Visualization comparison of feature maps for 12 channels after adding modules during the feature extraction stage in RT-DETR and SCP-DETR.

Visualization comparison of feature maps for 12 channels after adding modules during the feature extraction stage in RT-DETR and SCP-DETR.

<p>Visualization comparison of feature maps for 12 channels after adding modules during the feature extraction stage in RT-DETR and SCP-DETR.</p>

Saved in:

Bibliographic Details
Main Author:	Yuanyuan Wang (20181) (author)
Other Authors:	Tongtong Yin (14758525) (author), Xiuchuan Chen (22146211) (author), Yemeng Zhu (22146214) (author), Jibin Wang (373893) (author), Yonghao Ma (554708) (author), Luyue Liu (22146217) (author), Jiajun Wang (512021) (author)
Published:	2025
Subjects:	Neuroscience Sociology Space Science Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Chemical Sciences not elsewhere classified Information Systems not elsewhere classified storage seriously impact s3 feature layer s2 feature layer recently released yolov11m holds significant implications excessive background interference improves detection capability neck feature fusion ultimately enhancing small shaped convolution ), 4 %, representing subtle small targets specially designed co proposed method achieves original fusion module experimental results demonstrate div >< p target detection performance pcb defect detection method improves map results demonstrate proposed method fusion module defect targets detection accuracy small size pcb manufacturing efficient small depth convolution 3 %, proposed cspokm kernel module use spdconv superior performance paper proposes local scales limited due industrial production electronic components downsampling operations defects generated csp omni computational complexity better accommodates 6 %.
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_baqm4rtuqntsisq2fefjb64kvc