Title:
METHOD AND SYSTEM FOR RECOGNIZING MARINE OBJECT ON BASIS OF HYPERSPECTRAL DATA
Document Type and Number:
WIPO Patent Application WO/2022/139111
Kind Code:
A1
Abstract:
Disclosed in the present invention is a method for recognizing a marine object on the basis of hyperspectral data, comprising the steps of: collecting target hyperspectral data; pre-processing the target hyperspectral data; and detecting and identifying an object included in the target hyperspectral data on the basis of a marine object detection and identification model trained through learning about detection and identification of a marine object. According to the present invention, pre-processing and processing of hyperspectral data collected in real time depending on a communication state may be performed in the air or on the ground.
Inventors:
SEO DONGMIN (KR)
OH SANGWOO (KR)
KIM TAE-SUNG (KR)
OH SANGWOO (KR)
KIM TAE-SUNG (KR)
Application Number:
PCT/KR2021/011287
Publication Date:
June 30, 2022
Filing Date:
August 24, 2021
Export Citation:
Assignee:
KOREA INSTITUTE OF OCEAN SCIENCE ＆ TECH (KR)
International Classes:
G01J3/28; G01N21/3586; G01V8/10; G06K9/62; G06N3/08; G06N20/00
Foreign References:
| KR101621354B1 | 2016-05-20 | |||
| KR20180065411A | 2018-06-18 | |||
| US20160266246A1 | 2016-09-15 | |||
| KR102270834B1 | 2021-06-30 |
Other References:
NOH SI-HYEONG: "NOH, Si-Hyeong et al. A Study on Learning Medical Image Dataset and Analysis for Deep Learning", PROCEEDINGS OF ANNUAL SPRING CONFERENCE OF KIPS 2018 (ASK 2018), vol. 25, 1 May 2018 (2018-05-01), pages 350 - 351, XP055944799
LEE JEONGHWAN: "A Method of Color Image Segmentation Based on DBSCAN(Density Based Spatial Clustering of Applications with Noise) Using Compactness of Superpixels and Texture Information", JOURNAL OF THE KOREA SOCIETY OF DIGITAL INDUSTRY AND INFORMATION MANAGEMENT, vol. 11, no. 4, 30 December 2015 (2015-12-30), pages 89 - 97, XP055944836, ISSN: 1738-6667, DOI: 10.17662/ksdim.2015.11.4.089
LEE JEONGHWAN: "A Method of Color Image Segmentation Based on DBSCAN(Density Based Spatial Clustering of Applications with Noise) Using Compactness of Superpixels and Texture Information", JOURNAL OF THE KOREA SOCIETY OF DIGITAL INDUSTRY AND INFORMATION MANAGEMENT, vol. 11, no. 4, 30 December 2015 (2015-12-30), pages 89 - 97, XP055944836, ISSN: 1738-6667, DOI: 10.17662/ksdim.2015.11.4.089
Attorney, Agent or Firm:
DAE-A INTELLECTUAL PROPERTY CONSULTING (KR)
Download PDF: