Supporting Event-based Geospatial Anomaly Detection with Geovisual Analytics

2015 ◽  
Author(s):  
Orland Hoeber ◽  
Monjur Ul Hasan
Keyword(s):  
Anomaly Detection ◽  
Geovisual Analytics ◽  
Event Based

A geovisual analytics approach for analyzing event-based geospatial anomalies within movement data

2017 ◽  
Vol 17 (2) ◽  
pp. 91-107 ◽  
Author(s):  
Orland Hoeber ◽  
Monjur Ul Hasan
Keyword(s):  
Data Entry ◽  
Visual Exploration ◽  
Simple Task ◽  
Geographic Ranges ◽  
Movement Data ◽  
Geovisual Analytics ◽  
Monitoring And Enforcement ◽  
Exploratory Data ◽  
Event Based ◽  
Temporal Granularity

Comparing data collected on the movement of an entity to data on the location where the entity was reported to have been can be useful in monitoring and enforcement situations. Anomalies between these datasets may be indicative of illegal activity, systematic reporting errors, data entry errors, or equipment failure. While finding obvious anomalies may be a simple task, the discovery of more subtle inconsistencies can be challenging when there is a mismatch in the temporal granularity between the datasets, or when they cover large temporal and geographic ranges. We have developed a geovisual analytics approach called Visual Exploration of Movement-Event Anomalies (VEMEA) that automatically extracts potential anomalies from the data, visually encodes these on a map, and provides interactive filtering and exploration tools to allow expert analysts to investigate and evaluate the anomalies. Using two case studies from the fisheries enforcement domain, the value of VEMEA is illustrated for both confirmatory and exploratory data analysis tasks. Field trial evaluations conducted with expert fisheries data analysts further support the benefits of the approach.

scholarly journals EvAn: Neuromorphic Event-Based Sparse Anomaly Detection

2021 ◽  
Vol 15 ◽  
Author(s):  
Lakshmi Annamalai ◽  
Anirban Chakraborty ◽  
Chetan Singh Thakur
Keyword(s):  
Deep Learning ◽  
Anomaly Detection ◽  
Dynamic Range ◽  
High Dynamic Range ◽  
Careful Analysis ◽  
Motion Blur ◽  
Generative Adversarial Network ◽  
Adversarial Network ◽  
Unsupervised Deep Learning ◽  
Event Based

Event-based cameras are bio-inspired novel sensors that asynchronously record changes in illumination in the form of events. This principle results in significant advantages over conventional cameras, such as low power utilization, high dynamic range, and no motion blur. Moreover, by design, such cameras encode only the relative motion between the scene and the sensor and not the static background to yield a very sparse data structure. In this paper, we leverage these advantages of an event camera toward a critical vision application—video anomaly detection. We propose an anomaly detection solution in the event domain with a conditional Generative Adversarial Network (cGAN) made up of sparse submanifold convolution layers. Video analytics tasks such as anomaly detection depend on the motion history at each pixel. To enable this, we also put forward a generic unsupervised deep learning solution to learn a novel memory surface known as Deep Learning (DL) memory surface. DL memory surface encodes the temporal information readily available from these sensors while retaining the sparsity of event data. Since there is no existing dataset for anomaly detection in the event domain, we also provide an anomaly detection event dataset with a set of anomalies. We empirically validate our anomaly detection architecture, composed of sparse convolutional layers, on this proposed and online dataset. Careful analysis of the anomaly detection network reveals that the presented method results in a massive reduction in computational complexity with good performance compared to previous state-of-the-art conventional frame-based anomaly detection networks.

Anomaly Detection in Sensor Data

2018 ◽  
Vol 18 (1) ◽  
pp. 20-32 ◽  
Author(s):  
Jong-Min Kim ◽  
Jaiwook Baik
Keyword(s):  
Anomaly Detection ◽  
Sensor Data
Sign in / Sign up

Export Citation Format

Share Document