Hydra: an Ensemble of Convolutional Neural Networks for Geospatial Land Classification

Abstract

We describe in this paper Hydra, an ensemble of convolutional neural networks (CNN) for geospatial land classification. The idea behind Hydra is to create an initial CNN that is coarsely optimized but provides a good starting pointing for further optimization, which will serve as the Hydra's body. Then, the obtained weights are fine-tuned multiple times with different augmentation techniques, crop styles, and classes weights to form an ensemble of CNNs that represent the Hydra's heads. By doing so, we prompt convergence to different endpoints, which is a desirable aspect for ensembles. With this framework, we were able to reduce the training time while maintaining the classification performance of the ensemble. We created ensembles for our experiments using two state-of-the-art CNN architectures, ResNet and DenseNet. We have demonstrated the application of our Hydra framework in two datasets, FMOW and NWPU-RESISC45, achieving results comparable to the state-of-the-art for the former and the best reported performance so far for the latter. Code and CNN models are available at https://github.com/maups/hydra-fmow

arXiv comment12 pages, 14 figures, 5 tables

References in corpus (13)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Remote Sensing Image Scene Classification: Benchmark and State of the Art
• The Effectiveness of Data Augmentation in Image Classification using Deep Learning
• AID: A Benchmark Dataset for Performance Evaluation of Aerial Scene Classification
• Towards Better Exploiting Convolutional Neural Networks for Remote Sensing Scene Classification
• Rethinking the Inception Architecture for Computer Vision
• Trunk-Branch Ensemble Convolutional Neural Networks for Video-based Face Recognition
• Some Improvements on Deep Convolutional Neural Network Based Image Classification
• Xception: Deep Learning with Depthwise Separable Convolutions
• Learning to Compose Domain-Specific Transformations for Data Augmentation

Cited by in corpus (12)

• Remote Sensing Image Scene Classification Meets Deep Learning: Challenges, Methods, Benchmarks, and Opportunities
• Active Fire Detection in Landsat-8 Imagery: a Large-Scale Dataset and a Deep-Learning Study
• Ensembles of Convolutional Neural Networks models for pediatric pneumonia diagnosis
• MGML: Multi-Granularity Multi-Level Feature Ensemble Network for Remote Sensing Scene Classification
• Vehicle-Rear: A New Dataset to Explore Feature Fusion for Vehicle Identification Using Convolutional Neural Networks
• Measuring Human and Economic Activity from Satellite Imagery to Support City-Scale Decision-Making during COVID-19 Pandemic
• Embedded Self-Distillation in Compact Multi-Branch Ensemble Network for Remote Sensing Scene Classification
• FabricNet: A Fiber Recognition Architecture Using Ensemble ConvNets
• Anytime Inference with Distilled Hierarchical Neural Ensembles
• Remote Sensor Design for Visual Recognition with Convolutional Neural Networks
• Measuring economic activity from space: a case study using flying airplanes and COVID-19
• Towards a Universal Gating Network for Mixtures of Experts