scholarly journals System for an acoustic detection, localisation and classification

ACTA IMEKO ◽  
2021 ◽  
Vol 10 (2) ◽  
pp. 62
Author(s):  
Jakub Svatos ◽  
Jan Holub ◽  
Jan Belak
Keyword(s):  
Visual Detection ◽  
Detection System ◽  
False Alarms ◽  
Classification Algorithms ◽  
Acoustic Detection ◽  
Detection Techniques ◽  
Acoustic Detector ◽  
Military Applications ◽  
Transformation Methods

<p class="Abstract">Currently, acoustic detection techniques of gunshots (gunshot detection and its classification) are increasingly being used not only for military applications but also for civilian purposes. Detection, localisation, and classification of a dangerous event such as gunshots employing acoustic detection is a perspective alternative to visual detection, which is commonly used. In some situations, to detect and localise the source of a gunshot, an automatic acoustic detection system, which can classify the caliber, may be preferable. This paper presents a system for acoustic detection, which can detect, localise and classify acoustic events such as gunshots. The system has been tested in open and closed shooting ranges and tested firearms are 9 mm short gun, 6.35 mm short gun, .22 short gun, and .22 rifle gun with various subsonic and supersonic ammunition. As ‘false alarms’, sets of different impulse acoustic events like door slams, breaking glass, etc. have been used. Localisation and classification algorithms are also introduced. To successfully classify the tested acoustic signals, Continuous Wavelet and Mel Frequency Transformation methods have been used for the signal processing, and the fully two-layer connected neural network has been implemented. The results show that the acoustic detector can be used for reliable gunshot detection, localisation, and classification.</p>

scholarly journals Airborne Visual Detection and Tracking of Cooperative UAVs Exploiting Deep Learning

Sensors ◽  
2019 ◽  
Vol 19 (19) ◽  
pp. 4332 ◽  
Author(s):  
Roberto Opromolla ◽  
Giuseppe Inchingolo ◽  
Giancarmine Fasano
Keyword(s):  
Deep Learning ◽  
Visual Detection ◽  
Detection System ◽  
Flight Test ◽  
Target Range ◽  
Detection And Tracking ◽  
Military Applications ◽  
Technological Challenge ◽  
High Level ◽  
Processing Architecture

The performance achievable by using Unmanned Aerial Vehicles (UAVs) for a large variety of civil and military applications, as well as the extent of applicable mission scenarios, can significantly benefit from the exploitation of formations of vehicles able to fly in a coordinated manner (swarms). In this respect, visual cameras represent a key instrument to enable coordination by giving each UAV the capability to visually monitor the other members of the formation. Hence, a related technological challenge is the development of robust solutions to detect and track cooperative targets through a sequence of frames. In this framework, this paper proposes an innovative approach to carry out this task based on deep learning. Specifically, the You Only Look Once (YOLO) object detection system is integrated within an original processing architecture in which the machine-vision algorithms are aided by navigation hints available thanks to the cooperative nature of the formation. An experimental flight test campaign, involving formations of two multirotor UAVs, is conducted to collect a database of images suitable to assess the performance of the proposed approach. Results demonstrate high-level accuracy, and robustness against challenging conditions in terms of illumination, background and target-range variability.

scholarly journals Early Diagnosis of Coronary Artery Disease using UCI Data set

2019 ◽  
Vol 8 (9S4) ◽  
pp. 252-256
Keyword(s):  
Hybrid Approach ◽  
False Alarms ◽  
First Year ◽  
Classification Algorithms ◽  
Data Set ◽  
Statistical Measures ◽  
Artery Disease ◽  
Transformation Methods ◽  
Electrocardiogram Ecg ◽  
Advanced System

The focus of this project is based on both processing potential clinical features and implementing the classification architecture for detection of cardiac abnormality. The milestone of this first year involves analysis and investigation of different feature selection and transformation methods and theoretical modeling of single and hybrid systems by optimizing associated systematic parameters for better precision and recall. The importance of this paper is due to its clear objectives where an optimized and advanced system is designed and implemented for the cardiac disease utilizing computer aided diagnosis techniques for data and signal processing. The methodology is clear and trait forward using the hybrid approach of data mining techniques integrated to deliver enhanced performance on desired data set. In this paper comparative classification approaches are integrated to enhance system detection rate and decrease false alarms. The study focuses on feature preprocessing to select suitable feature subsets for classification algorithms like clustering (unsupervised learning) and SVM (supervised learning) which helps in generalizing the diagnosis system to detect unseen abnormality. For this study, will first apply statistical measures such as scoring ranking for clinical datasets consisting the electrocardiogram (ECG) features to reduce its dimension by eliminating irrelevant features. In the second part, will apply parametric tuned classification algorithms for selected feature subsets. The third part is to quantify the severity of CAD. At the last performance of the proposed system will be compared with other applied classification techniques in terms of accuracy, sensitivity and specificity.

