Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Satellite remote sensing image farmland block segmentation method based on improved Mask R-CNN model

A remote sensing image and satellite remote sensing technology, which is applied in the field of remote sensing image processing, can solve problems such as feature redundancy, weight ratio, and important feature omission, and achieve improved instance segmentation effect, high robustness, and enhanced multi-scale The effect of feature fusion

Pending Publication Date: 2021-12-31
广州华农大智慧农业科技有限公司
View PDF2 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method is to manually extract features, which may not only miss important features, but also may not take into account the redundancy and weight ratio between features.

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
  • Satellite remote sensing image farmland block segmentation method based on improved Mask R-CNN model
  • Satellite remote sensing image farmland block segmentation method based on improved Mask R-CNN model
  • Satellite remote sensing image farmland block segmentation method based on improved Mask R-CNN model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0042] A farmland block segmentation method for satellite remote sensing images based on the improved Mask R-CNN model, such as figure 1 As shown, the method includes the following steps:

[0043] S1: Acquire satellite remote sensing images, preprocess the remote sensing images to obtain normalized remote sensing images, and divide them into cloudy remote sensing images and cloudless remote sensing images according to whether there are clouds in the normalized remote sensing images;

[0044] S2: Use the improved Mask R-CNN model to segment the cloud in the cloudy remote sensing image to obtain the cloudless mask image, and then combine the cloudless mask image with the remote sensing image of the same area in different phases to synthesize the cloudless remote sensing image, thus Make all normalized remote sensing images cloud-free;

[0045] S3: Take the cloud-free remote sensing image as the training set, input it into the improved Mask R-CNN model for model training, and ob...

Embodiment 2

[0049] The flow chart of the satellite remote sensing image farmland block segmentation method based on the improved Mask R-CNN model is as follows figure 1 shown, including the following steps:

[0050] S1: Acquire satellite remote sensing images, preprocess the remote sensing images to obtain normalized remote sensing images, and divide them into cloudy remote sensing images and cloudless remote sensing images according to whether there are clouds in the normalized remote sensing images;

[0051] Described pretreatment comprises:

[0052] S11: Acquiring RPC information of satellite remote sensing images;

[0053] S12: Respectively perform orthorectification on multispectral and panchromatic data of satellite remote sensing images based on RPC information;

[0054] S13: registering the multispectral data based on the panchromatic data;

[0055] S14: Perform an image fusion operation on the registered images;

[0056] S15: performing radiometric calibration and atmospheric...

Embodiment 3

[0069] The flow chart of the satellite remote sensing image farmland block segmentation method based on the improved Mask R-CNN model is as follows figure 1 shown, including the following steps:

[0070] S1: Acquire satellite remote sensing images, preprocess the remote sensing images to obtain normalized remote sensing images, and divide them into cloudy remote sensing images and cloudless remote sensing images according to whether there are clouds in the normalized remote sensing images;

[0071] Described pretreatment comprises:

[0072] S11: Acquiring RPC information of satellite remote sensing images;

[0073] S12: Respectively perform orthorectification on multispectral and panchromatic data of satellite remote sensing images based on RPC information;

[0074] S13: registering the multispectral data based on the panchromatic data;

[0075] S14: Perform an image fusion operation on the registered images;

[0076] S15: performing radiometric calibration and atmospheric...

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 provides a satellite remote sensing image farmland block segmentation method based on an improved Mask R-CNN model, and relates to the technical field of remote sensing image processing. According to the method, the Mask R-CNN model is used for training, the model can automatically extract the farmland block texture features in the sensing image in the training process except the spectral features, the farmland boundary with high availability is obtained in combination with the spatial semantic information of the texture features and the spectral features, and the Mask R-CNN can automatically extract the features of the farmland blocks in the satellite remote sensing image, and compared with manual extraction of farmland block features, the method has the technical advantages of accuracy, high efficiency and high robustness. A PANet structure replaces an FPN structure to improve the Mask R-CNN model, multi-scale feature fusion is enhanced, and the instance segmentation effect of the Mask R-CNN model is improved.

Description

technical field [0001] The invention relates to the technical field of remote sensing image processing, and more specifically, to a method for segmenting farmland blocks of satellite remote sensing images based on an improved Mask R-CNN model. Background technique [0002] After years of development, satellite remote sensing technology has made great progress in terms of spectral resolution, spatial resolution, and temporal resolution. Earth observation capabilities. The main characteristics of high-resolution satellite remote sensing images are: rich texture information of ground objects, multiple imaging spectral bands, and short revisit time. The technology of realizing accurate detection of small areas under high resolution has been widely used in military and civilian fields at home and abroad. . With the acquisition of a large amount of satellite remote sensing data, the application of intelligent and precision agriculture has become a hot spot and trend, such as int...

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
IPC IPC(8): G06T7/11G06K9/62
CPCG06T7/11G06T2207/10032G06T2207/20081G06T2207/20084G06T2207/30188G06F18/253
Inventor 不公告发明人
Owner 广州华农大智慧农业科技有限公司
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com