A detection method for the changing area of buildings in remote sensing images
A technology of changing areas and detection methods, which is applied in the field of surveying and mapping remote sensing detection, can solve problems such as low detection accuracy and failure to consider multi-scale problems, and achieve the effect of increasing the size of the receptive field
Active Publication Date: 2022-04-26
武汉易米景科技有限公司
View PDF0 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0008] Aiming at the defects of the prior art, the purpose of the present invention is to provide a detection method of the changing area of the building in the remote sensing image, aiming to solve the problem of low detection accuracy in the prior art due to the failure to consider the multi-scale problem and the design of the decoder is too simple
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
[0041] In order to make the object, technical solution and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described here are only used to explain the present invention, not to limit the present invention.
[0042] In the surveying and mapping industry, it is necessary to draw topographic maps or other thematic maps for the geographical environment and ground objects. Generally, image data obtained from various sensors are used for vector editing to form maps. Among them, remote sensing images are used the most, and remote sensing images are divided into aerial images and satellite images. When performing vector editing of images, for two images of different time phases in the same area, the traditional method is to perform complete vector editing on both images of the two time phases. However, in the...
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 the technical field of surveying, mapping and remote sensing detection, and discloses a method for detecting a changing area of a building in a remote sensing image. The changed area detection model detects the changed area on the feature map to obtain the changed area map; S3 uses the semantic segmentation model to perform semantic segmentation on two images of the same area with different time phases, and then obtains the corresponding semantic segmentation results of the two images; S4 converts the changed The change detection result is obtained by performing dot product processing on the region map and the semantic segmentation result. The present invention uses the semantic segmentation data of the building to transfer the feature extraction layer of the change detection model, and detects the change area between two images through the change detection model, and then performs point multiplication with the semantic segmentation results of the two images, A more detailed change detection result between two images is obtained, and change information of overlapping buildings can be detected.
Description
technical field [0001] The invention belongs to the technical field of surveying, mapping and remote sensing detection, and more particularly relates to a detection method for a changing area of a building in a remote sensing image. Background technique [0002] With the current development of deep learning, the traditional method of feature extraction based on manual design and then using traditional machine learning models to detect changing regions has begun to be eliminated. With the improvement of computing resources, the abundance of labeled image data and the introduction of better deep learning models, deep learning has become the mainstream in the field of computer vision. In the field of change detection, various deep learning models have been proposed. The current deep learning models in the field of change detection are roughly divided into two genres: [0003] The first is to fuse the two time-phase images into one image and input it into the deep learning m...
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 Patents(China)
IPC IPC(8): G06V20/13G06V10/26G06V10/44G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/241
Inventor 李鹏飞高凯王嘉伟乐黎明
Owner 武汉易米景科技有限公司