Title:
MULTI-TARGET CONSTANT FALSE ALARM RATE DETECTION METHOD BASED ON DEEP NEURAL NETWORK
Document Type and Number:
WIPO Patent Application WO/2023/284698
Kind Code:
A1
Abstract:
A multi-target constant false alarm rate detection method based on a deep neural network. In the method, a simulated data set, which uses a data enhancement technique, is established, so as to train a pre-detector based on a deep neural network, and a peak value of a radar signal is classified, so as to distinguish whether a target or a clutter is present. A target that is detected by the pre-detector is removed from an original background sample, such that a reduced sample is formed. Background level estimation is performed on the basis of the reduced sample and by using an approximate maximum likelihood estimator based on a Taylor series, such that a false alarm adjustment threshold is obtained, and a target in a pre-detection result that is lower than the threshold is removed, such that a final detection result is output. By means of the method, a target is detected without the need to rely on a pre-estimated background level, such that an excellent detection performance can still be maintained in a scenario where the density of targets is high.
Inventors:
SONG CHUNYI (CN)
CAO ZHIHUI (CN)
SONG YUYING (CN)
AI FUYUAN (CN)
WU JINGXUAN (CN)
XU ZHIWEI (CN)
CAO ZHIHUI (CN)
SONG YUYING (CN)
AI FUYUAN (CN)
WU JINGXUAN (CN)
XU ZHIWEI (CN)
Application Number:
PCT/CN2022/105025
Publication Date:
January 19, 2023
Filing Date:
July 12, 2022
Export Citation:
Assignee:
UNIV ZHEJIANG (CN)
International Classes:
G01S13/02; G01S7/41; G06N3/04
Foreign References:
| CN113534120A | 2021-10-22 | |||
| CN112684428A | 2021-04-20 | |||
| CN111562569A | 2020-08-21 | |||
| CN109188388A | 2019-01-11 | |||
| CN107239740A | 2017-10-10 | |||
| CN104732243A | 2015-06-24 | |||
| CN108921029A | 2018-11-30 | |||
| CN108921030A | 2018-11-30 | |||
| CN112163450A | 2021-01-01 | |||
| CN106228124A | 2016-12-14 |
Other References:
CAO ZHIHUI; FANG WENWEI; SONG YUYING; HE LAI; SONG CHUNYI; XU ZHIWEI: "DNN-Based Peak Sequence Classification CFAR Detection Algorithm for High-Resolution FMCW Radar", IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, vol. 60, 24 September 2021 (2021-09-24), USA, pages 1 - 15, XP011899561, ISSN: 0196-2892, DOI: 10.1109/TGRS.2021.3113302
Attorney, Agent or Firm:
HANGZHOU AOCHUANG INTELLECTUAL PROPERTY AGENCY CO., LTD (CN)
Download PDF: