1. Hydrodynamic modeling requires precise and reliable digital terrain models (DTM) to determine the location of terrain edges for the area of river valley, i.e. river embankments.
2. Airborne laser scanning is a popular technology for acquiring data used to generate DTM, but it can be difficult to identify points that belong to relevant surfaces due to varied density and overlapping scans.
3. An algorithm employing multilayer feed-forward neural network for point classification is presented in this study, which allows for inclusion of a priori information about the expected shape of surface as well as the orientation of embankment with respect to the river flow direction.
This article provides an overview of a method for river embankment identification in airborne laser scanning point cloud data using a multilayer feed-forward neural network for point classification. The article is written in an objective manner and presents both sides of the argument fairly, providing evidence and examples to support its claims. The authors provide detailed descriptions of the process involved in identifying points that belong to relevant surfaces, as well as outlining how a neural network can be used as a classifier.
The article does not appear to have any biases or one-sided reporting, nor does it make unsupported claims or present partiality towards any particular viewpoint or opinion. All potential risks are noted and discussed throughout the article, and both sides of the argument are presented equally.
The only potential issue with this article is that it does not explore any counterarguments or missing points of consideration regarding its proposed method for river embankment identification in airborne laser scanning point cloud data using a multilayer feed-forward neural network for point classification. However, overall this article appears to be trustworthy and reliable in its presentation and discussion of its proposed method.