ca-fire-detector's Introduction

Wildfire Detection enabled camera using the NVIDIA Jetson TX2

Video in https://youtu.be/uWGVg6NTW6g

The 2017 California wildfire season was the most destructive wildfire season on record.

Early detection of ignition can result in faster response by fire agencies therefore minimizing destruction. In 2017 San Diego, CA residents were invited to watch strategically installed cameras and report fires: "Public can use webcams to watch for wildfires across San Diego County". This is a task that could potentially be automated.

NVIDIA Jetson TX2 is an ideal platform for this kind of applications.

ca-fire-detector's People

Contributors

Stargazers

Watchers

ca-fire-detector's Issues

Conversion to uff fails with dropout layers

if you add a Dropout layer fails to create uff

x = base_model.output
x = GlobalAveragePooling2D()(x)
x = Dropout(0.45)(x)
predictions = Dense(num_classes, activation='softmax')(x)

Error when training the model

Hi, I follow the fire-detection-jetson-save-ca.ipynb, and when I run to the line:

hist = model.fit(train_tensors, train_targets, batch_size=64, epochs=10,
      validation_data=(valid_tensors, valid_targets), callbacks=[checkpointer, tbCallback],
      verbose=2)  #, shuffle=True)

It reports the error as follows:

Train on 535 samples, validate on 87 samples
Epoch 1/10
 - 18s - loss: 2.6562 - acc: 0.8299 - val_loss: 2.7790 - val_acc: 0.8276

Epoch 00001: val_loss improved from inf to 2.77898, saving model to firemodel.weights.best.hdf5
Traceback (most recent call last):
  File "keras_fire.py", line 103, in <module>
    verbose=2)
  File "/usr/lib64/python2.7/site-packages/keras/engine/training.py", line 1042, in fit
    validation_steps=validation_steps)
  File "/usr/lib64/python2.7/site-packages/keras/engine/training_arrays.py", line 219, in fit_loop
    callbacks.on_epoch_end(epoch, epoch_logs)
  File "/usr/lib64/python2.7/site-packages/keras/callbacks.py", line 77, in on_epoch_end
    callback.on_epoch_end(epoch, logs)
  File "/usr/lib64/python2.7/site-packages/keras/callbacks.py", line 444, in on_epoch_end
    self.model.save(filepath, overwrite=True)
  File "/usr/lib64/python2.7/site-packages/keras/engine/network.py", line 1104, in save
    save_model(self, filepath, overwrite, include_optimizer)
  File "/usr/lib64/python2.7/site-packages/keras/engine/saving.py", line 122, in save_model
    save_weights_to_hdf5_group(model_weights_group, model_layers)
  File "/usr/lib64/python2.7/site-packages/keras/engine/saving.py", line 457, in save_weights_to_hdf5_group
    save_attributes_to_hdf5_group(g, 'weight_names', weight_names)
  File "/usr/lib64/python2.7/site-packages/keras/engine/saving.py", line 409, in save_attributes_to_hdf5_group
    group.attrs[name] = data
  File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
  File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
  File "/usr/lib64/python2.7/site-packages/h5py/_hl/attrs.py", line 95, in __setitem__
    self.create(name, data=value, dtype=base.guess_dtype(value))
  File "/usr/lib64/python2.7/site-packages/h5py/_hl/attrs.py", line 180, in create
    space = h5s.create_simple(shape)
  File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper
  File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper
  File "h5py/h5s.pyx", line 101, in h5py.h5s.create_simple
ValueError: Zero sized dimension for non-unlimited dimension (zero sized dimension for non-unlimited dimension)

Could you help me with this error and give me some suggestions? Many thanks!

Recommend Projects