RSandAI

Automatic-Road-Extraction-from-Historical-Maps-using-Deep-Learning-Techniques

This repository contains code for the paper "Automatic Road Extraction from Historical Maps using Deep Learning Techniques: A Regional Case Study of Turkey in a German World War II map"

DL-based-road-extraction-from-historical-maps

This repository contains the code, test patches and weights for the paper [Deep Learning based road extraction from historical maps]

HRPlanes

A benchmark dataset for deep learning-based airplane detection: HRPlanes

HRPlanes-HighResolution-Planes-Benchmark-Dataset

High Resolution Planes Benchmark Dataset-HRPlanes. This repo contains weights of YOLOv4 and Faster R-CNN networks trained with HRPlanes dataset. YOLOv4 training have been performed using Darknet (https://github.com/AlexeyAB/darknet). Faster R-CNN have been trained using TensorFlow Object Detection API v1.13 (https://github.com/tensorflow/models/tree/r1.13.0).

Istanbul-Building-Dataset-Benchmark-Building-Extraction-Dataset-and-DL-Models

This repo contains weights of Unet++ model with SE-ResNeXt101 encoder trained with Istanbul, Inria and Massachusetts datasets seperately. Trainings have been realized using PyTorch and segmentation models library (https://github.com/qubvel/segmentation_models.pytorch) We also provide an inference notebook to run prediction on GeoTiff images. This notebook also outputs prediction images as GeoTiff.

LULCMapping-WV3images-CORINE-DLMethods

Land Use and Land Cover Mapping Using Deep Learning Based Segmentation Approaches and VHR Worldview-3 Images

MTL-Based-DL-Framework-for-Building-Footprint-Segmentation

This repository contains the code for the paper "A MULTI-TASK DEEP LEARNING FRAMEWORK FOR BUILDING FOOTPRINT SEGMENTATION"

PanColorGAN-VHR-Satellite-Images

Rethinking CNN-Based Pansharpening: Guided Colorization of Panchromatic Images via GANs. Pretrained Weights and GAN training parts of the code can be found in this repo.

VHRShips

This study focuses on all stages of ship classification in the optical satellite images. The proposed “Hierarchical Design (HieD)” approach, which is based on deep learning techniques, performs Detection, Localization, Recognition and Identification (DLRI) of the ships in the optical satellite images. HieD is an end-to-end approach which allows the optimization of each stage of the DLRI independently. A unique and rich ship dataset (High Resolution Ships, HRShips), which is formed by the Google Earth Pro software, is used in this study. While Xception network is used in detection, recognition and identification stages; YOLOv4 is preferred for the localization of the ships.

VHRTrees

A benchmark dataset for deep learning-based tree detection: VHRTrees

RSandAI/MTL-Based-DL-Framework-for-Building-Footprint-Segmentation

This repository contains the code for the paper "A MULTI-TASK DEEP LEARNING FRAMEWORK FOR BUILDING FOOTPRINT SEGMENTATION"

RSandAI/DL-based-road-extraction-from-historical-maps

This repository contains the code, test patches and weights for the paper [Deep Learning based road extraction from historical maps]

RSandAI/LULCMapping-WV3images-CORINE-DLMethods

Land Use and Land Cover Mapping Using Deep Learning Based Segmentation Approaches and VHR Worldview-3 Images

RSandAI/Istanbul-Building-Dataset-Benchmark-Building-Extraction-Dataset-and-DL-Models

This repo contains weights of Unet++ model with SE-ResNeXt101 encoder trained with Istanbul, Inria and Massachusetts datasets seperately. Trainings have been realized using PyTorch and segmentation models library (https://github.com/qubvel/segmentation_models.pytorch) We also provide an inference notebook to run prediction on GeoTiff images. This notebook also outputs prediction images as GeoTiff.

RSandAI/Automatic-Road-Extraction-from-Historical-Maps-using-Deep-Learning-Techniques

This repository contains code for the paper "Automatic Road Extraction from Historical Maps using Deep Learning Techniques: A Regional Case Study of Turkey in a German World War II map"

RSandAI/HRPlanes

A benchmark dataset for deep learning-based airplane detection: HRPlanes

RSandAI/PanColorGAN-VHR-Satellite-Images

Rethinking CNN-Based Pansharpening: Guided Colorization of Panchromatic Images via GANs. Pretrained Weights and GAN training parts of the code can be found in this repo.

RSandAI/.github

RSandAI/aitlas

AiTLAS implements state-of-the-art AI methods for exploratory and predictive analysis of satellite images.

RSandAI/DL-LULC-Mapping

Land Use / Land Cover Mappings Using Deep Learning Methods

RSandAI/HRPlanes-HighResolution-Planes-Benchmark-Dataset

High Resolution Planes Benchmark Dataset-HRPlanes. This repo contains weights of YOLOv4 and Faster R-CNN networks trained with HRPlanes dataset. YOLOv4 training have been performed using Darknet (https://github.com/AlexeyAB/darknet). Faster R-CNN have been trained using TensorFlow Object Detection API v1.13 (https://github.com/tensorflow/models/tree/r1.13.0).

RSandAI/sewar

All image quality metrics you need in one package.

RSandAI/VHRShips

This study focuses on all stages of ship classification in the optical satellite images. The proposed “Hierarchical Design (HieD)” approach, which is based on deep learning techniques, performs Detection, Localization, Recognition and Identification (DLRI) of the ships in the optical satellite images. HieD is an end-to-end approach which allows the optimization of each stage of the DLRI independently. A unique and rich ship dataset (High Resolution Ships, HRShips), which is formed by the Google Earth Pro software, is used in this study. While Xception network is used in detection, recognition and identification stages; YOLOv4 is preferred for the localization of the ships.

RSandAI/VHRTrees

A benchmark dataset for deep learning-based tree detection: VHRTrees

RSandAI/HexaLCSeg

A Historical Benchmark Dataset from Hexagon Satellite Images for Land Cover Segmentation

RSandAI/HistRoadSegformer

RSandAI/PanColorGAN

Rethinking CNN-Based Pansharpening: Guided Colorization of Panchromatic Images via GANs.