Probabilistic Ship Detection and Classification Using Deep Learning
For an autonomous ship to navigate safely and avoid collisions with other ships, reliably detecting and classifying nearby ships under various maritime meteorological environments is essential. In this paper, a novel probabilistic ship detection and classification system based on deep learning is pr...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2018-06-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | http://www.mdpi.com/2076-3417/8/6/936 |
| _version_ | 1811339650837512192 |
|---|---|
| author | Kwanghyun Kim Sungjun Hong Baehoon Choi Euntai Kim |
| author_facet | Kwanghyun Kim Sungjun Hong Baehoon Choi Euntai Kim |
| author_sort | Kwanghyun Kim |
| collection | DOAJ |
| description | For an autonomous ship to navigate safely and avoid collisions with other ships, reliably detecting and classifying nearby ships under various maritime meteorological environments is essential. In this paper, a novel probabilistic ship detection and classification system based on deep learning is proposed. The proposed method aims to detect and classify nearby ships from a sequence of images. The method considers the confidence of a deep learning detector as a probability; the probabilities from the consecutive images are combined over time by Bayesian fusion. The proposed ship detection system involves three steps. In the first step, ships are detected in each image using Faster region-based convolutional neural network (Faster R-CNN). In the second step, the detected ships are gathered over time and the missed ships are recovered using the Intersection over Union of the bounding boxes between consecutive frames. In the third step, the probabilities from the Faster R-CNN are combined over time and the classes of the ships are determined by Bayesian fusion. To train and evaluate the proposed system, we collected thousands of ship images from Google image search and created our own ship dataset. The proposed method was tested with the collected videos and the mean average precision increased by 89.38 to 93.92% in experimental results. |
| first_indexed | 2024-04-13T18:30:00Z |
| format | Article |
| id | doaj.art-ef40a667016343df906f09813ed1a963 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-04-13T18:30:00Z |
| publishDate | 2018-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-ef40a667016343df906f09813ed1a9632022-12-22T02:35:07ZengMDPI AGApplied Sciences2076-34172018-06-018693610.3390/app8060936app8060936Probabilistic Ship Detection and Classification Using Deep LearningKwanghyun Kim0Sungjun Hong1Baehoon Choi2Euntai Kim3School of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, KoreaSchool of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, KoreaSchool of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, KoreaSchool of Electrical and Electronic Engineering, Yonsei University, 50 Yonsei-ro, Seodaemun-gu, Seoul 03722, KoreaFor an autonomous ship to navigate safely and avoid collisions with other ships, reliably detecting and classifying nearby ships under various maritime meteorological environments is essential. In this paper, a novel probabilistic ship detection and classification system based on deep learning is proposed. The proposed method aims to detect and classify nearby ships from a sequence of images. The method considers the confidence of a deep learning detector as a probability; the probabilities from the consecutive images are combined over time by Bayesian fusion. The proposed ship detection system involves three steps. In the first step, ships are detected in each image using Faster region-based convolutional neural network (Faster R-CNN). In the second step, the detected ships are gathered over time and the missed ships are recovered using the Intersection over Union of the bounding boxes between consecutive frames. In the third step, the probabilities from the Faster R-CNN are combined over time and the classes of the ships are determined by Bayesian fusion. To train and evaluate the proposed system, we collected thousands of ship images from Google image search and created our own ship dataset. The proposed method was tested with the collected videos and the mean average precision increased by 89.38 to 93.92% in experimental results.http://www.mdpi.com/2076-3417/8/6/936ship detectionship classificationship datasetdeep learningFaster R-CNNautonomous shipIntersection over Union (IoU) trackingBayesian fusion |
| spellingShingle | Kwanghyun Kim Sungjun Hong Baehoon Choi Euntai Kim Probabilistic Ship Detection and Classification Using Deep Learning Applied Sciences ship detection ship classification ship dataset deep learning Faster R-CNN autonomous ship Intersection over Union (IoU) tracking Bayesian fusion |
| title | Probabilistic Ship Detection and Classification Using Deep Learning |
| title_full | Probabilistic Ship Detection and Classification Using Deep Learning |
| title_fullStr | Probabilistic Ship Detection and Classification Using Deep Learning |
| title_full_unstemmed | Probabilistic Ship Detection and Classification Using Deep Learning |
| title_short | Probabilistic Ship Detection and Classification Using Deep Learning |
| title_sort | probabilistic ship detection and classification using deep learning |
| topic | ship detection ship classification ship dataset deep learning Faster R-CNN autonomous ship Intersection over Union (IoU) tracking Bayesian fusion |
| url | http://www.mdpi.com/2076-3417/8/6/936 |
| work_keys_str_mv | AT kwanghyunkim probabilisticshipdetectionandclassificationusingdeeplearning AT sungjunhong probabilisticshipdetectionandclassificationusingdeeplearning AT baehoonchoi probabilisticshipdetectionandclassificationusingdeeplearning AT euntaikim probabilisticshipdetectionandclassificationusingdeeplearning |