Spatial Distribution Of Features Research Articles

Efficient Crowd Anomaly Detection Using Sparse Feature Tracking and Neural Network

Crowd anomaly detection is crucial in enhancing surveillance and crowd management. This paper proposes an efficient approach that combines spatial and temporal visual descriptors, sparse feature tracking, and neural networks for efficient crowd anomaly detection. The proposed approach utilises diverse local feature extraction methods, including SIFT, FAST, and AKAZE, with a sparse feature tracking technique to ensure accurate and consistent tracking. Delaunay triangulation is employed to represent the spatial distribution of features in an efficient way. Visual descriptors are categorised into individual behaviour descriptors and interactive descriptors to capture the temporal and spatial characteristics of crowd dynamics and behaviour, respectively. Neural networks are then utilised to classify these descriptors and pinpoint anomalies, making use of their strong learning capabilities. A significant component of our study is the assessment of how dimensionality reduction methods, particularly autoencoders and PCA, affect the feature set’s performance. This assessment aims to balance computational efficiency and detection accuracy. Tests conducted on benchmark crowd datasets highlight the effectiveness of our method in identifying anomalies. Our approach offers a nuanced understanding of crowd movement and patterns by emphasising both individual and collective characteristics. The visual and local descriptors facilitate high-level analysis by closely relating to semantic information and crowd behaviour. The analysis observed shows that this approach offers an efficient framework for crowd anomaly detection, contributing to improved crowd management and public safety. The proposed model achieves accuracy of 99.5 %, 96.1%, 99.0% and 88.5% in the UMN scenes 1, 2, and 3 and violence in crowds datasets, respectively.

Feature disparity learning for weakly supervised object localization

Weakly supervised object localization (WSOL) aims to localize objects with only image-level labels. As a common WSOL method, adversarial erasing always masks the most discriminative region in the feature space to compel the network to localize more regions of the object. However, with the discriminative region vanishing, the localizer is confused when distinguishing the regions of object from the background. In this paper, we propose a new feature disparity learning (FDL), which encourages the network to learn more distinctive features from the object region with similarity measurement after feature enhancement. Specifically, we first introduce a Spatial Vector Cross Attention (SVCA) module. This module enhances responses in less discriminative region of erased feature maps by reintegrating the spatial distribution of features through the capture of interdependencies among spatial vectors on each channel. Furthermore, we propose a feature complementarity loss to measure the similarity between unerased features and erased features, guiding the network to learn feature disparities caused by adversarial erasing for improved localization and classification. Several experimental studies demonstrate a significant increase in localization performance over the existing state-of-the-art erasing methods on the CUB 200–2011 and ILSVRC 2016 datasets.

Big data-based method for automatic localization of power quality disturbance signal of PV access distribution network

At present, the conventional automatic localization method of the power quality disturbance signal of the distribution network mainly extracts the feature vector of the disturbance signal. It constructs the target localization function, which leads to poor localization accuracy because the noise part of the disturbance signal is ignored. In this regard, the automatic localization method of PV access power quality disturbance signal based on big data is proposed. By using the wavelet decomposition algorithm, the noise part of the power quality disturbance signal is removed, and the measurement matrix and reconstruction function are combined with compressing and reconstructing the disturbance signal. Finally, the automatic positioning of the disturbance signal is achieved by extracting the characteristics of the disturbance signal data and clustering analysis processing. In the experiment, the designed signal localization method is tested for the localization effect. The results can prove that when the proposed method is used to automatically localize the disturbed signal, the localization results are consistent with the spatial feature distribution of the signal and have a more desirable automatic localization effect.

