Color Night Light Remote Sensing Images Generation Using Dual-Transformation.

Area of Science:

• Remote Sensing
• Image Processing
• Geospatial Analysis

Background:

• Traditional night light images lack resolution and color, limiting their use in detailed analysis.
• Accurate urban electricity consumption estimation requires higher-resolution, informative night light data.

Purpose of the Study:

• To develop a novel fusion algorithm for generating high-quality color night light remote sensing images (color-NLRSIs).
• To enhance the spatial resolution and visual information of night light imagery for improved analysis.

Main Methods:

• A dual-transformation algorithm combining wavelet transform and IHS (Intensity, Hue, Saturation) color space transform.
• Merging Landsat multi-spectral bands with "NPP-VIIRS-like" night light data.
• Utilizing IHS modulation and discrete wavelet transform for component analysis and histogram matching.

Main Results:

• The dual-transformation algorithm outperforms traditional single-fusion methods in spatial resolution, detail contrast, and color fidelity.
• Fused color-NLRSIs show improved resolution from 500m to 40m, enabling analysis of smaller areas.
• Enhanced image quality metrics (MEAN, STD, EN, AG) indicate superior detail, texture, and clarity.

Conclusions:

• The dual-transformation algorithm provides the best overall performance for creating high-quality color night light remote sensing images.
• The fused images offer superior visualization and analytical capabilities for urban electricity estimation and ground feature analysis.