Underwater image enhancement (UIE) plays an essential role in improving the quality of raw images captured in an underwater environment. Existing UIE methods can be categorized into two types: handcraft-designed and deep learning-based methods. Generally, the handcraft-designed methods are more explainable due to the leverage of knowledge-based image priors, while the deep learning-based methods are usually criticized for their weak interpretability. In this study, we address this issue by integrating the merits of both handcraft-designed and deep learning-based methods. Specifically, a physical underwater imaging model-inspired deep CNN for UIE is designed. Instead of estimating a global background light magnitude and a transmission matrix separately in traditional image restoration-based UIE methods, we directly generate a single variable as the joint estimation of these two parameters within a deep CNN and directly recover the enhanced image as an output according to a reformulated physical underwater imaging model. The whole network is trained in an end-to-end manner and more importantly has good interpretability. The proposed method has been validated for the UIE task on a real-world underwater image dataset and the experimental results well demonstrate the superiority of our method over the existing ones for UIE.
24 January 2025 in International Journal of Machine Learning and Cybernetics
The sine cosine algorithm (SCA) was proposed for solving optimization tasks, of which the way to obtain the optimal solution is mainly through the continuous iteration of the sine and cosine update formulas. However, SCA also faces low population diversity and stagnation of locally optimal solutions. Hence, we try to eliminate these problems by proposing an enhanced version of SCA, named ESCA_PSO. ESCA_PSO is proposed based on hybrid SCA and particle swarm optimization (PSO) by incorporating multiple mutation strategies into the original SCA_PSO. To validate the effect of ESCA_PSO in handling global optimization problems, ESCA_PSO was compared with quality algorithms on various types of benchmark functions. In addition, the proposed ESCA_PSO was employed to tune the best parameters of support vector machines for dealing with medical diagnosis tasks. The results prove the efficiency of the proposed algorithms in solving optimization problems.
The Slime Mould Algorithm (SMA), renowned for its swarm-based approach, encounters challenges, particularly in maintaining a balance between exploration and exploitation, leading to a trade-off that impacts its optimization performance. The simple structure and limited hyperparameters of SMA contribute to difficulties in effectively navigating the exploration-exploitation trade-off, with a drawback being its poor ability for exploration. To address these challenges and enhance SMA, this paper introduces BSSMA, an improved variant that incorporates the Backtracking Search Algorithm (BSA). The introduction of the \(phaseratio\) parameter aims to synergize BSA and SMA, capitalizing on the strengths of both algorithms while mitigating their individual drawbacks, including SMA's poor exploration ability. BSA facilitates a thorough exploration, dispersing search agents widely across the solution space, ensuring significant diversity. These search agents then transition to SMA to further refine the search for optimal solutions while addressing SMA's exploration limitations. Evaluating the performance of BSSMA involves comparisons with 12 other meta-heuristic algorithms (MAs) and 10 advanced MAs using the CEC2017 benchmark functions. Experimental results showcase that the enhanced BSSMA outperforms SMA in terms of convergence speed and accuracy, specifically addressing the challenges associated with balancing exploration and exploitation trade-offs, including SMA's poor exploration ability. Additionally, to demonstrate BSSMA's effectiveness in practical engineering applications, a binary version (bBSSMA) is developed for feature selection (FS) using a V-shaped transfer function. Comparative experiments with seven other binary MA variants reveal that bBSSMA selects fewer features, attains higher classification accuracy, and demands less computational time. These results affirm the effectiveness of bBSSMA for practical feature selection applications.
31 October 2024 in Computers in Biology and Medicine
Techniques for identifying joinable or unionable tables in data lakes can yield valuable information for data scientists. However, more than half of their working time is spent familiarizing themselves with the metadata and correlations of datasets....