Characterization of aperiodic surfaces with mesh-based parameters

For technical surfaces, it is important to know their functional purpose and to characterize them accordingly. Therefore, ISO 21920–2 in 2D and ISO 25178–2 in 3D offer parameters that can assess surface functional properties. The topographic portions of a surface, for example hills and dales, can be classified as features and evaluated using feature parameters. However, no parameter exists to describe the spatial distribution of features with regard to the degree of homogeneity for aperiodic surfaces. Here we show the application of the Delaunay triangulation to quantify the spatial distribution respectively the geometric relationship of features. Therefore, the feature points are determined by watershed analysis and the resulting point cloud is meshed in 2D. Based on that mean and standard deviation of the triangle side lengths and the area disorder (AD) are calculated as new parameters. The method is demonstrated for sandblasted and chrome-plated specimens. In addition simulation is used to generate more data for analysis. With the proposed approach the distinction and extent of uniform, homogeneous or inhomogeneous spatial distributions of features with parameter AD can be determined.

Spatial Proximity Feature Selection With Residual Spatial–Spectral Attention Network for Hyperspectral Image Classification

Over the past few years, deep learning has been introduced to tackle hyperspectral image (HSI) classification and demonstrated good performance. In particular, the convolutional neural network (CNN) based methods have progressed. However, due to the high dimensionality of HSI and equal treatment of all bands, the performances of CNN based methods are hampered. The labels of land-covers often differ between edge and the center pixels in pixel-centered spatial information. These edge pixels may weaken the discrimination of spatial features and reduce classification accuracy. Motivated by the attention mechanism of the human visual system, the spatial proximity feature selection with residual spatial–spectral attention network is proposed in this article. It contains a residual spatial attention module, a residual spectral attention module, and a spatial proximity feature selection module. The residual spatial attention module aims to select the crucial spatial information, which assigns weights to different features by measuring the similarity between the surrounding elements and their central ones. The residual spectral attention module is designed for spectral bands which are selected from raw input data by emphasizing the valuable bands and suppressing the valueless. According to the spatial distribution of features, the spatial proximity feature selection module is used to filter features effectively. Experiments on three public data sets demonstrate that the proposed network outperforms the state-of-the-art methods in comparison.

TCD: Task-Collaborated Detector for Oriented Objects in Remote Sensing Images

Oriented object detection in remote sensing image interpretation is challenging due to the difficulty of locating objects with arbitrary orientations. Existing methods have made considerable progress based on oriented heads or anchors. However, most of them follow the classical detection paradigm, such as assigning samples based on Intersection-over-Unions (IoU) and predicting through two independent tasks. These fixed strategies impair the consistency between classification and localization predictions, resulting in the prediction with optimal localization accuracy being suppressed by the non-optimal ones during Non-Maximum Suppression (NMS). In order to address this problem, a Task-Collaborated Detector (TCD) is proposed. Compared with current single-stage methods, its improvements include two aspects: Task-Collaborated Assignment (TCA) and Task-Collaborated Head (TCH). Specifically, in order to better pull closer the best anchors for two tasks, TCA introduces classification and localization confidence into sample assignment and tends to select the anchors with accurate and consistent predictions as positive during training. TCH provides a better balance for learning interactive and discriminative features. It can flexibly adjust the spatial feature distribution of classification and localization tasks by learning the joint features from the aggregation layer. Extensive experiments are conducted on HRSC2016, DOTA, and DIOR-R, and the proposed TCD achieves state-of-the-art performance (90.60, 80.89, and 65.04 mAP, respectively). Consistency analysis also demonstrates that TCD can significantly improve prediction consistency.

Towards a Whole Sample Imaging Approach Using Diffusion Tensor Imaging to Examine the Foreign Body Response to Explanted Medical Devices.

