Texture-aware and structure-preserving superpixel segmentation

Wu, Jihuai; Liu, Chunxiao; Li, Biao

doi:10.1016/j.cag.2020.12.002

Cited by 13 publications

(7 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The reason is measure remains same no matter if there is a path along the pixels. The path along the pixels will result in smoother and content relevant pixels [ 16 , 36 ]. The Euclidean distance overlays a segmentation map over the image without having relevance to the actual content present in the image.…”

Section: Methodsmentioning

confidence: 99%

Content-Aware SLIC Super-Pixels for Semi-Dark Images (SLIC++)

Hashmani¹,

Memon²,

Raza

et al. 2022

Sensors

View full text Add to dashboard Cite

Super-pixels represent perceptually similar visual feature vectors of the image. Super-pixels are the meaningful group of pixels of the image, bunched together based on the color and proximity of singular pixel. Computation of super-pixels is highly affected in terms of accuracy if the image has high pixel intensities, i.e., a semi-dark image is observed. For computation of super-pixels, a widely used method is SLIC (Simple Linear Iterative Clustering), due to its simplistic approach. The SLIC is considerably faster than other state-of-the-art methods. However, it lacks in functionality to retain the content-aware information of the image due to constrained underlying clustering technique. Moreover, the efficiency of SLIC on semi-dark images is lower than bright images. We extend the functionality of SLIC to several computational distance measures to identify potential substitutes resulting in regular and accurate image segments. We propose a novel SLIC extension, namely, SLIC++ based on hybrid distance measure to retain content-aware information (lacking in SLIC). This makes SLIC++ more efficient than SLIC. The proposed SLIC++ does not only hold efficiency for normal images but also for semi-dark images. The hybrid content-aware distance measure effectively integrates the Euclidean super-pixel calculation features with Geodesic distance calculations to retain the angular movements of the components present in the visual image exclusively targeting semi-dark images. The proposed method is quantitively and qualitatively analyzed using the Berkeley dataset. We not only visually illustrate the benchmarking results, but also report on the associated accuracies against the ground-truth image segments in terms of boundary precision. SLIC++ attains high accuracy and creates content-aware super-pixels even if the images are semi-dark in nature. Our findings show that SLIC++ achieves precision of 39.7%, outperforming the precision of SLIC by a substantial margin of up to 8.1%.

show abstract

Section: Methodsmentioning

confidence: 99%

Content-Aware SLIC Super-Pixels for Semi-Dark Images (SLIC++)

Hashmani¹,

Memon²,

Raza

et al. 2022

Sensors

View full text Add to dashboard Cite

show abstract

“…For instance, Intrinsic Manifold SLIC (IMSLIC) [21] maps every pixel to a 2-dimensional manifold and measures the superpixel density through its area. Other examples, like Content-Adaptive Superpixels (CAS) [36] and Texture-Aware and Structure-Preserving (TASP) [35], generate content-sensitive superpixels by improving the separability in the feature space. However, differently from IMSLIC, CAS and TASP require a post-processing step for ensuring superpixel connectivity.…”

Section: Classical Methodsmentioning

confidence: 99%

Efficient Multiscale Object-based Superpixel Framework

Belém¹,

Cousty²,

Guimarães³

et al. 2022

Preprint

View full text Add to dashboard Cite

Superpixel segmentation can be used as an intermediary step in many applications, often to improve object delineation and reduce computer workload. However, classical methods do not incorporate information about the desired object. Deep-learning-based approaches consider object information, but their delineation performance depends on data annotation. Additionally, the computational time of object-based methods is usually much higher than desired. In this work, we propose a novel superpixel framework, named Superpixels through Iterative CLEarcutting (SICLE), which exploits object information being able to generate a multiscale segmentation onthe-fly. SICLE starts off from seed oversampling and repeats optimal connectivity-based superpixel delineation and object-based seed removal until a desired number of superpixels is reached. It generalizes recent superpixel methods, surpassing them and other state-of-the-art approaches in efficiency and effectiveness according to multiple delineation metrics.

show abstract

“…Classification can be divided into two categories: supervised and unsupervised. In general, supervised classification is called classification, while unsupervised classification is called clustering [7]. Classification is the most basic statistical analysis method in the field of pattern recognition, and its purpose is to classify images by finding points, curves, and surfaces in the image feature space using labeled training samples.…”

Section: Related Workmentioning

confidence: 99%

An Image Segmentation Method Based on the Law of Universal Gravitation and Adaptive Affinity Propagation Spectrum Clustering

Liu¹,

Zhu

2022

Mathematical Problems in Engineering

View full text Add to dashboard Cite

There are many problems of division in natural and social sciences, and with the development of science and technology, the requirement for division is also increasing. It is difficult to divide accurately by experience and expertise alone, and the most important research branch of the division problem is the clustering algorithm. It is to group similar samples into one class and divide the elements with large differences into different classes. Due to the simplicity and efficiency of the clustering algorithm, it is widely used in image segmentation. The conventional Spectral Clustering (SC) algorithm cannot recognize nonconvex data and has the disadvantage of strong dependence on biased parameter values. To address this problem, the Gravity-based Adaptive Spectral Clustering (GASC) algorithm is proposed in this study. Based on the conventional SC algorithm, the algorithm uses gravity to calculate the similarity (gravity) between data, and uses information entropy and adaptive enhancement (AdaBoost) algorithms to obtain the weights of correct cluster sampling points and wrong cluster sampling points in each cluster, so as to reduce the dependence of the algorithm on bias parameters and reduce the number of wrongly divided sample points. Meanwhile, the GASC algorithm is applied to image segmentation. The implementation process includes three stages: image preprocessing, feature extraction, and clustering. The comparison experiments show that the mean values of normalization and accuracy of the GASC algorithm are improved compared with other clustering algorithms, and the segmentation accuracy for images is higher.

show abstract

Texture-aware and structure-preserving superpixel segmentation

Cited by 13 publications

References 15 publications

Content-Aware SLIC Super-Pixels for Semi-Dark Images (SLIC++)

Content-Aware SLIC Super-Pixels for Semi-Dark Images (SLIC++)

Efficient Multiscale Object-based Superpixel Framework

An Image Segmentation Method Based on the Law of Universal Gravitation and Adaptive Affinity Propagation Spectrum Clustering

Contact Info

Product

Resources

About