Detailed Description
The following description of the technical solutions according to the embodiments of the present application will be given with reference to the accompanying drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the application without making any inventive effort, are intended to be within the scope of the application.
Unless defined otherwise, all technical and scientific terms used in the embodiments of the application have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used in the present application is for the purpose of describing particular embodiments only and is not intended to limit the scope of the present application.
In describing embodiments of the present application, unless otherwise indicated and limited thereto, the term "connected" should be construed broadly, for example, it may be an electrical connection, or may be a communication between two elements, or may be a direct connection, or may be an indirect connection via an intermediate medium, and it will be understood by those skilled in the art that the specific meaning of the term may be interpreted according to circumstances.
It should be noted that, the term "first\second\third" related to the embodiment of the present application is merely to distinguish similar objects, and does not represent a specific order for the objects, it is to be understood that "first\second\third" may interchange a specific order or sequence where allowed. It is to be understood that the "first\second\third" distinguishing objects may be interchanged where appropriate such that embodiments of the application described herein may be practiced in sequences other than those illustrated or described herein.
In one embodiment of the present application, fig. 1 is a block diagram of a real-time positioning command processing system based on traffic information of a congested road segment according to an embodiment of the present application. As shown in fig. 1, a real-time positioning command processing system 100 based on traffic information of a congested road segment according to an embodiment of the present application includes: the road section monitoring module 110 is used for acquiring road section monitoring images of monitored congestion road sections acquired by the unmanned aerial vehicle; a traffic flow monitoring module 120, configured to obtain traffic flow values of a plurality of road segments communicating with the monitored congestion road segment at a plurality of predetermined time points within a predetermined time period; the image block segmentation module 130 is configured to segment the road segment monitoring image of the monitored congestion road segment along the extending direction of the road segment to obtain a plurality of road segment sub-images; the road section semantic understanding module 140 is configured to pass the plurality of road section sub-images through a ViT model including an embedded layer to obtain a road section global semantic feature vector; the data structuring module 150 is configured to arrange the traffic flow values of the plurality of road segments at a plurality of predetermined time points within a predetermined time period into a traffic flow global input matrix according to a time dimension and a road segment sample dimension; the multi-scale sensing module 160 is configured to pass the global traffic flow input matrix through a multi-scale sensor including a first convolution layer and a second convolution layer to obtain a multi-scale traffic flow associated feature vector; the association encoding module 170 is configured to perform association encoding on the multi-scale traffic flow association feature vector and the road section global semantic feature vector to obtain a decoding feature matrix; and a decoding regression module 180 for passing the decoding feature matrix through a decoder to obtain a decoded value representing a predicted time value for passing through the congested road segment.
Specifically, in the embodiment of the present application, the road segment monitoring module 110 and the traffic flow monitoring module 120 are configured to acquire a road segment monitoring image of a monitored congested road segment acquired by an unmanned aerial vehicle; and acquiring traffic flow values of a plurality of road segments communicated with the monitored congestion road segment at a plurality of preset time points in a preset time period.
In view of the above technical problems, the technical concept of the present application is to determine the current congestion situation based on the monitored image of the monitored congestion road section by using deep learning and artificial intelligence technology, and predict the vehicle passing time of the congestion road section based on the traffic flow characteristics of the road section communicated with the monitored road section. Compared with the traditional fixed sensor equipment, the scheme has the advantages of wide monitoring range, high data updating frequency and the like, and can better adapt to the real-time dynamic requirements of urban traffic management. Therefore, the traffic condition of the congested road section can be effectively monitored and predicted, so that the road traffic efficiency is improved, and better service is brought to urban traffic management and resident traveling.
Specifically, in the technical scheme of the application, firstly, a road section monitoring image of a monitored congestion road section collected by an unmanned aerial vehicle is obtained. Compared with the traditional fixed sensor equipment, the unmanned aerial vehicle can fly and shoot more flexibly, can acquire a wider monitoring range, and can better reflect the traffic condition of a road section. And simultaneously, acquiring traffic flow values of a plurality of road sections communicated with the monitored congestion road section at a plurality of preset time points in a preset time period. Here, the traffic flow is an important index, and can reflect the traffic capacity and traffic conditions of the road. By acquiring the traffic flow values of a plurality of road segments communicating with the monitored congested road segments, the real-time traffic conditions of these road segments can be known.