Analysing the composition and organisation of the fibrous capsule formed as a result of the Foreign Body Response (FBR) to medical devices, is imperative for medical device improvement and biocompatibility. Typically, analysis is performed using histological techniques which often involve random sampling strategies. This method is excellent for acquiring representative values but can miss the unique spatial distribution of features in 3D, especially when analysing devices used in large animal studies. To overcome this limitation, we demonstrate a non-destructive method for high-resolution large sample imaging of the fibrous capsule surrounding human-sized implanted devices using diffusion tensor imaging (DTI). In this study we analyse the fibrous capsule surrounding two unique macroencapsulation devices that have been implanted in a porcine model for 21 days. DTI is used for 3D visualisation of the microstructural organisation and validated using the standard means of fibrous capsule investigation; histological analysis and qualitative micro computed tomography (microCT) and scanning electron microscopy (SEM) imaging. DTI demonstrated the ability to distinguish microstructural differences in the fibrous capsules surrounding two macroencapsulation devices made from different materials and with different surface topographies. DTI-derived metrics yielded insight into the microstructural organisation of both capsules which was corroborated by microCT, SEM and histology. The non-invasive characterisation of the integration of implants in the body has the potential to positively influence analysis methods in pre-clinical studies and accelerate the clinical translation of novel implantable devices.

Histogram Layers for Texture Analysis

An essential aspect of texture analysis is the extraction of features that describe the distribution of values in local, spatial regions. We present a localized histogram layer for artificial neural networks. Instead of computing global histograms as done previously, the proposed histogram layer directly computes the local, spatial distribution of features for texture analysis and parameters for the layer are estimated during backpropagation. We compare our method with state-of-the-art texture encoding methods such as the Deep Encoding Network Pooling, Deep Texture Encoding Network, Fisher Vector convolutional neural network, and Multi-level Texture Encoding and Representation on three material/texture datasets: (1) the Describable Texture Dataset; (2) an extension of the ground terrain in outdoor scenes; (3) and a subset of the Materials in Context dataset. Results indicate that the inclusion of the proposed histogram layer improves performance. The source code for the histogram layer is publicly available: https://github.com/GatorSense/Histogram_Layer.

Detection of spreading depolarization events and spatiotemporal analysis for advancing stroke therapy.

While the presence of spreading depolarization (SD) and associated spreading depression have been well studied and known to be associated with post-ischemic brain damage, the spatiotemporal spread of these events from the site of injury is not well understood. With the recent development of high-density micro-electrocorticographic (ECoG) electrode arrays, monitoring the spread of the depolarizing events and associated depression is possible. The goal of this work is to define the electrocorticographic features of SD and associated depression across the multichannel array and search for patterns in these features that emerge across both space and time. We present the spatial distribution of features found from chronic ECoG recordings acquired from awake behaving rats induced with a rodent model of stroke. SD events were detected with an unsupervised algorithm that searched for a stereotyped pattern in the first derivative of the ECoG. The algorithm yielded a 58% correct detection rate on average across four rats, and a 36% false positive rate. We defined key electrophysiological features and mapped them onto the physical brain regions using MATLAB, such as the peak-to-peak amplitude of each SD event, the width (or duration) of the SD event, direct current (DC) level, and average rate of decline in the signal baseline. We performed k-means clustering to the activity in this feature space which yielded three contiguous regions in physical space. The elbow optimization method was applied to a distortion metric and indicated that 3 clusters was optimal. These findings motivate us to conduct future studies that would verify whether these 3 clusters in electrode-space correspond to immunohistochemically defined regions of tissue health, namely, infarct, penumbra, and healthy tissue. Clinical Relevance- The extent and severity of damage that stroke ultimately causes is suspected to be related to the progression of spreading depolarization and associated depression. An understanding of how the features of these electrophysiological events progress across the brain and over time is an important step toward eventual development of closed-loop therapies which limit and minimize the long-term effects of stroke.

