@Article{WagnerSSHFLMYS:2022:KtSeHa,
author = "Wagner, Fabien Hubert and Silva, Ricardo Dalagnol and S{\'a}nchez
Ipia, Alber Hamersson and Hirye, Mayumi C. M. and Favrichon,
Samuel and Lee, Jake H. and Mauceri, Steffen and Yang, Yan and
Saatchi, Sassan",
affiliation = "{Instituto Nacional de Pesquisas Espaciais (INPE)} and {Instituto
Nacional de Pesquisas Espaciais (INPE)} and {Instituto Nacional de
Pesquisas Espaciais (INPE)} and {Universidade de S{\~a}o Paulo
(USP)} and {NASA-Jet Propulsion Laboratory} and {NASA-Jet
Propulsion Laboratory} and {NASA-Jet Propulsion Laboratory} and
{NASA-Jet Propulsion Laboratory} and {NASA-Jet Propulsion
Laboratory}",
title = "K-textures, a self-supervised hard clustering deep learning
algorithm for satellite image segmentation",
journal = "Frontiers in Environmental Science",
year = "2022",
volume = "10",
pages = "e946729",
month = "Sept.",
keywords = "deep learning - artificial neural network, discrete optimization
algorithm, landcover, planetscope satellite, segmentation (image
processing), self-supervised segmentation, tensorflow (2),
tropical forest.",
abstract = "Deep learning self-supervised algorithms that can segment an image
in a fixed number of hard clusters such as the k-means algorithm
and with an end-to-end deep learning approach are still lacking.
Here, we introduce the k-textures algorithm which provides
self-supervised segmentation of a 4-band image (RGB-NIR) for a k
number of classes. An example of its application on
high-resolution Planet satellite imagery is given. Our algorithm
shows that discrete search is feasible using convolutional neural
networks (CNN) and gradient descent. The model detects k hard
clustering classes represented in the model as k discrete binary
masks and their associated k independently generated textures,
which combined are a simulation of the original image. The
similarity loss is the mean squared error between the features of
the original and the simulated image, both extracted from the
penultimate convolutional block of Keras imagenet pre-trained
VGG-16 model and a custom feature extractor made with Planet data.
The main advances of the k-textures model are: first, the k
discrete binary masks are obtained inside the model using gradient
descent. The model allows for the generation of discrete binary
masks using a novel method using a hard sigmoid activation
function. Second, it provides hard clustering classeseach pixel
has only one class. Finally, in comparison to k-means, where each
pixel is considered independently, here, contextual information is
also considered and each class is not associated only with similar
values in the color channels but with a texture. Our approach is
designed to ease the production of training samples for satellite
image segmentation and the k-textures architecture could be
adapted to support different numbers of bands and for more complex
self-segmentation tasks, such as object self-segmentation. The
model codes and weights are available at
https://doi.org/10.5281/zenodo.6359859.",
doi = "10.3389/fenvs.2022.946729",
url = "http://dx.doi.org/10.3389/fenvs.2022.946729",
issn = "2296-665X",
language = "en",
targetfile = "fenvs-10-946729.pdf",
urlaccessdate = "11 maio 2024"
}