scholarly journals SSNet: Learning Mid-Level Image Representation Using Salient Superpixel Network

2019 ◽  
Vol 10 (1) ◽  
pp. 140
Author(s):  
Zhihang Ji ◽  
Fan Wang ◽  
Xiang Gao ◽  
Lijuan Xu ◽  
Xiaopeng Hu
Keyword(s):  
Saliency Detection ◽  
Image Representation ◽  
Detection Algorithm ◽  
High Order ◽  
Order Information ◽  
Visual Words ◽  
Foreground Region ◽  
Gestalt Grouping ◽  
Benchmark Datasets ◽  
The Impact

In the standard bag-of-visual-words (BoVW) model, the burstiness problem of features and the ignorance of high-order information often weakens the discriminative power of image representation. To tackle them, we present a novel framework, named the Salient Superpixel Network, to learn the mid-level image representation. For reducing the impact of burstiness occurred in the background region, we use the salient regions instead of the whole image to extract local features, and a fast saliency detection algorithm based on the Gestalt grouping principle is proposed to generate image saliency maps. In order to introduce the high-order information, we propose a weighted second-order pooling (WSOP) method, which is capable of exploiting the high-order information and further alleviating the impact of burstiness in the foreground region. Then, we conduct experiments on six image classification benchmark datasets, and the results demonstrate the effectiveness of the proposed framework with either the handcrafted or the off-the-shelf CNN features.

scholarly journals Intuitionistic Fuzzy Laplacian Twin Support Vector Machine for Semi-supervised Classification

2021 ◽  
Author(s):  
Jia-Bin Zhou ◽  
Yan-Qin Bai ◽  
Yan-Ru Guo ◽  
Hai-Xiang Lin
Keyword(s):  
Support Vector Machine ◽  
Negative Impact ◽  
Twin Support Vector Machine ◽  
Fuzzy Membership ◽  
Support Vector ◽  
Membership Functions ◽  
Fuzzy Membership Functions ◽  
Intuitionistic Fuzzy ◽  
Benchmark Datasets ◽  
The Impact

AbstractIn general, data contain noises which come from faulty instruments, flawed measurements or faulty communication. Learning with data in the context of classification or regression is inevitably affected by noises in the data. In order to remove or greatly reduce the impact of noises, we introduce the ideas of fuzzy membership functions and the Laplacian twin support vector machine (Lap-TSVM). A formulation of the linear intuitionistic fuzzy Laplacian twin support vector machine (IFLap-TSVM) is presented. Moreover, we extend the linear IFLap-TSVM to the nonlinear case by kernel function. The proposed IFLap-TSVM resolves the negative impact of noises and outliers by using fuzzy membership functions and is a more accurate reasonable classifier by using the geometric distribution information of labeled data and unlabeled data based on manifold regularization. Experiments with constructed artificial datasets, several UCI benchmark datasets and MNIST dataset show that the IFLap-TSVM has better classification accuracy than other state-of-the-art twin support vector machine (TSVM), intuitionistic fuzzy twin support vector machine (IFTSVM) and Lap-TSVM.

scholarly journals Seasonal variability of SST fronts and winds on the southeastern continental shelf of Brazil

2019 ◽  
Vol 69 (11-12) ◽  
pp. 1387-1399 ◽  
Author(s):  
Huan-Huan Chen ◽  
Yiquan Qi ◽  
Yuntao Wang ◽  
Fei Chai
Keyword(s):  
Continental Shelf ◽  
Wind Stress ◽  
Seasonal Variability ◽  
Detection Algorithm ◽  
Coupling Coefficients ◽  
Seafloor Topography ◽  
Wind Stress Curl ◽  
Large Coupling ◽  
The Impact ◽  
Sst Fronts

Abstract Fourteen years (September 2002 to August 2016) of high-resolution satellite observations of sea surface temperature (SST) data are used to describe the frontal pattern and frontogenesis on the southeastern continental shelf of Brazil. The daily SST fronts are obtained using an edge-detection algorithm, and the monthly frontal probability (FP) is subsequently calculated. High SST FPs are mainly distributed along the coast and decrease with distance from the coastline. The results from empirical orthogonal function (EOF) decompositions reveal strong seasonal variability of the coastal SST FP with maximum (minimum) in the astral summer (winter). Wind plays an important role in driving the frontal activities, and high FPs are accompanied by strong alongshore wind stress and wind stress curl. This is particularly true during the summer, when the total transport induced by the alongshore component of upwelling-favorable winds and the wind stress curl reaches the annual maximum. The fronts are influenced by multiple factors other than wind forcing, such as the orientation of the coastline, the seafloor topography, and the meandering of the Brazil Current. As a result, there is a slight difference between the seasonality of the SST fronts and the wind, and their relationship was varying with spatial locations. The impact of the air-sea interaction is further investigated in the frontal zone, and large coupling coefficients are found between the crosswind (downwind) SST gradients and the wind stress curl (divergence). The analysis of the SST fronts and wind leads to a better understanding of the dynamics and frontogenesis off the southeastern continental shelf of Brazil, and the results can be used to further understand the air-sea coupling process at regional level.