Controlled Glioma Cell Migration and Confinement Using Biomimetic‐Patterned Hydrogels

Cancer cell migration is significantly influenced by the physical properties (e.g., mechanics, topography) of the local extracellular matrix (ECM). However, how cells guide their movement in a complex, heterogeneous landscape while interpreting various physical factors, especially those with sizes on the subcellular length scale, is not completely understood. Herein, a photodegradable hydrogel is fabricated with control over the spatial distribution of features with different physical characteristics. On the surface of this hydrogel, patterns of features with sizes and spacings on the subcellular to cellular length scale are generated using a digital micromirror device, and the influence of the organization of these patterns on cellular migratory behavior is investigated. The size and spatial arrangement of hard (or stiff) and soft domains, more so than their topography, is found to have a significant impact on the migration speed. Using this approach, the maximum migration speed is observed on hydrogels with 10 μm pattern sizes spaced 10 μm apart. In this regime, the cells are confined and interact with both hard and soft regions at their periphery. Together, this study of cell–ECM interactions at the interface of hard and soft domains provides insights into how cancer cells interact with their environment.

Location Matters: A Framework to Investigate the Spatial Characteristics of Distributed Flood Attenuation

Distributed attenuation in flood management relies on small and low-impact runoff attenuating features variously distributed within a catchment. Distributed systems of reservoirs, natural flood management, and green infrastructure are practical examples of distributed attenuation. The effectiveness of attenuating features lies in their ability to work in concert, by reducing and slowing runoff in strategic parts of the catchment, and desynchronizing flows. The spatial distribution of attenuating features plays an essential role in the process. This article proposes a framework to place features in a hydrologic network, group them into spatially distributed systems, and analyze their flood attenuation effects. The framework is applied to study distributed systems of reservoirs in a rural watershed in Iowa, USA. The results show that distributed attenuation can be an effective alternative to a single centralized flood mitigation approach. The different flow peak attenuation of considered distributed systems suggest that the spatial distribution of features significantly influences flood magnitude at the catchment scale. The proposed framework can be applied to examine the effectiveness of distributed attenuation, and its viability as a widespread flood attenuation strategy in different landscapes and at multiple scales.

Surface finish classification using depth camera data

We propose a novel approach for surface finish classification of digitally fabricated structures using an industrial depth camera. Data collected at different viewpoints are jointly processed to derive the spatial distribution of features describing the reflectance, which is in turn related to the surface finish. The features can be used to classify the surfaces according to their finish e.g., for assessing the homogeneity or conformance. We apply the method to four sprayed plaster specimens of similar visual appearance but different roughness. Using nearest neighbor classification we achieve an accuracy of 97% for the plaster samples. The approach is a contribution towards real-time quality inspection in digital fabrication.

Interactive Estimation of Heterogeneity from Mass Spectrometry Imaging.

In this work, we demonstrate a new approach for interactively assessing hyperspectral data spatial structures for heterogeneity using mass spectrometry imaging. This approach is based on the visualization of the cosine distance as the similarity levels between mass spectra of a chosen region and the rest of the image (sample). The applicability of the method is demonstrated on a set of mass spectrometry images of frontal mouse brain slices. Selection of the reference pixel of the mass spectrometric image and a further view of the corresponding cosine distance map helps to prepare supporting vectors for further analysis, select features, and carry out biological interpretation of different tissues in the mass spectrometry context with or without histological annotation. Visual inspection of the similarity maps reveals the spatial distribution of features in tissue samples, which can serve as the molecular histological annotation of a slide.

A local feature extraction method for UAV-based image registration based on virtual line descriptors

Image local feature extraction is extensively utilized in the field of photogrammetry where the spatial distribution of features is important in high-quality image matching, particularly in high-resolution unmanned aerial vehicle (UAV) image registration. Presently, the spatial distribution problems are considered in some local feature extraction methods, though these methods are designed for point descriptors. Line descriptors are more robust to repetitive patterns compared to point descriptors and have attracted extensive attention in recent years. Hence, a feature extraction method is designed in this paper for line descriptors based on the K-connected virtual line descriptors matching method. Using the four measures, the quality of local features is quantified, and a regular gridding strategy based on the quality of local features is applied in the feature selection procedure. The proposed feature extraction method was successfully applied to match various simulated and real UAV-based images. Based on the experimental results using real images, it is indicated that two evaluation criteria, namely the spatial distribution quality of features and the number of correct matches, are improved to at least 12% and 15%, respectively, for verifying the capability of the proposed method to enhance matching performance.

