Salient object detection aims to locate objects that capture human attention within images. Previous approaches often pose this as a problem of image contrast analysis. In this work, we model an ...image as a hyper graph that utilizes a set of hyper edges to capture the contextual properties of image pixels or regions. As a result, the problem of salient object detection becomes one of finding salient vertices and hyper edges in the hyper graph. The main advantage of hyper graph modeling is that it takes into account each pixel's (or region's) affinity with its neighborhood as well as its separation from image background. Furthermore, we propose an alternative approach based on center-versus-surround contextual contrast analysis, which performs salient object detection by optimizing a cost-sensitive support vector machine (SVM) objective function. Experimental results on four challenging datasets demonstrate the effectiveness of the proposed approaches against the state-of-the-art approaches to salient object detection.
Forward-looking sonar (FLS) imagery system plays a significant role in oceanic object recognition and detection since it can overcome the limitation of lighting conditions and reflect the real ...situation of the underwater environment. However, object detection algorithms for FLS images remain challenging for two main reasons: 1) the noise caused by the coherent characteristic of the scattering phenomenon impairs the detector capture of target information and 2) the scene prior based on the uneven target scale distribution is generally neglected, which leads to the detector generating redundant anchors and slows down detection efficiency. Confronting such challenges, this article characterizes the noise and the uneven target scale distribution in FLS images as multiplicative speckle noise and scene prior, respectively. Therefore, we propose a novel underwater FLS image detection network, namely UFIDNet, to further improve detection performance by considering speckle noise reduction and scene prior in FLS images. More specifically, a speckle reduction auxiliary branch (SRAB) is designed to introduce additional despeckled supervision information to encourage the feature extractor to produce clean features and share them with the detection pipeline during the training phase. In particular, the noise distribution of FLS images is excavated for synthetic dataset construction and despeckle network (DSN) design to obtain despeckled supervision images. In addition, a feature selection strategy (FSS) embedded in detection branch is designed to screen out feature levels that do not match the target size, thus significantly reducing the generation of redundant anchors and improving detection speed. Experimental results show that our UFIDNet achieves 70.5% and 47.3% average precision (AP), 81.3% and 54.6% average recall (AR) (<inline-formula> <tex-math notation="LaTeX">\text {AR}_{\text {max=10}} </tex-math></inline-formula>), 27.0 and 26.1 FPS on two real FLS datasets, respectively, outperforming many state-of-the-art general detectors and sonar image detectors.
We developed spin valve tunneling magnetoresistance devices based on MgO barrier and two compositions of CoFeB electrodes capable of sensing magnetic field in tunable ranges with high sensitivity and ...low nonlinearity. The tunable field ranges are due to varying strength of perpendicular anisotropy in a sensing electrode induced by changing its thickness. The sensing field ranges span from (plus-or-minus sign)0.1 mT to (plus-or-minus sign)100 mT. In the narrowest field range devices showed sensitivity up to 91%/mT and nonlinearity below 1.5% of full scale and in the widest field range sensitivity up to 0.076%/mT and nonlinearity below 2% of full scale. The sensing characteristics and their dependence on the electrode thickness suggest that these device structures are useful for design low to medium magnetic field sensors.
Sea ice change detection from synthetic aperture radar (SAR) images can be regarded as a classification procedure, in which pixels are classified into changed and unchanged classes. However, existing ...methods usually suffer from the intrinsic speckle noise of multitemporal SAR images. To solve the problem, this letter presents a change detection method based on convolutional-wavelet neural networks (CWNNs). In CWNN, dual-tree complex wavelet transform is introduced into convolutional neural networks for changed and unchanged pixels' classification, and then, the effect of speckle noise is effectively reduced. In addition, a virtual sample generation scheme is employed to create samples for CWNN training, and the problem of limited samples is alleviated. Experimental results on two real SAR image data sets demonstrate the effectiveness and robustness of the proposed method.
Intrusion detection systems (IDSs) play a pivotal role in computer security by discovering and repealing malicious activities in computer networks. Anomaly-based IDS, in particular, rely on ...classification models trained using historical data to discover such malicious activities. In this paper, an improved IDS based on hybrid feature selection and two-level classifier ensembles are proposed. A hybrid feature selection technique comprising three methods, i.e., particle swarm optimization, ant colony algorithm, and genetic algorithm, is utilized to reduce the feature size of the training datasets (NSL-KDD and UNSW-NB15 are considered in this paper). Features are selected based on the classification performance of a reduced error pruning tree (REPT) classifier. Then, a two-level classifier ensemble based on two meta learners, i.e., rotation forest and bagging, is proposed. On the NSL-KDD dataset, the proposed classifier shows 85.8% accuracy, 86.8% sensitivity, and 88.0% detection rate, which remarkably outperform other classification techniques recently proposed in the literature. The results regarding the UNSW-NB15 dataset also improve the ones achieved by several state-of-the-art techniques. Finally, to verify the results, a two-step statistical significance test is conducted. This is not usually considered by the IDS research thus far and, therefore, adds value to the experimental results achieved by the proposed classifier.
