Underwater Image Analysis Research Articles

Contextual embedding generation of underwater images using deep learning techniques

<p>This article delves into the cutting-edge realm of artificial intelligence, specifically focusing on its application in marine research via underwater image analysis. It introduces an innovative, integrated approach that combines object detection with image captioning tailored for the aquatic domain. Central to this approach is the advanced technique of image feature extraction, complemented by the strategic implementation of attention mechanisms within neural networks. These mechanisms are key in enhancing the precision and contextual understanding of underwater imagery. The efficacy of this method is underscored by extensive experiments on diverse underwater datasets. Results show notable improvements in detecting and describing complex underwater scenes, thereby providing invaluable insights for marine biologists, environmentalists, and the broader scientific community. This exploration marks a significant advancement in marine research, offering a new lens through which the underwater world can be understood and preserved.</p>

Improving the Quality and Performance of Underwater Image Classification using the CLAHE-CNN Method

Research on underwater image analysis is critical because of challenges such as color distortion, low contrast, and noise in images. Various methods have been proposed to overcome this problem. To improve the quality and classification of underwater photos, this study aims to ensure improved performance using Contrast Limited Adaptive Histogram Equalization (CLAHE) and Convolutional Neural Networks (CNN) on underwater image datasets. The dataset consists of 500 JPG image with RGB channel and dimensions of 512 × 512 collected from online sources. There are classifications of sharks, eels, dolphins, sea rays, and whales in the underwater imagery dataset. The experiment was conducted using Python programming language on a computer that had 24GB RAM, Intel® Core™ i7-10510U CPU and hardware properties of Intel® HD Graphics graphics card. The results of this study show how CLAHE improved the CNN classification of underwater imagery by 0.91% in training data, 0.45% in validation data, and 2.02% in test data.

CNN based color balancing and denoising technique for underwater images: CNN-CBDT

Underwater image deteriorates due to the scattering effect of light in the water. The challenging factor in underwater image analysis is the blurring of the image and color distortion. Various techniques were explored to offer solutions for underwater image restoration. But, still, it has some regression exists in these techniques. This proposal introduces the deep CNN-based color balancing and denoising technique (CNN-CBDT) to enhance underwater images. One of the advantages of underwater characteristics is color, mostly green and blue. Due to its low color contrast, the image exists in fuzzy nature. CNN-based CBDT restores the image with the help ReLU unit in the CNN. Lastly, the suggested method's cutting-edge performance is validated by comparing experimental findings to GLNet, Histeq, and ACE algorithms in conditions of structural similarity (SSIM), peak signal-to-noise ratio (PSNR), structural similarity (SSIM), UCIQE, and UIQM. The suggested approach eliminates the impact of underwater elements, enhancing the color of the picture. It enhances PSNR by 17% having the highest value of 19.580 and SSIM by 15% having a value of 0.952. To make it applicable to real robots, the computation speed is calculated. As a result, the proposed method achieved a computation speed of 9.868 frames per second.

Shrimp Body Weight Estimation in Aquaculture Ponds Using Morphometric Features Based on Underwater Image Analysis and Machine Learning Approach

Shrimp is a marine culture found globally due to the ability of its yields to boost a country's economy. It is imperative to monitor its size to determine the condition of the shrimp underwater with complex noise using a non-invasive method. Therefore, this study aims to develop a new method for measuring the body weight of shrimp using morphometric features based on underwater image analysis and a machine learning approach. The method used consists of several steps, data collection using an underwater camera, image analysis using image grayscale, image binary, edge detection, region of interest detection, shrimp image morphometric features extraction, camera calibration using Triangle Similarity (TS), and Correction Factor (CF), calculation of morphometric features value, create machine learning model, training data, and testing data for estimation of underwater shrimp body weight. After testing the model, get the best accuracy value is RMSE = 0.05, MAE = 0.04, and R2 = 0.96 from the MLR method. In conclusion, the results showed that the hybrid method TS-CF-MLR is the best method for measuring underwater shrimp body weight estimation with the lowest error rate and highest coefficient of determination.

Prediction of Suspended Sediment Concentration Based on the Turbidity-Concentration Relationship Determined via Underwater Image Analysis

Sediment measurement data are essential for sediment transport analysis and therefore highly important in overall river planning. Extant sediment measurement methods consume considerable manpower and time and are limited by factors including economic reasons and worker risks. This study primarily aimed to predict the changes in SSC (Suspended Sediment Concentration) and turbidity by examining the change in color in underwater images. While maintaining a constant flow in a channel, the turbidity and concentration were measured under different SSC. Multiple regression models were developed using turbidity measurement results, and they exhibited high explanatory powers (adjusted R2 > 0.91). Furthermore, upon verification using the verification dataset of the experimental results, an excellent predictive power (RMSE ≈ 0.4 NTU) was demonstrated. The model with the highest predictive power, which was inclusive of red and green bands and showed no underlying multicollinearity was used to predict turbidity. Finally, the turbidity and suspended sediment concentration relationship determined from the experimental results was used to estimate the sediment concentration from the color changes in the underwater images. The concentrations that were predicted by the model showed satisfactory results, compared to the measurements (RMSE ≈ 21 ppm). This study indicated the feasibility of continuous SSC monitoring using underwater images as a new measurement method.

In an octopus's garden in the shade: Underwater image analysis of litter use by benthic octopuses

Benthic octopuses have been widely documented in artificial shelters for decades, and this use is apparently increasing. Despite any possible positive effects, the use of litter as shelter could have negative implications. In this work, we aimed to elucidate the interactions of octopuses with marine litter, identifying types of interactions and affected species and regions. To achieve this, we obtained 261 underwater images from ‘citizen science’ records, and identified 8 genera and 24 species of benthic octopuses interacting with litter. Glass objects were present in 41.6% of interactions, and plastic in 24.7%. Asia presented the highest number of images, and most records were from 2018 to 2021. Citizen science provided important evidence on octopus/marine litter interactions, highlighting its value and the need for more investigations on the subject. This information is fundamental to help prevent and mitigate the impacts of litter on octopuses, and identify knowledge gaps that require attention.

An optical reconstruction based underwater image analysis

An optical reconstruction based underwater image analysis

Underwater target recognition methods based on the framework of deep learning: A survey

The accuracy of underwater target recognition by autonomous underwater vehicle (AUV) is a powerful guarantee for underwater detection, rescue, and security. Recently, deep learning has made significant improvements in digital image processing for target recognition and classification, which makes the underwater target recognition study becoming a hot research field. This article systematically describes the application of deep learning in underwater image analysis in the past few years and briefly expounds the basic principles of various underwater target recognition methods. Meanwhile, the applicable conditions, pros and cons of various methods are pointed out. The technical problems of AUV underwater dangerous target recognition methods are analyzed, and corresponding solutions are given. At the same time, we prospect the future development trend of AUV underwater target recognition.

Local-CycleGAN: a general end-to-end network for visual enhancement in complex deep-water environment

Underwater image analysis is crucial for many applications such as seafloor survey, biological and environment monitoring, underwater vehicle navigation, inspection and maintenance of underwater infrastructure etc. However, due to light absorption and scattering, the images acquired underwater are always blurry and distorted in color. Most existing image enhancement algorithms typically focus on a few features of the imaging environments, and enhanced results depend on the characteristics of original images. In this study, a local cycle-consistent generative adversarial network is proposed to enhance images acquired in a complex deep-water environment. The proposed network uses a combination of a local discriminator and a global discriminator. Additionally, quality-monitor loss is adopted to evaluate the effect of the generated images. Experimental results show that the local cycle-consistent generative adversarial network is robust and can be generalized for many different image enhancement tasks in different types of complex deep-water environment with varied turbidity.

Fishing Net Health State Estimation Using Underwater Imaging

Fishing net cleanliness plays a critical role for aquaculture industry as bio-fouled nets restrict the flow of water through the net leading to a build-up of toxins and reduced oxygen levels within the pen, thereby putting the fish under increased stress. In this paper, we proposed an underwater fishing Net Health State Estimation (NHSE) method, which can automatically analyze the degree of fouling on the net through underwater image analysis using remotely operated vehicles (ROV) images, and calculate a blocking percentage metric of each net opening. The level of fouling estimated through this method help the operators decide on the need of cleaning or maintenance schedule. There are mainly six modules in the proposed NHSE method, namely user interaction, distortion correction, underwater image dehazing, marine growth segmentation, net-opening structure analysis, and blocked percentage estimation. To evaluate the proposed NHSE method, we collected and labeled several underwater images in Mulroy Bay, Ireland with pixel-wise annotations. In order to verify the universality and robustness of the algorithm, we simulated and built a virtual fishing farm, and, on this basis, collected and labeled fishing net images under different environmental conditions. Seven evaluation metrics are introduced to demonstrate the effectiveness and advantages of the proposed method.

L2,p-norm sequential bilateral 2DPCA: a novel robust technology for underwater image classification and representation

Many recently proposed robust two-dimensional principal component analysis (2DPCA) approaches can suppress the sensitivity to outliers in images to some extent. However, most approaches can neither perfectly minimize the reconstruction error nor use fewer coefficients to conveniently represent image information. To alleviate these deficiencies, we developed a novel robust 2DPCA approach for underwater image analysis, called l2,p-sequential bilateral-2DPCA (l2,p-SB-2DPCA). The outstanding advantages of l2,p-SB-2DPCA are as follows. First, our model uses the l2,p-norm as the metric criterion of the objective function, which not only improves the robustness of the algorithm but also preserves the basic properties of 2DPCA. Second, we establish the relationship between the variance of the projection data and the corresponding input data in both the row and column directions, which makes the model achieve good recognition ability while using fewer coefficients and further improves the interpretability of the model. Finally, to obtain the optimal value of l2,p-SB-2DPCA, we present an iterative algorithm. The experimental results show that the proposed algorithm achieves the best performance on three underwater datasets and one extended face dataset compared with other robust 2DPCA approaches.

