Skip To Content

An overview of the Deep Learning toolset

Mit der Image Analyst-Lizenz verfügbar.

The Deep Learning toolset contains tools to detect specific features in an image or to classify pixels in a raster dataset.

Deep learning is a type of machine learning artificial intelligence that detects features in imagery using multiple layers in neural networks where each layer is capable of extracting one or more unique features in the image. The tools in the Deep Learning toolset take advantage of GPU processing to perform analysis in a timely manner.

These ArcGIS Pro tools consume the models that have been trained to detect specific features in third-party deep learning frameworks—such as TensorFlow, CNTK and PyTorch—and output features or class maps.

The following table lists the available deep learning tools and provides a brief description of each:

ToolDescription

Classify Pixels Using Deep Learning

Führt ein trainiertes Deep-Learning-Modell auf einem Eingabe-Raster zur Erstellung eines klassifizierten Rasters aus, wobei jedem gültigen Pixel eine Klassenbeschriftung zugewiesen ist.

Detect Objects Using Deep Learning

Führt ein trainiertes Deep-Learning-Modell auf einem Eingabe-Raster zur Erstellung einer Feature-Class aus, die die gefundenen Objekte enthält. Bei den Features kann es sich um Rahmen oder Polygone um die gefundenen Objekte oder um Punkte im Mittelpunkt der Objekte handeln.

Non Maximum Suppression

Identifies duplicate features from the output of the Detect Objects Using Deep Learning tool as a postprocessing step and creates a new output with no duplicate features. The Detect Objects Using Deep Learning tool can return more than one bounding box or polygon for the same object, especially as a tiling side effect. If two features overlap more than a given maximum ratio, the feature with the lower confidence value will be removed.

Tools of the Deep Learning toolset

Verwandte Themen