15 March 2023 in IEEE Journal of Oceanic Engineering
In ocean engineering, an underwater vehicle is widely used as an important equipment to explore the ocean. However, due to the reflection and attenuation of light when propagating in water, the images captured by the visual system of an underwater vehicle in the complex underwater environment usually suffer from low visibility, blurred details, and color distortion. To solve this problem, in this article, we present an underwater image enhancement framework based on transfer learning, which consists of a domain transformation module and an image enhancement module. The two modules, respectively, perform color correction and image enhancement, effectively transferring in-air image dehazing to underwater image enhancement. To maintain the physical properties of an underwater image, we embed the physical model into the domain transformation module which ensures that the transformed image complies with the physical model. To effectively remove the color deviation, a coarse-grained similarity calculation is added to the domain transformation module to improve the model performance. The experimental results on real-world underwater images of different scenes show that the presented method is superior to some advanced underwater image enhancement algorithms both qualitatively and quantitatively. Furthermore, we conduct ablation experiments to indicate the contribution of each component and further validate the effectiveness of the presented method through application tests.
11 July 2022 in IEEE Transactions on Image Processing
Serving as an essential step for many applications of image processing, superpixel generation has attracted a lot of attentions. Most existing superpixel generation algorithms focus on the boundary adherence and compactness of the superpixels, but ignore the topological consistency between the superpixels, which severely limites their applications in the subsequent tasks, especially in the CNN based image processing tasks. In this paper, we present a fast lattice superpixel generation algorithm, which can generate superpixels with lattice topology like the original pixels. We also propose a local similarity loss function to improve the segmentation accuracy of the generated lattice superpixels. The whole algorithm is parallelly implemented on GPU. We perform extensive experiments on three datasets (i.e., BSDS500, NYUv2 and VOC) to verify the efficacy of our algorithm. The experimental results show that our method achieves competitive results compared to the state-of-the-art methods.
25 January 2019 in IEEE Transactions on Multimedia
14 November 2022 in IEEE Transactions on Circuits and Systems for Video Technology
Superpixel segmentation provides a way to capture object boundaries unsupervised and has benefited many compute vision applications. However, under-segmentation for weak boundaries and poor compatibility with image feature representations often limit its wide application. In this paper, we propose a new weak-boundary sensitive superpixel generation method and provide an all-in-one solution for images with different feature representations. We first design a local adaptive distance (LAD) to be more sensitive to feature changes in low-contrast regions. LAD leverages image local standard deviation as region contrast clues. It adaptively increases the feature distances in low-contrast regions to avoid feature space distances of weak boundaries being inundated by regularity constraints. LAD is scale-invariant that can be compatible with high bit-depth and multi-feature images. Then, based on LAD, we introduce a novel morphological contour evolution model to generate superpixels iteratively. Leveraging morphological dilation of superpixel shapes, the new model is more conducive to the boundary detection of irregular or slender objects. Extensive experiments demonstrate that our method favorably outperforms state-of-the-art methods, especially regarding the under-segmentation error and segmentation accuracy.
27 November 2025 in Neural Networks