# Labelling

To train our CNN we need labelled data of forest paths. Most data online however 
is meant for autnomous driving, usually on solid (asphalt/compacted gravel)
roads/paths.

## Freiburg

One (near) exception to this is the 
[Freiburg deepscene forest dataset](http://deepscene.cs.uni-freiburg.de/segment_random/FOREST_RGB). While this dataset still mostly contains 
compacted gravel park roads, it also contains a lot of smaller soft paths.

As such we adapted and used it for the training of our network.

See also [Dataprep > Datasets > Freiburg](../datasets#freiburg).

## Giusti

Another excellent dataset of forest paths is the data set from Giusti et al.'s [A Machine Learning Approach to Visual Perception of Forest Trails for Mobile Robots](https://people.idsia.ch/~guzzi/DataSet.html).

This data however is unfortunately only labelled as "left", "straight" or "center". 

See also [Dataprep > Datasets > Giusti](../datasets#giusti).

## Own data

We also collected our own data in a forest in flanders. See
 [Dataprep > Datasets > Steentjes](../datasets#steentjes) for more info.

## Crowd Sourced Labelling

To label the Giusti and our own data we employed the help of our 
fellow students in a crowd sourced labelling effort.

We set up a custom version of 
[Django Labeller](https://github.com/Britefury/django-labeller) 
on a Google Cloud VPS and used a 
[collaborative spreadsheet](https://docs.google.com/spreadsheets/d/1F_qDVuE1kOOzWPlyxjIwmBufMYoukAm_Qs4766otick) 
and a simple 
[instructional document](https://hackmd.io/eotmVGgfR6CO1zPifdQnyw)
to coordinate the labelling.

![Our Django Labeller instance](media/labeller.png)
*Our Django Labeller instance*

<br>

![Instructional document](media/document.png)
*Instructional document to coordinate labelling*