Return
Learning Gabor layer for edge detection network
DOI:10.1016/j.dsp.2025.105438.png)
Abstract
En 中文
• Propose novel LGLNet model for high-accuracy edge detection with fewer parameters and lower computational complexity. • Design Learning Gabor Layer (LGL) with 150 params. It generates position-specific filters for efficient extraction of fine edge features. • Use pixel-difference gradients to dynamically compute Gabor angles per spatial location in real-time.
Journal
IF:
3.6
Papers:
9.9K
Citations:
1.7W
Organization
No organization information available

