Search results for: similarity value
365 The Conceptual Design Model of an Automated Supermarket
Authors: V. Sathya Narayanan, P. Sidharth, V. R. Sanal Kumar
Abstract:
The success of any retail business is predisposed by its swift response and its knack in understanding the constraints and the requirements of customers. In this paper a conceptual design model of an automated customer-friendly supermarket has been proposed. In this model a 10-sided, space benefited, regular polygon shaped gravity shelves have been designed for goods storage and effective customer-specific algorithms have been built-in for quick automatic delivery of the randomly listed goods. The algorithm is developed with two main objectives, viz., delivery time and priority. For meeting these objectives the randomly listed items are reorganized according to the critical-path of the robotic arm specific to the identified shop and its layout and the items are categorized according to the demand, shape, size, similarity and nature of the product for an efficient pick-up, packing and delivery process. We conjectured that the proposed automated supermarket model reduces business operating costs with much customer satisfaction warranting a win-win situation.Keywords: automated supermarket, electronic shopping, polygon-shaped rack, shortest path algorithm for shopping
Procedia PDF Downloads 399364 Using Analytical Hierarchy Process and TOPSIS Approaches in Designing a Finite Element Analysis Automation Program
Authors: Ming Wen, Nasim Nezamoddini
Abstract:
Sophisticated numerical simulations like finite element analysis (FEA) involve a complicated process from model setup to post-processing tasks that require replication of time-consuming steps. Utilizing FEA automation program simplifies the complexity of the involved steps while minimizing human errors in analysis set up, calculations, and results processing. One of the main challenges in designing FEA automation programs is to identify user requirements and link them to possible design alternatives. This paper presents a decision-making framework to design a Python based FEA automation program for modal analysis, frequency response analysis, and random vibration fatigue (RVF) analysis procedures. Analytical hierarchy process (AHP) and technique for order preference by similarity to ideal solution (TOPSIS) are applied to evaluate design alternatives considering the feedback received from experts and program users.Keywords: finite element analysis, FEA, random vibration fatigue, process automation, analytical hierarchy process, AHP, TOPSIS, multiple-criteria decision-making, MCDM
Procedia PDF Downloads 104363 Debts and Debt-Based Sukuk Related to Risk Shifting Behavior
Authors: Siti Raihana Hamzah
Abstract:
This paper elaborates risk shifting in debt financing system as the ultimate cause of the global financial crisis. In contrast, risk sharing in equity financing like sukuk helps the economic system to be better sustained. Nevertheless, some types of sukuk are haunted by the issue of imitation with bonds. The critics on the imitation issue not only have raised doubt on the ability of sukuk to diminish risk shifting behavior but also the ability of this Islamic financial instrument to ensure better future financial stability. Through that, this paper provides discussion on the possibility of sukuk to induce risk shifting and how equity financing may help sukuk to be free from risk shifting. This paper is important in the sense that sukuk receives a significant demand from investors throughout the world. For this instrument to be supportive in the future economic stability, the issue of imitation needs to be identified and addressed. Furthermore, critics cannot be focused on debts and its ability to gauge the financial flux but also to sukuk due to their structures similarity.Keywords: global financial crisis, debt, risk-shifting, risk sharing, equity, sukuk, bonds
Procedia PDF Downloads 377362 A Greedy Alignment Algorithm Supporting Medication Reconciliation
Authors: David Tresner-Kirsch
Abstract:
