In this paper, we describe two algorithms suitable for the detection of Atrial Fibrillation episodes in very long terms (weeks) ECG monitoring, were the need of onboard implementation requires the development of relia...
详细信息
In this paper, we describe two algorithms suitable for the detection of Atrial Fibrillation episodes in very long terms (weeks) ECG monitoring, were the need of onboard implementation requires the development of reliable but simple and easy-to-implement methods. The proposed algorithms are based on the extraction of simple geometric features from the histogram of RR prematurity and delta RR. On the MIT Atrial fibrillation database, the RR prematurity algorithm provides the following performances: episodes sensitivity (S) 91%, episode positive Predictivity (P+) 92%, duration S 93%, duration P+ 97%. For the delta-RR algorithm the results were: episodes S 92%, episode P+ 78%, duration S 89%, duration P+ 90%
Tree search based detection algorithms provide a promising approach to solve the detection problems in MIMO systems. Depth-first, Breadth-first or Metric-first search strategies provide near max-log detection at reduc...
详细信息
Tree search based detection algorithms provide a promising approach to solve the detection problems in MIMO systems. Depth-first, Breadth-first or Metric-first search strategies provide near max-log detection at reduced but still significant complexity. In this paper we show how the incurred complexity can be reduced substantially. In order to reduce the number of metric calculations to a minimum, we propose a novel relative determination of search sequences for QAM constellations, usable inexpensively independent of the underlying constellation size and search strategy and moreover also usable for soft-in soft-out detection. Based on its application to a sphere detector, we will demonstrate the impact on complexity and performance of the detection as well as on the detector structure. Building on the results, we propose refinements of the resulting detector providing a very good performance at minimized complexity, making the resulting detector particularly favorable for implementation.
Pattern detection algorithms may be used as part of safety-relevant processes employed by industrial systems. Current approaches to functional safety mainly focus on random faults in hardware and the avoidance of syst...
详细信息
Pattern detection algorithms may be used as part of safety-relevant processes employed by industrial systems. Current approaches to functional safety mainly focus on random faults in hardware and the avoidance of systematic faults in both software and hardware. In this paper we build on the concepts of the international standard for functional safety IEC 61508 to extend safety-relevant notions to numerical and logical processes (algorithms) employed in pattern detection systems. In particular, we target the uncertainty pertaining to face detection systems where incorrect detection affects the overall system performance. We discuss a dual channel system that comprises two of the most commonly used and widely available face detection algorithms, Viola-Jones and Kienzle et al. We present a method for deriving the probability of failure in demand (PFD) from the combination of these two channels using both: 1oo2 and 2oo2 voting schemes. Finally, we compare experimental results from both the perspectives of availability and safety, and present conclusions with respect to the appropriate choice of information combination schemes and system architectures.
This study focuses on development of the algorithms that can be applied in video-oculograph method to find a center of a pupil area in captured eye image, "where the pupil area is partially covered by eyelashes a...
详细信息
This study focuses on development of the algorithms that can be applied in video-oculograph method to find a center of a pupil area in captured eye image, "where the pupil area is partially covered by eyelashes and eyelids. The eye image captured by monochrome charge-coupled device camera and frame grabber has spatial resolution of 640 pixels by 480 pixels and 8-bit gray levels per pixel. We assumed that the pupil area in the captured eye image was a perfect circle. The center of pupil area could be obtained by applying the proposed algorithms for thresholding of captured gray-scale image, size filtering and noise removing techniques even though the pupil area was partially covered by eyelashes and eyelids.
In recent years, the number of images and videos shared online increased and people have easy ways to access such content. “DeepFake” refers to any multimedia content created using deep learning technology in order ...
详细信息
In recent years, the number of images and videos shared online increased and people have easy ways to access such content. “DeepFake” refers to any multimedia content created using deep learning technology in order to make it appear realistic. The creation of deepfake videos and images using deep learning techniques leads to very realistic “DeepFake” videos and images by changing the digital content of images and videos. Deepfake is widely recognized as one of artificial intelligence’s most dangerous uses. Deepfake makes it possible to place a person in a totally imaginary situation since it is used to imitate an activity that the person did not perform. Deepfakes have been becoming increasingly dangerous to democracy, society’s security and people’s privacy. The distribution of such deepfake content on various platforms urged the international community to revaluate the threat to social security posed by such content. It encouraged the researchers around the world to develop effective deepfake detection methods. In this paper we have discussed such approaches of deepfake detection in videos and images that are available in recent studies and have provided comparative review of research on deepfake detection algorithms. It also compares the different detection techniques and examines their limitations and advantages.
