Overlap Problem Research Articles

Random networks are heterogeneous exhibiting a multi-scaling law

Unlike the scale-free (SF) architecture, random networks of the Erdos–Renyi (ER) are also called homogeneous networks consisting of the same kind of nodes because these nodes approximately have same degrees, which follow a Poisson distribution. This paper tries to demonstrate that this random network is actually heterogeneous, i.e., it is composed of different kinds of nodes by introducing a new metric, called B-index to quantify a node, defined as the extent that the node’s removal can break down its neighborhood. One interesting phenomenon is observed on random networks, i.e., the distribution of B-index can be roughly divided into many sub-distributions with different scalings exhibiting a multi-scaling law. This phenomenon appears, fluctuates and disappears as the changes of random networks. In addition, the analysis of spreading dynamics on the susceptible infected recovered (SIR) model suggests that the nodes with the highest B-index can alleviate the overlap problem of infected area of multiple origins, and are more influential for the spreading of epidemics, especially for the small number of origins.

Reconciling Instructors’ and Students’ Course Overlap Perspectives via Linked Data Visualization

Research on the use of modeling and mapping tools in curriculum management is thriving, often focusing on the perspectives of the faculty alone. However, scholarly works that also incorporate the students curriculum concerns are rare. A recurring theme in students curriculum concerns is the perceived overlap among courses, usually expressed at the level of common concepts across courses. This common concept emphasis imposes a challenge for modelers, who often focus on the course level comparisons because they usually lack tools with sub-course (concept) level granularity. This paper investigates how to model and represent curriculum information to help in reconciling the gap between instructors' and students' views of cross-course overlap. The proposed approach involves the design and development of a digital environment to 1) model a curriculum via linked data through an ontology representing concept-level granularity; 2) offer instructors aid in populating course content; and 3) facilitate the visualization and manipulation of data. The visualization tools are designed to offer functions for perceived common concept overlap identification and rectification. This digital environment was deployed and evaluated in the context of a curriculum review process, in which 25 course instructors employed the visualization tools to address a perceived course overlap problem. The preliminary results demonstrate, first, the usefulness of the approach in reconciling the views of instructors and students regarding perceived course overlap. Second, the results highlight that the approach contributes to transforming course overlap from a fuzzy notion to a more concrete and actionable construct defined as either repetition or reinforcement.

Efficient image segmentation based on deep learning for mineral image classification

Mineral image segmentation plays a vital role in the realization of machine vision based intelligent ore sorting equipment. However, the existing image segmentation methods still cannot effectively solve the problem of adhesion and overlap between mineral particles, and the segmentation performance of small and irregular particles still needs to be improved. To overcome these bottlenecks, we propose a deep learning based image segmentation method to segment the key areas in mineral images using morphological transformation to process mineral image masks. This investigation explores four aspects of the deep learning-based mineral image segmentation model, including backbone selection, module configuration, loss function construction, and its application in mineral image classification. Specifically, referring to the designs of U-Net, FCN, Seg Net, PSP Net, and DeepLab Net, this experiment uses different backbones as Encoder to building ten mineral image segmentation models with different layers, structures, and sampling methods. Simultaneously, we propose a new loss function suitable for mineral image segmentation and compare CNNs-based segmentation models' training performance under different loss functions. The experiment results show that the proposed mineral image segmentation has excellent segmentation performance, effectively solves adhesion and overlap between adjacent particles without affecting the classification accuracy. By using the Mobile Net as backbone, the PSP Net and DeepLab can achieve a high segmentation performance in mineral image segmentation tasks, and the 15 × 15 is the most suitable size for erosion element structure to process the mask images of the segmentation models.

