Improved YOLOv7 pronucleus and cleavage sphere detection method
The invention belongs to the technical field of artificial intelligence, and particularly relates to an improved YOLOv7 pronucleus and blastosphere detection method, which comprises a YOLOv7 network model, and is characterized in that the YOLOv7 network model comprises a network trunk part, a networ...
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 belongs to the technical field of artificial intelligence, and particularly relates to an improved YOLOv7 pronucleus and blastosphere detection method, which comprises a YOLOv7 network model, and is characterized in that the YOLOv7 network model comprises a network trunk part, a network neck part and a network detection head; partial convolution, a plug-and-play space attention module and a WI oU v3 loss function are adopted to further improve the recognition and positioning of the network framework on the blastocyte and the pronucleus in the embryo image; the method has a big data basis; according to the method, a large number of embryo images are collected, multiple embryo scholars are organized to complete data set labeling, and AI model training is completed based on labeled data; the AI model for cleavage and pronucleus positioning prediction is an end-to-end model, cleavage and pronucleus positioning prediction can be automatically completed, and manual intervention is not needed in the pr |
---|