{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/hydra-an-ensemble-of-convolutional-neural","title":"Hydra: an Ensemble of Convolutional Neural Networks for Geospatial Land Classification","arxiv_id":"1802.03518","date":"2018-02-10","proceeding":null,"authors":["Rodrigo Minetto","Mauricio Pamplona Segundo","Sudeep Sarkar"],"abstract":"We describe in this paper Hydra, an ensemble of convolutional neural networks\n(CNN) for geospatial land classification. The idea behind Hydra is to create an\ninitial CNN that is coarsely optimized but provides a good starting pointing\nfor further optimization, which will serve as the Hydra's body. Then, the\nobtained weights are fine-tuned multiple times with different augmentation\ntechniques, crop styles, and classes weights to form an ensemble of CNNs that\nrepresent the Hydra's heads. By doing so, we prompt convergence to different\nendpoints, which is a desirable aspect for ensembles. With this framework, we\nwere able to reduce the training time while maintaining the classification\nperformance of the ensemble. We created ensembles for our experiments using two\nstate-of-the-art CNN architectures, ResNet and DenseNet. We have demonstrated\nthe application of our Hydra framework in two datasets, FMOW and NWPU-RESISC45,\nachieving results comparable to the state-of-the-art for the former and the\nbest reported performance so far for the latter. Code and CNN models are\navailable at https://github.com/maups/hydra-fmow","url_abs":"http://arxiv.org/abs/1802.03518v2","url_pdf":"http://arxiv.org/pdf/1802.03518v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"hydra-an-ensemble-of-convolutional-neural","repo_url":"https://github.com/maups/hydra-fmow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"concatenated-skip-connection","method_name":"Concatenated Skip Connection"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-block","method_name":"Dense Block"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"kaiming-initialization","method_name":"Kaiming Initialization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1802.03518","atlas_url":"https://app.syntology.ai/?focus=1802.03518","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}