Image landmark identification method based on semantic segmentation
The invention provides an image landmark recognition method based on semantic segmentation. The image landmark recognition method comprises the steps of collecting a picture containing a landmark building, constructing a sample data set, training an image semantic segmentation model by using a deep...
Gespeichert in:
Hauptverfasser: | , , , , , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention provides an image landmark recognition method based on semantic segmentation. The image landmark recognition method comprises the steps of collecting a picture containing a landmark building, constructing a sample data set, training an image semantic segmentation model by using a deep learning algorithm, and obtaining mask pictures of different surface feature types; performing semantic segmentation on the sample data set by using the trained semantic segmentation model to obtain semantic segmentation maps of different types of ground features, mask foreground and background, and obtain a picture only containing building information; dividing the masked picture into training data and verification data, and training and optimizing a model structure to obtain an optimal landmark recognition deep learning model; and inputting a landmark image to be recognized, performing semantic segmentation and masking, performing landmark recognition on the masked image, and obtaining and outputting the name inf |
---|