Transfer Learning Models and Datasets for a Reliable Emergency Landing Field Identification

The file data.tar.gz compromises three HDF5 datasets. This file has been split into 100 files. The files can be merged, decompressed and unpacked with the following commands:

cat data* > data.tar.gz
tar -xzf data.tar.gz

Afterwards, the three files: train_test_data_ss8_supervised_new.hdf5, train_test_data_ss16_supervised_new.hdf5, train_test_data_ss32_supervised_new.hdf5 are ready to get processed. 

Internal structure of the datasets:
Search Window (SW) 8 m^2:
HDF5 "train_test_data_ss8_supervised_new.hdf5" {
GROUP "/" {
   GROUP "test" {
      DATASET "fus_data" {
         DATATYPE  H5T_IEEE_F32LE
         DATASPACE  SIMPLE { ( 76288, 40, 40, 8 ) / ( 76382, 40, 40, 8 ) }
      }
      DATASET "labels" {
         DATATYPE  H5T_STD_I64LE
         DATASPACE  SIMPLE { ( 76382, 1 ) / ( 76382, 1 ) }
      }
   }
   GROUP "train" {
      DATASET "fus_data" {
         DATATYPE  H5T_IEEE_F32LE
         DATASPACE  SIMPLE { ( 380928, 40, 40, 8 ) / ( 380998, 40, 40, 8 ) }
      }
      DATASET "labels" {
         DATATYPE  H5T_STD_I64LE
         DATASPACE  SIMPLE { ( 380998, 1 ) / ( 380998, 1 ) }
      }
   }
}}

SW 16 m^2:
HDF5 "train_test_data_ss16_supervised_new.hdf5" {
GROUP "/" {
   GROUP "test" {
      DATASET "fus_data" {
         DATATYPE  H5T_IEEE_F32LE
         DATASPACE  SIMPLE { ( 17024, 80, 80, 8 ) / ( 17054, 80, 80, 8 ) }
      }
      DATASET "labels" {
         DATATYPE  H5T_STD_I64LE
         DATASPACE  SIMPLE { ( 17054, 1 ) / ( 17054, 1 ) }
      }
   }
   GROUP "train" {
      DATASET "fus_data" {
         DATATYPE  H5T_IEEE_F32LE
         DATASPACE  SIMPLE { ( 84992, 80, 80, 8 ) / ( 85068, 80, 80, 8 ) }
      }
      DATASET "labels" {
         DATATYPE  H5T_STD_I64LE
         DATASPACE  SIMPLE { ( 85068, 1 ) / ( 85068, 1 ) }
      }
   }
}}

SW 32 m^2:
HDF5 "train_test_data_ss32_supervised_new.hdf5" {
GROUP "/" {
   GROUP "test" {
      DATASET "fus_data" {
         DATATYPE  H5T_IEEE_F32LE
         DATASPACE  SIMPLE { ( 3328, 160, 160, 8 ) / ( 3359, 160, 160, 8 ) }
      }
      DATASET "labels" {
         DATATYPE  H5T_STD_I64LE
         DATASPACE  SIMPLE { ( 3359, 1 ) / ( 3359, 1 ) }
      }
   }
   GROUP "train" {
      DATASET "fus_data" {
         DATATYPE  H5T_IEEE_F32LE
         DATASPACE  SIMPLE { ( 16768, 160, 160, 8 ) / ( 16793, 160, 160, 8 ) }
      }
      DATASET "labels" {
         DATATYPE  H5T_STD_I64LE
         DATASPACE  SIMPLE { ( 16793, 1 ) / ( 16793, 1 ) }
      }
   }
}}

The sample count of the various generated dataset is as follows: 
SW 8 m^2: {train: 380,928 with {0: 190,464, 1: 190,464}, test: 76,288 with {0: 38,152, 1: 38,136}}
SW 16 m^2: {train: 84,992 with {0: 42,498, 1: 42,494}, test: 17,024 with {0: 8,516, 1: 8,508}}
SW 32 m^2: {train: 16,768 with {0: 8,424, 1: 8,344}, test: 3,328 with {0: 1,672, 1: 1,656}}

Each sample is composed as follows:
RGB = sample[:,:,:3]; Theoretically: [0, 1] per color channel
NIR = sample[:,:,3]; Theoretically: [0, 1]
Slope = sample[:,:,4]; Theoretically: [0, 90]
Roughness = sample[:,:,5]; Theoretically: [0, 78.78]
NDVI = sample[:,:,6]; Theoretically: [-1, 1]
DOM = sample[:,:,7]; Theoretically: [0, 429.90]

====================================================================================================

The following three files compromise the model and optimizer state variable of our PyTorch models trained on the aforementioned datasets: best_alexnet_final.pth, best_resnet18_final.pth, best_wide_resnet50_2_final.pth

Below find a more precise description of each model:
best_resnet18_final.pth

Model: ResNet-18

Dataset: SW 8

Input: RGB-NIR-Slope -> R: [0,224,224], G: [1,224,224], B: [2,224,224], NIR: [3,224,224], Slope: [4,224,224]

best_wide_resnet50_2_final.pth

Model: Wide-ResNet-50-2

Dataset: SW 16

Input: NDVI-Slope -> NDVI: [0,224,224], Slope: [1,224,224]

best_alexnet_final.pth

Model: AlexNet

Dataset: SW 32

Input: RGB-Slope -> R: [0,224,224], G: [1,224,224], B: [2,224,224], Slope: [3,224,224]

Each model is capable of performing a binary classification, distinguishing between landable and unlandable samples