Unlock instant, AI-driven research and patent intelligence for your innovation.

Aerial image-based road image data set preparation method and device

An image data set and aerial image technology, applied in the field of image processing, can solve the problems of low color fidelity, time-consuming and laborious, and manual labeling workload, and achieve the effect of improving resolution

Active Publication Date: 2021-10-22
ROCKET FORCE UNIV OF ENG +1
View PDF11 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] There are some problems in the deep learning method: 1. It is usually necessary to manually label the original image with the help of labeling tools to obtain label data. , the process is complicated. Taking the labeling of a road segmentation data set with a range of 10 square kilometers as an example, it takes about one and a half months for one person to complete the labeling process, and it is necessary to carefully check the road information of the original image, which is a time-consuming and laborious process. , the efficiency is very low; 2. When the original image data is insufficient, the deep learning method augments the original image data through data augmentation, which may cause the original image road to break and affect the quality of the original image data, thus affecting The accuracy of the prepared road image data set; 3. Satellite image data is usually acquired by satellite remote sensing equipment, which has low resolution, low color fidelity, is easily affected by the environment, and the image is unstable

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Aerial image-based road image data set preparation method and device
  • Aerial image-based road image data set preparation method and device
  • Aerial image-based road image data set preparation method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0025] See figure 1 , figure 1 It is a schematic flow chart of a method for preparing a road image data set based on aerial images provided by an embodiment of the present invention, and the method includes:

[0026] Step 101, acquiring a set of aerial images and original satellite images within a preset road area, wherein the set of aerial images includes multiple frames of aerial images, and the original satellite images correspond to original road network raster images.

[0027] The aerial images have the characteristics of high resolution, high color fidelity, and strong stability. The road image data set prepared based on the aerial images has the advantages of high precision, high resolution, high color fidelity, and strong stability. .

[0028] The road network raster image is a bitmap that reflects the geographic features and topological relationships of satellite images, also known as a bitmap. The road image data set prepared based on the road network raster image ...

Embodiment 2

[0128] See Figure 10 , Figure 10 It is a schematic structural diagram of a road image data set preparation device based on aerial images provided by an embodiment of the present invention, and the device includes:

[0129] The acquiring module 1001 is configured to acquire a set of aerial images and original satellite images in a preset road area, wherein the set of aerial images includes multiple frames of aerial images, and the original satellite images correspond to original road network raster images.

[0130] The first projection module 1002 is configured to perform a first projection operation on each frame of the aerial image and the original satellite image to obtain a first satellite image.

[0131] The second projection module 1003 is configured to perform a second projection operation on each frame of the aerial image and the original road network raster image to obtain a target road network raster image corresponding to each frame of the aerial image.

[0132] ...

Embodiment 3

[0154] The embodiment of the present invention also provides an electronic device, such as Figure 11 As shown, it includes a processor 1101, a communication interface 1102, a memory 1103, and a communication bus 1104, wherein the processor 1101, the communication interface 1102, and the memory 1103 complete mutual communication through the communication bus 1104,

[0155] memory 1103, for storing computer programs;

[0156] When the processor 1101 is used to execute the program stored on the memory 1103, the following steps are implemented:

[0157] Acquiring a set of aerial images and original satellite images in a preset road area, wherein the set of aerial images includes multiple frames of aerial images, and the original satellite images correspond to original road network raster images;

[0158] performing a first projection operation on each frame of the aerial image and the original satellite image to obtain a first satellite image;

[0159] performing a second proje...

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

No PUM Login to View More

Abstract

The invention discloses an aerial image-based road image data set preparation method, and the method comprises the steps of obtaining an aerial image set and an original satellite image in a preset road region, the aerial image set comprising multiple frames of aerial images, and the original satellite image corresponding to an original road network raster image; performing first projection operation on each frame of aerial image and the original satellite image to obtain a first satellite image; performing second projection operation on each frame of aerial image and the original road network raster image to obtain a target road network raster image corresponding to each frame of aerial image; and determining each frame of aerial image and the target road network raster image corresponding to each frame of aerial image as a group of target road image data to obtain a target road image data set. According to the invention, the precision and efficiency of road image data set preparation can be improved.

Description

technical field [0001] The invention belongs to the field of image processing, and in particular relates to a method and device for preparing a road image data set based on aerial images. Background technique [0002] As an important infrastructure, roads play an important role in urban planning, smart travel, image registration, automatic navigation, and geographic information system construction. Therefore, the extraction of road information has very important practical significance. [0003] In the prior art, the method of deep learning is usually used to realize the preparation of road data sets. [0004] There are some problems in the deep learning method: 1. It is usually necessary to manually label the original image with the help of labeling tools to obtain label data. , the process is complicated. Taking the labeling of a road segmentation data set with a range of 10 square kilometers as an example, it takes about one and a half months for one person to complete t...

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
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F16/51G06F16/29G06F16/587G06F17/16
CPCG06F16/51G06F16/29G06F16/587G06F17/16
Inventor 杨东方孟德宇赵家玮肖鹏杨晶岚
Owner ROCKET FORCE UNIV OF ENG