Reconciling patient medication lists from multiple sources is a critical task supporting the safe delivery of patient care. Manual reconciliation is a time-consuming and error-prone process, and recently attempts have been made to develop efficiency- and safety-oriented automated support for professionals performing the task. An important capability of any such support system is automated alignment – finding which medications from a list correspond to which medications from a different source, regardless of misspellings, naming differences (e.g. brand name vs. generic), or changes in treatment (e.g. switching a patient from one antidepressant class to another). This work describes a new algorithmic solution to this alignment task, using a greedy matching approach based on string similarity, edit distances, concept extraction and normalization, and synonym search derived from the RxNorm nomenclature. The accuracy of this algorithm was evaluated against a gold-standard corpus of 681 medication records; this evaluation found that the algorithm predicted alignments with 99% precision and 91% recall. This performance is sufficient to support decision support applications for medication reconciliation.Keywords: clinical decision support, medication reconciliation, natural language processing, RxNorm
Procedia PDF Downloads 277361 Fast and Scale-Adaptive Target Tracking via PCA-SIFT
Authors: Yawen Wang, Hongchang Chen, Shaomei Li, Chao Gao, Jiangpeng Zhang
Abstract:
As the main challenge for target tracking is accounting for target scale change and real-time, we combine Mean-Shift and PCA-SIFT algorithm together to solve the problem. We introduce similarity comparison method to determine how the target scale changes, and taking different strategies according to different situation. For target scale getting larger will cause location error, we employ backward tracking to reduce the error. Mean-Shift algorithm has poor performance when tracking scale-changing target due to the fixed bandwidth of its kernel function. In order to overcome this problem, we introduce PCA-SIFT matching. Through key point matching between target and template, that adjusting the scale of tracking window adaptively can be achieved. Because this algorithm is sensitive to wrong match, we introduce RANSAC to reduce mismatch as far as possible. Furthermore target relocating will trigger when number of match is too small. In addition we take comprehensive consideration about target deformation and error accumulation to put forward a new template update method. Experiments on five image sequences and comparison with 6 kinds of other algorithm demonstrate favorable performance of the proposed tracking algorithm.Keywords: target tracking, PCA-SIFT, mean-shift, scale-adaptive
Procedia PDF Downloads 425360 LGG Architecture for Brain Tumor Segmentation Using Convolutional Neural Network
Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan
Abstract:
The most aggressive form of brain tumor is called glioma. Glioma is kind of tumor that arises from glial tissue of the brain and occurs quite often. A fully automatic 2D-CNN model for brain tumor segmentation is presented in this paper. We performed pre-processing steps to remove noise and intensity variances using N4ITK and standard intensity correction, respectively. We used Keras open-source library with Theano as backend for fast implementation of CNN model. In addition, we used BRATS 2015 MRI dataset to evaluate our proposed model. Furthermore, we have used SimpleITK open-source library in our proposed model to analyze images. Moreover, we have extracted random 2D patches for proposed 2D-CNN model for efficient brain segmentation. Extracting 2D patched instead of 3D due to less dimensional information present in 2D which helps us in reducing computational time. Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.77 for complete, 0.76 for core, 0.77 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.Keywords: brain tumor segmentation, convolutional neural networks, deep learning, LGG
Procedia PDF Downloads 175359 Evaluation of QSRR Models by Sum of Ranking Differences Approach: A Case Study of Prediction of Chromatographic Behavior of Pesticides
Authors: Lidija R. Jevrić, Sanja O. Podunavac-Kuzmanović, Strahinja Z. Kovačević
Abstract:
The present study deals with the selection of the most suitable quantitative structure-retention relationship (QSRR) models which should be used in prediction of the retention behavior of basic, neutral, acidic and phenolic pesticides which belong to different classes: fungicides, herbicides, metabolites, insecticides and plant growth regulators. Sum of ranking differences (SRD) approach can give a different point of view on selection of the most consistent QSRR model. SRD approach can be applied not only for ranking of the QSRR models, but also for detection of similarity or dissimilarity among them. Applying the SRD analysis, the most similar models can be found easily. In this study, selection of the best model was carried out on the basis of the reference ranking (“golden standard”) which was defined as the row average values of logarithm of retention time (logtr) defined by high performance liquid chromatography (HPLC). Also, SRD analysis based on experimental logtr values as reference ranking revealed similar grouping of the established QSRR models already obtained by hierarchical cluster analysis (HCA).Keywords: chemometrics, chromatography, pesticides, sum of ranking differences