scholarly journals An IoT-Focused Intrusion Detection System Approach Based on Preprocessing Characterization for Cybersecurity Datasets

Sensors ◽  
2021 ◽  
Vol 21 (2) ◽  
pp. 656
Author(s):  
Xavier Larriva-Novo ◽  
Víctor A. Villagrá ◽  
Mario Vega-Barbas ◽  
Diego Rivera ◽  
Mario Sanz Rodrigo
Keyword(s):  
Machine Learning ◽  
Intrusion Detection ◽  
High Performance ◽  
Learning Algorithm ◽  
Detection System ◽  
Machine Learning Algorithms ◽  
Statistical Characteristics ◽  
Detection Techniques ◽  
Traffic Characteristics ◽  
Benchmark Datasets

Security in IoT networks is currently mandatory, due to the high amount of data that has to be handled. These systems are vulnerable to several cybersecurity attacks, which are increasing in number and sophistication. Due to this reason, new intrusion detection techniques have to be developed, being as accurate as possible for these scenarios. Intrusion detection systems based on machine learning algorithms have already shown a high performance in terms of accuracy. This research proposes the study and evaluation of several preprocessing techniques based on traffic categorization for a machine learning neural network algorithm. This research uses for its evaluation two benchmark datasets, namely UGR16 and the UNSW-NB15, and one of the most used datasets, KDD99. The preprocessing techniques were evaluated in accordance with scalar and normalization functions. All of these preprocessing models were applied through different sets of characteristics based on a categorization composed by four groups of features: basic connection features, content characteristics, statistical characteristics and finally, a group which is composed by traffic-based features and connection direction-based traffic characteristics. The objective of this research is to evaluate this categorization by using various data preprocessing techniques to obtain the most accurate model. Our proposal shows that, by applying the categorization of network traffic and several preprocessing techniques, the accuracy can be enhanced by up to 45%. The preprocessing of a specific group of characteristics allows for greater accuracy, allowing the machine learning algorithm to correctly classify these parameters related to possible attacks.

scholarly journals Ganoderma boninense Disease Detection by Near-Infrared Spectroscopy Classification: A Review

Sensors ◽  
2021 ◽  
Vol 21 (9) ◽  
pp. 3052
Author(s):  
Mas Ira Syafila Mohd Hilmi Tan ◽  
Mohd Faizal Jamlos ◽  
Ahmad Fairuz Omar ◽  
Fatimah Dzaharudin ◽  
Suramate Chalermwisutkul ◽  
...  
Keyword(s):  
Infrared Spectroscopy ◽  
Early Detection ◽  
Near Infrared Spectroscopy ◽  
Near Infrared ◽  
Predictive Analytics ◽  
Detection System ◽  
Imaging Techniques ◽  
Ganoderma Boninense ◽  
Machine Learning Classification ◽  
Detection Techniques

Ganoderma boninense (G. boninense) infection reduces the productivity of oil palms and causes a serious threat to the palm oil industry. This catastrophic disease ultimately destroys the basal tissues of oil palm, causing the eventual death of the palm. Early detection of G. boninense is vital since there is no effective treatment to stop the continuing spread of the disease. This review describes past and future prospects of integrated research of near-infrared spectroscopy (NIRS), machine learning classification for predictive analytics and signal processing towards an early G. boninense detection system. This effort could reduce the cost of plantation management and avoid production losses. Remarkably, (i) spectroscopy techniques are more reliable than other detection techniques such as serological, molecular, biomarker-based sensor and imaging techniques in reactions with organic tissues, (ii) the NIR spectrum is more precise and sensitive to particular diseases, including G. boninense, compared to visible light and (iii) hand-held NIRS for in situ measurement is used to explore the efficacy of an early detection system in real time using ML classifier algorithms and a predictive analytics model. The non-destructive, environmentally friendly (no chemicals involved), mobile and sensitive leads the NIRS with ML and predictive analytics as a significant platform towards early detection of G. boninense in the future.