Model scale investigation of aspects influencing the ice resistance of ships sailing ahead in level ice

ABSTRACTFor a reliable prediction of the total resistance in ice in an early ship design stage, it is of great importance to have knowledge on the different influencing aspects. Three of the quite poorly investigated aspects of ship resistance in ice are evaluated for a contemporary common ice-breaking ship type with model tests in ice: The different components of the total resistance in ice are investigated by tests in pre-sawn ice. The additional resistance of a skeg in the bow area is analysed with model tests in ice with an instrumented skeg. The ship bottom ice coverage is determined by means of an underwater image analysis methodology. The results of these investigations are given and discussed.

Underwater image segmentation based on fast level set method

Image segmentation is a fundamental process in image processing that has found application in many fields, such as neural image analysis, and underwater image analysis. In this paper, we propose a novel fast level set method (FLSM)-based underwater image segmentation method to improve the traditional level set methods by avoiding the calculation of signed distance function (SDF). The proposed method can speed up the computational complexity without re-initialisation. We also provide a fast semi-implicit additive operator splitting (AOS) algorithm to improve the computational complex. The experiments show that the proposed FLSM performs well in selecting local or global segmentation regions.

An adaptive method of multi-scale edge detection for underwater image

This paper presents a new approach for underwater image analysis using the bi-dimensional empirical mode decomposition (BEMD) technique and the phase congruency information. The BEMD algorithm, fully unsupervised, it is mainly applied to texture extraction and image filtering, which are widely recognized as a difficult and challenging machine vision problem. The phase information is the very stability feature of image. Recent developments in analysis methods on the phase congruency information have received large attention by the image researchers. In this paper, the proposed method is called the EP model that inherits the advantages of the first two algorithms, so this model is suitable for processing underwater image. Moreover, the receiver operating characteristic (ROC) curve is presented in this paper to solve the problem that the threshold is greatly affected by personal experience when underwater image edge detection is performed using the EP model. The EP images are computed using combinations of the Canny detector parameters, and the binaryzation image results are generated accordingly. The ideal EP edge feature extractive maps are estimated using correspondence threshold which is optimized by ROC analysis. The experimental results show that the proposed algorithm is able to avoid the operation error caused by manual setting of the detection threshold, and to adaptively set the image feature detection threshold. The proposed method has been proved to be accuracy and effectiveness by the underwater image processing examples.

Fully automated image segmentation for benthic resource assessment of poly-metallic nodules

Underwater image analysis is a new field for computational pattern recognition. In academia as well as in the industry, it is more and more common to use camera-equipped stationary landers, autonomous underwater vehicles, ocean floor observatory systems or remotely operated vehicles for image based monitoring and exploration. The resulting image collections create a bottleneck for manual data interpretation owing to their size.In this paper, the problem of measuring size and abundance of poly-metallic nodules in benthic images is considered. A foreground/background separation (i.e. separating the nodules from the surrounding sediment) is required to determine the targeted quantities. Poly-metallic nodules are compact (convex), but vary in size and appear as composites with different visual features (color, texture, etc.).Methods for automating nodule segmentation have so far relied on manual training data. However, a hand-drawn, ground-truthed segmentation of nodules and sediment is difficult (or even impossible) to achieve for a sufficient number of images. The new ES4C algorithm (Evolutionary tuned Segmentation using Cluster Co-occurrence and a Convexity Criterion) is presented that can be applied to a segmentation task without a reference ground truth. First, a learning vector quantization groups the visual features in the images into clusters. Secondly, a segmentation function is constructed by assigning the clusters to classes automatically according to defined heuristics. Using evolutionary algorithms, a quality criterion is maximized to assign cluster prototypes to classes. This criterion integrates the morphological compactness of the nodules as well as feature similarity in different parts of nodules. To assess its applicability, the ES4C algorithm is tested with two real-world data sets. For one of these data sets, a reference gold standard is available and we report a sensitivity of 0.88 and a specificity of 0.65.Our results show that the applied heuristics, which combine patterns in the feature domain with patterns in the spatial domain, lead to good segmentation results and allow full automation of the resource-abundance assessment for benthic poly-metallic nodules.

A method for underwater image analysis using bi-dimensional empirical mode decomposition technique

Recent developments in underwater image recognition methods have received large attention by the ocean engineering researchers. In this paper, an improved bi-dimensional empirical mode decomposition (BEMD) approach is employed to decompose the given underwater image into intrinsic mode functions (IMFs) and residual. We developed a joint algorithm based on BEMD and Canny operator to extract multi-pixel edge features at multiple scales in IMFs sub-images. So the multiple pixel edge extraction is an advantage of our approach; the other contribution of this method is the realization of the bi-dimensional sifting process, which is realized utilizing regional-based operators to detect local extreme points and constructing radial basis function for curve surface interpolation. The performance of the multi-pixel edge extraction algorithm for processing underwater image is demonstrated in the contrast experiment with both the proposed method and the phase congruency edge detection.

Detection and removal of jellyfish using underwater image analysis

Detection and removal of jellyfish using underwater image analysis