scholarly journals Anomalous Event Recognition in Videos Based on Joint Learning of Motion and Appearance with Multiple Ranking Measures

2021 ◽  
Vol 11 (3) ◽  
pp. 1344
Author(s):  
Shikha Dubey ◽  
Abhijeet Boragule ◽  
Jeonghwan Gwak ◽  
Moongu Jeon
Keyword(s):  
Moving Objects ◽  
Area Under The Curve ◽  
Multiple Instance Learning ◽  
Context Dependency ◽  
False Alarms ◽  
Joint Learning ◽  
Flow Maps ◽  
Anomalous Event ◽  
Ranking Measures ◽  
Motion Flow

Given the scarcity of annotated datasets, learning the context-dependency of anomalous events as well as mitigating false alarms represent challenges in the task of anomalous activity detection. We propose a framework, Deep-network with Multiple Ranking Measures (DMRMs), which addresses context-dependency using a joint learning technique for motion and appearance features. In DMRMs, the spatial-time-dependent features are extracted from a video using a 3D residual network (ResNet), and deep motion features are extracted by integrating the motion flow maps’ information with the 3D ResNet. Afterward, the extracted features are fused for joint learning. This data fusion is then passed through a deep neural network for deep multiple instance learning (DMIL) to learn the context-dependency in a weakly-supervised manner using the proposed multiple ranking measures (MRMs). These MRMs consider multiple measures of false alarms, and the network is trained with both normal and anomalous events, thus lowering the false alarm rate. Meanwhile, in the inference phase, the network predicts each frame’s abnormality score along with the localization of moving objects using motion flow maps. A higher abnormality score indicates the presence of an anomalous event. Experimental results on two recent and challenging datasets demonstrate that our proposed framework improves the area under the curve (AUC) score by 6.5% compared to the state-of-the-art method on the UCF-Crime dataset and shows AUC of 68.5% on the ShanghaiTech dataset.

scholarly journals Detection of signals of moving objects based on the time selection method

2021 ◽  
Vol 66 (3) ◽  
pp. 335-342
Author(s):  
V. M. Artemiev ◽  
S. M. Kostromitsky ◽  
A. O. Naumov
Keyword(s):  
Moving Objects ◽  
Average Duration ◽  
Threshold Level ◽  
Selection Method ◽  
Probability Of Detection ◽  
False Alarms ◽  
Threshold Method ◽  
Detection Of Signals ◽  
Two Stages ◽  
Time Selection

To increase the efficiency of detecting moving objects in radiolocation, additional features are used, associated with the characteristics of trajectories. The authors assumed that trajectories are correlated, that allows extrapolation of the coordinate values taking into account their increments over the scanning period. The detection procedure consists of two stages. At the first, detection is carried out by the classical threshold method with a low threshold level, which provides a high probability of detection with high values of the probability of false alarms. At the same time uncertainty in the selection of object trajectory embedded in false trajectories arises. Due to the statistical independence of the coordinates of the false trajectories in comparison with the correlated coordinates of the object, the average duration of the first of them is less than the average duration of the second ones. This difference is used to solve the detection problem at the second stage based on the time-selection method. The obtained results allow estimation of the degree of gain in the probability of detection when using the proposed method.

scholarly journals IoT Based Human Intrusion Detection System using Lab View

2019 ◽  
Vol 8 (6S4) ◽  
pp. 557-560 ◽  
Keyword(s):  
Intrusion Detection ◽  
Intrusion Detection System ◽  
Moving Objects ◽  
Detection System ◽  
Modern World ◽  
False Alarms ◽  
Recent Past ◽  
Gray Scale ◽  
Intruder Detection ◽  
The Status

Recent times there is an increase in thefts in the recent past. This creates a very bad environment for people to live in fear. The problem with home security in the modern world is a cause for concern. The conventional intruder detection system now we are using are highly expensive and there can be a possibility of false alarms. This problem is fixed by building a home intruder detection system that can accurately detect a human intruder, while filtering out movements that are caused due to any other moving objects using LabVIEW and Python. The images that were acquired and analyzed through frame comparisons are converted to gray scale images and then processed to detect an intruder. Here LabVIEW works as server and Python works as a client. At client video is acquired continuously, video is converted into images. Images are processed and information is send to the server. Server displays the status of the intruder with date and time. If the intruder is present then the system compares the intruder’s data with the data in the system. The images that were acquired and analyzed through frame comparisons converted to gray scale images that represent change, and then filtered through a series of image refining VI’s, helping to enhance our change detection results. If the data matches then the processing stops, if not then the system alerts the user through SMS or email if any intruder has been detected and sends the image to the user app. Through app we can make an alarm.

scholarly journals Artificial Intelligence Sepsis Prediction Algorithm Learns to Say "I don't know"

2021 ◽  
Author(s):  
Supreeth P. Shashikumar ◽  
Gabriel Wardi ◽  
Atul Malhotra ◽  
Shamim Nemati
Keyword(s):  
Area Under The Curve ◽  
The United States ◽  
Prediction Algorithm ◽  
False Alarms ◽  
High Area ◽  
Life Saving ◽  
Sequential Prediction ◽  
Sepsis Risk ◽  
Deep Learning Model ◽  
Timely Administration

Sepsis is a leading cause of morbidity and mortality worldwide. Early identification of sepsis is important as it allows timely administration of potentially life-saving resuscitation and antimicrobial therapy. We present COMPOSER (COnformal Multidimensional Prediction Of SEpsis Risk), a deep learning model for the early prediction of sepsis, specifically designed to reduce false alarms by detecting unfamiliar patients/situations arising from erroneous data, missingness, distributional shift and data drifts. COMPOSER flags these unfamiliar cases as 'indeterminate' rather than making spurious predictions. Six patient cohorts (515,720 patients) curated from two healthcare systems in the United States across intensive care units (ICU) and emergency departments (ED) were used to train and externally and temporally validate this model. In a sequential prediction setting, COMPOSER achieved a consistently high area under the curve (AUC) (ICU: 0.925-0.953; ED: 0.938-0.945). Out of over 6 million prediction windows roughly 20% and 8% were identified as 'indeterminate' amongst non-septic and septic patients, respectively. COMPOSER provided early warning within a clinically actionable timeframe (ICU: 12.2 [3.2 22.8] and ED: 2.1 [0.8 4.5] hours prior to first antibiotics order) across all six cohorts, thus allowing for identification and prioritization of patients at high risk for sepsis.

scholarly journals METHODS OF IMAGE RECOGNITION IN A VIDEO STREAM

2021 ◽  
Vol 8 (1) ◽  
pp. 042-049
Author(s):  
D. I. Ivanov ◽  
Keyword(s):  
Video Processing ◽  
Moving Objects ◽  
Video Stream ◽  
Multiple Instance Learning ◽  
Learning Tools ◽  
Tracking Accuracy ◽  
Correlation Filters ◽  
Advantages And Disadvantages ◽  
Kernelized Correlation Filters ◽  
Tracking Objects

The article examines the problem of automatic object recognition using a video stream as a digital image. Algorithms for recognizing and tracking objects in the video stream are considered, methods used in video processing are analyzed, and the use of machine learning tools in working with video is described.The main approaches to solving the problem of recognizing moving objects in a video stream are investigated: the detection-based approach and the tracking-based approach. Arguments are made in favor of the tracking-based approach, and, in addition, modern methods of tracking objects in the video stream are considered. In particular, the algorhythms: Online Boosting Tracker - one of the first object tracking algorithms with high tracking accuracy, MIL Tracker (Multiple Instance Learning Tracker), which is a development of the idea of learning with a teacher and the Online Boosting algorithm and the KCF Tracker algorithm (Kernelized Correlation Filters Tracker) - a method that uses the mathematical properties of overlapping areas of positive examples.As a result, the advantages and disadvantages of the considered methods and algorithms for recognizing and tracking objects for various applications are highlighted.

scholarly journals Supervised Distance-Based Feature Selection for Hyperspectral Target Detection

2019 ◽  
Vol 11 (17) ◽  
pp. 2049 ◽  
Author(s):  
Moeini Rad ◽  
Abkar ◽  
Mojaradi
Keyword(s):  
Target Detection ◽  
Area Under The Curve ◽  
Image Clustering ◽  
False Alarms ◽  
Detection Accuracy ◽  
Operating Characteristics ◽  
Autocorrelation Matrix ◽  
Target Energy ◽  
Roc Area ◽  
The Impact

Feature/band selection (FS/BS) for target detection (TD) attempts to select features/bands that increase the discrimination between the target and the image background. Moreover, TD usually suffers from background interference. Therefore, bands that help detectors to effectively suppress the background and magnify the target signal are considered to be more useful. In this regard, three supervised distance-based filter FS methods are proposed in this paper. The first method is based on the TD concept. It uses the image autocorrelation matrix and the target signature in the detection space (DS) for FS. Features that increase the first-norm distance between the target energy and the mean energy of the background in DS are selected as optimal. The other two methods use background modeling via image clustering. The cluster mean spectra, along with the target spectrum, are then transferred into DS. Orthogonal subspace projection distance (OSPD) and first-norm distance (FND) are used as two FS criteria to select optimal features. Two datasets, HyMap RIT and SIM.GA, are used for the experiments. Several measures, i.e., true positives (TPs), false alarms (FAs), target detection accuracy (TDA), total negative score (TNS), and the receiver operating characteristics (ROC) area under the curve (AUC) are employed to evaluate the proposed methods and to investigate the impact of FS on the TD performance. The experimental results show that our proposed FS methods, as compared with five existing FS methods, have improving impacts on common target detectors and help them to yield better results.

scholarly journals Remote sensing detection enhancement

2021 ◽  
Vol 8 (1) ◽  
Author(s):  
Tian J. Ma
Keyword(s):  
Remote Sensing ◽  
Moving Objects ◽  
Signal To Noise Ratio ◽  
Detection Threshold ◽  
False Alarms ◽  
Camera Motion ◽  
Low Contrast ◽  
Low Snr ◽  
Number Of False Alarms ◽  
Remote Sensor

AbstractBig Data in the area of Remote Sensing has been growing rapidly. Remote sensors are used in surveillance, security, traffic, environmental monitoring, and autonomous sensing. Real-time detection of small moving targets using a remote sensor is an ongoing, challenging problem. Since the object is located far away from the sensor, the object often appears too small. The object’s signal-to-noise-ratio (SNR) is often very low. Occurrences such as camera motion, moving backgrounds (e.g., rustling leaves), low contrast and resolution of foreground objects makes it difficult to segment out the targeted moving objects of interest. Due to the limited appearance of the target, it is tough to obtain the target’s characteristics such as its shape and texture. Without these characteristics, filtering out false detections can be a difficult task. Detecting these targets, would often require the detector to operate under a low detection threshold. However, lowering the detection threshold could lead to an increase of false alarms. In this paper, the author will introduce a new method that improves the probability to detect low SNR objects, while decreasing the number of false alarms as compared to using the traditional baseline detection technique.

scholarly journals A transferrable and interpretable multiple instance learning model for microsatellite instability prediction based on histopathology images

2020 ◽  
Author(s):  
Rui Cao ◽  
Fan Yang ◽  
Si-Cong Ma ◽  
Li Liu ◽  
Yan Li ◽  
...  
Keyword(s):  
Deep Learning ◽  
Microsatellite Instability ◽  
Transfer Learning ◽  
Area Under The Curve ◽  
Learning Model ◽  
Multiple Instance Learning ◽  
The Cancer Genome Atlas ◽  
Data Set ◽  
Histopathology Images ◽  
Deep Learning Model

ABSTRACTBackgroundMicrosatellite instability (MSI) is a negative prognostic factor for colorectal cancer (CRC) and can be used as a predictor of success for immunotherapy in pan-cancer. However, current MSI identification methods are not available for all patients. We propose an ensemble multiple instance learning (MIL)-based deep learning model to predict MSI status directly from histopathology images.DesignTwo cohorts of patients were collected, including 429 from The Cancer Genome Atlas (TCGA-COAD) and 785 from a self-collected Asian data set (Asian-CRC). The initial model was developed and validated in TCGA-COAD, and then generalized in Asian-CRC through transfer learning. The pathological signatures extracted from the model are associated with genotypes for model interpretation.ResultsA model called Ensembled Patch Likelihood Aggregation (EPLA) was developed in the TCGA-COAD training set based on two consecutive stages: patch-level prediction and WSI-level prediction. The EPLA model achieved an area-under-the -curve (AUC) of 0.8848 in the TCGA-COAD test set, which outperformed the state-of-the-art approach, and an AUC of 0.8504 in the Asian-CRC after transfer learning. Furthermore, the five pathological imaging signatures identified using the model are associated with genomic and transcriptomic profiles, which makes the MIL model interpretable. Results show that our model recognizes pathological signatures related to mutation burden, DNA repair pathways, and immunity.ConclusionOur MIL-based deep learning model can effectively predict MSI from histopathology images and are transferable to a new patient cohort. The interpretability of our model by association with genomic and transcriptomic biomarkers lays the foundation for prospective clinical research.

Detection of Atrial Fibrillation and Normal Sinus Rhythm Using Multiple Machine Learning Classifiers

2021 ◽  
Vol 11 (5) ◽  
pp. 1453-1462
Author(s):  
Fiaz Majeed ◽  
Muhammad Asim ◽  
Syed Ali Abbas ◽  
Abdul Jaleel ◽  
Abdul Majid ◽  
...  
Keyword(s):  
Atrial Fibrillation ◽  
Sinus Rhythm ◽  
Predictive Value ◽  
Normal Sinus Rhythm ◽  
Area Under The Curve ◽  
High Ratio ◽  
Classification Model ◽  
False Alarms ◽  
Normal Sinus ◽  
Statistical Measures

In this era of eHealth, healthcare data has gained a significant importance due to having human survival information. Detection of the atrial fibrillation (AF) from electrocardiogram (ECG) rhythm is a promising area of research because of its critical impact on mortality ratio in all over the world. Although, most of the studies have shown more than 90% results accuracy in terms of specificity (Sp) and sensitivity (Se), yet this accuracy is not sufficient and cannot be considered reliable for AF continuous monitoring due to high ratio of false alarms. Existing works scarcely compare the accuracy of the results generated by a classifier with those of other robust classifiers. Further, the results are needed to be verified with more statistical measures. In this paper, a multiple classifiers-based model is proposed in which the transition between normal sinus rhythm (NSR) and AF are performed on the basis of ventricle activities of the heart. The proposed scheme first extracts several features that are pertinent to AF diagnosis from the ECG data. Later, the classification model categorizes rhythms into classes using statistical techniques. The experimental evaluation is performed on five datasets which include AF Challenge 2017 database, NSR database (NSRDB), NSR RR interval database (NSRDB-2), AF database (AFDB) and long-term AF database (LTAFDB). Based on verification of results using different measures, the proposed scheme outperforms in comparison to the existing systems in terms of area under the curve (AUC), Se, Sp, positive predictive value (PPV), accuracy (ACC), and negative predictive value (NPV) for all datasets. Specifically, the decision tree (DT) obtains 99% AUC, 92% Se and 98% Sp for the AF Challenge 2017 database, which are improved than parallel systems.