Analysing Temporal Effects on Classification of SAR and Optical Images

2021 ◽  
Author(s):  
Ahmet Batuhan Polat ◽  
Ozgun Akcay ◽  
Fusun Balik Sanli
Keyword(s):  
Rural Areas ◽  
Classification Accuracy ◽  
Winter Season ◽  
Support Vector ◽  
Classification Algorithms ◽  
Sar Images ◽  
Optical Images ◽  
Object Based ◽  
Training Samples

&lt;p&gt;Obtaining high accuracy in land cover classification is a non-trivial problem in geosciences for monitoring urban and rural areas. In this study, different classification algorithms were tested with different types of data, and besides the effects of seasonal changes on these classification algorithms and the evaluation of the data used are investigated. In addition, the effect of increasing classification training samples on classification accuracy has been revealed as a result of the study. Sentinel-1 Synthetic Aperture Radar (SAR) images and Sentinel-2 multispectral optical images were used as datasets. Object-based approach was used for the classification of various fused image combinations. The classification algorithms Support Vector Machines (SVM), Random Forest (RF) and K-Nearest Neighborhood (kNN) methods were used for this process. In addition, Normalized Difference Vegetation Index (NDVI) was examined separately to define the exact contribution to the classification accuracy. &amp;#160;As a result, the overall accuracies were compared by classifying the fused data generated by combining optical and SAR images. It has been determined that the increase in the number of training samples improve the classification accuracy. Moreover, it was determined that the object-based classification obtained from single SAR imagery produced the lowest classification accuracy among the used different dataset combinations in this study. In addition, it has been shown that NDVI data does not increase the accuracy of the classification in the winter season as the trees shed their leaves due to climate conditions.&lt;/p&gt;

scholarly journals Decision Tree: A Machine Learning for Intrusion Detection

2019 ◽  
Vol 8 (6S4) ◽  
pp. 1126-1130
Keyword(s):  
Machine Learning ◽  
Intrusion Detection ◽  
Detection System ◽  
Research Work ◽  
Machine Learning Techniques ◽  
Data Sets ◽  
Legitimate User ◽  
Learning Techniques ◽  
Three Stages

The Intrusion is a major threat to unauthorized data or legal network using the legitimate user identity or any of the back doors and vulnerabilities in the network. IDS mechanisms are developed to detect the intrusions at various levels. The objective of the research work is to improve the Intrusion Detection System performance by applying machine learning techniques based on decision trees for detection and classification of attacks. The methodology adapted will process the datasets in three stages. The experimentation is conducted on KDDCUP99 data sets based on number of features. The Bayesian three modes are analyzed for different sized data sets based upon total number of attacks. The time consumed by the classifier to build the model is analyzed and the accuracy is done.

scholarly journals Machine learning classification of new asteroid families members

2020 ◽  
Author(s):  
Valerio Carruba
Keyword(s):  
Machine Learning ◽  
Family Members ◽  
Classification Algorithms ◽  
New Members ◽  
Machine Learning Classification ◽  
New Family ◽  
Robotic Telescope ◽  
Traditional Approaches ◽  
Asteroid Families

&lt;p&gt;Asteroid families are groups of asteroids that are the product of collisions or of the rotational fission of a parent object. &amp;#160;These groups are mainly identified in proper elements or frequencies domains. &amp;#160; Because of robotic telescope surveys, the number of known asteroids has increased from about 10,000 in the early 90's to more than 750,000 nowadays. Traditional approaches for identifying new members of asteroid families, like the hierarchical clustering method (HCM), may &amp;#160; struggle to keep up with the growing rate of new discoveries. Here we used machine learning classification algorithms to identify new family members based on the orbital distribution in proper (a,e,sin(i)) of previously known family constituents. We compared the outcome of nine classification algorithms from stand alone and ensemble approaches. &amp;#160;The Extremely Randomized Trees (ExtraTree) method had the highest precision, enabling to&amp;#160; retrieve up to 97% of family members identified with standard HCM.&lt;/p&gt;

Sign in / Sign up

Export Citation Format

Share Document