H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network
Ship detection from high-resolution optical satellite images is still an important task that deserves optimal solutions. This paper introduces a novel high-resolution image network-based approach based on the preselection of a region of interest (RoI). This pre-selected network first identifies and...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/24/4192 |
| _version_ | 1797543926358343680 |
|---|---|
| author | Gang Tang Shibo Liu Iwao Fujino Christophe Claramunt Yide Wang Shaoyang Men |
| author_facet | Gang Tang Shibo Liu Iwao Fujino Christophe Claramunt Yide Wang Shaoyang Men |
| author_sort | Gang Tang |
| collection | DOAJ |
| description | Ship detection from high-resolution optical satellite images is still an important task that deserves optimal solutions. This paper introduces a novel high-resolution image network-based approach based on the preselection of a region of interest (RoI). This pre-selected network first identifies and extracts a region of interest from input images. In order to efficiently match ship candidates, the principle of our approach is to distinguish suspected areas from the images based on hue, saturation, value (HSV) differences between ships and the background. The whole approach is the basis of an experiment with a large ship dataset, consisting of Google Earth images and HRSC2016 datasets. The experiment shows that the H-YOLO network, which uses the same weight training from a set of remote sensing images, has a 19.01% higher recognition rate and a 16.19% higher accuracy than applying the you only look once (YOLO) network alone. After image preprocessing, the value of the intersection over union (IoU) is also greatly improved. |
| first_indexed | 2024-03-10T13:52:23Z |
| format | Article |
| id | doaj.art-87d19b028f6e439c90f417b01609c1ac |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T13:52:23Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-87d19b028f6e439c90f417b01609c1ac2023-11-21T01:58:20ZengMDPI AGRemote Sensing2072-42922020-12-011224419210.3390/rs12244192H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected NetworkGang Tang0Shibo Liu1Iwao Fujino2Christophe Claramunt3Yide Wang4Shaoyang Men5Logistics Engineering College, Shanghai Maritime University, Shanghai 201306, ChinaLogistics Engineering College, Shanghai Maritime University, Shanghai 201306, ChinaSchool of Information and Telecommunication Engineering, Tokai University, Tokyo 1088619, JapanNaval Academy Research Institute, F-29240 Lanvéoc, FranceInstitut d’Électronique et des Technologies du numérique (IETR), UMR CNTS 6164, Polytech Nantes-Site de la Chantrerie, 44306 Nantes, FranceSchool of Medical Information Engineering, Guangzhou University of Chinese Medicine, Guangzhou 510006, ChinaShip detection from high-resolution optical satellite images is still an important task that deserves optimal solutions. This paper introduces a novel high-resolution image network-based approach based on the preselection of a region of interest (RoI). This pre-selected network first identifies and extracts a region of interest from input images. In order to efficiently match ship candidates, the principle of our approach is to distinguish suspected areas from the images based on hue, saturation, value (HSV) differences between ships and the background. The whole approach is the basis of an experiment with a large ship dataset, consisting of Google Earth images and HRSC2016 datasets. The experiment shows that the H-YOLO network, which uses the same weight training from a set of remote sensing images, has a 19.01% higher recognition rate and a 16.19% higher accuracy than applying the you only look once (YOLO) network alone. After image preprocessing, the value of the intersection over union (IoU) is also greatly improved.https://www.mdpi.com/2072-4292/12/24/4192ship detectionYOLOv3remote sensing |
| spellingShingle | Gang Tang Shibo Liu Iwao Fujino Christophe Claramunt Yide Wang Shaoyang Men H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network Remote Sensing ship detection YOLOv3 remote sensing |
| title | H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network |
| title_full | H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network |
| title_fullStr | H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network |
| title_full_unstemmed | H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network |
| title_short | H-YOLO: A Single-Shot Ship Detection Approach Based on Region of Interest Preselected Network |
| title_sort | h yolo a single shot ship detection approach based on region of interest preselected network |
| topic | ship detection YOLOv3 remote sensing |
| url | https://www.mdpi.com/2072-4292/12/24/4192 |
| work_keys_str_mv | AT gangtang hyoloasingleshotshipdetectionapproachbasedonregionofinterestpreselectednetwork AT shiboliu hyoloasingleshotshipdetectionapproachbasedonregionofinterestpreselectednetwork AT iwaofujino hyoloasingleshotshipdetectionapproachbasedonregionofinterestpreselectednetwork AT christopheclaramunt hyoloasingleshotshipdetectionapproachbasedonregionofinterestpreselectednetwork AT yidewang hyoloasingleshotshipdetectionapproachbasedonregionofinterestpreselectednetwork AT shaoyangmen hyoloasingleshotshipdetectionapproachbasedonregionofinterestpreselectednetwork |