In one embodiment of the application, the location and extent of the monitored congested road segments are determined; calling an acquisition program of the unmanned aerial vehicle, and setting acquisition parameters including shooting angles, resolution, frame rate and the like; and starting the unmanned aerial vehicle, enabling the unmanned aerial vehicle to fly to the space above the monitored congestion road section, and starting to collect images. In this way, road segment monitoring images of monitored congested road segments acquired by the unmanned aerial vehicle are acquired.
Further, determining adjacent road sections of the monitored congestion road sections, and installing a traffic flow detector; setting parameters of a traffic flow detector, including a detection time period, a detection time point, a traffic flow calculation method and the like; at a plurality of time points over a predetermined period of time, a traffic flow detector is activated to record traffic flow data. In this way, the vehicle flow values of a plurality of road sections communicated with the monitored congestion road section at a plurality of preset time points in a preset time period are obtained
Specifically, in the embodiment of the present application, the image block segmentation module 130 is configured to segment the road segment monitoring image of the monitored congestion road segment along the extending direction of the road segment to obtain a plurality of road segment sub-images. In practice, the monitored congested road section may be a longer road section, and the traffic conditions on this road section may be uneven, for example, there may be local congestion. In order to better reflect the traffic condition of the road section, in the technical scheme of the application, the road section monitoring image of the monitored congestion road section is subjected to image block segmentation along the extending direction of the road section so as to obtain a plurality of road section sub-images. Thus, each road segment sub-image obtained after image segmentation corresponds to a local area on the road segment, and the traffic condition of the local area can be highlighted to a certain extent.
The image block segmentation is to segment a large image into a plurality of small images according to a certain rule so as to analyze and process each small image. In one embodiment of the application, the road section monitoring image of the monitored congestion road section is divided equally, and the road section monitoring image of the monitored congestion road section is divided equally into a plurality of image blocks with equal size along the extending direction of the road section, and each image block corresponds to one road section sub-image.
In another embodiment of the present application, the road segment monitoring image of the monitored congestion road segment is segmented based on road segment segmentation, the monitored congestion road segment is segmented into a plurality of small road segments, the road segment segmentation result is mapped onto the road segment monitoring image to obtain the positions and the ranges of the small road segments on the image, and finally the image is segmented according to the positions and the ranges of the small road segments to obtain a plurality of road segment sub-images.
Specifically, in the embodiment of the present application, the road segment semantic understanding module 140 is configured to pass the plurality of road segment sub-images through a ViT model including an embedded layer to obtain a road segment global semantic feature vector. The plurality of road segment sub-images are then passed through a ViT model that includes an embedded layer to obtain a road segment global semantic feature vector. It should be understood that, since the multiple road segment sub-images are not structured data, in the technical solution of the present application, a ViT model including an embedded layer is used for analyzing and processing the road segment sub-images. Wherein the embedding layer is capable of converting the image information into feature vectors. The ViT model is an image processing model based on a transducer, and can perform global semantic understanding on feature vectors corresponding to a plurality of road segment sub-images, so as to obtain the road segment global semantic feature vector.
Fig. 2 is a block diagram of the link semantic understanding module in the real-time positioning command processing system based on traffic information of a congested link according to an embodiment of the present application, as shown in fig. 2, the link semantic understanding module 140 includes: an embedding unit 141, configured to perform vector embedding on each road segment sub-image in the plurality of road segment sub-images by using the embedding layer including the ViT model of the embedding layer to obtain a sequence of image embedding vectors; and a transform coding unit 142, configured to input the sequence of image embedded vectors into the converter of the ViT model including the embedded layer to obtain the road segment global semantic feature vector.
Fig. 3 is a block diagram of the transcoding unit in the real-time positioning command processing system based on traffic information of a congested road segment according to an embodiment of the present application, as shown in fig. 3, the transcoding unit 142 includes: a vector construction subunit 1421, configured to perform one-dimensional arrangement on the sequence of the image embedded vectors to obtain a global image feature vector; a self-attention subunit 1422, configured to calculate products between the global image feature vector and transpose vectors of respective image embedding vectors in the sequence of image embedding vectors to obtain a plurality of self-attention correlation matrices; a normalization subunit 1423, configured to perform normalization processing on each of the plurality of self-attention correlation matrices to obtain a plurality of normalized self-attention correlation matrices; a attention calculating subunit 1424, configured to obtain a plurality of probability values by using a Softmax classification function from each normalized self-attention correlation matrix in the plurality of normalized self-attention correlation matrices; and an attention applying subunit 1425, configured to weight each image embedding vector in the sequence of image embedding vectors with each probability value in the plurality of probability values as a weight to obtain the global semantic feature vector of the road segment.