VERY HIGH-ORDER SPATIALLY IMPLICIT SCHEMES FOR COMPUTATIONAL ACOUSTICS ON CURVILINEAR MESHES

2001 ◽  
Vol 09 (04) ◽  
pp. 1259-1286 ◽  
Author(s):  
MIGUEL R. VISBAL ◽  
DATTA V. GAITONDE
Keyword(s):  
Three Dimensional ◽  
Radiation Condition ◽  
Higher Order ◽  
High Order ◽  
High Frequencies ◽  
Decomposition Strategies ◽  
The Impact ◽  
Spurious Modes ◽  
Very High ◽  
Computational Strategy

A high-order compact-differencing and filtering algorithm, coupled with the classical fourth-order Runge–Kutta scheme, is developed and implemented to simulate aeroacoustic phenomena on curvilinear geometries. Several issues pertinent to the use of such schemes are addressed. The impact of mesh stretching in the generation of high-frequency spurious modes is examined and the need for a discriminating higher-order filter procedure is established and resolved. The incorporation of these filtering techniques also permits a robust treatment of outflow radiation condition by taking advantage of energy transfer to high-frequencies caused by rapid mesh stretching. For conditions on the scatterer, higher-order one-sided filter treatments are shown to be superior in terms of accuracy and stability compared to standard explicit variations. Computations demonstrate that these algorithmic components are also crucial to the success of interface treatments created in multi-domain and domain-decomposition strategies. For three-dimensional computations, special metric relations are employed to assure the fidelity of the scheme in highly curvilinear meshes. A variety of problems, including several benchmark computations, demonstrate the success of the overall computational strategy.

scholarly journals State model based face mask detection

2018 ◽  
Vol 7 (2.22) ◽  
pp. 35
Author(s):  
Kavitha M ◽  
Mohamed Mansoor Roomi S ◽  
K Priya ◽  
Bavithra Devi K
Keyword(s):  
Face Mask ◽  
Search Space ◽  
Gaussian Mixture ◽  
Detection Algorithm ◽  
State Model ◽  
Face Region ◽  
Automatic Teller Machine ◽  
Foreground Region ◽  
The Face ◽  
Surveillance Mechanism

The Automatic Teller Machine plays an important role in the modern economic society. ATM centers are located in remote central which are at high risk due to the increasing crime rate and robbery.These ATM centers assist with surveillance techniques to provide protection. Even after installing the surveillance mechanism, the robbers fool the security system by hiding their face using mask/helmet. Henceforth, an automatic mask detection algorithm is required to, alert when the ATM is at risk. In this work, the Gaussian Mixture Model (GMM) is applied for foreground detection to extract the regions of interest (ROI) i.e. Human being. Face region is acquired from the foreground region through  the torso partitioning and applying Viola-Jones algorithm in this search space. Parts of the face such as Eye pair, Nose, and Mouth are extracted and a state model is developed to detect  mask.  

scholarly journals Multiple Saliency and Channel Sensitivity Network for Aggregated Convolutional Feature

2019 ◽  
Vol 33 ◽  
pp. 9013-9020
Author(s):  
Xuanlu Xiang ◽  
Zhipeng Wang ◽  
Zhicheng Zhao ◽  
Fei Su
Keyword(s):  
State Of The Art ◽  
Image Representation ◽  
Training Data ◽  
Gram Matrix ◽  
Redundant Information ◽  
Deep Architecture ◽  
Benchmark Datasets ◽  
Supervised Methods ◽  
Ranking Loss ◽  
Effective Channel

In this paper, aiming at two key problems of instance-level image retrieval, i.e., the distinctiveness of image representation and the generalization ability of the model, we propose a novel deep architecture - Multiple Saliency and Channel Sensitivity Network(MSCNet). Specifically, to obtain distinctive global descriptors, an attention-based multiple saliency learning is first presented to highlight important details of the image, and then a simple but effective channel sensitivity module based on Gram matrix is designed to boost the channel discrimination and suppress redundant information. Additionally, in contrast to most existing feature aggregation methods, employing pre-trained deep networks, MSCNet can be trained in two modes: the first one is an unsupervised manner with an instance loss, and another is a supervised manner, which combines classification and ranking loss and only relies on very limited training data. Experimental results on several public benchmark datasets, i.e., Oxford buildings, Paris buildings and Holidays, indicate that the proposed MSCNet outperforms the state-of-the-art unsupervised and supervised methods.