An “on-off” signal-switchable electrochemiluminescence biosensor for ultrasensitive detection of dual microRNAs based on DNAzyme-powered DNA walker

Herein, an “on-off” signal-switchable electrochemiluminescence (ECL) biosensor for dual miRNAs (miRNA-21 and miRNA-155) detection was constructed based on carbon dots (CDs)-labeled DNA walker, which can effectively avoid problems of the signal overlap or limited available labels that general exist in multiplex assays. In the presence of miRNA-21, the CDs-DNA walker were captured on the electrode via nucleic acid hybridization, producing a strong ECL signal corresponding to miRNA-21 (signal on). For miRNA-155, the biosensor was pre-incubated with CDs-DNA walker and assistant DNA (DNA-21). Target miRNA-155 could trigger metal ion-dependent DNAzyme cleavage. With the assistance of Mn2+, the substrate strands were cleaved and CDs detached from the electrode, resulting in a reduced ECL signal (signal off). The interference of miRNA-21 was eliminated because no more CDs-DNA walker was added. As a result, benefiting from the signal amplification offered by the CDs-DNA walker and the inherent properties of ECL technique, this biosensor achieved low detection limits of 33 fM for miRNA-21 and 33 aM for miRNA-155. This paper also demonstrates the biosensor’s bioanalytical application for multiplex detection of miRNA-21/155 in human serum samples and the total RNA extraction solution from the human cancer cell lines.

Reliable self-healing FBG sensor network for improvement of multipoint strain sensing

In this paper, we proposed and designed reliable self-healing fiber Bragg grating (FBG) sensor network for improvement of multipoint strain sensing. We use an optical switch (OSW) to reconfigure the sensor network and improve the self-healing function during fault happen in the sensor network. In this work, to prove and validate the detection performance of our proposed Bragg wavelength detection technique even when the number of overlap spectra of FBGs increases, we conduct three experiments and collect three different cases of strain data (namely case 1, case 2, and case 3) by applying different strain steps to FBGs. In case 1 experiment, strain was applied to only FBG1 sensor, while the other four FBGs keeps fixed. In case 2 experiment, strain was applied to FBG1 and FBG2 sensors at the same time and with different strain steps. In case 3 experiment, strain was applied to FBG1, FBG2 and FBG3 sensors at the same time and with different strain steps. As a result, in all cases of the experiment, three situations of spectra were introduced between FBGs like non-overlapped, partially overlapped and completely overlapped spectra. To solve this overlap problem, we used deep learning technique to accurately identify the Bragg wavelength of FBGs in the condition of the partially or fully overlapped spectra. Therefore, our proposed FBG sensor system can improve the reliability and detection accuracy of the sensor system even the number of overlaps FBGs spectra increases.

Optical ranging system based on multiple pulse train interference using soliton microcomb

The multiple pulse train interference (MPTI) ranging system takes advantage of the high accuracy and absolute length measurement capability, but traditional scheme has a large measurement dead zone. In this Letter, we propose a MPTI ranging scheme based on chip-based soliton microcomb to achieve high-precision but no dead-zone measurement. The measurement optical path is improved to solve the problem of cross correlation interference fringe overlap, and a peak fitting algorithm is employed to further improve the measurement accuracy of the system. Using a commercial He–Ne interferometer, a 384 nm precision is obtained in 1.5 m distance measurements. Advantages of the simple optical path, high precision, and no measurement dead zone are expected to realize on-chip integration and provide a solution for precision measurement.

Community-based [formula omitted]-shell decomposition for identifying influential spreaders

How to identify the most influential nodes in a network for the maximization of influence spread is a great challenge. Known methods like k-shell decomposition determine core nodes who individually might be the most influential spreaders for the spreading originating in a single origin. However, these techniques are not suitable for determining multiple origins that together lead to the most effective spreading. The reason is that core nodes are often found to be located closely to each other, which results in large overlapping regions rather than spreading far across the network. In this paper, we propose a new algorithm, called community-basedk-shell decomposition, by which a network can be viewed as multiple hierarchically ordered structures each branching off from the innermost shell to the periphery shell. To alleviate the overlap problem, our algorithm pursues a greedy strategy that preferably selects core nodes from different communities in the network, thus maximizing the joint influence of multiple origins. We systematically evaluate our algorithm against competing algorithms on multiple networks with varying network characteristics, and find that our algorithm outperforms other algorithms on networks that exhibit community structures, and the stronger communities, the better performance.

Lifetime-Associated Two-Dimensional Infrared Spectroscopy Reveals the Hydrogen-Bond Structure of Supercooled Water in Soft Confinement.

