Search results for: multi-scale feature extraction

Authors: Hu Ding, Kai Liu, Guoan Tang

Abstract:

The Chinese Loess Plateau suffers from serious gully erosion induced by natural and human causes. Gully features detection including gully-affected area and its two dimension parameters (length, width, area et al.), is a significant task not only for researchers but also for policy-makers. This study aims at gully-affected area detection in three catchments of Chinese Loess Plateau, which were selected in Changwu, Ansai, and Suide by using unmanned aerial vehicle (UAV). The methodology includes a sequence of UAV data generation, image segmentation, feature calculation and selection, and random forest classification. Two experiments were conducted to investigate the influences of segmentation strategy and feature selection. Results showed that vertical and horizontal root-mean-square errors were below 0.5 and 0.2 m, respectively, which were ideal for the Loess Plateau region. The segmentation strategy adopted in this paper, which considers the topographic information, and optimal parameter combination can improve the segmentation results. Besides, the overall extraction accuracy in Changwu, Ansai, and Suide achieved was 84.62%, 86.46%, and 93.06%, respectively, which indicated that the proposed method for detecting gully-affected area is more objective and effective than traditional methods. This study demonstrated that UAV can bridge the gap between field measurement and satellite-based remote sensing, obtaining a balance in resolution and efficiency for catchment-scale gully erosion research.

Keywords: unmanned aerial vehicle (UAV), object-analysis image analysis, gully erosion, gully-affected area, Loess Plateau, random forest

Procedia PDF Downloads 194

Authors: Stefan Wagenpfeil, Felix Engel, Paul McKevitt, Matthias Hemmje

Abstract:

Multimedia Indexing and Retrieval is generally designed and implemented by employing feature graphs. These graphs typically contain a significant number of nodes and edges to reflect the level of detail in feature detection. A higher level of detail increases the effectiveness of the results but also leads to more complex graph structures. However, graph-traversal-based algorithms for similarity are quite inefficient and computation intensive, especially for large data structures. To deliver fast and effective retrieval, an efficient similarity algorithm, particularly for large graphs, is mandatory. Hence, in this paper, we define a graph-projection into a 2D space (Graph Code) as well as the corresponding algorithms for indexing and retrieval. We show that calculations in this space can be performed more efficiently than graph-traversals due to a simpler processing model and a high level of parallelization. In consequence, we prove that the effectiveness of retrieval also increases substantially, as Graph Codes facilitate more levels of detail in feature fusion. Thus, Graph Codes provide a significant increase in efficiency and effectiveness (especially for Multimedia indexing and retrieval) and can be applied to images, videos, audio, and text information.

Keywords: indexing, retrieval, multimedia, graph algorithm, graph code

Procedia PDF Downloads 133

Authors: Y. Xu, L. Xiong, Z. Xu

Abstract:

In order to protect data privacy, image with sensitive or private information needs to be encrypted before being outsourced to the cloud. However, this causes difficulties in image retrieval and data management. A secure image retrieval method based on orthogonal decomposition is proposed in the paper. The image is divided into two different components, for which encryption and feature extraction are executed separately. As a result, cloud server can extract features from an encrypted image directly and compare them with the features of the queried images, so that the user can thus obtain the image. Different from other methods, the proposed method has no special requirements to encryption algorithms. Experimental results prove that the proposed method can achieve better security and better retrieval precision.

Keywords: secure image retrieval, secure search, orthogonal decomposition, secure cloud computing

Procedia PDF Downloads 459

Authors: Tomyslav Sledevič, Artūras Serackis, Gintautas Tamulevičius, Dalius Navakauskas

Abstract:

This paper presents a comparative evaluation of features extraction algorithm for a real-time isolated word recognition system based on FPGA. The Mel-frequency cepstral, linear frequency cepstral, linear predictive and their cepstral coefficients were implemented in hardware/software design. The proposed system was investigated in the speaker-dependent mode for 100 different Lithuanian words. The robustness of features extraction algorithms was tested recognizing the speech records at different signals to noise rates. The experiments on clean records show highest accuracy for Mel-frequency cepstral and linear frequency cepstral coefficients. For records with 15 dB signal to noise rate the linear predictive cepstral coefficients give best result. The hard and soft part of the system is clocked on 50 MHz and 100 MHz accordingly. For the classification purpose, the pipelined dynamic time warping core was implemented. The proposed word recognition system satisfies the real-time requirements and is suitable for applications in embedded systems.

Keywords: isolated word recognition, features extraction, MFCC, LFCC, LPCC, LPC, FPGA, DTW

Procedia PDF Downloads 471

Authors: Delphine Vendryes, Dushyanth Sekhar, Baojia Tong, Matthew Theisen, Chester Curme

Abstract:

Much of the data that inform the decisions of governments, corporations and individuals are harvested from unstructured documents. Data extraction is defined here as a process that turns non-machine-readable information into a machine-readable format that can be stored, for instance, in a database. In financial services, introducing more automation in data extraction pipelines is a major challenge. Information sought by financial data consumers is often buried within vast bodies of unstructured documents, which have historically required thorough manual extraction. Automated solutions provide faster access to non-machine-readable datasets, in a context where untimely information quickly becomes irrelevant. Data quality standards cannot be compromised, so automation requires high data integrity. This multifaceted task is broken down into smaller steps: ingestion, table parsing (detection and structure recognition), text analysis (entity detection and disambiguation), schema-based record extraction, user feedback incorporation. Selected intermediary steps are phrased as machine learning problems. Solutions leveraging cutting-edge approaches from the fields of computer vision (e.g. table detection) and natural language processing (e.g. entity detection and disambiguation) are proposed.

Keywords: computer vision, entity recognition, finance, information retrieval, machine learning, natural language processing

Procedia PDF Downloads 90

Authors: Zuzanna Miodonska, Michal Krecichwost, Pawel Badura

Abstract:

Sigmatism (lisping) is a speech disorder in which sibilant consonants are mispronounced. The diagnosis of this phenomenon is usually based on the auditory assessment. However, the progress in speech analysis techniques creates a possibility of developing computer-aided sigmatism diagnosis tools. The aim of the study is to statistically verify whether specific acoustic features of sibilant sounds may be related to pronunciation correctness. Such knowledge can be of great importance while implementing classifiers and designing novel tools for automatic sibilants pronunciation evaluation. The study covers analysis of various speech signal measures, including features proposed in the literature for the description of normative sibilants realization. Amplitudes and frequencies of three fricative formants (FF) are extracted based on local spectral maxima of the friction noise. Skewness, kurtosis, four normalized spectral moments (SM) and 13 mel-frequency cepstral coefficients (MFCC) with their 1st and 2nd derivatives (13 Delta and 13 Delta-Delta MFCC) are included in the analysis as well. The resulting feature vector contains 51 measures. The experiments are performed on the speech corpus containing words with selected sibilant sounds (/ʃ, ʒ/) pronounced by 60 preschool children with proper pronunciation or with natural pathologies. In total, 224 /ʃ/ segments and 191 /ʒ/ segments are employed in the study. The Mann-Whitney U test is employed for the analysis of stigmatism and normative pronunciation. Statistically, significant differences are obtained in most of the proposed features in children divided into these two groups at p < 0.05. All spectral moments and fricative formants appear to be distinctive between pathology and proper pronunciation. These metrics describe the friction noise characteristic for sibilants, which makes them particularly promising for the use in sibilants evaluation tools. Correspondences found between phoneme feature values and an expert evaluation of the pronunciation correctness encourage to involve speech analysis tools in diagnosis and therapy of sigmatism. Proposed feature extraction methods could be used in a computer-assisted stigmatism diagnosis or therapy systems.

Keywords: computer-aided pronunciation evaluation, sigmatism diagnosis, speech signal analysis, statistical verification

Procedia PDF Downloads 276

Authors: Pei-Jyun Hsieh, Cheng-Hsuan Li, Bor-Chen Kuo

Abstract:

For hyperspectral image classification, feature reduction is an important pre-processing for avoiding the Hughes phenomena due to the difficulty for collecting training samples. Hence, lots of researches developed feature selection methods such as F-score, HSIC (Hilbert-Schmidt Independence Criterion), and etc., to improve hyperspectral image classification. However, most of them only consider the class separability in the original space, i.e., a linear class separability. In this study, we proposed a nonlinear class separability measure based on kernel trick for selecting an appropriate feature subset. The proposed nonlinear class separability was formed by a generalized RBF kernel with different bandwidths with respect to different features. Moreover, it considered the within-class separability and the between-class separability. A genetic algorithm was applied to tune these bandwidths such that the smallest with-class separability and the largest between-class separability simultaneously. This indicates the corresponding feature space is more suitable for classification. In addition, the corresponding nonlinear classification boundary can separate classes very well. These optimal bandwidths also show the importance of bands for hyperspectral image classification. The reciprocals of these bandwidths can be viewed as weights of bands. The smaller bandwidth, the larger weight of the band, and the more importance for classification. Hence, the descending order of the reciprocals of the bands gives an order for selecting the appropriate feature subsets. In the experiments, three hyperspectral image data sets, the Indian Pine Site data set, the PAVIA data set, and the Salinas A data set, were used to demonstrate the selected feature subsets by the proposed nonlinear feature selection method are more appropriate for hyperspectral image classification. Only ten percent of samples were randomly selected to form the training dataset. All non-background samples were used to form the testing dataset. The support vector machine was applied to classify these testing samples based on selected feature subsets. According to the experiments on the Indian Pine Site data set with 220 bands, the highest accuracies by applying the proposed method, F-score, and HSIC are 0.8795, 0.8795, and 0.87404, respectively. However, the proposed method selects 158 features. F-score and HSIC select 168 features and 217 features, respectively. Moreover, the classification accuracies increase dramatically only using first few features. The classification accuracies with respect to feature subsets of 10 features, 20 features, 50 features, and 110 features are 0.69587, 0.7348, 0.79217, and 0.84164, respectively. Furthermore, only using half selected features (110 features) of the proposed method, the corresponding classification accuracy (0.84168) is approximate to the highest classification accuracy, 0.8795. For other two hyperspectral image data sets, the PAVIA data set and Salinas A data set, we can obtain the similar results. These results illustrate our proposed method can efficiently find feature subsets to improve hyperspectral image classification. One can apply the proposed method to determine the suitable feature subset first according to specific purposes. Then researchers can only use the corresponding sensors to obtain the hyperspectral image and classify the samples. This can not only improve the classification performance but also reduce the cost for obtaining hyperspectral images.

Keywords: hyperspectral image classification, nonlinear feature selection, kernel trick, support vector machine

Procedia PDF Downloads 244

Authors: Mingjun Xu, Man Pun Wan

Abstract:

This study proposed an effective smoke control strategy for the large flat space with a low ceiling to achieve the requirement of perpetual tenability. For the large flat space with a low ceiling, the depth of the smoke reservoir is very shallow, and it is difficult to perpetually constrain the smoke within a limited space. A series of numerical tests were conducted to determine the smoke strategy. A zonal design i.e., the fire zone and two adjacent zones was proposed and validated to be effective in controlling smoke. Once a fire happens in a compartment space, the Engineered Smoke Control (ESC) system will be activated in three zones i.e., the fire zone, in which the fire happened, and two adjacent zones. The smoke can be perpetually constrained within the three smoke zones. To further improve the extraction efficiency, sequential activation of the ESC system within the 3 zones turned out to be more efficient than simultaneous activation. Additionally, the proposed zonal and sequential extraction design can reduce the mechanical extraction flow rate by up to 40.7 % as compared to the conventional method, which is much more economical than that of the conventional method.

Keywords: performance-based design, perpetual tenability, smoke control, fire plume

Procedia PDF Downloads 47

Authors: A. N. Paithane, D. S. Bormane, S. D. Shirbahadurkar

Abstract:

It has been seen that emotion recognition is an important research topic in the field of Human and computer interface. A novel technique for Feature Extraction (FE) has been presented here, further a new method has been used for human emotion recognition which is based on HHT method. This method is feasible for analyzing the nonlinear and non-stationary signals. Each signal has been decomposed into the IMF using the EMD. These functions are used to extract the features using fission and fusion process. The decomposition technique which we adopt is a new technique for adaptively decomposing signals. In this perspective, we have reported here potential usefulness of EMD based techniques.We evaluated the algorithm on Augsburg University Database; the manually annotated database.

Keywords: intrinsic mode function (IMF), Hilbert-Huang transform (HHT), empirical mode decomposition (EMD), emotion detection, electrocardiogram (ECG)

Procedia PDF Downloads 549

Authors: Arwa Alzughaibi

Abstract:

Human Motion detection is a challenging task due to a number of factors including variable appearance, posture and a wide range of illumination conditions and background. So, the first need of such a model is a reliable feature set that can discriminate between a human and a non-human form with a fair amount of confidence even under difficult conditions. By having richer representations, the classification task becomes easier and improved results can be achieved. The Aim of this paper is to investigate the reliable and accurate human motion detection models that are able to detect the human motions accurately under varying illumination levels and backgrounds. Different sets of features are tried and tested including Histogram of Oriented Gradients (HOG), Deformable Parts Model (DPM), Local Decorrelated Channel Feature (LDCF) and Aggregate Channel Feature (ACF). However, we propose an efficient and reliable human motion detection approach by combining Histogram of oriented gradients (HOG) and local phase quantization (LPQ) as the feature set, and implementing search pruning algorithm based on optical flow to reduce the number of false positive. Experimental results show the effectiveness of combining local phase quantization descriptor and the histogram of gradient to perform perfectly well for a large range of illumination conditions and backgrounds than the state-of-the-art human detectors. Areaunder th ROC Curve (AUC) of the proposed method achieved 0.781 for UCF dataset and 0.826 for CDW dataset which indicates that it performs comparably better than HOG, DPM, LDCF and ACF methods.

Keywords: human motion detection, histograms of oriented gradient, local phase quantization, local phase quantization

Procedia PDF Downloads 234

Authors: Aliaksandr Huminski

Abstract:

Semantic Role Labeling (SRL) as shallow semantic parsing approach includes recognition and labeling arguments of a verb in a sentence. Verb participants are linked with specific semantic roles (Agent, Patient, Instrument, Location, etc.). Thus, SRL can answer on key questions such as ‘Who’, ‘When’, ‘What’, ‘Where’ in a text and it is widely applied in dialog systems, question-answering, named entity recognition, information retrieval, and other fields of NLP. However, SRL has the following flaw: Two sentences with identical (or almost identical) meaning can have different semantic role structures. Let consider 2 sentences: (1) John put butter on the bread. (2) John buttered the bread. SRL for (1) and (2) will be significantly different. For the verb put in (1) it is [Agent + Patient + Goal], but for the verb butter in (2) it is [Agent + Goal]. It happens because of one of the most interesting and intriguing features of a verb: Its ability to capture participants as in the case of the verb butter, or their features as, say, in the case of the verb drink where the participant’s feature being liquid is shared with the verb. This capture looks like a total fusion of meaning and cannot be decomposed in direct way (in comparison with compound verbs like babysit or breastfeed). From this perspective, SRL looks really shallow to represent semantic structure. If the key point in semantic representation is an opportunity to use it for making inferences and finding hidden reasons, it assumes by default that two different but semantically identical sentences must have the same semantic structure. Otherwise we will have different inferences from the same meaning. To overcome the above-mentioned flaw, the following approach is suggested. Assume that: P is a participant of relation; F is a feature of a participant; Vcp is a verb that captures a participant; Vcf is a verb that captures a feature of a participant; Vpr is a primitive verb or a verb that does not capture any participant and represents only a relation. In another word, a primitive verb is a verb whose meaning does not include meanings from its surroundings. Then Vcp and Vcf can be decomposed as: Vcp = Vpr +P; Vcf = Vpr +F. If all Vcp and Vcf will be represented this way, then primitive verbs Vpr can be considered as a canonical form for SRL. As a result of that, there will be no hidden participants caught by a verb since all participants will be explicitly unfolded. An obvious example of Vpr is the verb go, which represents pure movement. In this case the verb drink can be represented as man-made movement of liquid into specific direction. Extraction and using primitive verbs for SRL create a canonical representation unique for semantically identical sentences. It leads to the unification of semantic representation. In this case, the critical flaw related to SRL will be resolved.

Keywords: decomposition, labeling, primitive verbs, semantic roles

Procedia PDF Downloads 344

Authors: Marina Stramarkou, Sofia Papadaki, Konstantina Kyriakopoulou, Magdalini Krokida

Abstract:

Nowadays, microalgae represent the diverse branch of microorganism that is used not only in fish farming, but also in food, cosmetics, pharmaceuticals and biofuel production as they can produce a wide range of unique functional ingredients. In the present work, a remarkable microalga Chlorella vulgaris (CV) was selected as a raw material for the recovery of multifunctional extracts. First of all, the drying of raw biomass was examined with freeze-drying showing the best behavior. Ultrasonic-assisted extraction (UAE) using different solvents was applied under the specific optimized conditions. In case of raw biomass, ethanol was the suitable solvent, whereas on dried samples water performed better. The total carotenoid, β-carotene, chlorophyll and protein content in the raw materials, extracts and extraction residues was determined using UV-Vis spectrometry. The microalgae biomass and the extracts were evaluated regarding their antiradical activity using the DPPH method.

