The MaSTr1325 dataset for training deep USV obstacle detection models

Bovcon, Borja; Muhovič, Jon; Perš, Janez; Kristan, Matej

doi:10.1109/iros40897.2019.8967909

Cited by 66 publications

(95 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Currently there are only a limited number of publicly available datasets. For example, the MaSTr1325 [39] is collected for ASVs' object detection. The Multi-modal Marine Obstacle Detection Dataset 2 (MODD2) [36] has been intensively used for cross-validation due to the enriched scenes included.…”

Section: B Obstacle Detection For Asv In Maritime Environmentsmentioning

confidence: 99%

“…The WODIS is trained on the MaSTr1325 dataset [39], which is a new large-scale marine semantic segmentation training dataset for the development of obstacle detection methods for small-sized coastal ASVs [37]. The dataset contains 1325 high resolution images taken in realistic conditions and all images are per-pixel labelled into three types: sea, sky and obstacles.…”

Section: A Dataset and Evaluation Metricsmentioning

confidence: 99%

See 1 more Smart Citation

WODIS: Water Obstacle Detection Network Based on Image Segmentation for Autonomous Surface Vehicles in Maritime Environments

Chen

Liu

Achuthan

2021

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

A reliable obstacle detection system is crucial for Autonomous Surface Vehicles (ASVs) to realise fully autonomous navigation with no need of human intervention. However, the current detection methods have particular drawbacks such as poor detection for small objects, low estimation accuracy caused by water surface reflection and a high rate of false-positive on water-sky interference. Therefore, we propose a new encoderdecoder structured deep semantic segmentation network, which is Water Obstacle Detection network based on Image Segmentation (WODIS), to solve above mentioned problems. The first design feature of WODIS utilises the use of an encoder network to extract high-level data based on different sampling rates. In order to improve obstacle detection at sea-sky-line areas, an Attention Refine Module (ARM) activated by both global average pooling and max pooling to capture high-level information has been designed and integrated into WODIS. In addition, a Feature Fusion Module (FFM) is introduced to help concatenate the multi-dimensional high-level features in the decoder network. The WODIS is tested and cross validated using four different types of maritime datasets with the results demonstrating that mIoU of WODIS can achieve superior segmentation effects for sea level obstacles to values as high as 91.3%.

show abstract

Section: B Obstacle Detection For Asv In Maritime Environmentsmentioning

confidence: 99%

Section: A Dataset and Evaluation Metricsmentioning

confidence: 99%

WODIS: Water Obstacle Detection Network Based on Image Segmentation for Autonomous Surface Vehicles in Maritime Environments

Chen

Liu

Achuthan

2021

IEEE Trans. Instrum. Meas.

View full text Add to dashboard Cite

show abstract

“…Bovcon et al [154] introduced the MaSTr1325 dataset for training deep USV obstacle detection models in small-sized coastal USV.…”

Section: Marine Datasets Comparison Moosbauer Et Al [144]mentioning

confidence: 99%

Survey on Deep Learning-Based Marine Object Detection

Zhang

et al. 2021

Journal of Advanced Transportation

View full text Add to dashboard Cite

We present a survey on marine object detection based on deep neural network approaches, which are state-of-the-art approaches for the development of autonomous ship navigation, maritime surveillance, shipping management, and other intelligent transportation system applications in the future. The fundamental task of maritime transportation surveillance and autonomous ship navigation is to construct a reachable visual perception system that requires high efficiency and high accuracy of marine object detection. Therefore, high-performance deep learning-based algorithms and high-quality marine-related datasets need to be summarized. This survey focuses on summarizing the methods and application scenarios of maritime object detection, analyzes the characteristics of different marine-related datasets, highlights the marine detection application of the YOLO series model, and also discusses the current limitations of object detection based on deep learning and possible breakthrough directions. The large-scale, multiscenario industrialized neural network training is an indispensable link to solve the practical application of marine object detection. A widely accepted and standardized large-scale marine object verification dataset should be proposed.

show abstract

“…The second approach is to learn richer features using deep convolutional neural networks, inspired by the recent developments in deep learning. These deep neural networks are able to learn rich features and achieve desirable visionbased semantic segmentation results [9]. Bovcon et al [10] proposed a decoder-encoder network and achieved the stateof-the-art performance on the public marine environment Multi-modal Marine Obstacle Detection Dataset 2 (MODD2) dataset [11].…”

Section: Introductionmentioning

confidence: 99%

ShorelineNet: An Efficient Deep Learning Approach for Shoreline Semantic Segmentation for Unmanned Surface Vehicles

Yao

Kanoulas

et al. 2021

2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

View full text Add to dashboard Cite

This paper introduces a novel deep learning approach to the semantic segmentation of shoreline environments with a high frames-per-second (fps) performance, making the approach readily applicable to autonomous navigation for Unmanned Surface Vehicles (USV). The proposed ShorelineNet is an efficient deep neural network of high performance relying only on visual input. ShorelineNet uses monocular visual input to produce accurate shoreline separation and obstacle detection compared to the state-of-the-art, and achieves this with realtime performance. Experimental validation on a challenging multi-modal maritime obstacle detection dataset, the MODD2 dataset, achieves a much faster inference (25fps on an NVIDIA Tesla K80 and 6fps on a CPU) with respect to the recent stateof-the-art methods, while keeping the performance equally high (68.8% F-score). This makes ShorelineNet a robust and effective model to be used for a reliable USV navigation that require realtime and high-performance semantic segmentation of maritime environments.

show abstract

The MaSTr1325 dataset for training deep USV obstacle detection models

Cited by 66 publications

References 28 publications

WODIS: Water Obstacle Detection Network Based on Image Segmentation for Autonomous Surface Vehicles in Maritime Environments

WODIS: Water Obstacle Detection Network Based on Image Segmentation for Autonomous Surface Vehicles in Maritime Environments

Survey on Deep Learning-Based Marine Object Detection

ShorelineNet: An Efficient Deep Learning Approach for Shoreline Semantic Segmentation for Unmanned Surface Vehicles

Contact Info

Product

Resources

About