We demonstrate a method to address the problem of spectral overlap in multidimensional vibrational spectroscopy and use it to investigate supercooled aqueous sorbitol solutions. The absence of crystallization in these solutions has been attributed to “soft” confinement of water in subnanometer voids in the sorbitol matrix, but the details of the hydrogen-bond structure are still largely unknown. 2D-IR spectroscopy of the OH-stretch mode is an excellent tool to investigate hydrogen bonding, but in this case it seems difficult because of the overlapping water and sorbitol contributions to the 2D-IR spectrum. Using the difference in OH-stretch lifetimes of water and sorbitol we can cleanly separate these contributions. Surprisingly, the separated 2D-IR spectra show that the hydrogen-bond disorder of soft-confined water is independent of temperature and decoupled from its orientational order. We believe the approach we use to separate overlapping 2D-IR spectra will enhance the applicability of 2D-IR spectroscopy to study multicomponent systems.

Occlusion Robust Wheat Ear Counting Algorithm Based on Deep Learning.

Counting the number of wheat ears in images under natural light is an important way to evaluate the crop yield, thus, it is of great significance to modern intelligent agriculture. However, the distribution of wheat ears is dense, so the occlusion and overlap problem appears in almost every wheat image. It is difficult for traditional image processing methods to solve occlusion problem due to the deficiency of high-level semantic features, while existing deep learning based counting methods did not solve the occlusion efficiently. This article proposes an improved EfficientDet-D0 object detection model for wheat ear counting, and focuses on solving occlusion. First, the transfer learning method is employed in the pre-training of the model backbone network to extract the high-level semantic features of wheat ears. Secondly, an image augmentation method Random-Cutout is proposed, in which some rectangles are selected and erased according to the number and size of the wheat ears in the images to simulate occlusion in real wheat images. Finally, convolutional block attention module (CBAM) is adopted into the EfficientDet-D0 model after the backbone, which makes the model refine the features, pay more attention to the wheat ears and suppress other useless background information. Extensive experiments are done by feeding the features to detection layer, showing that the counting accuracy of the improved EfficientDet-D0 model reaches 94%, which is about 2% higher than the original model, and false detection rate is 5.8%, which is the lowest among comparative methods.

A label-free SERS sensor for the detection of Hg2+ based on phenylacetylene functionalized Ag nanoparticles

The development of a facile, rapid, precise, and ultrasensitive assay for the detection of mercury ions (Hg2+) is an urgent problem to be solved because excessive Hg2+ poses a great threat to human health and the ecological environment. Herein, a novel surface-enhanced Raman scattering (SERS) sensor which employed phenylacetylene (PA) as a label-free probe was established. The PA functionalized silver nanoparticles (PA-AgNPs) produced an intense SERS signal in the Raman-silent region, which effectively avoided the interference of solvent and overcame the signal overlap problem. In the presence of Hg2+, the PA molecules separated from the Ag surface due to the strong coordination between the alkynyl group and Hg2+, which resulted in a significant SERS signal attenuation. Basing on the proposed SERS platform, a linear relationship with a wide range from 10–10 to 10–8 M and a low detection limit of 87.6 pM were obtained. Besides, the selectivity and reproducibility of the SERS system were also investigated. What’s more, we applied the SERS sensor to analyze Hg2+ in lake water with high accuracy and excellent recovery. The obtained results demonstrated that the established SERS platform is expected to be a promising alternative tool for the determination of Hg2+ in practical application.

When components collide: Spatiotemporal overlap of the N400 and P600 in language comprehension

The problem of spatiotemporal overlap between event-related potential (ERP) components is generally acknowledged in language research. However, its implications for the interpretation of experimental results are often overlooked. In a previous experiment on the functional interpretation of the N400 and P600, it was argued that a P600 effect to implausible words was largely obscured – in one of two implausible conditions – by an overlapping N400 effect of semantic association. In the present ERP study, we show that the P600 effect of implausibility is uncovered when the critical condition is tested against a proper baseline condition which elicits a similar N400 amplitude, while it is obscured when tested against a baseline condition producing an N400 effect. Our findings reveal that component overlap can result in the apparent absence or presence of an effect in the surface signal and should therefore be carefully considered when interpreting ERP patterns. Importantly, we show that, by factoring in the effects of spatiotemporal overlap between the N400 and P600 on the surface signal, which we reveal using rERP analysis, apparent inconsistencies in previous findings are easily reconciled, enabling us to draw unambiguous conclusions about the functional interpretation of the N400 and P600 components. Overall, our results provide compelling evidence that the N400 reflects lexical retrieval processes, while the P600 indexes compositional integration of word meaning into the unfolding utterance interpretation.

