A prior-knowledge-based threshold segmentation method of forward-looking sonar images for underwater linear object detection

Liu, Lixin; Bian, Hongyu; Yagi, Shinichi; Yang, Xiaodong

doi:10.7567/jjap.55.07kg06

Cited by 9 publications

(6 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another limitation is the accurate segmentation of the area saturated by the camera flash. The saturated value influences the result, and this problem can be overcome using more complicated methods, including the energy minimization method [ 48 ], Gaussian mixture model [ 49 ], prior knowledge-based method [ 50 , 51 ], multiscale-based method [ 52 ], and random walk method [ 53 ]. However, these approaches increase the computation time.…”

Section: Discussionmentioning

confidence: 99%

Intra-Oral Photograph Analysis for Gingivitis Screening in Orthodontic Patients

Kim

Lee

2023

IJERPH

View full text Add to dashboard Cite

This study aimed to confirm the presence of gingival inflammation through image analysis of the papillary gingiva using intra-oral photographs (IOPs) before and after orthodontic treatment and to confirm the possibility of using gingival image analysis for gingivitis screening. Five hundred and eighty-eight (n = 588) gingival sites from the IOPs of 98 patients were included. Twenty-five participants who had completed their orthodontic treatments and were aged between 20 and 37 were included. Six points on the papillary gingiva were selected in the maxillary and mandibular anterior incisors. The red/green (R/G) ratio values were obtained for the selected gingival images and the modified gingival index (GI) was compared. The change in the R/G values during the orthodontic treatment period appeared in the order of before orthodontic treatment (BO), mid-point of orthodontic treatment (MO), three-quarters of the way through orthodontic treatment (TO), and immediately after debonding (IDO), confirming that it was similar to the change in the GI. The R/G value of the gingiva in the image correlated with the GI. Therefore, it could be used as a major index for gingivitis diagnosis using images.

show abstract

Section: Discussionmentioning

confidence: 99%

Intra-Oral Photograph Analysis for Gingivitis Screening in Orthodontic Patients

Kim

Lee

2023

IJERPH

View full text Add to dashboard Cite

show abstract

“…Lampert use image detection method to realize the tracking of spectral lines [7]. Reference [8] discussed the threshold segmentation method of underwater linear object detection based on prior knowledge. Mukherjee proposed a symbol analysis method to detect targets in low resolution sonar images under the condition of very restricted data [9].…”

Section: Introductionmentioning

confidence: 99%

A Spatial Detection Method of Underwater Acoustic Target Based on Radiated Noise Features

Chen

Zhu

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

In order to solve the problem that the traditional sonar detection method based on beam energy is not able to detect weak targets, a feature-based spatial beam target detection method is proposed in this paper. Firstly, the feature is extracted from the pre-formed beams in the whole spatial domain, and then using the relevance vector machine (RVM) for training, the probability output of target detection is obtained. Compared with traditional methods, this method can obtain excellent detection performance under the condition of low signal-to-noise ratio or signal-to-interference ratio. The probability output of this method can better represent the uncertainty of target detection, and it is also convenient for information fusion and other processing, which is different from the output of traditional binary detection. At the end of the paper, simulated data and experimental data both proved that this method is effective.

show abstract

“…At present, sonar image segmentation methods can be roughly divided into five categories based on: (1) thresholding, (2) edge detection, (3) Markov random field models, (4) clustering algorithms, and (5) artificial neural networks [ 20 ]. Liu et al [ 21 ] proposed a threshold segmentation method for underwater linear target detection based on prior knowledge, and they achieved good segmentation quality and computation time by analyzing the threshold variation. Wu et al [ 22 ] introduced a fractal coding algorithm for regional segmentation of sonar images, which improved the segmentation speed.…”

Section: Introductionmentioning

confidence: 99%

Feature Pyramid U-Net with Attention for Semantic Segmentation of Forward-Looking Sonar Images

Zhao

Chen

et al. 2022

Sensors

View full text Add to dashboard Cite

Forward-looking sonar is a technique widely used for underwater detection. However, most sonar images have underwater noise and low resolution due to their acoustic properties. In recent years, the semantic segmentation model U-Net has shown excellent segmentation performance, and it has great potential in forward-looking sonar image segmentation. However, forward-looking sonar images are affected by noise, which prevents the existing U-Net model from segmenting small objects effectively. Therefore, this study presents a forward-looking sonar semantic segmentation model called Feature Pyramid U-Net with Attention (FPUA). This model uses residual blocks to improve the training depth of the network. To improve the segmentation accuracy of the network for small objects, a feature pyramid module combined with an attention structure is introduced. This improves the model’s ability to learn deep semantic and shallow detail information. First, the proposed model is compared against other deep learning models and on two datasets, of which one was collected in a tank environment and the other was collected in a real marine environment. To further test the validity of the model, a real forward-looking sonar system was devised and employed in the lake trials. The results show that the proposed model performs better than the other models for small-object and few-sample classes and that it is competitive in semantic segmentation of forward-looking sonar images.

show abstract

A prior-knowledge-based threshold segmentation method of forward-looking sonar images for underwater linear object detection

Cited by 9 publications

References 37 publications

Intra-Oral Photograph Analysis for Gingivitis Screening in Orthodontic Patients

Intra-Oral Photograph Analysis for Gingivitis Screening in Orthodontic Patients

A Spatial Detection Method of Underwater Acoustic Target Based on Radiated Noise Features

Feature Pyramid U-Net with Attention for Semantic Segmentation of Forward-Looking Sonar Images

Contact Info

Product

Resources

About