Keywords: antioxidant activity, pigments, proteins, ultrasound assisted extraction

Procedia PDF Downloads 312

Authors: A. Abdi, A. Barari, A. Hojatollah Nikbakht, Khosro Ebrahim

Abstract:

Aim: The purpose of the current study was to investigate the effect of a high intensity treadmill running training (8 weeks) with or without aqueous extraction of Crataegus pentagyna on heart and plasma LDH and CK. Design: Thirty-two Wistar male rats (4-6 weeks old, 125-135 gr weight) were used. Animals were randomly assigned into training (n = 16) and control (n = 16) groups and further divided into saline-control (SC, n = 8), saline-training (ST, n = 8), red Crataegus pentagyna extraction -control (CPEC, n = 8), and red Crataegus pentagyna extraction -training (CPET, n = 8) groups. Training groups have performed a high-intensity running program 34 m/min on 0% grade, 60 min/day, 5 days/week) on a motor-driven treadmill for 8 weeks. Animals were fed orally with Crataegus extraction and saline solution (500mg/kg body weight/or 10ml/kg body weight) for last six weeks. Seventy- two hours after the last training session, rats were sacrificed; plasma and heart were excised and immediately frozen in liquid nitrogen. LDH and CK levels were measured by colorimetric method. Statistical analysis was performed using a one way analysis of variance and Tukey test. Significance was accepted at P = 0.05. Results: Result showed that consumption crataegus lowers LDH and CK in heart and plasma. Also the heart LDH and CK were lower in the CPET compared to the ST, while plasma LDH and CK in CPET was higher than the ST. The results of ANOVA showed that the due high-intensity exercise and consumption crataegus, there are significant differences between levels of hearth LDH (P < 0/001), plasma (P < 0/006) and hearth (P < 0/001) CK. Conclusion: It appears that high-intensity exercise led to increased tissue damage and inflammatory factors in plasma. In other hand, consumption aqueous extraction of Red Crataegus maybe inhibits these factors and prevents muscle and heart damage.

Keywords: LDH, CK, crataegus, intensity

Procedia PDF Downloads 416

Authors: Yulan Wu

Abstract:

The spread of fake news on social media has posed significant societal harm to the public and the nation, with its threats spanning various domains, including politics, economics, health, and more. News on social media often covers multiple domains, and existing models studied by researchers and relevant organizations often perform well on datasets from a single domain. However, when these methods are applied to social platforms with news spanning multiple domains, their performance significantly deteriorates. Existing research has attempted to enhance the detection performance of multi-domain datasets by adding single-domain labels to the data. However, these methods overlook the fact that a news article typically belongs to multiple domains, leading to the loss of domain knowledge information contained within the news text. To address this issue, research has found that news records in different domains often use different vocabularies to describe their content. In this paper, we propose a fake news detection framework that combines domain knowledge and expert knowledge. Firstly, it utilizes an unsupervised domain discovery module to generate a low-dimensional vector for each news article, representing domain embeddings, which can retain multi-domain knowledge of the news content. Then, a feature extraction module uses the domain embeddings discovered through unsupervised domain knowledge to guide multiple experts in extracting news knowledge for the total feature representation. Finally, a classifier is used to determine whether the news is fake or not. Experiments show that this approach can improve multi-domain fake news detection performance while reducing the cost of manually labeling domain labels.

Keywords: fake news, deep learning, natural language processing, multiple domains

Procedia PDF Downloads 44

Authors: Lamyaa Gamal El-Deen Taha, Ashraf Sharawi

Abstract:

China launched satellite GF-2 in 2014. This study deals with comparing nearest neighbor object-based classification and neural network classification methods for classification of the fused GF-2 image. Firstly, rectification of GF-2 image was performed. Secondly, a comparison between nearest neighbor object-based classification and neural network classification for classification of fused GF-2 was performed. Thirdly, the overall accuracy of classification and kappa index were calculated. Results indicate that nearest neighbor object-based classification is better than neural network classification for urban mapping.

Keywords: GF-2 images, feature extraction-rectification, nearest neighbour object based classification, segmentation algorithms, neural network classification, multilayer perceptron

Procedia PDF Downloads 360

Authors: A. N. Kurbanova, G. K. Sugurbekova, N. K. Akhmetov

Abstract:

