Functional Description

This tool is used for object extraction of remote sensing image service data, that is, deep learning algorithm is used for image object extraction, that is, object contours are identified at the pixel level.

The task of object detection is to classify a single object and locate it with the minimum external matrix. Semantic segmentation only classifies each pixel and does not distinguish individual objects. Object extraction (instance segmentation) requires correct identification of all objects in the image and simultaneous segmentation of different object instances, combining object detection and semantic segmentation techniques in computer vision tasks.

It has been widely used in image analysis, and is often used to solve segmentation and extraction problems in object-intensive scenes, such as urban building segmentation, stadium segmentation, port ship segmentation, etc.

Supported Data Sources

• iPortal Service Resource

Supported Formats: Image Service

Functional Entrance

Two methods are available to execute image object extraction analysis.

Direct Command Input in Dialog Box

• Locate the Dialogue panel on the left side of the main interface.

Select Command via Instruction Center

• Navigate to Instruction-> Remote Sensing Analysis Tools on the left panel of the main interface.

Operation Steps

Direct Command Input in Dialog Box

Enter the image object extraction analysis command in the dialog box, then click the Execute button.

For example:"Search for service resources named "sensing," explore and utilize the image dataset labeled "object-extract-4326," conduct image object extraction analysis, and visualize the analysis results on a map within original resource as base image.".

Select Command via Instruction Center

In the Remote Sensing Analysis Tools section, click the Image Object Extraction Analysis command box. The predefined command content will be automatically added to the dialog box. Click the Execute button to proceed.