Procedia PDF Downloads 369358 Reducing the Computational Overhead of Metaheuristics Parameterization with Exploratory Landscape Analysis
Authors: Iannick Gagnon, Alain April
Abstract:
The performance of a metaheuristic on a given problem class depends on the class itself and the choice of parameters. Parameter tuning is the most time-consuming phase of the optimization process after the main calculations and it often nullifies the speed advantage of metaheuristics over traditional optimization algorithms. Several off-the-shelf parameter tuning algorithms are available, but when the objective function is expensive to evaluate, these can be prohibitively expensive to use. This paper presents a surrogate-like method for finding adequate parameters using fitness landscape analysis on simple benchmark functions and real-world objective functions. The result is a simple compound similarity metric based on the empirical correlation coefficient and a measure of convexity. It is then used to find the best benchmark functions to serve as surrogates. The near-optimal parameter set is then found using fractional factorial design. The real-world problem of NACA airfoil lift coefficient maximization is used as a preliminary proof of concept. The overall aim of this research is to reduce the computational overhead of metaheuristics parameterization.Keywords: metaheuristics, stochastic optimization, particle swarm optimization, exploratory landscape analysis
Procedia PDF Downloads 147357 Synthesis of Balanced 3-RRR Planar Parallel Manipulators
Authors: Arakelian Vigen, Geng Jing, Le Baron Jean-Paul
Abstract:
The paper deals with the design of parallel manipulators with balanced inertia forces and moments. The balancing of the resultant of the inertia forces of 3-RRR planar parallel manipulators is carried out through mass redistribution and centre of mass acceleration minimization. The proposed balancing technique is achieved in two steps: at first, optimal redistribution of the masses of input links is accomplished, which ensures the similarity of the end-effector trajectory and the manipulator’s common centre of mass trajectory, then, optimal trajectory planning of the end-effector by 'bang-bang' profile is reached. In such a way, the minimization of the magnitude of the acceleration of the centre of mass of the manipulator brings about a minimization of shaking force. To minimize the resultant of the inertia moments (shaking moment), the active balancing via inertia flywheel is applied. However, in this case, the active balancing is quite different from previous applications because it provides only a partial cancellation of the shaking moment due to the incomplete balancing of shaking force.Keywords: dynamic balancing, inertia force minimization, inertia moment minimization, 3-RRR planar parallel manipulator
Procedia PDF Downloads 453356 Failure Mode Effect and Criticality Analysis Based Maintenance Planning through Traditional and Multi-Criteria Decision Making Approach for Aluminium Wire Rolling Mill Plant
Authors: Nilesh Pancholi, Mangal Bhatt
Abstract:
This paper highlights comparative results of traditional FMECA and multi-factor decision-making approach based on “Technique for Order Preference by Similarity to Ideal Solution (TOPSIS)” for aluminum wire rolling mill plant. The suggested study is carried out to overcome the limitations of FMECA by assigning the scores against each failure modes in crisp values to evaluate the criticalities of the failure modes without uncertainty. The primary findings of the paper are that sudden impact on the rolls seems to be most critical failure cause and high contact stresses due to rolling & sliding action of mesh to be least critical failure cause. It is suggested to modify the current control practices with proper maintenance strategy based on achieved maintainability criticality index (MCI). The outcome of the study will be helpful in deriving optimized maintenance plan to maximize the performance of continuous process industry.Keywords: reliability, maintenance, FMECA, TOPSIS, process industry
Procedia PDF Downloads 273355 Content-Based Mammograms Retrieval Based on Breast Density Criteria Using Bidimensional Empirical Mode Decomposition
Authors: Sourour Khouaja, Hejer Jlassi, Nadia Feddaoui, Kamel Hamrouni
Abstract:
Most medical images, and especially mammographies, are now stored in large databases. Retrieving a desired image is considered of great importance in order to find previous similar cases diagnosis. Our method is implemented to assist radiologists in retrieving mammographic images containing breast with similar density aspect as seen on the mammogram. This is becoming a challenge seeing the importance of density criteria in cancer provision and its effect on segmentation issues. We used the BEMD (Bidimensional Empirical Mode Decomposition) to characterize the content of images and Euclidean distance measure similarity between images. Through the experiments on the MIAS mammography image database, we confirm that the results are promising. The performance was evaluated using precision and recall curves comparing query and retrieved images. Computing recall-precision proved the effectiveness of applying the CBIR in the large mammographic image databases. We found a precision of 91.2% for mammography with a recall of 86.8%.Keywords: BEMD, breast density, contend-based, image retrieval, mammography
Procedia PDF Downloads 227354 Synthesis and Pharmaco-Potential Evaluation of Quinoline Hybrids
Authors: Paul Awolade, Parvesh Singh
Abstract:
The global threat of pathogenic resistance to available therapeutic agents has become a menace to clinical practice, public health and man’s existence inconsequential. This has therefore led to an exigency in the development of new molecular scaffolds with profound activity profiles. In this vein, a versatile synthetic tool for accessing new molecules by incorporating two or more pharmacophores into a single entity with the unique ability to be recognized by multiple receptors hence leading to an improved bioactivity, known as molecular hybridization, has been explored with tremendous success. Accordingly, aware of the similarity in pharmacological activity spectrum of quinoline and 1,2,3-triazole pharmacophores such as; anti-Alzheimer, anticancer, anti-HIV, antimalarial and antimicrobial to mention but a few, the present study sets out to synthesize hybrids of quinoline and 1,2,3-triazole. The hybrids were accessed via click chemistry using copper catalysed azide-alkyne 1,3-dipolar cycloaddition reaction. All synthesized compounds were evaluated for their pharmaco-potential in an antimicrobial assay out of which the 3-OH derivative emerged as the most active with MIC value of 4 μg/mL against Cryptococcus neoformans; a value superior to standard Fluconazole and comparable to Amphotericin B. Structures of synthesized hybrids were elucidated using appropriate spectroscopic techniques (1H, 13C and 2D NMR, FT-IR and HRMS).Keywords: bioisostere, click chemistry, molecular hybridization, quinoline, 1, 2, 3-triazole
Procedia PDF Downloads 120353 British English vs. American English: A Comparative Study
Authors: Halima Benazzouz
Abstract:
It is often believed that British English and American English are the foremost varieties of the English Language serving as reference norms for other varieties;that is the reason why they have obviously been compared and contrasted.Meanwhile,the terms “British English” and “American English” are used differently by different people to refer to: 1) Two national varieties each subsuming regional and other sub-varieties standard and non-standard. 2) Two national standard varieties in which each one is only part of the range of English within its own state, but the most prestigious part. 3) Two international varieties, that is each is more than a national variety of the English Language. 4) Two international standard varieties that may or may not each subsume other standard varieties.Furthermore,each variety serves as a reference norm for users of the language elsewhere. Moreover, without a clear identification, as primarily belonging to one variety or the other, British English(Br.Eng) and American English (Am.Eng) are understood as national or international varieties. British English and American English are both “variants” and “varieties” of the English Language, more similar than different.In brief, the following may justify general categories of difference between Standard American English (S.Am.E) and Standard British English (S.Br.e) each having their own sociolectic value: A difference in pronunciation exists between the two foremost varieties, although it is the same spelling, by contrast, a divergence in spelling may be recognized, eventhough the same pronunciation. In such case, the same term is different but there is a similarity in spelling and pronunciation. Otherwise, grammar, syntax, and punctuation are distinctively used to distinguish the two varieties of the English Language. Beyond these differences, spelling is noted as one of the chief sources of variation.Keywords: Greek, Latin, French pronunciation expert, varieties of English language
Procedia PDF Downloads 492352 Prediction for the Pressure Drop of Gas-Liquid Cylindrical Cyclone in Sub-Sea Production System
Authors: Xu Rumin, Chen Jianyi, Yue Ti, Wang Yaan
Abstract:
With the rapid development of subsea oil and gas exploitation, the demand for the related underwater process equipment is increasing fast. In order to reduce the energy consuming, people tend to separate the gas and oil phase directly on the seabed. Accordingly, an advanced separator is needed. In this paper, the pressure drop of a new type of separator named Gas Liquid Cylindrical Cyclone (GLCC) which is used in the subsea system is investigated by both experiments and numerical simulation. In the experiments, the single phase flow and gas-liquid two phase flow in GLCC were tested. For the simulation, the performance of GLCC under both laboratory and industrial conditions was calculated. The Eulerian model was implemented to describe the mixture flow field in the GLCC under experimental conditions and industrial oil-natural gas conditions. Furthermore, a relationship among Euler number (Eu), Reynolds number (Re), and Froude number (Fr) is generated according to similarity analysis and simulation data, which can present the GLCC separation performance of pressure drop. These results can give reference to the design and application of GLCC in deep sea.Keywords: dimensionless analysis, gas-liquid cylindrical cyclone, numerical simulation, pressure drop
Procedia PDF Downloads 164351 Comparative Analysis of Edge Detection Techniques for Extracting Characters
Authors: Rana Gill, Chandandeep Kaur
Abstract:
Segmentation of images can be implemented using different fundamental algorithms like edge detection (discontinuity based segmentation), region growing (similarity based segmentation), iterative thresholding method. A comprehensive literature review relevant to the study gives description of different techniques for vehicle number plate detection and edge detection techniques widely used on different types of images. This research work is based on edge detection techniques and calculating threshold on the basis of five edge operators. Five operators used are Prewitt, Roberts, Sobel, LoG and Canny. Segmentation of characters present in different type of images like vehicle number plate, name plate of house and characters on different sign boards are selected as a case study in this work. The proposed methodology has seven stages. The proposed system has been implemented using MATLAB R2010a. Comparison of all the five operators has been done on the basis of their performance. From the results it is found that Canny operators produce best results among the used operators and performance of different edge operators in decreasing order is: Canny>Log>Sobel>Prewitt>Roberts.Keywords: segmentation, edge detection, text, extracting characters
Procedia PDF Downloads 423350 Identification and Characterization of 18S rRNA Gene of Demodex Canis From the Dog Population of Mizoram, India
Authors: Moneesh Thakur, Hridayesh Prasad, Nikitasha Bora, Parimal Roy Choudhary, A. K. Samanta, Sanjeev Kumar
Abstract:
Canine demodicosis is a common parasitic condition which involves dog skin. Demodicosis in dogs is due the prominent growth of Demodex. Out of various canine Demodex spp., Demodex canis is the most often involved species. Canine demodicosis can occur as either a localized or generalized form of demodicosis severely affect the dogs and in non-treated dogs may cause death. This study was planned with the aim to screen and characterize the 18S rRNA gene of isolated Demodex canis. A total of 1200 dogs were screened during this study period. The skin scrapings of all the suspected dogs were examined under a microscope at 100X magnification for the presence of Demodex canis. The skin scrapings positive for Demodex canis were examined using PCR for confirmation. A total of 35 dogs were confirmed a positive result for D. canis based on 18S rRNA gene amplification by PCR. Further, the 18S rRNA gene of isolated Demodex canis was cloned and sequenced for genome analysis. On the sequence analysis, it was found that isolated sequence (GenBank Accession No. MK177513) had close similarity (99.7%) to that of D. canis genotype of China (Accession No. MG372254).Keywords: PCR, phylogenetic analysis, cloning and sequening, Demodex canis
Procedia PDF Downloads 88349 Evaluation of Genetic Diversity in Iranian Native Silkworm Bombyx mori Using RAPD (Random Amplification of Polymorphic DNA) Molecular Marker
Authors: Rouhollah Radjabi, Mojtaba Zarei, Elham Sanatgar, Hossein Shouhani
Abstract:
RAPD molecular markers in order to discrimination of the Iranian native Bombyx mori silkworm breeds were used. DNA extraction using phenol - chloroform was and the qualitative and quantitative measurements of extracted DNA and its dilution, the obtained bands on agarose gel 1.5 percent were marked and analyzed. Results showed that the bands are observed between 250-2500 bp and most bands have been observed as Gilani-orange, the lowest bands observed are Khorasani-lemon. Primer 3 with 100% polymorphism with the highest polymorphism and primer 2 with 61.5 polymorphism had the lowest percentage of polymorphism. Cluster analysis of races and placed them in three main groups, races Gilani - orange, Baghdad and Khorasani -pink if the first group, camel's thorn, Herati - yellow race was alone in the second group and Khorasani – lemon was alone in the third group. The greatest similarity between the races, between Khorasani- pink and Baghdad (0.64). RAPD markers have been determined different silkworm races based on various morphological or economic characteristics except geographic origin.Keywords: silkworm, molecular marker, RAPD, Iran
Procedia PDF Downloads 427348 Multimodal Direct Neural Network Positron Emission Tomography Reconstruction
Authors: William Whiteley, Jens Gregor
Abstract:
In recent developments of direct neural network based positron emission tomography (PET) reconstruction, two prominent architectures have emerged for converting measurement data into images: 1) networks that contain fully-connected layers; and 2) networks that primarily use a convolutional encoder-decoder architecture. In this paper, we present a multi-modal direct PET reconstruction method called MDPET, which is a hybrid approach that combines the advantages of both types of networks. MDPET processes raw data in the form of sinograms and histo-images in concert with attenuation maps to produce high quality multi-slice PET images (e.g., 8x440x440). MDPET is trained on a large whole-body patient data set and evaluated both quantitatively and qualitatively against target images reconstructed with the standard PET reconstruction benchmark of iterative ordered subsets expectation maximization. The results show that MDPET outperforms the best previously published direct neural network methods in measures of bias, signal-to-noise ratio, mean absolute error, and structural similarity.Keywords: deep learning, image reconstruction, machine learning, neural network, positron emission tomography
Procedia PDF Downloads 106347 Human Action Retrieval System Using Features Weight Updating Based Relevance Feedback Approach
Authors: Munaf Rashid
Abstract:
For content-based human action retrieval systems, search accuracy is often inferior because of the following two reasons 1) global information pertaining to videos is totally ignored, only low level motion descriptors are considered as a significant feature to match the similarity between query and database videos, and 2) the semantic gap between the high level user concept and low level visual features. Hence, in this paper, we propose a method that will address these two issues and in doing so, this paper contributes in two ways. Firstly, we introduce a method that uses both global and local information in one framework for an action retrieval task. Secondly, to minimize the semantic gap, a user concept is involved by incorporating features weight updating (FWU) Relevance Feedback (RF) approach. We use statistical characteristics to dynamically update weights of the feature descriptors so that after every RF iteration feature space is modified accordingly. For testing and validation purpose two human action recognition datasets have been utilized, namely Weizmann and UCF. Results show that even with a number of visual challenges the proposed approach performs well.Keywords: relevance feedback (RF), action retrieval, semantic gap, feature descriptor, codebook
Procedia PDF Downloads 461346 The Dynamics of Unsteady Squeezing Flow between Parallel Plates (Two-Dimensional)
Authors: Jiya Mohammed, Ibrahim Ismail Giwa
Abstract:
Unsteady squeezing flow of a viscous fluid between parallel plates is considered. The two plates are considered to be approaching each other symmetrically, causing the squeezing flow. Two-dimensional rectangular Cartesian coordinate is considered. The Navier-Stokes equation was reduced using similarity transformation to a single fourth order non-linear ordinary differential equation. The energy equation was transformed to a second order coupled differential equation. We obtained solution to the resulting ordinary differential equations via Homotopy Perturbation Method (HPM). HPM deforms a differential problem into a set of problem that are easier to solve and it produces analytic approximate expression in the form of an infinite power series by using only sixth and fifth terms for the velocity and temperature respectively. The results reveal that the proposed method is very effective and simple. Comparisons among present and existing solutions were provided and it is shown that the proposed method is in good agreement with Variation of Parameter Method (VPM). The effects of appropriate dimensionless parameters on the velocity profiles and temperature field are demonstrated with the aid of comprehensive graphs and tables.Keywords: coupled differential equation, Homotopy Perturbation Method, plates, squeezing flow
Procedia PDF Downloads 466345 Meyer Wavelet Transform and Jaccard Deep Q-Net for Small Object Classification Using Multi-Modal Images
Authors: Mian Muhammad Kamal
Abstract:
Accurate detection of small objects is extremely essential in critical applications like military reconnaissance and emergency rescue. However, owing to the low resolution, occlusion, and background interference, small object detection is a tedious process. One of the most appropriate approaches is to combine the data available in multimodal images to enhance the detection ability. This paper proposes a small object detection technique using three kinds of multimodal images, such as Hyperspectral-Multispectral (HS-MS), HS-Synthetic Aperture Radar (HS-SAR), and HS-SAR-Digital Surface Model (HS-SAR-DSM). The detection is accomplished by utilizing the Jaccard Deep Q-Net (JDQN) that is created by the incorporation of the Jaccard similarity measure and Deep Q-Network (DQN) using Regression modeling. Further, a Deep Maxout Network (DMN) is used for fusing the detected outputs obtained from each modality so as to generate the final output. Moreover, the supremacy of the proposed JDQN in detecting small objects is established by the utilization of metrics, like accuracy, Mean Squared Error (MSE), precision, and Root MSE (RMSE), and experimentation reveals that the JDQN recorded superior accuracy of 0.907, normalized MSE of 0.448, precision of 0.904, and normalized RMSE of 0.670.Keywords: small object detection, Multimodality, deep learning, Jaccard deep q-net, deep maxout network
Procedia PDF Downloads 8344 Hybridized Approach for Distance Estimation Using K-Means Clustering
Authors: Ritu Vashistha, Jitender Kumar
Abstract:
Clustering using the K-means algorithm is a very common way to understand and analyze the obtained output data. When a similar object is grouped, this is called the basis of Clustering. There is K number of objects and C number of cluster in to single cluster in which k is always supposed to be less than C having each cluster to be its own centroid but the major problem is how is identify the cluster is correct based on the data. Formulation of the cluster is not a regular task for every tuple of row record or entity but it is done by an iterative process. Each and every record, tuple, entity is checked and examined and similarity dissimilarity is examined. So this iterative process seems to be very lengthy and unable to give optimal output for the cluster and time taken to find the cluster. To overcome the drawback challenge, we are proposing a formula to find the clusters at the run time, so this approach can give us optimal results. The proposed approach uses the Euclidian distance formula as well melanosis to find the minimum distance between slots as technically we called clusters and the same approach we have also applied to Ant Colony Optimization(ACO) algorithm, which results in the production of two and multi-dimensional matrix.Keywords: ant colony optimization, data clustering, centroids, data mining, k-means
Procedia PDF Downloads 124343 Obstacle Classification Method Based on 2D LIDAR Database
Authors: Moohyun Lee, Soojung Hur, Yongwan Park
Abstract:
In this paper is proposed a method uses only LIDAR system to classification an obstacle and determine its type by establishing database for classifying obstacles based on LIDAR. The existing LIDAR system, in determining the recognition of obstruction in an autonomous vehicle, has an advantage in terms of accuracy and shorter recognition time. However, it was difficult to determine the type of obstacle and therefore accurate path planning based on the type of obstacle was not possible. In order to overcome this problem, a method of classifying obstacle type based on existing LIDAR and using the width of obstacle materials was proposed. However, width measurement was not sufficient to improve accuracy. In this research, the width data was used to do the first classification; database for LIDAR intensity data by four major obstacle materials on the road were created; comparison is made to the LIDAR intensity data of actual obstacle materials; and determine the obstacle type by finding the one with highest similarity values. An experiment using an actual autonomous vehicle under real environment shows that data declined in quality in comparison to 3D LIDAR and it was possible to classify obstacle materials using 2D LIDAR.Keywords: obstacle, classification, database, LIDAR, segmentation, intensity
