Remote Sensing Landslide Detection With Auxiliary Image Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing remote sensing landslide detection technologies face challenges in complex environments due to factors like vegetation cover and meteorological changes, leading to low detection accuracy and incomplete utilization of available data.
Innovation Solution
A remote sensing landslide object detection model is developed using a training method that fuses visible light and auxiliary images, incorporating an embedding module, attention feature extraction, and Mask-RCNN model to enhance feature extraction and detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only single type of remote sensing data is used for landslide detection, then the detection process is simple, but the detection accuracy is low and potential information is not fully utilized
Solution Approach 1:
The patent combines multiple types of remote sensing data (SAR images, optical images, digital elevation models, vegetation indices) into a unified data fusion framework. The multi-source data are integrated through feature extraction and fusion modules to create comprehensive landslide detection, resolving the contradiction by merging diverse data types to improve accuracy while managing system complexity through structured integration.
Solution Approach 2:
The patent creates a composite data structure that integrates features from different remote sensing sources (radar backscatter, optical reflectance, elevation data, vegetation coverage) similar to how composite materials combine different substances. This composite approach allows the system to leverage the strengths of each data type while maintaining a unified detection framework, improving overall detection accuracy without linearly increasing complexity.
2Loss of information
If comprehensive data fusion and feature extraction are performed, then more potential information is extracted, but the processing complexity and computational cost increase
Solution Approach 1:
The patent segments the feature extraction process into distinct modules: SAR feature extraction, optical image feature extraction, DEM feature extraction, and vegetation index feature extraction. Each module processes specific data types independently, extracting relevant features before fusion. This segmentation reduces overall complexity by breaking down the comprehensive extraction task into manageable, specialized components while ensuring complete information utilization.
Solution Approach 2:
The patent performs preliminary feature extraction and preprocessing on each remote sensing data type before fusion. By pre-processing individual data sources (normalizing SAR imagery, correcting optical images, extracting elevation features, calculating vegetation indices), the system prepares data in advance for efficient fusion, reducing the complexity of the main detection process while ensuring complete information extraction.
3Measurement precision
If auxiliary images with multiple attributes are used, then detection accuracy is improved, but data processing requirements increase
Solution Approach 1:
The patent extracts and uses only the most relevant attributes from auxiliary images (e.g., vegetation coverage from NDVI, elevation from DEM, surface water from SAR backscatter characteristics). By selectively extracting key features rather than processing all possible image attributes, the system improves detection accuracy while controlling computational resource consumption through focused feature selection.
Data Source
AI summary
The present invention relates to the field of remote sensing image object detection technology, particularly a remote sensing landslide object detection model, a method, a system and a readable medium. The remote sensing landslide object detection model provided by the present invention, firstly, the model pre-trains the embedding module, the location encoding module, and the attention feature extraction module on the first training set to realize the learning of the knowledge attributes associated with the auxiliary images; and then the attention feature extraction module and the Mask-RCNN model are further trained on the complete data set, so as to realize the fusion of the knowledge features and the visible image features, and comprehensively describe the characteristics of landslides, and the deep learning model is adopted to automatically extract the complex features to improve the detection capability of the landslide area.