The asphaltenes are heavy fraction of crude oil. Asphaltenes on oilfield is known for its ability to plug wells, surface equipment and pores of the geologic formations. The present research is devoted to the deasphalting of crude oil as the initial stage refining oil. Solvent deasphalting was conducted by extraction with organic solvents (cyclohexane, carbon tetrachloride, chloroform). Analysis of availability of metals was conducted by ICP-MS and spectral feature at deasphalting was achieved by FTIR. High contents of asphaltenes in crude oil reduce the efficiency of refining processes. Moreover, high distribution heteroatoms (e.g., S, N) were also suggested in asphaltenes cause some problems: environmental pollution, corrosion and poisoning of the catalyst. The main objective of this work is to study the effect of deasphalting process crude oil to improve its properties and improving the efficiency of recycling processes. Experiments of solvent extraction are using organic solvents held in the crude oil JSC “Pavlodar Oil Chemistry Refinery. Experimental results show that deasphalting process also leads to decrease Ni, V in the composition of the oil. One solution to the problem of cleaning oils from metals, hydrogen sulfide and mercaptan is absorption with chemical reagents directly in oil residue and production due to the fact that asphalt and resinous substance degrade operational properties of oils and reduce the effectiveness of selective refining of oils. Deasphalting of crude oil is necessary to separate the light fraction from heavy metallic asphaltenes part of crude oil. For this oil is pretreated deasphalting, because asphaltenes tend to form coke or consume large quantities of hydrogen. Removing asphaltenes leads to partly demetallization, i.e. for removal of asphaltenes V/Ni and organic compounds with heteroatoms. Intramolecular complexes are relatively well researched on the example of porphyinous complex (VO2) and nickel (Ni). As a result of studies of V/Ni by ICP MS method were determined the effect of different solvents-deasphalting – on the process of extracting metals on deasphalting stage and select the best organic solvent. Thus, as the best DAO proved cyclohexane (C6H12), which as a result of ICP MS retrieves V-51.2%, Ni-66.4%? Also in this paper presents the results of a study of physical and chemical properties and spectral characteristics of oil on FTIR with a view to establishing its hydrocarbon composition. Obtained by using IR-spectroscopy method information about the specifics of the whole oil give provisional physical, chemical characteristics. They can be useful in the consideration of issues of origin and geochemical conditions of accumulation of oil, as well as some technological challenges. Systematic analysis carried out in this study; improve our understanding of the stability mechanism of asphaltenes. The role of deasphalted crude oil fractions on the stability asphaltene is described.

Keywords: asphaltenes, deasphalting, extraction, vanadium, nickel, metalloporphyrins, ICP-MS, IR spectroscopy

Procedia PDF Downloads 221

Authors: Gong Zhilin, Jing Yang, Jian Yin

Abstract:

The purpose of this paper is to develop a novel system for credit card fraud detection based on sequential modeling of data using hybrid deep learning models. The projected model encapsulates five major phases are pre-processing, imbalance-data handling, feature extraction, optimal feature selection, and fraud detection with an ensemble classifier. The collected raw data (input) is pre-processed to enhance the quality of the data through alleviation of the missing data, noisy data as well as null values. The pre-processed data are class imbalanced in nature, and therefore they are handled effectively with the K-means clustering-based SMOTE model. From the balanced class data, the most relevant features like improved Principal Component Analysis (PCA), statistical features (mean, median, standard deviation) and higher-order statistical features (skewness and kurtosis). Among the extracted features, the most optimal features are selected with the Self-improved Arithmetic Optimization Algorithm (SI-AOA). This SI-AOA model is the conceptual improvement of the standard Arithmetic Optimization Algorithm. The deep learning models like Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and optimized Quantum Deep Neural Network (QDNN). The LSTM and CNN are trained with the extracted optimal features. The outcomes from LSTM and CNN will enter as input to optimized QDNN that provides the final detection outcome. Since the QDNN is the ultimate detector, its weight function is fine-tuned with the Self-improved Arithmetic Optimization Algorithm (SI-AOA).

Keywords: credit card, data mining, fraud detection, money transactions

Procedia PDF Downloads 104

Authors: Nidhal K. Azawi, John M. Gauch

Abstract:

Colorectal cancer is one of the leading causes of cancer death in the US and the world, which is why millions of colonoscopy examinations are performed annually. Unfortunately, noise, specular highlights, and motion artifacts corrupt many images in a typical colonoscopy exam. The goal of our research is to produce automated techniques to detect and correct or remove these noninformative images from colonoscopy videos, so physicians can focus their attention on informative images. In this research, we first automatically extract features from images. Then we use machine learning and deep neural network to classify colonoscopy images as either informative or noninformative. Our results show that we achieve image classification accuracy between 92-98%. We also show how the removal of noninformative images together with image alignment can aid in the creation of image panoramas and other visualizations of colonoscopy images.

Keywords: colonoscopy classification, feature extraction, image alignment, machine learning

Procedia PDF Downloads 233

Authors: Amesh P, Suneesh A S, Venkatesan K A

Abstract:

The magnetic solid-phase extraction is a relatively new method among the other solid-phase extraction techniques for the separating of metal ions from aqueous solutions, such as mine water and groundwater, contaminated wastes, etc. However, the bare magnetic particles (Fe3O4) exhibit poor selectivity due to the absence of target-specific functional groups for sequestering the metal ions. The selectivity of these magnetic particles can be remarkably improved by covalently tethering the task-specific ligands on magnetic surfaces. The magnetic particles offer a number of advantages such as quick phase separation aided by the external magnetic field. As a result, the solid adsorbent can be prepared with the particle size ranging from a few micrometers to the nanometer, which again offers the advantages such as enhanced kinetics of extraction, higher extraction capacity, etc. Conventionally, the magnetite (Fe3O4) particles were prepared by the hydrolysis and co-precipitation of ferrous and ferric salts in aqueous ammonia solution. Since the covalent linking of task-specific functionalities on Fe3O4 was difficult, and it is also susceptible to redox reaction in the presence of acid or alkali, it is necessary to modify the surface of Fe3O4 by silica coating. This silica coating is usually carried out by hydrolysis and condensation of tetraethyl orthosilicate over the surface of magnetite to yield a thin layer of silica-coated magnetite particles. Since the silica-coated magnetite particles amenable for further surface modification, it can be reacted with task-specific functional groups to obtain the functionalized magnetic particles. The surface area exhibited by such magnetic particles usually falls in the range of 50 to 150 m2.g-1, which offer advantage such as quick phase separation, as compared to the other solid-phase extraction systems. In addition, the magnetic (Fe3O4) particles covalently linked on mesoporous silica matrix (MCM-41) and task-specific ligands offer further advantages in terms of extraction kinetics, high stability, longer reusable cycles, and metal extraction capacity, due to the large surface area, ample porosity and enhanced number of functional groups per unit area on these adsorbents. In view of this, the present paper deals with the synthesis of uranium specific diethylenetriamine ligand (DETA) ligand anchored on silica-coated magnetite (Fe-DETA) as well as on magnetic mesoporous silica (MCM-Fe-DETA) and studies on the extraction of uranium from aqueous solution spiked with uranium to mimic the mine water or groundwater contaminated with uranium. The synthesized solid-phase adsorbents were characterized by FT-IR, Raman, TG-DTA, XRD, and SEM. The extraction behavior of uranium on the solid-phase was studied under several conditions like the effect of pH, initial concentration of uranium, rate of extraction and its variation with pH and initial concentration of uranium, effect of interference ions like CO32-, Na+, Fe+2, Ni+2, and Cr+3, etc. The maximum extraction capacity of 233 mg.g-1 was obtained for Fe-DETA, and a huge capacity of 1047 mg.g-1 was obtained for MCM-Fe-DETA. The mechanism of extraction, speciation of uranium, extraction studies, reusability, and the other results obtained in the present study suggests Fe-DETA and MCM-Fe-DETA are the potential candidates for the extraction of uranium from mine water, and groundwater.