It should be understood that since the transducer structure proposed by Google in 2017, a wave of hot surge is rapidly initiated, and for the NLP field, the self-attention mechanism replaces the conventional cyclic neural network structure adopted when processing sequence data, so that not only is parallel training realized, but also the training efficiency is improved, and meanwhile, good results are obtained in application. In NLP, a sequence is input into a transducer, but in the field of vision, how to convert a 2d picture into a 1d sequence needs to be considered, and the most intuitive idea is to input pixels in the picture into the transducer, but the complexity is too high.
While the ViT model can reduce the complexity of input, the picture is cut into image blocks, each image block is projected as a fixed length vector into the transducer, and the operation of the subsequent encoder is identical to that of the original transducer. However, because the pictures are classified, a special mark is added into the input sequence, and the output corresponding to the mark is the final class prediction. ViT exhibits quite excellent performance over many visual tasks, but the lack of inductive biasing allows ViT to be applied to small data sets with very much dependence on model regularization (model regularization) and data augmentation (data augmentation) compared to CNN (Convolutional Neural Network ).
Specifically, in the embodiment of the present application, the data structuring module 150 is configured to arrange the traffic flow values of the plurality of road segments at a plurality of predetermined time points in a predetermined time period into a traffic flow global input matrix according to a time dimension and a road segment sample dimension. Considering that there is a complex correlation between different road segments in traffic flow data in traffic management, for example, a change in traffic flow of a road segment may affect a change in traffic flow of an adjacent road segment. The traffic flow of the same link also has a dynamic change characteristic in time series. In the technical scheme of the application, in order to mine the implicit association relation, firstly, the traffic flow values of a plurality of road sections at a plurality of preset time points in a preset time period are arranged into a traffic flow global input matrix according to a time dimension and a road section sample dimension so as to convert traffic flow data into a computable matrix form, thereby facilitating subsequent data processing and analysis.
Wherein, arrange in time dimension, include: and arranging a plurality of preset time points in a preset time period according to a time sequence to serve as a column of a global input matrix of the traffic flow, namely taking a traffic flow value corresponding to each time point as a column of data.
Further, the arrangement according to the road section sample dimension comprises: the monitored congestion road sections and a plurality of road sections communicated with the monitored congestion road sections are arranged according to a certain sequence and used as rows of a global input matrix of the traffic flow, namely the traffic flow value corresponding to each road section is used as one row of data.
The traffic flow values are arranged in the manner, and the obtained traffic flow global input matrix can be used as the input of a follow-up traffic flow analysis and prediction model. Each element represents a traffic flow value of a certain road section at a certain time point, and can be used for analyzing a change trend of the traffic flow, predicting future traffic flow and the like.
Specifically, in the embodiment of the present application, the multi-scale sensing module 160 is configured to pass the global traffic flow input matrix through a multi-scale sensor including a first convolution layer and a second convolution layer to obtain a multi-scale traffic flow correlation feature vector. And then, the traffic flow global input matrix passes through a multi-scale sensor comprising a first convolution layer and a second convolution layer to obtain a multi-scale traffic flow associated feature vector. The first convolution layer and the second convolution layer of the multi-scale sensor have convolution kernels with different scales, and can effectively extract correlation features of traffic flow data in different time spans.
Fig. 4 is a block diagram of the multi-scale sensing module in the real-time positioning command processing system based on traffic information of a congested road segment according to an embodiment of the present application, as shown in fig. 4, the multi-scale sensing module 160 includes: a first convolution unit 161, configured to input the traffic flow global input matrix into a first convolution layer of the multi-scale sensor to obtain a first-scale traffic flow feature vector, where the first convolution layer has a convolution kernel of a first scale; a second convolution unit 162, configured to input the global traffic flow input matrix into a second convolution layer of the multi-scale sensor to obtain a second-scale traffic flow feature vector, where the second convolution layer has a convolution kernel of a second scale, and the first scale is different from the second scale; and a merging unit 163, configured to concatenate the first-scale traffic flow feature vector and the second-scale traffic flow feature vector to obtain the multi-scale traffic flow correlation feature vector.
It should be noted that the multi-scale sensor is essentially a deep neural network model based on deep learning, which can fit any function by a predetermined training strategy and has a higher feature extraction generalization capability than the conventional feature engineering.
The multi-scale sensor comprises a plurality of parallel one-dimensional convolution layers, wherein in the process of extracting the characteristics of the multi-scale sensor, the plurality of parallel one-dimensional convolution layers respectively carry out one-dimensional convolution coding on input data by one-dimensional convolution check with different scales so as to capture local implicit characteristics of a sequence.
Wherein the first convolution unit 161 is configured to: and carrying out convolution processing, pooling processing and nonlinear activation processing on the traffic flow global input matrix by using a first convolution layer of the multi-scale sensor so as to output the traffic flow characteristic vector with the first scale by the first convolution layer.
A second convolution unit 162 for: and carrying out convolution processing, pooling processing and nonlinear activation processing on the traffic flow global input matrix by using a second convolution layer of the multi-scale sensor so as to output the traffic flow characteristic vector with the second scale by the second convolution layer.
The convolutional neural network is a deep learning model and is generally used in the fields of image recognition, voice recognition, natural language processing and the like. In the analysis of traffic data, convolutional neural networks may be used to extract characteristics of the traffic data, such as traffic density, vehicle speed, and congestion level. The multi-scale sensor is a common structure in a convolutional neural network, and features at different scales can be extracted by applying convolution kernels with different sizes on the different scales, so that the performance of the model is improved.
In traffic data analysis, multiscale sensors may be used to process traffic data over different time spans, such as hours, half hours, and five minutes. By using convolution kernels with different scales, the multi-scale sensor can effectively extract the correlation characteristics of the vehicle flow data under different time spans, thereby improving the accuracy and the robustness of the model.
Specifically, in the embodiment of the present application, the association encoding module 170 is configured to perform association encoding on the multi-scale traffic flow association feature vector and the road section global semantic feature vector to obtain a decoding feature matrix. It should be understood that in traffic management, the characteristic information of a road segment has a great influence on the change of the traffic flow, for example, the width of the road segment, the road type, the surrounding environment, etc. all affect the change of the traffic flow. Therefore, in the technical scheme of the application, the multi-scale traffic flow associated feature vector and the road section global semantic feature vector are associated and encoded to fuse traffic flow data and road section feature information, so as to obtain a decoding feature matrix. In this way, the accuracy of congestion prediction is improved.
Wherein, the association encoding module 170 is configured to: performing association coding on the multi-scale traffic flow association feature vector and the road section global semantic feature vector by using the following coding formula to obtain a decoding feature matrix; wherein, the coding formula is:
Wherein V isa Representing the multi-scale vehicle flow correlation feature vector,transposed vector representing the multi-scale traffic flow associated feature vector, Vc Representing the global semantic feature vector of the road segment, M1 Representing the decoding feature matrix,/a>Representing matrix multiplication.
Specifically, in the embodiment of the present application, the decoding regression module 180 is configured to pass the decoding feature matrix through a decoder to obtain a decoded value, where the decoded value is used to represent a time value predicted to pass through the congested road segment. The decoding feature matrix is then passed through a decoder to obtain a decoded value representing a predicted time value for passing through the congested road segment. Wherein the decoder may generate the output values from the input feature matrix. That is, in the technical solution of the present application, the decoder is configured to restore the decoding feature matrix obtained by association encoding to a predicted value representing a time value passing through the congested road section, so as to provide an actual reference for traffic management and travel, for example, the decoder may be used to optimize control of traffic lights, adjust departure time of buses, etc., improve road traffic efficiency, and reduce travel time.
Wherein, the decoding regression module 180 is configured to: passing the decoding feature matrix through a decoder in a decoding formula to obtain a decoded value representing a predicted time value for passing through the congested road segment; wherein, the decoding formula is:wherein M isd Representing the decoding feature matrix, Y representing the decoded values, W representing a weight matrix, B representing a bias vector,>representing a matrix multiplication.
In convolutional neural networks, the decoder is a module for mapping high-level abstract features back into the original input space. Together with the encoder, a complete self-encoder or convolutional neural network is usually constructed. The main function of the decoder is to re-map the advanced features generated by the encoder back to the original input space, thereby reconstructing the original input data.
The decoder typically consists of a deconvolution layer or an upsampling layer. The deconvolution layer is a layer in a convolutional neural network that extends the low-dimensional feature map into a high-dimensional space, thereby enabling up-sampling of the image. The upsampling layer is a layer for enlarging the image size, and is usually implemented by interpolation or other techniques. In image processing, the decoder may be used for tasks such as image reconstruction, image denoising, image super resolution, etc. In traffic data analysis, a decoder may be used to map the encoder-generated traffic characteristics back to the original traffic data space, thereby reconstructing the original traffic data. Further, the congestion road section traffic information-based real-time positioning command processing system further comprises a training module for training the ViT model comprising an embedded layer, the multi-scale sensor comprising a first convolution layer and a second convolution layer and the decoder; fig. 5 is a block diagram of the training module in the real-time positioning command processing system based on traffic information of a congested road segment according to an embodiment of the present application, as shown in fig. 5, the training module 190 includes: a training data acquisition module 191, configured to acquire training data, where the training data includes a training road segment monitoring image of a monitored congestion road segment acquired by an unmanned aerial vehicle, a training vehicle flow value of a plurality of road segments communicated with the monitored congestion road segment at a plurality of predetermined time points within a predetermined time period, and a real value of a time value of the congestion road segment; the training image block segmentation module 192 is configured to segment the training road segment monitoring image of the monitored congestion road segment along the road segment extending direction to obtain a plurality of training road segment sub-images; a training road section semantic understanding module 193, configured to pass the plurality of training road section sub-images through the ViT model including the embedded layer to obtain a training road section global semantic feature vector; the training data structuring module 194 is configured to arrange training traffic flow values of the multiple road segments at multiple preset time points within a preset time period into a training traffic flow global input matrix according to a time dimension and a road segment sample dimension; a training multi-scale perception module 195, configured to pass the training traffic global input matrix through the multi-scale perceptron including the first convolution layer and the second convolution layer to obtain a training multi-scale traffic associated feature vector; the training association coding module 196 is configured to perform association coding on the training multi-scale traffic flow association feature vector and the training road section global semantic feature vector to obtain a training decoding feature matrix; a true value difference loss module 197 for passing the training decoding feature matrix through the decoder to obtain a true value difference loss function value; the pseudo-cycle difference module 198 is configured to calculate a pseudo-cycle difference penalty factor for the training road global semantic feature vector and the training multi-scale traffic flow associated feature vector; and a training module 199 for training the multi-scale perceptron and the decoder through the model ViT comprising the embedded layer, the first and second convolutional layers, with a weighted sum of the true value difference loss function value and the pseudo-cyclic difference penalty factor as a loss function value, and based on the direction of gradient descent propagation.
In the technical scheme of the application, the road section global semantic feature vector expresses the context associated coding feature of the image feature semantics of each road section sub-image, and the multi-scale traffic flow associated feature vector expresses the cross dimension associated feature of different scales in the time-sample dimension, so that the difference of the multi-scale traffic flow associated feature vector and the road section global semantic feature vector in the feature dimension and the feature semantics leads to unbalanced overall feature distribution relative to the distribution of the position-by-position associated codes, thereby influencing the training effect of a model and the accuracy of the decoding value of the decoding feature matrix obtained from the associated codes.
Based on this, the applicant of the present application proceeds beyond the true-to-try difference loss function for the decoded valuesStep-wise introduction of global semantic feature vectors, e.g. denoted V, for said road segments1 And the multiscale traffic flow associated feature vector, e.g. denoted V2 As a loss function, the pseudo-cyclic difference penalty factor of (a) is expressed in detail as: calculating pseudo-cycle difference penalty factors of the training road section global semantic feature vector and the training multi-scale vehicle flow associated feature vector according to the following optimization formula; wherein, the optimization formula is:
Wherein V is1 Is the global semantic feature vector of the training road section, V2 Is the training multi-scale vehicle flow associated feature vector, D (V1 ，V2 ) For the distance matrix between the training road section global semantic feature vector and the training multi-scale vehicle flow correlation feature vector, the distance matrix is IIF The Frobenius norm of the matrix, L is the length of the eigenvector, d (V1 ，V2 ) Is the Euclidean distance between the global semantic feature vector of the training road section and the training multi-scale vehicle flow associated feature vector, and is II2 Is the two norms of the vector, log represents a logarithmic function based on 2, and alpha and beta are weighted hyper-parameters,is the pseudo-cyclic difference penalty factor, +.>And->Respectively representing subtraction by position and addition by position.
Here, consider the road segment global semantic feature vector V1 And the multiscale traffic flow associated feature vector V2 The imbalance distribution between the two can lead to gradient propagation abnormality in the model training process based on the back propagation of gradient descent, thereby forming a modelA pseudo-loop of parameter updates, said pseudo-loop difference penalty factors treating the pseudo-loop of model parameter updates as a true loop in a model training process minimizing a loss function by introducing penalty factors for expressing both spatial and numerical relationships of closely related numerical pairs of feature values, to implement the road segment global semantic feature vector V by simulated activation of gradient propagation1 And the multiscale traffic flow associated feature vector V2 Progressive coupling of the respective feature distributions, thereby improving the training effect of the model and the accuracy of the decoded values of the decoded feature matrix.
