Sea-Surface Object Detection Based on Electro-Optical Sensors: A Review

Lyu, Hongguang; Shao, Zeyuan; Cheng, Tao; Yin, Yixin; Gao, Xiaowei

doi:10.1109/mits.2022.3198334

Cited by 7 publications

(4 citation statements)

References 165 publications

(197 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, uncertainty needs to be modeled and handled to provide confidence measures and uncertainty estimates about fusion results. The ultimate goal of multi-sensor information fusion is to support specific applications and decisions 11 . Fusion results can be used in environmental perception, target tracking, navigation and positioning, robot control, intelligent transportation and other fields to provide more accurate and comprehensive information support.…”

Section: Methods Of Information Combinationmentioning

confidence: 99%

A multi-visual information synthesis function model-based technique is being looked at for autonomous vehicle detection and monitoring

Zhang

2023

Third International Conference on Electronics, Electrical and Information Engineering (ICEEIE 2023)

View full text Add to dashboard Cite

With the development of autonomous driving technology, intelligent vehicle detection and tracking has become a key research field. The traditional method based on a single visual sensor faces challenges in complex scenes. Therefore, the method of multi-visual information fusion has become an important way to improve the performance of detection and tracking. This research aims to explore an intelligent vehicle detection and tracking method based on multi-visual information fusion function model. First, we propose a multi-visual information fusion functional model and combine them through a feature fusion module. In this way, we can comprehensively utilize the advantages of multiple vision sensors to improve the accuracy and robustness of detection and tracking. Extensive experimental evaluations are then performed on real datasets. The final experimental results show that the intelligent vehicle detection and tracking method based on the multi-visual information fusion function model has higher accuracy and robustness than the traditional method. By comprehensively utilizing information from multiple vision sensors, we can better deal with complex scenarios and challenges, and provide more reliable perception capabilities for autonomous driving systems. The results of this research are of great significance for promoting the development of intelligent vehicle detection and tracking technology, and also provide reference and reference for multi-sensor information fusion methods in other fields.

show abstract

Section: Methods Of Information Combinationmentioning

confidence: 99%

A multi-visual information synthesis function model-based technique is being looked at for autonomous vehicle detection and monitoring

Zhang

2023

Third International Conference on Electronics, Electrical and Information Engineering (ICEEIE 2023)

View full text Add to dashboard Cite

show abstract

“…The performance of ISMOD methods heavily depends on the dataset's quality, especially for deep learning and sensor fusion methods [127]. Currently, many deep learning maritime object detection methods are trained and tested on NRCID, making it challenging to directly promote them for ISMOD based on RGB cameras.…”

Section: 23mentioning

confidence: 99%

A review of intelligent ship marine object detection based on RGB camera

Yang,

Solihin,

Zhao

et al. 2023

IET Image Processing

View full text Add to dashboard Cite

The article presents a comprehensive summary of Intelligent Ship Marine Object Detection (ISMOD) based on the RGB Camera. Marine object detection plays a pivotal role in enabling intelligent ships to acquire crucial data and security assurances for autonomous navigation. Among the various detection sensors, the RGB Camera is an informative and cost‐effective tool with a wide range of civil applications. In the beginning, the ISMOD metrics based on the RGB camera is analyzed from three significant aspects, namely accuracy, speed, and robustness. Subsequently, the latest research status and comparative overview are presented, encompassing three mainstream detection methods: traditional detection, deep learning detection, and sensor fusion detection. Finally, the existing challenges of ISMOD are discussed and future development trends are recommended. The results demonstrate that forthcoming development will predominantly concentrate on deep learning approaches, complemented by other techniques. It is imperative to advance detection performance in domains such as deep fusion, multi‐feature extraction, multi‐fusion technology, and lightweight detection architecture.

show abstract

“…Traditional maritime object detection algorithms generally follow a three-phase detection framework, namely, horizon detection, staticbackground subtraction, and foreground segmentation (Lyu et al, 2022). In the first phase, Fefilatyev et al (2012) utilized Hough transform to detect the horizon position and thus reduced the object search space, and used threshold segmentation to obtain the maritime ship object after image registration.…”

Section: Introductionmentioning

confidence: 99%

Anchor boxes adaptive optimization algorithm for maritime object detection in video surveillance

Zheng,

Zhao,

et al. 2023

Front. Mar. Sci.

View full text Add to dashboard Cite

With the development of the marine economy, video surveillance has become an important technical guarantee in the fields of marine engineering, marine public safety, marine supervision, and maritime traffic safety. In video surveillance, maritime object detection (MOD) is one of the most important core technologies. Affected by the size of maritime objects, distance, day and night weather, and changes in sea conditions, MOD faces challenges such as false detection, missed detection, slow detection speed, and low accuracy. However, the existing object detection algorithms usually adopt predefined anchor boxes to search and locate for objects of interest, making it difficult to adapt to maritime objects’ complex features, including the varying scale and large aspect ratio difference. Therefore, this paper proposes a maritime object detection algorithm based on the improved convolutional neural network (CNN). Firstly, a differential-evolutionary-based K-means (DK-means) anchor box clustering algorithm is proposed to obtain adaptive anchor boxes to satisfy the maritime object characteristics. Secondly, an adaptive spatial feature fusion (ASFF) module is added in the neck network to enhance multi-scale feature fusion. Finally, focal loss and efficient intersection over union (IoU) loss are adopted to replace the original loss function to improve the network convergence speed. The experimental results on the Singapore maritime dataset show that our proposed algorithm improves the average precision by 7.1%, achieving 72.7%, with a detection speed of 113 frames per second, compared with You Only Look Once v5 small (YOLOv5s). Moreover, compared to other counterparts, it can achieve a better speed–accuracy balance, which is superior and feasible for the complex maritime environment.

show abstract

Sea-Surface Object Detection Based on Electro-Optical Sensors: A Review

Cited by 7 publications

References 165 publications

A multi-visual information synthesis function model-based technique is being looked at for autonomous vehicle detection and monitoring

A multi-visual information synthesis function model-based technique is being looked at for autonomous vehicle detection and monitoring

A review of intelligent ship marine object detection based on RGB camera

Anchor boxes adaptive optimization algorithm for maritime object detection in video surveillance

Contact Info

Product

Resources

About