AI-POWERED GEOLOGICAL INSIGHTS
Lithological Mapping Based on Multi-Source Fusion Data and Convolutional Neural Networks: A Case Study of the Guyang Area, Inner Mongolia, China
Remote sensing offers distinct advantages for lithological mapping, but its ability to detect underlying bedrock is limited in covered areas, whereas geochemical data are constrained by sparse sampling and low spatial resolution. To address these challenges, this study proposes a texture-guided adaptive data fusion framework combined with a Multi-scale Convolutional Neural Network (MCNN) for lithological mapping, using the Guyang area in Inner Mongolia as a case study. First, the non-linear relationships between geochemical components and remote sensing spatial textures are modeled to achieve complementary integration of heterogeneous multi-source data. Second, an MCNN model is constructed to extract multi-scale geological features, enabling improved discrimination of lithological units and more effective inference of concealed bedrock beneath Quaternary cover. Experimental results show that the proposed method overcomes the limitations of single data sources and achieves an overall accuracy (OA) of 0.95 on the fused dataset. Ablation experiments further demonstrate that the texture-guided fusion strategy significantly improves lithological identification performance. This study provides an effective framework for intelligent geological mapping and confirms the feasibility of inferring underlying bedrock in covered areas using multi-source surface information.
Executive Impact & Key Findings
This study proposes a texture-guided adaptive data fusion framework with a Multi-scale Convolutional Neural Network (MCNN) for lithological mapping, demonstrating significant improvements. The texture-guided fusion strategy improved Overall Accuracy (OA) by 7% (from 0.88 to 0.95). The parallel MCNN architecture effectively extracts multi-scale features, overcoming limitations of single-scale convolution and sample imbalance. The model successfully infers bedrock distribution and concealed structural features in Quaternary-covered areas, proving the potential of multi-source surface information for intelligent geological mapping.
0.00
Overall Accuracy
0.00
AUC Score
0.00
F1-Score (Avg.)
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
This section details the innovative texture-guided adaptive data fusion framework and the Multi-scale Convolutional Neural Network (MCNN) architecture developed for lithological mapping. Learn how heterogeneous multi-source data (remote sensing and geochemical) are integrated to overcome limitations of single data sources, and how the MCNN captures multi-scale geological features for enhanced discrimination.
Multi-Source Data Fusion Process
Image Decomposition (High/Low Freq.)
→
Scaling Transformation
→
Establish Correlation Function
→
Image Reconstruction (High Freq.)
→
Final Image Fusion (Low + High Freq.)
| Model Configuration | Overall Accuracy (OA) |
|---|---|
| CNN + Geochemical Data (Baseline) | 0.70 |
| MCNN + Geochemical Data | 0.88 |
| CNN + Multi-source Fused Data | 0.75 |
| MCNN + Multi-source Fused Data (Proposed) | 0.95 |
Explore the experimental results validating the proposed framework. This section covers classification performance, the geological interpretation of findings, and the effectiveness in inferring concealed bedrock in Quaternary-covered areas, highlighting the synergy between data sources and model architecture.
0.95
Overall Accuracy Achieved on Fused Data
The proposed MCNN model with multi-source fused data achieved an outstanding Overall Accuracy of 0.95, demonstrating superior performance in lithological classification.
Guyang Area: Inferring Concealed Bedrock
In the Guyang area, the model successfully inferred bedrock distribution and structural features beneath Quaternary cover. This demonstrates the potential of multi-source data fusion for geological mapping in regions with limited direct bedrock exposure.
- Geochemical signals provide compositional insights despite transport.
- RS-derived structural information restores spatial constraints.
- Improved prediction of concealed lithological units, aiding mineral exploration.
This section provides a detailed geological context of the Guyang area in Inner Mongolia. It covers the complex stratigraphy, lithology (Archean to Mesozoic), tectonic setting (Yinshan Fault Uplift), and the history of magmatic activity, which collectively create a rich environment for mineral resources and present challenges for traditional mapping.
The study concludes that the texture-guided adaptive data fusion framework combined with an improved MCNN significantly enhances lithological mapping, particularly in complex and covered geological terrains. The framework provides a robust solution for integrating heterogeneous geoscientific data, proving effective for intelligent geological mapping and inferring concealed bedrock.
Calculate Your Potential ROI with AI Mapping
Estimate the significant time and cost savings your enterprise could achieve by integrating AI-powered lithological mapping solutions.
Estimated Annual Savings
$0
Annual Hours Reclaimed
0
Your AI Mapping Implementation Roadmap
A typical phased approach to integrating advanced multi-source fusion and MCNN for geological mapping, tailored for enterprise deployment.
Data Acquisition & Preprocessing
Duration: 1-2 Weeks
Gathering and cleaning remote sensing imagery and geochemical data, including interpolation and ILR transformation.
Multi-Source Data Fusion
Duration: 2-3 Weeks
Implementing the texture-guided adaptive fusion framework to integrate heterogeneous data, creating a unified feature space.
MCNN Model Training & Optimization
Duration: 3-4 Weeks
Training the Multi-scale CNN on the fused dataset, including hyperparameter tuning, class imbalance handling with SMOTE and Focal Loss.
Lithological Mapping & Validation
Duration: 1-2 Weeks
Generating lithological maps, performing ablation studies, and validating the model's performance against geological ground truth.
Deployment & Reporting
Duration: 1 Week
Finalizing the mapping results, documenting the methodology, and preparing for integration into enterprise geological survey workflows.
Ready to Transform Your Geological Mapping?
Don't let complex terrains and data limitations hinder your exploration. Leverage our AI expertise to gain unprecedented clarity and efficiency.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat