PMID- 36560379
OWN - NLM
STAT- MEDLINE
DCOM- 20221226
LR  - 20221226
IS  - 1424-8220 (Electronic)
IS  - 1424-8220 (Linking)
VI  - 22
IP  - 24
DP  - 2022 Dec 19
TI  - Multi-Scale Strengthened Directional Difference Algorithm Based on the Human 
      Vision System.
LID - 10.3390/s222410009 [doi]
LID - 10009
AB  - The human visual system (HVS) mechanism has been successfully introduced into the 
      field of infrared small target detection. However, most of the current detection 
      algorithms based on the mechanism of the human visual system ignore the 
      continuous direction information and are easily disturbed by highlight noise and 
      object edges. In this paper, a multi-scale strengthened directional difference 
      (MSDD) algorithm is proposed. It is mainly divided into two parts: local 
      directional intensity measure (LDIM) and local directional fluctuation measure 
      (LDFM). In LDIM, an improved window is used to suppress most edge clutter, 
      highlights, and holes and enhance true targets. In LDFM, the characteristics of 
      the target area, the background area, and the connection between the target and 
      the background are considered, which further highlights the true target signal 
      and suppresses the corner clutter. Then, the MSDD saliency map is obtained by 
      fusing the LDIM map and the LDFM map. Finally, an adaptive threshold segmentation 
      method is employed to capture true targets. The experiments show that the 
      proposed method achieves better detection performance in complex backgrounds than 
      several classical and widely used methods.
FAU - Zhang, Yuye
AU  - Zhang Y
AUID- ORCID: 0000-0002-4118-6978
AD  - Information Science and Engineering Department, Xinjiang University, Urumqi 
      830017, China.
FAU - Zheng, Ying
AU  - Zheng Y
AD  - Information Science and Engineering Department, Xinjiang University, Urumqi 
      830017, China.
FAU - Li, Xiuhong
AU  - Li X
AD  - Information Science and Engineering Department, Xinjiang University, Urumqi 
      830017, China.
LA  - eng
GR  - 2020D01C026/Xinjiang Natural Science Foundation/
PT  - Journal Article
DEP - 20221219
PL  - Switzerland
TA  - Sensors (Basel)
JT  - Sensors (Basel, Switzerland)
JID - 101204366
SB  - IM
MH  - Humans
MH  - *Algorithms
MH  - *Optical Devices
MH  - Attention
PMC - PMC9783212
OTO - NOTNLM
OT  - human visual system (HVS)
OT  - infrared image
OT  - multi-scale strengthened directional difference (MSDD)
OT  - small target detection
COIS- The authors declare no conflicts of interest.
EDAT- 2022/12/24 06:00
MHDA- 2022/12/27 06:00
PMCR- 2022/12/19
CRDT- 2022/12/23 02:04
PHST- 2022/09/16 00:00 [received]
PHST- 2022/12/08 00:00 [revised]
PHST- 2022/12/15 00:00 [accepted]
PHST- 2022/12/23 02:04 [entrez]
PHST- 2022/12/24 06:00 [pubmed]
PHST- 2022/12/27 06:00 [medline]
PHST- 2022/12/19 00:00 [pmc-release]
AID - s222410009 [pii]
AID - sensors-22-10009 [pii]
AID - 10.3390/s222410009 [doi]
PST - epublish
SO  - Sensors (Basel). 2022 Dec 19;22(24):10009. doi: 10.3390/s222410009.