whose number of connected edges is greater than or equal to a preset number threshold are taken as a set to obtain corresponding vertex clusters; wherein any vertex cluster corresponds to a gathering place; A type confirmation module, used for obtaining the place type of the gathering place corresponding to the vertex cluster according to the working condition information of the construction equipment corresponding to each vertex in the vertex cluster and a pre-trained type recognition model; If the construction equipment is a crane, the types of the gathering places include at least parking lots and construction sites, and the construction sites include at least factory construction sites, subway construction sites, and bridge construction sites; The working condition information at least includes the arrangement mode, working time, lifting weight, lifting height, lifting amplitude, and rotation angle of the crane.

6. The device according to claim 5, characterized in that Also includes: Model pre-training module; The model pre-training module includes: A first result acquisition unit is used to acquire time window data features according to a training sample and an LSTM neural network; the training sample includes sample operating condition information; A second result acquisition unit, used to acquire the output result of the training sample according to the time window data features and the CNN neural network; An error determination unit, configured to obtain an error corresponding to each training sample based on an error between an output result of each training sample and a corresponding label; the label includes a sample location type corresponding to the training sample; A parameter adjustment unit is used to adjust the parameters of the LSTM neural network and / or the CNN neural network with the error corresponding to each training sample being within a preset range as a training target, so as to obtain the type recognition model.

7. A computer device comprising a memory, a processor, and a computer program stored in the memory and executed by the processor, characterized in that: When the processor executes the computer program, the steps of the venue type identification method according to any one of claims 1 to 4 are implemented.

8. A storage medium, characterized in that: include: The storage medium stores a computer program, and when the computer program is executed by the processor, the steps of the place type identification method according to any one of claims 1 to 4 are implemented.

Images