Sea-surface small target detection is always a difficult problem in high-resolution maritime ubiquitous radars for complex characteristics of sea clutter, weak target returns, and diversity of ...targets. Multiple features extracted from radar returns in different domains have ability but not enough to solely distinguish radar returns with target from sea clutter. Joint exploitation of multiple features becomes the key to improve detection performance. In this article, the K-nearest neighbor (KNN) algorithm and anomaly detection idea are cooperated to develop a novel sea-surface target detection method in the feature space spanned by the eight existing salient features. The detection is realized by the anomaly detection followed by a specially designed KNN-based classifier with a controllable false alarm rate. In the anomaly detection, a decision region is determined by the hyper-spherical coverage of the training set of sea clutter that is sufficient and ergodic in the feature space. The KNN-based classifier is designed based on the training sample set of sea clutter and the training sample set of simulated target returns plus sea clutter that is sufficient but nonergodic, by joint usage of feature weighting, neighbor weighting, and distance weighting. The novel method is validated by the two open and recognized IPIX and CSIR radar databases for sea-surface small target detection. The results show that it provides significant performance improvement in comparison with the existing multiple-feature-based detection methods, owing to the fact that the novel method avoids the dimension restriction and feature compression loss in the existing methods.
Since its inception, the Internet of Things (IoT) has witnessed mushroom growth as a breakthrough technology. In a nutshell, IoT is the integration of devices and data such that processes are ...automated and centralized to a certain extent. IoT is revolutionizing the way business is done and is transforming society as a whole. As this technology advances further, the need to exploit detection and weakness awareness increases to prevent unauthorized access to critical resources and business functions, thereby rendering the system unavailable. Denial of Service (DoS) and Distributed DoS attacks are all too common. In this paper, we propose a Protocol Based Deep Intrusion Detection (PB-DID) architecture, in which we created a data-set of packets from IoT traffic by comparing features from the UNSWNB15 and Bot-IoT data-sets based on flow and Transmission Control Protocol (TCP). We classify non-anomalous, DoS, and DDoS traffic uniquely by taking care of the problems like imbalanced and over-fitting. We have achieved a classification accuracy of 96.3% by using deep learning (DL) technique.
Change detection (CD) aims to identify surface changes from bitemporal images. In recent years, deep learning (DL)-based methods have made substantial breakthroughs in the field of CD. However, CD ...results can be easily affected by external factors, including illumination, noise, and scale, which leads to pseudo-changes and noise in the detection map. To deal with these problems and achieve more accurate results, a deeply supervised (DS) attention metric-based network (DSAMNet) is proposed in this article. A metric module is employed in DSAMNet to learn change maps by means of deep metric learning, in which convolutional block attention modules (CBAM) are integrated to provide more discriminative features. As an auxiliary, a DS module is introduced to enhance the feature extractor's learning ability and generate more useful features. Moreover, another challenge encountered by data-driven DL algorithms is posed by the limitations in change detection datasets (CDDs). Therefore, we create a CD dataset, Sun Yat-Sen University (SYSU)-CD, for bitemporal image CD, which contains a total of 20 000 aerial image pairs of size <inline-formula> <tex-math notation="LaTeX">256\times256 </tex-math></inline-formula>. Experiments are conducted on both the CDD and the SYSU-CD dataset. Compared to other state-of-the-art methods, our network achieves the highest accuracy on both datasets, with an F1 of 93.69% on the CDD dataset and 78.18% on the SYSU-CD dataset.
Existing methods of the small target detection from infrared videos are not effective with the complex background. It is mainly caused by: 1) the interference of strong edges and the similarity with ...other nontarget objects and 2) the lack of the context information of both the background and the target in a spatio-temporal domain. By considering these two points, we propose to slide a window in a single frame and form a spatio-temporal cube with the current frame patch and other frame patches in the spatio-temporal domain. Then, we establish a spatio-temporal tensor model based on these patches. According to the sparse prior of the target and the local correlation of the background, the separation of the target and the background can be cast as a low rank and sparse tensor decomposition problem. The target is obtained from the sparse tensor by the tensor decomposition. The experiments show that our method gains better detection performance in infrared videos with the complex background by making full use of the spatio-temporal context information.
Homophobia or Transphobia can be defined as the hatred, discomfort, or dislike of lesbian, gay, transgender or bisexual people. Studies have shown that these individuals were more likely to develop ...mental health issues, likely due to being subjected to more forms of abuse on social media. Hence there is an ardent need to develop automated abusive speech detection systems to tackle the abusive content on social media. There has been an elevation in hate speech or abuse and this paper focuses on the LGBTQIA+ community. Due to the shortage of resources in the said study area, we hypothesize that data augmentation via Pseudolabeling by transliterating the code-mixed text to the parent language will improve the models’ performances on the newly constructed dataset. We put our hypothesis into testing, and studied the performances of several multilingual language models for our cause.
PDF
