Image Classification is one of the most basic tasks in computer vision, and it is also used in many scenarios in the field of GIS, such as intelligent identification of electricity meter types, photo album classification, identification of animals/plants, etc. Based on high-level visual information, SuperMap uses EfficientNet deep network model to extract the abstract semantic information of images and classify them.

Image Classification Sample Management can perform preview, classification and category management on sample images, and can divide similar images into unified categories in batches, which is convenient and fast to operate. The main operation instructions are as follows:

1. View Sample Picture: Select a picture in the Sample Library List to view it in the window. You can switch the picture by pressing the Previous and Next buttons in the tabs or the Up and Down keys on the keyboard. At the same time, you can perform browsing operations such as zooming in, zooming out, and panning.
2. Image Classification: provides two quick classification methods:
   • Context Menu Category: Select one or more pictures in the Sample Library List, and click Right button-& gt; Move-& gt; Category to move the specified picture to the specified category
   ;
   • Category index number key: In the picture preview window, press the index number corresponding to the category to quickly classify the picture into the corresponding category. For example, the picture in the Sample Picture Window belongs to Type1 and its index number is 1. At this time, press the number key 1 in the keyboard to classify the picture into Type1.



3. The sample pictures can be classified in the above way in turn. The following figure shows before and after Image Classification.

Related topics

Sample library management

Object DetectionSample Management

Export Sample Library