As described above, the real-time positioning command processing system 100 based on the traffic information of the congested road segment according to the embodiment of the present application may be implemented in various terminal devices, for example, a server or the like for real-time positioning command processing based on the traffic information of the congested road segment. In one example, the congestion road segment traffic information based real-time location command processing system 100 according to an embodiment of the present application may be integrated into a terminal device as a software module and/or hardware module. For example, the congestion road segment traffic information-based real-time positioning command processing system 100 may be a software module in the operating system of the terminal device, or may be an application developed for the terminal device; of course, the real-time positioning command processing system 100 based on the traffic information of the congested road segment may also be one of a plurality of hardware modules of the terminal device.
Alternatively, in another example, the congestion road segment traffic information-based real-time positioning command processing system 100 and the terminal device may be separate devices, and the congestion road segment traffic information-based real-time positioning command processing system 100 may be connected to the terminal device through a wired and/or wireless network and transmit the interactive information in a contracted data format.
In one embodiment of the present application, fig. 6 is a flowchart of a method for processing real-time positioning command based on traffic information of a congested road segment according to an embodiment of the present application. Fig. 7 is a schematic diagram of a system architecture of a real-time positioning command processing method based on traffic information of a congested road segment according to an embodiment of the present application. As shown in fig. 6 and fig. 7, the method for processing traffic information real-time positioning command based on a congested road segment according to an embodiment of the present application includes: 210, acquiring a road section monitoring image of a monitored congestion road section acquired by an unmanned aerial vehicle; 220, acquiring traffic flow values of a plurality of road segments communicated with the monitored congestion road segment at a plurality of preset time points in a preset time period; 230, performing image block segmentation on the road section monitoring image of the monitored congestion road section along the extending direction of the road section to obtain a plurality of road section sub-images; 240, passing the plurality of road segment sub-images through a ViT model containing an embedded layer to obtain a road segment global semantic feature vector; 250, arranging the traffic flow values of the plurality of road sections at a plurality of preset time points in a preset time period into a traffic flow global input matrix according to a time dimension and a road section sample dimension; 260, passing the global traffic flow input matrix through a multi-scale sensor comprising a first convolution layer and a second convolution layer to obtain a multi-scale traffic flow associated feature vector; 270, performing association coding on the multi-scale traffic flow association feature vector and the road section global semantic feature vector to obtain a decoding feature matrix; and, passing 280 the decoding feature matrix through a decoder to obtain a decoded value representing a predicted time value for passing through the congested road segment.
It will be appreciated by those skilled in the art that the specific operations of the respective steps in the congestion section-based traffic information real-time location commander processing method described above have been described in detail in the description of the congestion section-based traffic information real-time location commander processing system described above with reference to fig. 1 to 5, and thus, repetitive descriptions thereof will be omitted.
Fig. 8 is an application scenario diagram of a real-time positioning command processing system based on traffic information of a congested road segment according to an embodiment of the present application. As shown in fig. 8, in this application scenario, first, a road segment monitoring image (e.g., C1 as shown in fig. 8) of a monitored congested road segment collected by an unmanned aerial vehicle (e.g., D as shown in fig. 8) is acquired, and vehicle flow values (e.g., C2 as shown in fig. 8) of a plurality of road segments communicating with the monitored congested road segment at a plurality of predetermined time points within a predetermined time period are acquired; the acquired link monitoring images are then input into a server (e.g., S as illustrated in fig. 8) deployed with a congestion link traffic information-based real-time location and command processing algorithm capable of processing the link monitoring images based on the congestion link traffic information to generate decoded values representing time values predicted to pass through the congested link.
It is also noted that in the apparatus, devices and methods of the present application, the components or steps may be disassembled and/or assembled. Such decomposition and/or recombination should be considered as equivalent aspects of the present application.
The previous 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 generic principles defined herein may be applied to other aspects without departing from the scope of the application. Thus, the present application is not intended to be limited to the aspects shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.
The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit embodiments of the application to the form disclosed herein. Although a number of example aspects and embodiments have been discussed above, a person of ordinary skill in the art will recognize certain variations, modifications, alterations, additions, and subcombinations thereof.