Deep learning model for seabed sediment classification based on fuzzy ranking feature optimization

Cui, Xiaodong; Yang, Fanlin; Wang, Xin; Ai, Bo; Luo, Yu; Ma, Dan

doi:10.1016/j.margeo.2020.106390

Cited by 32 publications

(24 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A useful confirmation from this research is that fuzzy set theory is highly flexible [14] and thus, has many other applications including the prediction of time dependent variables [26]. Fuzzy-based methods lend a flexibility to environmental modelling and assessment that crisp methods do not [34,35]. It can be incorporated into other types of modelling schemes such as those that consider connectivity in earth system processes, for example.…”

Section: Discussionmentioning

confidence: 68%

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

Thomas

Khan²,

Valeo

et al. 2021

Environments

View full text Add to dashboard Cite

Fuzzy set theory has shown potential for reducing uncertainty as a result of data sparsity and also provides advantages for quantifying gradational changes like those of pollutant concentrations through fuzzy clustering based approaches. The ability to lower the sampling frequency and perform laboratory analyses on fewer samples, yet still produce an adequate pollutant distribution map, would reduce the initial cost of new remediation projects. To assess the ability of fuzzy modeling to make spatial predictions using fewer sample points, its predictive ability was compared with the ordinary kriging (OK) and inverse distance weighting (IDW) methods under increasingly sparse data conditions. This research used a Takagi–Sugeno (TS) fuzzy modelling approach with fuzzy c-means (FCM) clustering to make spatial predictions of the lead concentrations in soil. The performance of the TS model was very dependent on the number of outliers in the respective validation set. For modeling under sparse data conditions, the TS fuzzy modeling approach using FCM clustering and constant width Gaussian shaped membership functions did not show any advantages over IDW and OK for the type of data tested. Therefore, it was not possible to speculate on a possible reduction in sampling frequency for delineating the extent of contamination for new remediation projects.

show abstract

Section: Discussionmentioning

confidence: 68%

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

Thomas

Khan²,

Valeo

et al. 2021

Environments

View full text Add to dashboard Cite

show abstract

“…In particular, deep learning networks such as convolutional neural networks (CNNs), have proven to greatly outperform traditional machine learning approaches in common computer vision tasks, including the semantic segmentation of images (Lateef and Ruichek, 2019). This has sparked interest in the marine scientific community to explore the potential of CNNs for marine habitat mapping (Cui et al, 2021;Qin et al, 2021;Anokye et al, 2023). The most-commonly used CNN for semantic segmentation in many fields is the U-Net network (Ronneberger et al, 2015;Leclerc et al, 2019) and its modified versions.…”

mentioning

confidence: 99%

Seabed classification of multibeam echosounder data into bedrock/non-bedrock using deep learning

Garone,

Birkenes Lønmo,

Schimel

et al. 2023

Front. Earth Sci.

View full text Add to dashboard Cite

The accurate mapping of seafloor substrate types plays a major role in understanding the distribution of benthic marine communities and planning a sustainable exploitation of marine resources. Traditionally, this activity has relied on the efforts of marine geology experts, who accomplish it manually by examining information from acoustic data along with the available ground-truth samples. However, this approach is challenging and time-consuming. Hence, it is important to explore automatic methods to replace this manual process. In this study, we investigated the potential of deep learning (U-Net) for classifying the seabed as either “bedrock” or “non-bedrock” using bathymetry and/or backscatter data, acquired with multibeam echosounders (MBES). Slope and hillshade data, derived from the bathymetry, were also included in the experiment. Several U-Net models, taking as input either one of these datasets or a combination of them, were trained using an expert delineated map as reference. The analysis revealed that U-Net has the ability to map bedrock and non-bedrock areas reliably. On our test set, the models using either bathymetry or slope data showed the highest performance metrics and the best visual match with the reference map. We also observed that they often identified topographically rough features as bedrock, which were not interpreted as such by the human expert. While such discrepancy would typically be considered an error of the model, the scale of the expert annotations as well as the different methods used by the experts to manually generate maps must be considered when evaluating the predictions quality. While encouraging results were obtained here, further research is necessary to explore the potential of deep learning in mapping other seabed types and evaluating the models’ generalization capabilities on similar datasets but different geographical locations.

show abstract

“…The study site is situated in the southern Irish Sea around the United Kingdom, which has become the hotspot area for seabed mapping and marine surveys (figure 1). Over these years, the geological conditions have ranged from rocky reefs to deep mud basins [8]. The experimental datasets contain the backscatter intensity data, bathymetry data, and groundtruth sediment samples.…”

Section: Introductionmentioning

confidence: 99%

Seabed sediment classification based on topography-based image partitioning: a case study from the southern Irish Sea

Wan,

Qin,

Cui

et al. 2024

Fourth International Conference on Geology, Mapping, and Remote Sensing (ICGMRS 2023)

View full text Add to dashboard Cite

Seabed sediment classification is meaningful for seafloor habitat mapping and marine resources exploration. Multibeam Echo-sounding System can acquire backscatter and topographic information, which has become the mainstream for detecting seabed sediment. Because topographic features are closely correlated with sediment distribution, and sediment types tend to be distributed in a continuous pattern, we propose a seabed sediment classification based on topographybased image partitioning. Multi-resolution segmentation and K-means clustering are utilized for image partition based on topographic features, then the classification maps of different subregions are integrated as one result map. The experimental results show that the partitioned methods considerably outperform the global methods in terms of accuracy. This method provides ideas for achieving high-resolution seafloor mapping in the future.

show abstract

Deep learning model for seabed sediment classification based on fuzzy ranking feature optimization

Cited by 32 publications

References 39 publications

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

An Investigation of Takagi-Sugeno Fuzzy Modeling for Spatial Prediction with Sparsely Distributed Geospatial Data

Seabed classification of multibeam echosounder data into bedrock/non-bedrock using deep learning

Seabed sediment classification based on topography-based image partitioning: a case study from the southern Irish Sea

Contact Info

Product

Resources

About