Target detection using features for sonar images

Robust object detection in sonar images is an important task for underwater exploration, navigation and mapping. Current methods make assumptions about the shape, highlight or shadow of an object, which may be invalid for some environments or targets. We focus on the area of feature extraction-based detection, which does not rely on information about the shape of the target, towards a robust framework for target detection for a variety of seabed structures and target types. The proposed framework first estimates the seabed type from the spatial distribution of features to determine the set of optimal parameters, and then obtains a set of features which are filtered according to intensity and distribution to yield a detection decision. The proposed method also provides a means to determine the seabed type, and a machine-learning based methodology to choose the feature detectors' parameters to match the evaluated seabed type. We report the performance of a variety of feature detectors for a simulated environment and of one feature detector for real sonar images. Results show the importance of choosing the parameters of the feature extractors based on the current environmental conditions and the proposed method obtains a favourable tradeoff between detection and false alarm rates.

A versatile technique for high-resolution three-dimensional imaging of human arterial segments using microcomputed tomography

BackgroundQuantitative methods for evaluating microstructure of arterial specimens typically rely on histologic techniques that involve random sampling, which cannot account for the unique spatial distribution of features in three dimensions. MethodsTo overcome this limitation, we demonstrate a nondestructive method for three-dimensional imaging of intact human blood vessels using microcomputed tomography (microCT). Human artery segments were dehydrated and stained in an iodine solution then imaged with a standard laboratory microCT scanner. Image visualization and segmentation was performed using commercially available and open source software. ResultsStaining of cadaveric vessels with iodine enabled clear visualization of the arterial wall with microCT, preserved tissue morphology, and generated high-resolution images with a voxel size of 5.4 μm. Various components of the arterial wall were segmented using a combination of manual and automatic thresholding algorithms. ConclusionsOur approach allows for spatial mapping of human artery tissue samples that can guide targeted histologic analysis of smaller tissue segments, provide geometric data to inform finite element models, quantify degree of atherosclerosis, and help to evaluate the foreign body response to intravascular medical implants. (JVS–Vascular Science 2021;2:13-19.) Clinical RelevanceIn this article, we describe a powerful technique for whole artery analysis of pathologic human tissue specimens that provides high-resolution spatial detail regarding composition of the blood vessel wall. The protocol described here is a valuable adjunct that can be used as a research tool to inform finite element modeling of arteries, quantify pathologic response (ie, neointimal hyperplasia and vascular calcification), and evaluate the tissue/device interface of implanted medical devices.

Similarity Analysis of 3D Structures of Proteins Based Tile-CNN

The 3D structure of a protein is closely related to its function, and the similarity analysis between their structures can help reveal the function of proteins. However, there exist two problems arising from the analysis of 3D structures of proteins. The proteins with a similar sequence may have different structures, while the proteins with a similar structure may have different sequences. In the analysis of similarity in 3D structures of proteins, it remains difficult for the traditional methods using the spatial feature distribution and geometry or topology features of proteins to solve these problems. In this paper, a Tile-CNN network is proposed to analyze the similarity of proteins in 3D structure. In order to capture the overall and the local features as exhibited by the 3D structures of proteins, it projects 3D protein models into 2D protein images from different views and then cuts these 2D projected images using the tile strategy. After the training of proteins with these images in the Tile-CNN, the test protein model can be expressed by an analysis matrix, and then the similarity between 3D structures of proteins is computed using the root mean square distance (RMSD) for the benchmark matrix and the analysis matrix. As revealed by the experimental results, the proposed algorithm is more robust in analyzing the similarity of 3D structures of proteins and produces a satisfactory performance in solving the two aforementioned problems.