Keywords: diethylenetriamine, magnetic mesoporous silica, magnetic solid-phase extraction, uranium extraction, wastewater treatment

Procedia PDF Downloads 140

Authors: Alexandra C. Blaga, Dan Caşcaval, Alexandra Tucaliuc, Madalina Poştaru, Anca I. Galaction

Abstract:

Amino acids are valuable chemical products used in in human foods, in animal feed additives and in the pharmaceutical field. Recently, there has been a noticeable rise of amino acids utilization throughout the world to include their use as raw materials in the production of various industrial chemicals: oil gelating agents (amino acid-based surfactants) to recover effluent oil in seas and rivers and poly(amino acids), which are attracting attention for biodegradable plastics manufacture. The amino acids can be obtained by biosynthesis or from protein hydrolysis, but their separation from the obtained mixtures can be challenging. In the last decades there has been a continuous interest in developing processes that will improve the selectivity and yield of downstream processing steps. The liquid-liquid extraction of amino acids (dissociated at any pH-value of the aqueous solutions) is possible only by using the reactive extraction technique, mainly with extractants of organophosphoric acid derivatives, high molecular weight amines and crown-ethers. The purpose of this study was to analyse the separation of nine amino acids of acidic character (l-aspartic acid, l-glutamic acid), basic character (l-histidine, l-lysine, l-arginine) and neutral character (l-glycine, l-tryptophan, l-cysteine, l-alanine) by reactive extraction with di-(2-ethylhexyl)phosphoric acid (D2EHPA) dissolved in butyl acetate. The results showed that the separation yield is controlled by the pH value of the aqueous phase: the reactive extraction of amino acids with D2EHPA is possible only if the amino acids exist in aqueous solution in their cationic forms (pH of aqueous phase below the isoeletric point). The studies for individual amino acids indicated the possibility of selectively separate different groups of amino acids with similar acidic properties as a function of aqueous solution pH-value: the maximum yields are reached for a pH domain of 2–3, then strongly decreasing with the pH increase. Thus, for acidic and neutral amino acids, the extraction becomes impossible at the isolelectric point (pHi) and for basic amino acids at a pH value lower than pHi, as a result of the carboxylic group dissociation. From the results obtained for the separation from the mixture of the nine amino acids, at different pH, it can be observed that all amino acids are extracted with different yields, for a pH domain of 1.5–3. Over this interval, the extract contains only the amino acids with neutral and basic character. For pH 5–6, only the neutral amino acids are extracted and for pH > 6 the extraction becomes impossible. Using this technique, the total separation of the following amino acids groups has been performed: neutral amino acids at pH 5–5.5, basic amino acids and l-cysteine at pH 4–4.5, l-histidine at pH 3–3.5 and acidic amino acids at pH 2–2.5.

Keywords: amino acids, di-(2-ethylhexyl) phosphoric acid, reactive extraction, selective extraction

Procedia PDF Downloads 402

Authors: Wei Li, Tuo Yang

Abstract:

In order to satisfy the real-time and accurate requirements of cigarette smoke detection in complex scenes, a cigarette smoke detection technology based on the combination of deep learning and color features was proposed. Firstly, based on the color features of cigarette smoke, the suspicious cigarette smoke area in the image is extracted. Secondly, combined with the efficiency of cigarette smoke detection and the problem of network overfitting, a network model for cigarette smoke detection was designed according to YOLOV3 algorithm to reduce the false detection rate. The experimental results show that the method is feasible and effective, and the accuracy of cigarette smoke detection is up to 99.13%, which satisfies the requirements of real-time cigarette smoke detection in complex scenes.

Keywords: deep learning, computer vision, cigarette smoke detection, YOLOV3, color feature extraction

Procedia PDF Downloads 61

Authors: Nassima Noufail, Sara Bouhali

Abstract:

In this work, we develop a semi-supervised solution for the purpose of action detection in videos and propose an efficient algorithm for video segmentation. The approach is divided into video segmentation, feature extraction, and classification. In the first part, a video is segmented into clips, and we used the K-means algorithm for this segmentation; our goal is to find groups based on similarity in the video. The application of k-means clustering into all the frames is time-consuming; therefore, we started by the identification of transition frames where the scene in the video changes significantly, and then we applied K-means clustering into these transition frames. We used two image filters, the gaussian filter and the Laplacian of Gaussian. Each filter extracts a set of features from the frames. The Gaussian filter blurs the image and omits the higher frequencies, and the Laplacian of gaussian detects regions of rapid intensity changes; we then used this vector of filter responses as an input to our k-means algorithm. The output is a set of cluster centers. Each video frame pixel is then mapped to the nearest cluster center and painted with a corresponding color to form a visual map. The resulting visual map had similar pixels grouped. We then computed a cluster score indicating how clusters are near each other and plotted a signal representing frame number vs. clustering score. Our hypothesis was that the evolution of the signal would not change if semantically related events were happening in the scene. We marked the breakpoints at which the root mean square level of the signal changes significantly, and each breakpoint is an indication of the beginning of a new video segment. In the second part, for each segment from part 1, we randomly selected a 16-frame clip, then we extracted spatiotemporal features using convolutional 3D network C3D for every 16 frames using a pre-trained model. The C3D final output is a 512-feature vector dimension; hence we used principal component analysis (PCA) for dimensionality reduction. The final part is the classification. The C3D feature vectors are used as input to a multi-class linear support vector machine (SVM) for the training model, and we used a multi-classifier to detect the action. We evaluated our experiment on the UCF101 dataset, which consists of 101 human action categories, and we achieved an accuracy that outperforms the state of art by 1.2%.

Keywords: video segmentation, action detection, classification, Kmeans, C3D

Procedia PDF Downloads 52

Authors: Narayana Murthy Akurathi, Vijaya Lakshmi Marella

Abstract:

The literature is reviewed with regard to online Super critical fluid extraction (SFE) coupled directly with supercritical fluid chromatography (SFC) -mass spectrometry that have typically more sensitive than conventional LC-MS/MS and GC-MS/MS. It is becoming increasingly interesting to use on-line techniques that combine sample preparation, separation and detection in one analytical set up. This provides less human intervention, uses small amount of sample and organic solvent and yields enhanced analyte enrichment in a shorter time. The sample extraction is performed under light shielding and anaerobic conditions, preventing the degradation of thermo labile analytes. It may be able to analyze compounds over a wide polarity range as SFC generally uses carbon dioxide which was collected as a by-product of other chemical reactions or is collected from the atmosphere as it contributes no new chemicals to the environment. The diffusion of solutes in supercritical fluids is about ten times greater than that in liquids and about three times less than in gases which results in a decrease in resistance to mass transfer in the column and allows for fast high resolution separations. The drawback of SFC when using carbon dioxide as mobile phase is that the direct introduction of water samples poses a series of problems, water must therefore be eliminated before it reaches the analytical column. Hundreds of compounds analysed simultaneously by simple enclosing in an extraction vessel. This is mainly applicable for pharmaceutical industry where it can analyse fatty acids and phospholipids that have many analogues as their UV spectrum is very similar, trace additives in polymers, cleaning validation can be conducted by putting swab sample in an extraction vessel, analysing hundreds of pesticides with good resolution.

Keywords: super critical fluid extraction (SFE), super critical fluid chromatography (SFC), LCMS/MS, GCMS/MS

Procedia PDF Downloads 369

Authors: Andriana Mkrtchyan, Vahe Khlghatyan

Abstract:

The choice of a profession is one of the most important decisions people make throughout their life. With the development of modern science, technologies, and all the spheres existing in the modern world, more and more professions are being arisen that complicate even more the process of choosing. Hence, there is a need for a guiding platform to help people to choose a profession and the right career path based on their interests, skills, and personality. This review aims at analyzing existing methods of comparing PDF format documents and suggests that a 3-stage approach is implemented for the comparison, that is – 1. text extraction from PDF format documents, 2. comparison of the extracted text via NLP algorithms, 3. comparison representation using special shape and color psychology methodology.

Keywords: color psychology, data acquisition/extraction, data augmentation, disambiguation, natural language processing, outlier detection, semantic similarity, text-mining, user evaluation, visual search

Procedia PDF Downloads 46

Authors: Maha Jazouli

Abstract:

Suicide is one of the most important causes of death in the prison environment, both in Canada and internationally. Rates of attempts of suicide and self-harm have been on the rise in recent years, with hangings being the most frequent method resorted to. The objective of this article is to propose a method to automatically detect in real time suicidal behaviors. We present a gesture recognition system that consists of three modules: model-based movement tracking, feature extraction, and gesture recognition using machine learning algorithms (MLA). Our proposed system gives us satisfactory results. This smart video surveillance system can help assist staff responsible for the safety and health of inmates by alerting them when suicidal behavior is detected, which helps reduce mortality rates and save lives.

Keywords: suicide detection, Kinect azure, RGB-D camera, SVM, machine learning, gesture recognition

Procedia PDF Downloads 158

Authors: Boukari Nassim

Abstract:

This paper describes the use of odd pair autoregressive coefficients (Yule _Walker and Burg) for the feature extraction of electroencephalogram (EEG) signals. In the classification: the radial basis function neural network neural network (RBFNN) is employed. The RBFNN is described by his architecture and his characteristics: as the RBF is defined by the spread which is modified for improving the results of the classification. Five types of EEG signals are defined for this work: Set A, Set B for normal signals, Set C, Set D for interictal signals, set E for ictal signal (we can found that in Bonn university). In outputs, two classes are given (AC, AD, AE, BC, BD, BE, CE, DE), the best accuracy is calculated at 99% for the combined odd pair autoregressive coefficients. Our method is very effective for the diagnosis of epileptic EEG signals.

