Road screening method, device, server and storage medium
A screening method and road technology, applied in the field of road identification, can solve the problems of manpower consumption, high cost, time delay or interference of regional conditions, and achieve the effect of improving efficiency, reducing cost, and improving accurate determination.
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0024] The road screening method provided in this embodiment is applicable to the situation of screening actual roads from various candidate roads, and the method can be performed by a road screening device, which can be implemented by software and / or hardware. Can be integrated in the server. see figure 1 , the method of this implementation specifically includes the following steps:
[0025] S110. Determine each candidate road included in the target satellite image.
[0026] Wherein, the target satellite image is the image data obtained by photographing or scanning the ground features through cameras, TV cameras, multi-spectral scanners and other equipment during the operation of the artificial earth satellite, including pictures or videos. Wherein, each candidate road may include road name, road length and width information, location information, satellite image, real-time road conditions and other information.
[0027] The candidate roads are extracted from the target sa...
Embodiment 2
[0039] This embodiment is optimized on the basis of the above embodiments, details not described in detail in this embodiment can be found in the above embodiments. see figure 2 , the road screening method provided in this embodiment includes:
[0040] S210. Using the target satellite image as an input of the road recognition model, extract each candidate road included in the target satellite image.
[0041]Wherein, the road recognition model is generated in the following manner: matching the basic road data with sample satellite images to determine the roads included in the sample satellite images; according to the roads included in the sample satellite images, the convolutional network model is trained to generate The road recognition model. Specifically, collect the determined existing road data as the basic road data, collect sample satellite images, and match the basic road data with the sample satellite images to construct training samples and test samples. Optionally...
Embodiment 3
[0050] This embodiment provides a road screening device, see image 3 , the device specifically includes:
[0051] A road determination module 310, configured to determine each candidate road included in the target satellite image;
[0052] Trajectory acquisition module 320, used to filter and obtain the driving trajectory from the trajectory data of each device;
[0053] The matching module 330 is configured to match each candidate road included in the target satellite image with the driving trajectory, and determine the road included in each candidate road according to the matching result.
[0054] Optionally, the road determination module 310 is specifically used for:
[0055] Using the target satellite image as the input of the road recognition model, extracting each candidate road included in the target satellite image;
[0056] Wherein, the road recognition model is generated as follows:
[0057] matching the underlying road data with the sample satellite imagery to ...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


