Traffic data filling method for lossy measurement based generating model
A technology for generating models and filling methods, applied in the field of intelligent transportation and deep learning, can solve the problem of not making full use of the temporal and spatial characteristics of traffic data and historical information of the data, and achieve the effect of improving applicability and recovery accuracy.
Active Publication Date: 2018-11-27
FUZHOU UNIV
View PDF3 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The purpose of the present invention is to provide a traffic data filling method based on a generation model based on lossy measurement. The method combines the generation model based on lossy measurement and 3D convolutional neural network, which can overcome the time-space of traffic data that cannot be fully utilized by existing methods. Disadvantages of historical information of features and data, the use of generative models can recover a variety of missing data and improve the accuracy of recovery
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0038] Attached below Figure 1-3 , the technical solution of the present invention is described in detail.
[0039] It should be pointed out that the following detailed description is exemplary and intended to provide further explanation to the present application. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.
[0040] It should be noted that the terminology used here is only for describing specific implementations, and is not intended to limit the exemplary implementations according to the present application. As used herein, unless the context clearly dictates otherwise, the singular is intended to include the plural, and it should also be understood that when the terms "comprising" and / or "comprising" are used in this specification, they mean There are features, steps, operations, means, components and / or combinations thereof.
...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The present invention relates to a traffic data filling method for a lossy measurement based generating model. Considering that approach time traffic data and a large amount of historical data can help to improve the accuracy of the recovery of traffic data. The method combines a 3D convolutional neural network with the lossy measurement based generating model to construct a novel network model torealize the filling of traffic data. The advantages of the present invention are that incomplete data can be used to perform training to realize the recovery of traffic flow loss data; and the present invention measures the difference between real data and generated data of a known point, obtains an optimal input of a generating network by using a minimum loss function, and obtains the optimal generated data so as to realize the recovery of the traffic flow data. The method overcomes the shortcomings that the traditional methods cannot use the incomplete data to perform training, and fully utilizes the historical traffic flow data and effectively extracts spatio-temporal characteristics of the traffic flow data, thereby improving the accuracy of the recovery of the traffic flow data.
Description
technical field [0001] The invention relates to the fields of intelligent transportation and deep learning, in particular to a traffic data filling method based on a generation model of lossy measurement. Background technique [0002] In the actual process of traffic flow data collection, due to problems such as signal loss and sensor damage, traffic flow data is sometimes missing. And how to use these missing data to fill and apply it to training is also a very important issue in the field of traffic flow research, and the present invention also studies this issue. [0003] In the original methods, most of them are based on vector filling or filling based on time and space features. Most of these methods cannot make full use of the space-time characteristics of the data, which affects the accuracy of recovery. Recently, a 3D_DCGAN method based on DCGAN and three-dimensional convolution is proposed to repair traffic flow data. However, this method cannot use incomplete dat...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06Q50/26G06N3/04
CPCG06Q50/26G06N3/045
Inventor 郑海峰李奥奇李智敏冯心欣
Owner FUZHOU UNIV