diff --git a/paper/paper.md b/paper/paper.md
index 7f36f4b..f0803d0 100644
--- a/paper/paper.md
+++ b/paper/paper.md
@@ -55,10 +55,10 @@ As remote sensing data and sophisticated processing tools become increasingly av
 
 In this paper, we introduce **scikit-eo**, a specialized library with tailored analysis capabilities designed to meet the unique demands of environmental studies. From statistical methods to machine learning algorithms, **scikit-eo** assists researchers in uncovering intricate spatial patterns, relationships, and trends, while also simplifying the evaluation and calibration of generated outputs.
 
-**scikit-eo** is an open-source package built entirely in Python through Object-Oriented Programming and Structured Programming that provides a helpful variety of remote sensing tools (see \autoref{fig:workflow}), from primary and exploratory functions to more advanced methods to classify, calibrate, or fuse satellite imagery. Depending on the users' needs, **scikit-eo** can provide the basic but essential land cover characterization mapping, including the confusion matrix and the required metrics such as user's accuracy, producer's accuracy, omission and commission errors. These required metrics can be combined as a pandas ```DataFrame``` object. Furthermore, a class prediction map is a result of land cover mapping, i.e., a land cover map, which represents the output of the classification algorithm or the output of the segmentation algorithm. These two outcomes must include uncertainties with a confidence level (e.g., at $95$% or $90$%). All required metrics from the confusion matrix can be easily computed and included confidence levels with **scikit-eo** following guidance proposed by @OLOFSSON201442. Other useful tools for remote sensing analysis can be found in this package; and for more information about the full list of the supported functions, tutorials as well as how to install and execute the package within a Python setting, visit the [scikit-eo](https://yotarazona.github.io/scikit-eo/) website.
-
 ![Workflow of main functionalities of the *scikit-eo* python package as well as outputs that can be obtained by using the tools developed. \label{fig:workflow}](workflow_updated.png){ width=100% }
 
+**scikit-eo** is an open-source package built entirely in Python through Object-Oriented Programming and Structured Programming that provides a helpful variety of remote sensing tools (see \autoref{fig:workflow}), from primary and exploratory functions to more advanced methods to classify, calibrate, or fuse satellite imagery. Depending on the users' needs, **scikit-eo** can provide the basic but essential land cover characterization mapping, including the confusion matrix and the required metrics such as user's accuracy, producer's accuracy, omission and commission errors. These required metrics can be combined as a pandas ```DataFrame``` object. Furthermore, a class prediction map is a result of land cover mapping, i.e., a land cover map, which represents the output of the classification algorithm or the output of the segmentation algorithm. These two outcomes must include uncertainties with a confidence level (e.g., at $95$% or $90$%). All required metrics from the confusion matrix can be easily computed and included confidence levels with **scikit-eo** following guidance proposed by @OLOFSSON201442. Other useful tools for remote sensing analysis can be found in this package; and for more information about the full list of the supported functions, tutorials as well as how to install and execute the package within a Python setting, visit the [scikit-eo](https://yotarazona.github.io/scikit-eo/) website.
+
 # Audience
 
 **scikit-eo** is an adaptable Python package that covers multiple users, including students, remote sensing professionals, environmental analysis researchers, and organizations looking for satellite image analysis. Its tools and algorithms implemented make it well-suited for various applications, such as university teaching, that includes technical and practical sessions and cutting-edge research using the most recent machine learning and deep learning techniques applied to remote sensing.