RICA-MD: A Refined ICA Algorithm for Motion Detection

2021 ◽  
Vol 17 (1s) ◽  
pp. 1-17
Author(s):  
Chao Zhang ◽  
Xiaopei Wu ◽  
Jianchao Lu ◽  
Xi Zheng ◽  
Alireza Jolfaei ◽  
...  
Keyword(s):  
Motion Detection ◽  
Moving Objects ◽  
Rapid Development ◽  
False Alarms ◽  
Analysis Method ◽  
Detection Algorithms ◽  
Complex Scenes ◽  
Statistical Analysis Method ◽  
Conventional Methods ◽  
Ica Algorithm

With the rapid development of various computing technologies, the constraints of data processing capabilities gradually disappeared, and more data can be simultaneously processed to obtain better performance compared to conventional methods. As a standard statistical analysis method that has been widely used in many fields, Independent Component Analysis (ICA) provides a new way for motion detection by extracting the foreground without precisely modeling the background. However, most existing ICA-based motion detection algorithms use only two-channel data for source separation and simply generate the observation vectors by decomposing and reconstructing the images by row, hence they cannot obtain an integrated and accurate shape of the moving objects in complex scenes. In this article, we propose a refined ICA algorithm for motion detection (RICA-MD), which fuses a larger number of channels than conventional ICA-based motion detection algorithms to provide more effective information for foreground extraction. Meanwhile, we propose four novel methods for generating observation vectors to further cover the diverse motion styles of the moving objects. These improvements enable RICA-MD to effectively deal with slowly moving objects, which are difficult to detect using conventional methods. Our quantitative evaluation in multiple scenes shows that our proposed method is able to achieve a better performance at an acceptable cost of false alarms.

Learning relevant H&E slide morphologies for prediction of colorectal cancer tumor mutation burden using weakly supervised deep learning.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e15244-e15244
Author(s):  
Rohan P. Joshi ◽  
Andrew J. Kruger ◽  
Lingdao Sha ◽  
Madhavi Kannan ◽  
Aly A. Khan ◽  
...  
Keyword(s):  
Colorectal Cancer ◽  
Deep Learning ◽  
High Performance ◽  
Cross Validation ◽  
Area Under The Curve ◽  
Multiple Instance Learning ◽  
Tumor Mutation Burden ◽  
Synonymous Mutations ◽  
Mutation Burden ◽  
Weakly Supervised

e15244 Background: Tumor mutation burden (TMB) is the number of non-synonymous mutations present in a cancer exome. In colorectal cancer (CRC), high TMB is associated with microsatellite instability (MSI), POLE mutations, and response to immunotherapy. TMB prediction from whole-slide images (WSIs) could aid workflows that determine MSI and POLE status. Deep learning has previously been used to predict MSI status from WSIs. This approach assumed the morphologies of all regions within the tumor are equally associated with MSI. Here, we predict TMB using a weakly supervised deep learning framework that relaxes this assumption and automatically learns relevant regions within the tumor that are most associated with TMB, potentially uncovering morphological associations. Methods: Weakly supervised learning methods facilitate classification of samples that contain many individual instances, only some of which are related to the sample label. Here, a given WSI has a single TMB-high or -low label and contains individual regions that may or may not be associated with TMB status. We implemented a ResNet18 attention-based, multiple-instance learning (MIL), convolutional neural network to simultaneously learn which tiles are important for prediction of the slide-level TMB and the tile features that are associated TMB-high and -low. We determined performance through 8-fold cross-validation within a Tempus dataset using a 75%-12.5%-12.5% split of ~940 WSIs for training, validation, and testing folds. Results: In the cross-validation, we observed a receiver operating characteristic area under the curve of 0.854 (95% CI 0.776-0.932), an average precision of 0.723 (95% CI 0.580-0.865), and an accuracy of 0.889 (95% CI 0.833-0.945) in the held-out test sets. Morphologies predicted as irrelevant for TMB include adipose tissue and WSI artifacts. Visualizations of model weights show morphologies determined to be most associated with TMB-high and -low, such as high tumor/lymphocyte content and vasculature/red blood cells, respectively. Conclusions: Attention-MIL shows high performance for the prediction of TMB in CRC from H&E images and potentially reveals the morphologies of CRC that are most associated with TMB. Future directions include further investigation of morphological associations, generalizing this model beyond Tempus acquired data, and re-training on the entire Tempus dataset.

Sign in / Sign up

Export Citation Format

Share Document