Study on Visual Detection Algorithm of Sea Surface Targets Based on Improved YOLOv3

Liu, Tao; Pang, Bo; Ai, Shangmao; Sun, Xiaoqiang

doi:10.3390/s20247263

Cited by 29 publications

(21 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Liu at al. [43] show that the sea-surface target detection based on improved YOLOv3 has high accuracy. In forward-looking imaging, Zhang et al [44] use the Rayleigh distribution to model the sea clutter.…”

Section: Introductionmentioning

confidence: 90%

Forward-Looking Super-Resolution Imaging for Sea-Surface Target with Multi-Prior Bayesian Method

Zuo

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

Traditional forward-looking super-resolution methods mainly concentrate on enhancing the resolution with ground clutter or no clutter scenes. However, sea clutter exists in the sea-surface target imaging, as well as ground clutter when the imaging scene is a seacoast.Meanwhile, restoring the contour information of the target has an important effect, for example, in the autonomous landing on a ship. This paper aims to realize the forward-looking imaging of a sea-surface target. In this paper, a multi-prior Bayesian method, which considers the environment and fuses the contour information and the sparsity of the sea-surface target, is proposed. Firstly, due to the imaging environment in which more than one kind of clutter exists, we introduce the Gaussian mixture model (GMM) as the prior information to describe the interference of the clutter and noise. Secondly, we fuse the total variation (TV) prior and Laplace prior, and propose a multi-prior to model the contour information and sparsity of the target. Third, we introduce the latent variable to simplify the logarithm likelihood function. Finally, to solve the optimal parameters, the maximum posterior-expectation maximization (MAP-EM) method is utilized. Experimental results illustrate that the multi-prior Bayesian method can enhance the azimuth resolution, and preserve the contour information of the sea-surface target.

show abstract

Section: Introductionmentioning

confidence: 90%

Forward-Looking Super-Resolution Imaging for Sea-Surface Target with Multi-Prior Bayesian Method

Zuo

et al. 2021

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…The essence of mooring area detection is to detect a circular anchor area, which can be obtained by the method of Euclidean space distance, image identification [38,39] and so on. In [40], the adaptive genetic algorithm was applied to optimize the automatic detection results of mooring area based on the gray system and an artificial neural network, and the final detection results of mooring area were output while the artificial intelligence system completed the regional detection of the mooring area. However, the anchorage circle radius and safety distance model were inaccurate.…”

Section: Methods Of Mooring Area Detectionmentioning

confidence: 99%

Research on Intelligent Detection Algorithm of the Single Anchored Mooring Area for Maritime Autonomous Surface Ships

et al. 2022

View full text Add to dashboard Cite

Mooring area detection represents one of the key technological problems that must be solved in the development of Maritime Autonomous Surface Ships (MASS). In view of the lack of research on the current detection methods for ship mooring area, a new intelligent detection algorithm of the single anchored mooring area for MASS was proposed in this study, aiming at improving the detection ability and accuracy of the MASS mooring area. Firstly, the laws of short period swinging motion, long period circumferential motion and reciprocating motion in the radial direction of an anchoring ship were summarized. Secondly, an anchorage circle radius model and safety distance model between the anchor positions were established and various constrains were considered including ship type, ship particulars, draft, safety impact caused by other ships passing through the anchoring ship. Thirdly, the Monte-Carlo stochastic simulation method was used to measure the mooring area, which can detect the anchor position intelligently. Finally, a case study on MATLAB demonstrated that the proposed intelligent detection algorithm for MASS is effective under various marine scenarios. The results enrich the theory of MASS mooring area detection; therefore, the algorithm has great potential to be equipped on MASS in the future.

show abstract

“…However, these methods require considerable computational resources. Tao et al [28] proposed a crossfeature graph feature fusion structure based on YOLO V3 for combining different anchoring methods with a focus loss. Han et al [29] optimized the backbone network of YOLO V4 and designed an amplified feeling field module for small-target detection for improving the acquisition of spatial small-target ship information and the robustness of the spatial displacement.…”

Section: Related Workmentioning

confidence: 99%

A Study on Small-Scale Ship Detection Based on Attention Mechanism

Zheng

Liu

2022

IEEE Access

View full text Add to dashboard Cite

show abstract

Study on Visual Detection Algorithm of Sea Surface Targets Based on Improved YOLOv3

Cited by 29 publications

References 32 publications

Forward-Looking Super-Resolution Imaging for Sea-Surface Target with Multi-Prior Bayesian Method

Forward-Looking Super-Resolution Imaging for Sea-Surface Target with Multi-Prior Bayesian Method

Research on Intelligent Detection Algorithm of the Single Anchored Mooring Area for Maritime Autonomous Surface Ships

A Study on Small-Scale Ship Detection Based on Attention Mechanism

Contact Info

Product

Resources

About