Keywords: epilepsy, EEG signals classification, combined odd pair autoregressive coefficients, radial basis function neural network

Procedia PDF Downloads 326

Authors: Aliya Kurbanova, Nurlan Akhmetov, Abilmansur Yeshmuratov, Yerzhigit Sugurbekov, Ramiz Zulkharnay, Gulzat Demeuova, Murat Baisariyev, Gulnar Sugurbekova

Abstract:

Last decades crude oils have tended to become more challenge to process due to increasing amounts of sour and heavy crude oils. Some crude oils contain high vanadium and nickel content, for example Pavlodar LLP crude oil, which contains more than 23.09 g/t nickel and 58.59 g/t vanadium. In this study, we used two types of metal removing methods such as solvent extraction and electrochemical. The present research is conducted for comparative analysis of the deasphalting with organic solvents (cyclohexane, carbon tetrachloride, chloroform) and electrochemical method. Applying the cyclic voltametric analysis (CVA) and Inductively coupled plasma mass spectrometry (ICP MS), these mentioned types of metal extraction methods were compared in this paper. Maximum efficiency of deasphalting, with cyclohexane as the solvent, in Soxhlet extractor was 66.4% for nickel and 51.2% for vanadium content from crude oil. Percentage of Ni extraction reached maximum of approximately 55% by using the electrochemical method in electrolysis cell, which was developed for this research and consists of three sections: oil and protonating agent (EtOH) solution between two conducting membranes which divides it from two capsules of 10% sulfuric acid and two graphite electrodes which cover all three parts in electrical circuit. Ions of metals pass through membranes and remain in acid solutions. The best result was obtained in 60 minutes with ethanol to oil ratio 25% to 75% respectively, current fits into the range from 0.3A to 0.4A, voltage changed from 12.8V to 17.3V.

Keywords: demetallization, deasphalting, electrochemical removal, heavy metals, petroleum engineering, solvent extraction

Procedia PDF Downloads 283

Authors: M. Budakoglu, M. Karaman, D. Kiran, Z. Doner, B. Zeytuncu, B. Tanç, M. Kumral

Abstract:

Lake Acıgöl is one of the large saline lacustrine environment in Turkey. Eleven trace metals (Cr, Mn, Fe, Al, Co, Ni, Cu, Zn, Cd, Pb and As) in 9 surface and subsurface sediment samples from the Lake Acıgöl were analyzed with the bulk and sequential extraction analysis methods by ICP-MS to obtain the metal distribution patterns in this extreme environment. Five stepped sequential extraction technique (1- exchangeable, 2- bond to carbonates, 3- bond to iron and manganese oxides/hydroxides, 4- bond to organic matter and sulphides, and 5- residual fraction incorporated into clay and silicate mineral lattices) was used to characterize the various forms of metals in the <63μ size sediments. The metal contents (ppm) and their percentages for each extraction step were reported and compared with the results obtained from the total digestion. Results indicate that sum of the four fraction are in good agreement with the total digestion results of Ni, Cd, As, Zn, Cu and Fe with the satisfactory recoveries (94.04–109.0%) and the method used is reliable and repeatable for these elements. It was found that there were high correlations between Fe vs. Ni loads in the fraction of F2 and F4 with R2= 0,91 and 0,81, respectively. Comparison of totally 135 chemical analysis results in three sampling location and for 5 fraction between Fe-Co, Co-Ni and Fe-Ni element couples were presented elevated correlations with R2=0,98, 0,92 and 0,91, respectively.

Keywords: Lake Acigol, sequancial extraction, recent lake sediment, geochemical speciation of heavy metals

Procedia PDF Downloads 385

Authors: Parvaneh Hajeb, Shahram Shakibazadeh, Md. Zaidul Islam Sarker

Abstract:

High-quality fish oil for human consumption requires low levels of toxic elements. The aim of this study was to develop a method to extract oil from fish wastes with the least toxic elements contamination. Supercritical fluid extraction (SFE) was applied to detoxify fish oils from toxic elements. The SFE unit used consisted of an intelligent HPLC pump equipped with a cooling jacket to deliver CO2. The freeze-dried fish waste sample was extracted by heating in a column oven. Under supercritical conditions, the oil dissolved in CO2 was separated from the supercritical phase using pressure reduction. The SFE parameters (pressure, temperature, CO2 flow rate, and extraction time) were optimized using response surface methodology (RSM) to extract the highest levels of toxic elements. The results showed that toxic elements in fish oil can be reduced using supercritical CO2 at optimum pressure 40 MPa, temperature 61 ºC, CO2 flow rate 3.8 MPa, and extraction time 4.25 hr. There were significant reductions in the mercury (98.2%), cadmium (98.9%), arsenic (96%), and lead contents (99.2%) of the fish oil. The fish oil extracted using this method contained elements at levels that were much lower than the accepted limits of 0.1 μg/g. The reduction of toxic elements using the SFE method was more efficient than that of the conventional methods due to the high selectivity of supercritical CO2 for non-polar compounds.

Keywords: food safety, toxic elements, fish oil, supercritical carbon dioxide

Procedia PDF Downloads 403

Authors: Majid Habib Khan, Jinwei Zhao, Xinhong Hei, Liu Jiedong, Rana Shahzad Noor, Muhammad Imran

Abstract:

Deep learning is increasingly employed in image classification, yet the scarcity and high cost of labeled data for training remain a challenge. Limited samples often lead to overfitting due to biased sample distribution. This paper introduces a dynamic distribution calibration method for few-shot learning. Initially, base and new class samples undergo normalization to mitigate disparate feature magnitudes. A pre-trained model then extracts feature vectors from both classes. The method dynamically selects distribution characteristics from base classes (both adjacent and remote) in the embedding space, using a threshold value approach for new class samples. Given the propensity of similar classes to share feature distributions like mean and variance, this research assumes a Gaussian distribution for feature vectors. Subsequently, distributional features of new class samples are calibrated using a corrected hyperparameter, derived from the distribution features of both adjacent and distant base classes. This calibration augments the new class sample set. The technique demonstrates significant improvements, with up to 4% accuracy gains in few-shot classification challenges, as evidenced by tests on miniImagenet and CUB datasets.

Keywords: deep learning, computer vision, image classification, few-shot learning, threshold

Procedia PDF Downloads 37