Multi-Variable, High Order, Performance Models (2005C)

2005 ◽  
Author(s):  
David Japikse ◽  
Oleg Dubitsky ◽  
Kerry N. Oliphant ◽  
Robert J. Pelton ◽  
Daniel Maynes ◽  
...  
Keyword(s):  
Data Processing ◽  
Large Data ◽  
High Order ◽  
Large Data Sets ◽  
Data Sets ◽  
Performance Models ◽  
Statistical Accuracy ◽  
Evaluation Methodologies ◽  
New Models ◽  
The Impact

In the course of developing advanced data processing and advanced performance models, as presented in companion papers, a number of basic scientific and mathematical questions arose. This paper deals with questions such as uniqueness, convergence, statistical accuracy, training, and evaluation methodologies. The process of bringing together large data sets and utilizing them, with outside data supplementation, is considered in detail. After these questions are focused carefully, emphasis is placed on how the new models, based on highly refined data processing, can best be used in the design world. The impact of this work on designs of the future is discussed. It is expected that this methodology will assist designers to move beyond contemporary design practices.

scholarly journals Pre-Processing Filter Reflecting Human Visual Perception to Improve Saliency Detection Performance

Electronics ◽  
2021 ◽  
Vol 10 (23) ◽  
pp. 2892
Author(s):  
Kyungjun Lee ◽  
Seungwoo Wee ◽  
Jechang Jeong
Keyword(s):  
Saliency Detection ◽  
Visual Saliency ◽  
Ground Truth ◽  
Bilateral Filter ◽  
Input Image ◽  
Human Visual Perception ◽  
Difference Of Gaussians ◽  
Surrounding Environment ◽  
Benchmark Datasets ◽  
Previous State

Salient object detection is a method of finding an object within an image that a person determines to be important and is expected to focus on. Various features are used to compute the visual saliency, and in general, the color and luminance of the scene are widely used among the spatial features. However, humans perceive the same color and luminance differently depending on the influence of the surrounding environment. As the human visual system (HVS) operates through a very complex mechanism, both neurobiological and psychological aspects must be considered for the accurate detection of salient objects. To reflect this characteristic in the saliency detection process, we have proposed two pre-processing methods to apply to the input image. First, we applied a bilateral filter to improve the segmentation results by smoothing the image so that only the overall context of the image remains while preserving the important borders of the image. Second, although the amount of light is the same, it can be perceived with a difference in the brightness owing to the influence of the surrounding environment. Therefore, we applied oriented difference-of-Gaussians (ODOG) and locally normalized ODOG (LODOG) filters that adjust the input image by predicting the brightness as perceived by humans. Experiments on five public benchmark datasets for which ground truth exists show that our proposed method further improves the performance of previous state-of-the-art methods.

scholarly journals The Tharsis mantle source of depleted shergottites revealed by 90 million impact craters

2021 ◽  
Vol 12 (1) ◽  
Author(s):  
A. Lagain ◽  
G. K. Benedix ◽  
K. Servis ◽  
D. Baratoux ◽  
L. S. Doucet ◽  
...  
Keyword(s):  
Thermal Anomaly ◽  
Detection Algorithm ◽  
Impact Craters ◽  
Early History ◽  
Volcanic Province ◽  
Crater Detection ◽  
Depleted Mantle ◽  
History Of ◽  
Secondary Crater ◽  
The Impact

AbstractThe only martian rock samples on Earth are meteorites ejected from the surface of Mars by asteroid impacts. The locations and geological contexts of the launch sites are currently unknown. Determining the impact locations is essential to unravel the relations between the evolution of the martian interior and its surface. Here we adapt a Crater Detection Algorithm that compile a database of 90 million impact craters, allowing to determine the potential launch position of these meteorites through the observation of secondary crater fields. We show that Tooting and 09-000015 craters, both located in the Tharsis volcanic province, are the most likely source of the depleted shergottites ejected 1.1 million year ago. This implies that a major thermal anomaly deeply rooted in the mantle under Tharsis was active over most of the geological history of the planet, and has sampled a depleted mantle, that has retained until recently geochemical signatures of Mars’ early history.

Sign in / Sign up

Export Citation Format

Share Document