Frequency dependence of attenuation and backscatter coefficient of ex vivo human lymphedema dermis.

Radio-frequency (RF) signals from the most dominant scatterer in a dermis, i.e., collagen fibers, are collected as backscattered signals. We aim to confirm the frequency dependence of the spatial distribution of features in ultrasound images, as well as the attenuation coefficient (AC) and backscatter coefficient (BSC) of skin tissue without [LE (-)] and with lymphedema [LE (+)]. Measurement samples (n = 13) were excised from human skin tissue with LE (-) and middle severity LE (+). A laboratory-made scanner and single-element concave transducers (range 9-47MHz) were used to measure RF data. A localized AC was computed from the normalized power spectrum using the linear least squares technique. The reflector method and compensation technique of the attenuation of tissue were applied to calculate the BSC. In addition, effective scatterer diameter (ESD), effective acoustic concentration (EAC), and integrated BSC (IBS) were calculated from the BSC as the benchmark to differentiate LE (-) and LE (+) tissues. High-frequency ultrasound displayed different echogenicity and texture compared between LE (-) and LE (+) in all transducers. The AC for LE (-) (0.22dB/mm/MHz) and LE (+) (0.29dB/mm/MHz) was comparable. BSC in LE (-) and LE (+) increased linearly with each transducer. The difference of intercept of the BSC between LE (-) and LE (+) indicated that both EAC and IBS of LE (+) were higher than that of LE (-). In contrast, ESD correlated with the slope of the BSC demonstrated the same tendency for both LE (-) and LE (+). These tendencies appeared for each transducer independent of the frequency bandwidth. Frequency independence of AC and BSC in LE (-) and LE (+) was confirmed. Several 9- to 19-MHz ultrasound beams are sufficient for BSC analysis to discriminate LE (-) and LE (+) in terms of the penetration depth of the ultrasound.

Cannabis consumers in Poland

The purpose of this study is an attempt to answer the following questions: (1) Is there a regional differentiation of cannabis consumers in Poland? (2) What is the scale of this differentiation? (3) What factors affect this differentiation? Our empirical study is based on the analysis of an anonymous online survey. The coefficient of variation, taxonomic measures of similarity, synthetic index and Spearman correlation coefficient are used in the analysis. It is observed that cannabis consumption in Poland is most popular among men, youth and young adults (Generation Y), as well as among those who are well-educated, relatively well-off and live in cities. The structure of selected demographic and socio-economic features of the respondents is poorly differentiated regionally. Regional differentiation in the structure of cannabis consumption frequency is at an average level. The spatial distribution of features characterizing the respondents is highly consistent with the spatial distribution of city dwellers aged 15–44, their sex and 5-year age ranges, the level of education, the size of cities and the level of gross domestic product.

Spatially Regularized Structural Support Vector Machine for Robust Visual Tracking

Structural support vector machine (SSVM) is popular in the visual tracking field as it provides a consistent target representation for both learning and detection. However, the spatial distribution of feature is not considered in standard SSVM-based trackers, therefore leading to limited performance. To obtain a robust discriminative classifier, this paper proposes a novel tracking framework that spatially regularizes SSVM, which yields a new spatially regularized SSVM (SRSSVM). We utilize the spatial regularization prior to penalize the learning classifier with the same size as the target region. The location of classifier spatially located far from the center of region is assigned large weight and vice versa. Then, it is introduced into the SSVM model as a regularization factor to learn the robust discriminative model. Furthermore, an optimizing algorithm with dual coordination descent is presented to efficiently solve the SRSSVM tracking model. Our proposed SRSSVM tracking method has low computational cost like the traditional linear SSVM tracker while can significantly improve the robustness of the discriminative classifier. The experimental results on three popular tracking benchmark data sets show that the proposed SRSSVM tracking method performs favorably against the state-of-the-art trackers.