This repository contains the full source code of the paper:
Trong-An Bui, Pei-Jun Lee, Chun-Sheng Liang, Pei-Hsiang Hsu, Shiuan-Hal Shiu and Chen-Kai Tsai, "Edge-Computing-Enabled Deep Learning Approach for Low-Light Satellite Image Enhancement," in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 17, pp. 4071-4083, 2024, doi: 10.1109/JSTARS.2024.3357093.
Abstract: Edge computing enables rapid data processing and decision-making on satellite payloads. Deploying deep learning-based techniques for low-light image enhancement improves early detection and tracking accuracy on satellite platforms, but it faces challenges due to limited computational resources. This article proposes an edge-computing-enabled inference model specifically designed onboard satellites. The proposed model follows an encoder–decoder architecture to generate the illumination map with low multiplication matrix complexity, 25.52 GMac of
keywords: {Satellites;Computational modeling;Image edge detection;Image enhancement;Edge computing;Payloads;Deep learning;Deep learning;edge computing;edge-computing-enabled;image enhancement;low light satellite images;onboard}, URL: https://ieeexplore.ieee.org/abstract/document/10412123
Step 1: Training Proposed Lightweight Model lowlight_train.py
Step 2: Compress the training weight Post-training quantization compress_trained_weights.py
Step 3: Inference with the Proposed Model, including Quantized Convolution and Piece-Wise sigmoid function main.cpp
Experiment Results
Please reference this paper in your manuscript when you use this source code.