A Novel Feature Points Tracking Algorithm in Terms of IMU-Aided Information Fusion

Feature tracking plays a vital role in a monocular visual-inertial system (VINS) or a visual task based on feature points. However, in terms of feature points tracking, most of the existing VINS solutions adopt the classical method where the feature extraction and matching are carried out independently. Due to such nonintegrated working manner, the matching performs traversal operation globally rather than in a reasonable search space, which increases the probability of false matches, and reduces the accuracy. In this article, a novel feature points tracking algorithm in terms of inertial measurement unit (IMU)-aided information fusion is presented, which can reduce the search space to improve accuracy, and boost efficiency. This method starts with a preintegration-based predictor which can predict the position of the feature points in the current frame according to the feature points that need to be matched in the previous frame, and the measurements of IMU between two frames. Then, a variable-sized search window, in which the feature extraction and matching are locally carried out, is built at the predicted location. Furthermore, to solve the convergence and overlap problem of feature points in the tracking process, a feature update module of the bionic population is attached to the local matcher. Finally, the comparison experiments are performed on the public datasets to show the effectiveness and superiority of our method. It should be emphasized that the proposed algorithm is a universal framework of solution, which can meet various task requirements by choosing different feature points extraction algorithms, and improve the efficiency of matching in an integrated way.

Color-UNet++: A resolution for colorization of grayscale images using improved UNet++

Colorization is the computer-assisted application of color to a gray scale image, which presents two problems to modern deep learning-based approaches. One is to provide colorization models with both high expressibility and strong learning ability, as current models have difficulty both excelling at coloring and being easy to train. The other is to return a picture without uneven overlap. This paper proposes a deep convolutional network framework called Color-UNet++ for the end-to-end solution of these colorization problems. Color-UNet++ is adjusted to settle gradient dispersion and explosion by capturing more transfer and intermediate results during backpropagation. We adjust the de-convolution structure to solve the problem of uneven overlap. We design the model in YUV instead of RGB color space, with an objective function that is appropriate to the coloring problem and can capture a wide range of colors. A large number of experimental results on LFW and LSUN datasets confirm the method’s superiority.

ResCaps: an improved capsule network and its application in ultrasonic image classification of thyroid papillary carcinoma

Ultrasonic image examination is the first choice for the diagnosis of thyroid papillary carcinoma. However, there are some problems in the ultrasonic image of thyroid papillary carcinoma, such as poor definition, tissue overlap and low resolution, which make the ultrasonic image difficult to be diagnosed. Capsule network (CapsNet) can effectively address tissue overlap and other problems. This paper investigates a new network model based on capsule network, which is named as ResCaps network. ResCaps network uses residual modules and enhances the abstract expression of the model. The experimental results reveal that the characteristic classification accuracy of ResCaps3 network model for self-made data set of thyroid papillary carcinoma was 81.06%. Furthermore, Fashion-MNIST data set is also tested to show the reliability and validity of ResCaps network model. Notably, the ResCaps network model not only improves the accuracy of CapsNet significantly, but also provides an effective method for the classification of lesion characteristics of thyroid papillary carcinoma ultrasonic images.

Remote Intimacy

Remote Intimacy

Non-overlapping coverage in random feeding

Can random deposition create dense non-overlapping material feeding? The question is very fundamental for the research of particle packing, while the answer is of great importance for any industrial process that applies single object operation. To gain an insight into this issue, we studied the overlap problems of convex particles in the manner of uniformly random deposition. The overlap probability of two convex particles with arbitrary shapes and sizes is formulated, and the coverage fractions of free particles and sticking particles (particles of the bottom layer) are precisely predicted. Simulations with rectangular particles verified the theory. Surprisingly, free particles can only occupy less than 7.5% of the plane area, much smaller than what is intuitively expected. Sticking particles, however, can easily cover 19%, a factor of 2.5 times larger. The finding is of great value for applications that need to create dense non-overlapping feeding.

