Comparison of possibilistic fuzzy local information C-means and possibilistic K-nearest neighbors for synthetic aperture sonar image segmentation

Peeples, Joshua; Cook, Matthew; Suen, Daniel; Zare, Alina; Keller, James M.

doi:10.1117/12.2519484

Cited by 8 publications

(4 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recent research on SSS target recognition has shown that the target recognition method based on CNNs has outperformed traditional machine learning methods, including the fuzzy logic method, K nearest neighbor, support vector machines, etc., [15][16][17]. With an increasing scale, the recognition network can extract deeper features from images to obtain richer feature information.…”

Section: Literature Reviewmentioning

confidence: 99%

Multi-Mode Channel Position Attention Fusion Side-Scan Sonar Transfer Recognition

Wang

Huo

et al. 2023

Electronics

View full text Add to dashboard Cite

Side-scan sonar (SSS) target recognition is an important part of building an underwater detection system and ensuring a high-precision perception of underwater information. In this paper, a novel multi-channel multi-location attention mechanism is proposed for a multi-modal phased transfer side-scan sonar target recognition model. Optical images from the ImageNet database, synthetic aperture radar (SAR) images and SSS images are used as the training datasets. The backbone network for feature extraction is transferred and learned by a staged transfer learning method. The head network used to predict the type of target extracts the attention features of SSS through a multi-channel and multi-position attention mechanism, and subsequently performs target recognition. The proposed model is tested on the SSS test dataset and evaluated using several metrics, and compared with different recognition algorithms as well. The results show that the model has better recognition accuracy and robustness for SSS targets.

show abstract

Section: Literature Reviewmentioning

confidence: 99%

Multi-Mode Channel Position Attention Fusion Side-Scan Sonar Transfer Recognition

Wang

Huo

et al. 2023

Electronics

View full text Add to dashboard Cite

show abstract

“…Descriptions of the seafloor are critical for many applications, especially mine countermeasures where awareness of environmental changes improves mine detection [1]. Many scene understanding methods for SAS have been developed which rely upon superpixel segmentations [2,3]. While metrics have been implemented to score superpixel oversegmentations in reference to crisp ground truth [4], we develop a labeling scheme and scoring metric designed for imagery with soft boundaries and complex transitions between textures and compare our approach to traditional crisp methods.…”

Section: Introductionmentioning

confidence: 99%

The Weakly-Labeled Rand Index

Stewart¹,

Hampton²,

Zare³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Synthetic Aperture Sonar (SAS) surveys produce imagery with large regions of transition between seabed types. Due to these regions, it is difficult to label and segment the imagery and, furthermore, challenging to score the image segmentations appropriately. While there are many approaches to quantify performance in standard crisp segmentation schemes, drawing hard boundaries in remote sensing imagery where gradients and regions of uncertainty exist is inappropriate. These cases warrant weak labels and an associated appropriate scoring approach. In this paper, a labeling approach and associated modified version of the Rand index for weakly-labeled data is introduced to address these issues. Results are evaluated with the new index and compared to traditional segmentation evaluation methods. Experimental results on a SAS data set containing must-link and cannot-link labels show that our Weakly-Labeled Rand index scores segmentations appropriately in reference to qualitative performance and is more suitable than traditional quantitative metrics for scoring weaklylabeled data.

show abstract

“…To overcome those deficits in pixel clustering procedure, we adopt Possibilistic C-Means (PCM) clustering strategy. PCM has long history of theoretical background [22,23] and many successful applications in engineering [24,25] and medical domain [26,27] and many constraint relaxed models exist [28]. We also relax some part of standard PCM as proposed in [22] to adapt the domain constraints and due to the low intensity contrast between the target area and the background, we need fuzzy stretching [2] to enhance the contrast.…”

Section: Introductionmentioning

confidence: 99%

Automatic segmentation of ceramic materials with relaxed possibilistic C-Means clustering for defect detection

Kim

Song

Park

2020

IJEECS

View full text Add to dashboard Cite

<span>Auromatic inspection system is necessary for reliable quality control if ceramic materials to avoid operator subjectivity and fatigue in visual inspection. Automatic segmentation from material’s image is then the most important process to develop such an inspection system. In this paper, we propose a Possibilistic C-Means pixel clustering algorithm with fuzzy stretching to form the defect object in segmentation. In experiment using 50 images containing a certain amount of defects, the proposed method was successful in 49 cases or 98% of opportunities. That performance is roughly twice better than that of standard K-means clustering in defect object formation</span><span>Auromatic inspection system is necessary for reliable quality control if ceramic materials to avoid operator subjectivity and fatigue in visual inspection. Automatic segmentation from material’s image is then the most important process to develop such an inspection system. In this paper, we propose a Possibilistic C-Means pixel clustering algorithm with fuzzy stretching to form the defect object in segmentation. In experiment using 50 images containing a certain amount of defects, the proposed method was successful in 49 cases or 98% of opportunities. That performance is roughly twice better than that of standard K-means clustering in defect object formation.</span>

show abstract

Comparison of possibilistic fuzzy local information C-means and possibilistic K-nearest neighbors for synthetic aperture sonar image segmentation

Cited by 8 publications

References 24 publications

Multi-Mode Channel Position Attention Fusion Side-Scan Sonar Transfer Recognition

Multi-Mode Channel Position Attention Fusion Side-Scan Sonar Transfer Recognition

The Weakly-Labeled Rand Index

Automatic segmentation of ceramic materials with relaxed possibilistic C-Means clustering for defect detection

Contact Info

Product

Resources

About