[Hard and soft classification method of multi-spectral remote sensing image based on adaptive thresholds].

Area of Science:

• Remote Sensing
• Image Analysis
• Geospatial Science

Context:

• Traditional hard and soft classification methods for satellite image analysis have limitations.
• Accurate land cover and land use mapping is crucial for environmental monitoring and resource management.

Purpose:

• To develop and evaluate a novel Hard and Soft Classification Model (HSCM) for improved satellite image classification.
• To leverage the advantages of both hard and soft classification techniques through adaptive thresholding.

Summary:

• The proposed Hard and Soft Classification Model (HSCM) integrates traditional hard classification methods (HCM) and soft classification models (SCM).
• HSCM utilizes adaptive threshold calculation for enhanced classification performance.
• In land cover mapping, HSCM achieved an overall accuracy of 71.10% and a kappa coefficient of 60.07%, outperforming HCM (71.06% accuracy, 60.03% kappa) and SCM (67.86% accuracy, 56.12% kappa).

Impact:

• The HSCM demonstrates a significant improvement in land cover and land use classification accuracy.
• This new method offers a more robust approach for analyzing satellite imagery.
• Enhanced classification accuracy supports better environmental and land resource management decisions.