A Reliable Separation Algorithm of ADS-B Signal Based on Time Domain

With the development of Aerospace Science and technology, automatic dependent surveillance-broadcast has become a core technique in the field of air surveillance. Installing ADS-B receiver on LEO satellite can solve the problem of small coverage of ground receiver, to realize global coverage and monitoring. However, the satellite ADS-B system is faced with serious collision and overlap problems, which has a serious impact on the signal decoding, leading to the wrong decoding or even loss of important information. In this paper, a time-domain ADS-B blind signal separation algorithm is proposed. When there is a certain power difference between the two source signals, the overlap signals are offset by the high-power signal and low-power signal to get the corresponding cancellation signals. According to the superposition mode of different pulses, different bit decision results are obtained according to the amplitude, to recover the source signal. Simulations demonstrate that the proposed algorithm is feasible and has a lower bit error rate.

Hierarchical Annotation Event Extraction Method in Multiple Scenarios

In the event extraction task, considering that there may be multiple scenarios in the corpus and an argument may play different roles under different triggers, the traditional tagging scheme can only tag each word once, which cannot solve the problem of argument overlap. A hierarchical tagging pipeline model for Chinese corpus based on the pretrained model Bert was proposed, which can obtain the relevant arguments of each event in a hierarchical way. The pipeline structure is selected in the model, and the event extraction task is divided into event trigger classification and argument recognition. Firstly, the pretrained model Bert is used to generate the feature vector and transfer it to bidirectional gated recurrent unit+conditional random field (BiGRU+CRF) model for trigger classification; then, the marked event type features are spliced into the corpus as known features and then passed into BiGRU+CRF for argument recognition. We evaluated our method on DUEE, combined with data enhancement and mask operation. Experimental results show that our method is improved compared with other baselines, which prove the effectiveness of the model in Chinese corpus.

Bloom’s Learning Outcomes’ Automatic Classification Using LSTM and Pretrained Word Embeddings

Bloom’s taxonomy is a popular model to classify educational learning objectives into different learning levels for three domains including cognitive, affective and psycho motor. Each domain is further detailed into different levels. The cognitive domain includes knowledge, comprehension, application, analysis, synthesis and evaluation levels. In educational institutions, designing course learning outcomes (CLOs) as per different levels of Bloom and mapping of assessment items on designed CLOs is an important task — every semester, faculty and administrators read thousands of statements to complete the tedious task of such mapping of CLOs and assessment items into Bloom’s levels for an improved student learning. This paper proposes LSTM based deep learning model to perform classification of CLOs and assessment items in different levels of Bloom in cognitive domain. Although, there has been some attempts in the literature to automatically assign Bloom’s taxonomy category using keywords-based approach but it suffers from the problem of low accuracy and overlapping of keywords. Initially, when we performed keywords-based approach on our datasets we achieved an overall accuracy of 55% for classification of CLOs and assessment items into Bloom’s taxonomy. The proposed model predicts Bloom’s level for CLO and assessment question item, respectively. The proposed model is simple in terms of the architecture as compared to other deep learning models reported in literature and achieves classification accuracy of 87% and 74% on CLOs and assessment question items, respectively. The proposed model obtained 3% increase in overall accuracy comparing to an existing study for the same task. To the best of our knowledge, this is first attempt towards applying deep learning on classifying educational objectives in Bloom’s levels.

Road Obstacle Object Detection Based on Improved YOLO V4

Abstract In recent years, as one of the important technical tasks in the field of deep learning, object detection has broad prospects and applications in the field of road obstacle detection. However, in the real driving scene, there are many obstacles, serious occlusion, overlap and other problems, so that the existing obstacle detection algorithm can not effectively detect the obstacles on the road, so it can not guarantee the driving safety. In order to solve the above problems, this paper improves on the basis of Yolo V4 algorithm. Firstly, kmeans + + clustering is used to generate a priori box suitable for the data set to enhance the scale adaptability; Then, the ciou is used as the loss function of coordinate prediction to evaluate the coincidence degree of prediction frame and truth value frame more reasonably. Finally, a suitable target detection data set is constructed by preprocessing the public data set cityccaps. The experimental results show that the improved algorithm can achieve more than 90% accuracy for obstacles with large number of targets in the training set. Compared with the original Yolo V4, the average detection accuracy of the improved algorithm is improved by 2.03%.