Doodleverse/Segmentation Zoo/Seg2Map Res-UNet models for segmentation of AAAI/flooded buildings in RGB 1024x1024 high-res. images

<em><strong>Doodleverse/Segmentation Zoo/Seg2Map Res-UNet models for segmentation of AAAI/flooded buildings in RGB 1024x1024 high-res. images</strong></em> Models have been created using Segmentation Gym* using the following dataset**: https://github.com/FrontierDevelopmentLab/multi3net These Residual-UNet model data are based on 1m spatial footprint images and associated labels of flooded buildings in Houston. Imagery made available through DigitalGlobe*** Image size used by model: 1024 x 1024 x 3 pixels classes:<br> other<br> flooded building <br> File descriptions For each model, there are 5 files with the same root name: 1. '.json' config file: this is the file that was used by Segmentation Gym* to create the weights file. It contains instructions for how to make the model and the data it used, as well as instructions for how to use the model for prediction. It is a handy wee thing and mastering it means mastering the entire Doodleverse. 2. '.h5' weights file: this is the file that was created by the Segmentation Gym* function `train_model.py`. It contains the trained model's parameter weights. It can called by the Segmentation Gym* function `seg_images_in_folder.py`. Models may be ensembled. 3. '_modelcard.json' model card file: this is a json file containing fields that collectively describe the model origins, training choices, and dataset that the model is based upon. There is some redundancy between this file and the `config` file (described above) that contains the instructions for the model training and implementation. The model card file is not used by the program but is important metadata so it is important to keep with the other files that collectively make the model and is such is considered part of the model 4. '_model_history.npz' model training history file: this numpy archive file contains numpy arrays describing the training and validation losses and metrics. It is created by the Segmentation Gym function `train_model.py` 5. '.png' model training loss and mean IoU plot: this png file contains plots of training and validation losses and mean IoU scores during model training. A subset of data inside the .npz file. It is created by the Segmentation Gym function `train_model.py` Additionally, BEST_MODEL.txt contains the name of the model with the best validation loss and mean IoU References<br> *Segmentation Gym: Buscombe, D., &amp; Goldstein, E. B. (2022). A reproducible and reusable pipeline for segmentation of geoscientific imagery. Earth and Space Science, 9, e2022EA002332. https://doi.org/10.1029/2022EA002332 See: https://github.com/Doodleverse/segmentation_gym **Rudner, T. G. J.; Rußwurm, M.; Fil, J.; Pelich, R.; Bischke, B.; Kopačková, V.; Biliński, P. Segmenting Flooded Buildings via Fusion of Multiresolution, Multisensor, and Multitemporal Satellite Imagery. In AAAI 2019. https://arxiv.org/pdf/1812.01756.pdf ***DigitalGlobe. 2018. DigitalGlobe Open Data Program. https://www.digitalglobe.com/opendata. Online; accessed 2018-09-01.### 模型说明 本仓库包含**Doodleverse/Segmentation Zoo/Seg2Map Res-UNet模型**，用于对美国人工智能协会（AAAI）竞赛中的洪涝建筑（flooded buildings）进行1024×1024高分辨率RGB图像的语义分割。 本系列模型通过分割训练工具包（Segmentation Gym）构建，所用数据集链接为：https://github.com/FrontierDevelopmentLab/multi3net。 本批残差UNet（Res-UNet）模型基于休斯顿地区1米空间分辨率的遥感影像及对应的洪涝建筑标注标签构建。所用影像由数字地球（DigitalGlobe）开放数据项目提供。模型输入图像尺寸为1024×1024×3像素。 模型分类类别包括： 1. 其他（other） 2. 洪涝建筑（flooded building） ### 文件说明 针对每个模型，均包含5个同名根文件，具体如下： 1. `.json`配置文件：该文件为分割训练工具包（Segmentation Gym）生成权重文件时所用的配置文件，其中包含模型构建、所用数据集的相关参数，以及模型推理的操作指南。该文件是核心配置文件，掌握该文件即可完整使用整个Doodleverse工具链。 2. `.h5`权重文件：该文件由分割训练工具包（Segmentation Gym）的`train_model.py`脚本生成，存储了训练完成的模型参数权重，可通过分割训练工具包的`seg_images_in_folder.py`脚本调用。支持多模型集成。 3. `_modelcard.json`模型卡片文件：该JSON文件包含描述模型起源、训练配置与所用数据集的元数据字段。该文件与上述配置文件存在部分冗余内容，但该文件不被程序调用，仅作为重要元数据留存，需与其他模型文件一并保存，属于模型套件的组成部分。 4. `_model_history.npz`模型训练历史文件：该NumPy归档文件存储了训练与验证集的损失函数值及评估指标数组，由分割训练工具包的`train_model.py`脚本生成。 5. `.png`训练可视化文件：该PNG图像展示了模型训练过程中的训练与验证损失曲线、平均交并比（mean IoU）得分曲线，为上述`.npz`文件中数据的可视化子集，由分割训练工具包的`train_model.py`脚本生成。 此外，`BEST_MODEL.txt`文件存储了在验证集上取得最优损失与平均交并比得分的模型名称。 ### 参考文献 * 分割训练工具包（Segmentation Gym）：Buscombe, D., & Goldstein, E. B. (2022). 可复现且可复用的地球科学影像语义分割流程. *Earth and Space Science*, 9, e2022EA002332. https://doi.org/10.1029/2022EA002332 相关仓库：https://github.com/Doodleverse/segmentation_gym ** Rudner, T. G. J.; Rußwurm, M.; Fil, J.; Pelich, R.; Bischke, B.; Kopačková, V.; Biliński, P. 基于多分辨率、多传感器、多时相卫星影像融合的洪涝建筑分割方法. 发表于美国人工智能协会2019年会（AAAI 2019）. https://arxiv.org/pdf/1812.01756.pdf *** 数字地球（DigitalGlobe）. 2018. 数字地球开放数据项目. https://www.digitalglobe.com/opendata. 在线资源；访问时间：2018-09-01.