The segmentation effect of different algorithms on building objects and vegetation (the photos were taken by the author and there are no copyright issues).

The segmentation effect of different algorithms on building objects and vegetation (the photos were taken by the author and there are no copyright issues).

<p>The segmentation effect of different algorithms on building objects and vegetation (the photos were taken by the author and there are no copyright issues).</p>

Saved in:

Bibliographic Details
Main Author:	Hong Li (20183) (author)
Other Authors:	Norriza Hussin (19797176) (author), Dandan He (49161) (author), Zexun Geng (19797179) (author), Shengpu Li (19797182) (author)
Published:	2024
Subjects:	Biochemistry Space Science Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Information Systems not elsewhere classified made great progress isprs vaihingen dataset deep learning technology convolutional neural networks 44 %, 92 12 %, 94 scale target images ordinary scene images main bird images system model ran 41 %, respectively 24 %, respectively traditional image segmentation traditional model images 1 image segmentation model completed segmentation accuracy xlink "> three models still unable spn method spatial continuity significantly improved significantly better results showed results show residual connections residual connection practical applications overall effect otsu method net method information loss feature fusion f1 value existing technologies complex scenes certain degree background complexity average accuracy 84th iterations
Tags:	Add Tag No Tags, Be the first to tag this record!

Be the first to leave a comment!

Cannot write session to /tmp/vufind_sessions/sess_7r4569scjv0b9th7sh1osgd7mi