Face recognition is an easy task for humans but it is tedious and complex for computers. Currently, Eigenfaces, Local Binary pattern Histograms (LBPH) and Fisherfaces algorithms are considered as state-of-the-art and ...
详细信息
ISBN:
(纸本)9789380544342
Face recognition is an easy task for humans but it is tedious and complex for computers. Currently, Eigenfaces, Local Binary pattern Histograms (LBPH) and Fisherfaces algorithms are considered as state-of-the-art and are widely used algorithms for face detection. Eigenfaces and Fisherfaces uses Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA) respectively while LBPH uses histogram to recognize the face in the image. In this paper we have analyzed and compared the Eigenfaces, Local Binary pattern Histograms (LBPH) and Fisherfaces algorithms for different scenarios where there is high probability of getting errors. From the simulation results it has been observed that for these algorithms there is increases in the error rate whenever there is a change in the environment, lighting condition, hair and moustache change etc. in the face. Moreover, from the results it has been found that Eigenfaces and Fisherfaces are more prone to the errors than the LBPH.
the following topics are dealt with: learning (artificial intelligence); neural nets; computer aided instruction; Internet; text analysis; Internet of things; data mining; feature extraction; diseases; and natural lan...
详细信息
ISBN:
(数字)9781728173498
ISBN:
(纸本)9781728173504
the following topics are dealt with: learning (artificial intelligence); neural nets; computer aided instruction; Internet; text analysis; Internet of things; data mining; feature extraction; diseases; and natural language processing.
the proceedings contain 7 papers. the special focus in this conference is on Mining Ubiquitous and Social Environments. the topics include: Analyzing big data streams with apache SAMOA;Multimodal behavioral mobility p...
ISBN:
(纸本)9783030339067
the proceedings contain 7 papers. the special focus in this conference is on Mining Ubiquitous and Social Environments. the topics include: Analyzing big data streams with apache SAMOA;Multimodal behavioral mobility pattern mining and analysis using topic modeling on GPS data;sequential monte carlo inference based on activities for overlapping community models;results of a survey about the perceived task similarities in micro task crowdsourcing systems;Provenance of explicit and implicit interactions on social media with W3C PROV-DM.
the timely identification of pathogens is vital in order to effectively control diseases and avoid antimicrobial resistance. Non-invasive point-of-care diagnostic tools are recently trending in identification of the p...
详细信息
ISBN:
(纸本)9781450372992
the timely identification of pathogens is vital in order to effectively control diseases and avoid antimicrobial resistance. Non-invasive point-of-care diagnostic tools are recently trending in identification of the pathogens and becoming a helpful tool especially for rural areas. Machine learning approaches have been widely applied on biological markers for predicting diseases and pathogens. However, there are few studies in the literature that have utilized volatile organic compounds (VOCs) as non-invasive biological markers to identify bacterial pathogens. Furthermore, there is no comprehensive study investigating the effect of different distance and similarity metrics for pathogen classification based on VOC data. In this study, we compared various non-Euclidean distance and similarity metrics with Euclidean metric to identify significantly contributing VOCs to predict pathogens. In addition, we also utilized backward feature elimination (BFE) method to accurately select the best set of features. the dataset we utilized for experiments was composed from the publications published between 1977 and 2016, and consisted of associations in between 703 VOCs and 11 pathogens. We performed extensive set of experiments with five different distance metrics in both uniform and weighted manner. Comprehensive experiments showed that it is possible to correctly predict pathogens by using 68 VOCs among 703 with 78.6% accuracy using k-nearest neighbour classifier and Sorensen distance metric.
Identifying risky driving behavior is of central importance for increasing traffic safety. this paper tackles the task of analyzing real (naturalistic) driving data captured by in-vehicle sensors using interpretable d...
详细信息
Faced with a large number of fault management data, whether there are some important and interesting but unknown patterns in the event data has always attracted the attention of fault management engineer. Based on the...
详细信息
ISBN:
(纸本)9781728105215
Faced with a large number of fault management data, whether there are some important and interesting but unknown patterns in the event data has always attracted the attention of fault management engineer. Based on the analysis of fault management process and daily fault management data of engineering machinery and equipment, a fault management information event model is put forward, which includes six elements: description elements, object elements, time elements, space elements, modification elements and condition elements. Based on information event model we propose a fault pattern mining method, and we also tested the proposed approach on the practical production data of a construction machinery data service provider, the experimental results show that our approach is correct and effective.
Multi-Directional Ternary pattern (M-DTP) is the face descriptor which reduces the noise and extracts the key expression related characteristics of the input face image to overcome the recent challenges encountered in...
详细信息
Nowadays, we can use immersive interaction and display technologies in collaborative analytical reasoning and decision making scenarios. In order to support heterogeneous professional communities of practice in their ...
详细信息
ISBN:
(纸本)9783030218164;9783030218171
Nowadays, we can use immersive interaction and display technologies in collaborative analytical reasoning and decision making scenarios. In order to support heterogeneous professional communities of practice in their digital transformation, it is necessary not only to provide the technologies but to understand the work practices under transformations as well as the security, privacy and other concerns of the communities. Our approach is a comprehensive and evolutionary socio-technological learning analytics and design process leading to a flexible infrastructure where professional communities can co-create their wearable enhanced learning solution. In the core, we present a multi-sensory fusion recorder and player that allows the recordings of multi-actor activity sequences by human activity recognition and the computational support of immersive learning analytics to support training scenarios. Our approach enables cross-domain collaboration by fusing, aggregating and visualizing sensor data coming from wearables and modern production systems. the software is open source and based on the outcomes of several national and international funded projects.
Space debris is a threat to manned spacecraft due to its high velocity, a method based on BP neural network is presented to estimate the degree of hypervelocity impact damage. In this research, projectiles were fired ...
详细信息
ISBN:
(纸本)9781728157122
Space debris is a threat to manned spacecraft due to its high velocity, a method based on BP neural network is presented to estimate the degree of hypervelocity impact damage. In this research, projectiles were fired by two stage light gas gun to impact single aluminum plate, the damage formed on target plate was a crater or a perforating hole. the impact signals were obtained by ultrasonic sensors. According to the signal characteristics of different damage mode, signal amplitude, S2 modal energy, proportion of high-frequency energy and signal propagation distance were chosen as the input characteristic parameters to build a neural network. Experimental results showed that the trained network can be used to identify damage patterns, the correct rate was higher than 85%. It can also predict the depth of crater and hole diameter.
暂无评论