A low illumination target detection method based on a dynamic gradient gain allocation strategy

Zhiqiang Li, Jian Xiang, Jiawen Duan · 2024

Read A low illumination target detection method based on a dynamic gradient gain allocation strategy by Zhiqiang Li, Jian Xiang, Jiawen Duan (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11585611.txt.

Description

Li, Xiang, and Duan built this detection method in twenty twenty-four for finding objects in near-dark images. Standard detectors miss targets or fire falsely when light is low. Their model, DimNet, reworks several parts of a baseline detector at once. The aim is sharper features and cleaner boxes under poor light.

A new neck design mixes fine spatial detail with high-level meaning. A fusion module blends channel, spatial, local, and global cues together. The detection head was rebuilt and slimmed by sharing parameters. A fresh loss function focuses on target centers and downweights poor guesses. On a public dark-image dataset, DimNet beat both its baseline and the prior best method by clear margins.

Catalog record

CREATORZhiqiang Li, Jian Xiang, Jiawen Duan
YEAR2024
COLLECTIONPAPERS
FORMATTXT
FILE SIZE73 KB
VERSION1.0
LICENSECC BY 4.0
RIGHTSCC BY
SOURCEEurope PMC

Source: https://europepmc.org/articles/PMC11585611

Browse the full library · About the Tower of Babel