IGARSS 2019
258 papers
The 2019 IEEE International Geoscience and Remote Sensing Symposium (IGARSS) featured 288 papers focusing on advancements in multimodal data fusion and deep learning applications for Earth observation. A primary research direction involved the integration of radar and optical imagery to overcome environmental challenges, such as cloud cover and signal saturation in high-biomass regions. Methodological trends highlighted the use of parallel multi-branch architectures and temporal attention mechanisms to process heterogeneous data sequences and improve continuous satellite monitoring.
Technical contributions emphasized high-resolution image analysis, including road network extraction, landslide geohazard assessment, and ship classification using Sentinel-1 and Sentinel-2 data. Notable work explored boundary adherence in segmentation and the extraction of urban functional zones through self-adaptive scaling. Prolific contributors included Licheng Jiao, Yanfei Zhong, and Liangpei Zhang, whose research addressed spectral-spatial classification and object detection in optical remote sensing videos. These efforts collectively focused on improving the accuracy of automated feature extraction and ecosystem analysis from complex geospatial datasets.
Top Research Directions
Research directions most represented by IGARSS 2019 papers.
Top Authors
Impactful Papers
Papers from IGARSS 2019 that shaped research clusters.
Influential
Semi-Supervised Learning with Graphs: Covariance Based Superpixels For Hyperspectral Image Classific
influential
Superpixel Based Graph Reduction
A Fast and Precise Method for Large-Scale Land-Use Mapping Based on Deep Learning
influential
Distant Supervision for Scaling Emotion
BigEarthNet: A Large-Scale Benchmark Archive For Remote Sensing Image Understanding
influential
Bridging the Natural to Remote Sensing D
Landslide Geohazard Assessment with Convolutional Neural Networks Using Sentinel-2 Imagery Data
influential
Cloud Native Architectures for Planetary
Crop Phenology Classification Using A Representation Learning Network From Sentinel-1 SAR Data
influential
Mitigating Temporal Data Gaps in Optical
Landslide Geohazard Assessment with Convolutional Neural Networks Using Sentinel-2 Imagery Data
influential
3D Convolutions for Spatiotemporal Featu