The great presumption of change detection has led to rapid development of diverse change detection algorithms. Unsupervised change detection has a vital role in a wide variety of applications like remote sensing, moti...
详细信息
The great presumption of change detection has led to rapid development of diverse change detection algorithms. Unsupervised change detection has a vital role in a wide variety of applications like remote sensing, motion detection, environmental monitoring, medical diagnosis, damage assessment, agricultural surveys, surveillance etc. In this paper a systematic survey of the commonly used methodologies for unsupervised change detection is presented.
Several pelwise motion detectors are reviewed in this paper. They are compared in the context of intrusion detection in indoor scenes. The presented motion detection algorithms are based on a pelwise detection of chan...
详细信息
Several pelwise motion detectors are reviewed in this paper. They are compared in the context of intrusion detection in indoor scenes. The presented motion detection algorithms are based on a pelwise detection of changes in the observed input frame with respect to a recursively updated background. The same global decision module is applied to the outputs of the respective pelwise change detectors. The overall performance is shown to depend on the type of the pelwise temporal filter, and on the image features applied to it.
Synthetic aperture interferometric radiometers require phase coherent detection to allow the visibility function to be measured and subsequently converted to brightness temperature images. Products of signals between ...
详细信息
Synthetic aperture interferometric radiometers require phase coherent detection to allow the visibility function to be measured and subsequently converted to brightness temperature images. Products of signals between channels produce the complex "visibility" terms of the spatial spectrum of the brightness temperature scene. Today's technology offers the choice of either analog or digital detection methods for relatively low microwave frequencies. Digital detection may offer advantages, particularly in certain instrument configurations. This paper reviews the results of an evaluation of a particular digital method and a variety of FIR filter designs. A programmable, real time processor was developed to study various Hilbert Transform algorithms for digitally forming the in-phase (I) and quadrature (Q) components of the received signals. An overview of the test set-up and test results using several FIR filters is presented, allowing a trade between numerical simplicity and the gain and phase balance between I and Q.
We investigate statistical anomaly detection algorithms for detecting SYN flooding, which is the most common type of denial of service (DoS) attack. The two algorithms considered are an adaptive threshold algorithm an...
详细信息
We investigate statistical anomaly detection algorithms for detecting SYN flooding, which is the most common type of denial of service (DoS) attack. The two algorithms considered are an adaptive threshold algorithm and a particular application of the cumulative sum (CUSUM) algorithm for change point detection. The performance is investigated in terms of the detection probability, the false alarm ratio, and the detection delay. Particular emphasis is on investigating the tradeoffs among these metrics and how they are affected by the parameters of the algorithm and the characteristics of the attacks. Such an investigation can provide guidelines to effectively tune the parameters of the detection algorithm to achieve specific performance requirements in terms of the above metrics.
The Internet-of-things (IoT) networks are witnessing a drastic increase over the years. Twenty billion devices connected to the Internet are expected in 2022. The need for identifying communities within such networks ...
详细信息
ISBN:
(数字)9781728140582
ISBN:
(纸本)9781728140599
The Internet-of-things (IoT) networks are witnessing a drastic increase over the years. Twenty billion devices connected to the Internet are expected in 2022. The need for identifying communities within such networks can serve as a strong complexity reduction mean for many discovery and identification services. The idea of communities in IoT networks is also motivated by the emerging concept of socializing IoT devices. In this paper, we investigate the application of two community detection algorithms, namely Louvain and Bron-Kerbosch algorithms, on IoT networks usually represented by large-scale graphs. The objective is to convert the complex IoT network into multiple overlapping and non-overlapping communities where its elements share common characteristics. Starting from a real-world IoT networks, we use its dataset to extract community-structured IoT network based on different types of relationships such as co-location, owner social relationships, and autonomously build object relationships among objects. Our analysis showcases how community detection algorithms structure the IoT network into communities based on the different relationships established between objects.
暂无评论