Procedia PDF Downloads 337342 Fast Prediction Unit Partition Decision and Accelerating the Algorithm Using Cudafor Intra and Inter Prediction of HEVC
Authors: Qiang Zhang, Chun Yuan
Abstract:
Since the PU (Prediction Unit) decision process is the most time consuming part of the emerging HEVC (High Efficient Video Coding) standardin intra and inter frame coding, this paper proposes the fast PU decision algorithm and speed up the algorithm using CUDA (Compute Unified Device Architecture). In intra frame coding, the fast PU decision algorithm uses the texture features to skip intra-frame prediction or terminal the intra-frame prediction for smaller PU size. In inter frame coding of HEVC, the fast PU decision algorithm takes use of the similarity of its own two Nx2N size PU's motion vectors and the hierarchical structure of CU (Coding Unit) partition to skip some modes of PU partition, so as to reduce the motion estimation times. The accelerate algorithm using CUDA is based on the fast PU decision algorithm which uses the GPU to make the motion search and the gradient computation could be parallel computed. The proposed algorithm achieves up to 57% time saving compared to the HM 10.0 with little rate-distortion losses (0.043dB drop and 1.82% bitrate increase on average).Keywords: HEVC, PU decision, inter prediction, intra prediction, CUDA, parallel
Procedia PDF Downloads 391341 Deep Graph Embeddings for the Analysis of Short Heartbeat Interval Time Series
Authors: Tamas Madl
Abstract:
Sudden cardiac death (SCD) constitutes a large proportion of cardiovascular mortalities, provides little advance warning, and the risk is difficult to recognize based on ubiquitous, low cost medical equipment such as the standard, 12-lead, ten second ECG. Autonomic abnormalities have been shown to be strongly predictive of SCD risk; yet current methods are not trivially applicable to the brevity and low temporal and electrical resolution of standard ECGs. Here, we build horizontal visibility graph representations of very short inter-beat interval time series, and perform unsuper- vised representation learning in order to convert these variable size objects into fixed-length vectors preserving similarity rela- tions. We show that such representations facilitate classification into healthy vs. at-risk patients on two different datasets, the Mul- tiparameter Intelligent Monitoring in Intensive Care II and the PhysioNet Sudden Cardiac Death Holter Database. Our results suggest that graph representation learning of heartbeat interval time series facilitates robust classification even in sequences as short as ten seconds.Keywords: sudden cardiac death, heart rate variability, ECG analysis, time series classification
Procedia PDF Downloads 227340 Ensemble of Deep CNN Architecture for Classifying the Source and Quality of Teff Cereal
Authors: Belayneh Matebie, Michael Melese
Abstract:
The study focuses on addressing the challenges in classifying and ensuring the quality of Eragrostis Teff, a small and round grain that is the smallest cereal grain. Employing a traditional classification method is challenging because of its small size and the similarity of its environmental characteristics. To overcome this, this study employs a machine learning approach to develop a source and quality classification system for Teff cereal. Data is collected from various production areas in the Amhara regions, considering two types of cereal (high and low quality) across eight classes. A total of 5,920 images are collected, with 740 images for each class. Image enhancement techniques, including scaling, data augmentation, histogram equalization, and noise removal, are applied to preprocess the data. Convolutional Neural Network (CNN) is then used to extract relevant features and reduce dimensionality. The dataset is split into 80% for training and 20% for testing. Different classifiers, including FVGG16, FINCV3, QSCTC, EMQSCTC, SVM, and RF, are employed for classification, achieving accuracy rates ranging from 86.91% to 97.72%. The ensemble of FVGG16, FINCV3, and QSCTC using the Max-Voting approach outperforms individual algorithms.Keywords: Teff, ensemble learning, max-voting, CNN, SVM, RF
Procedia PDF Downloads 40339 Image Analysis for Obturator Foramen Based on Marker-controlled Watershed Segmentation and Zernike Moments
Authors: Seda Sahin, Emin Akata
Abstract:
Obturator foramen is a specific structure in pelvic bone images and recognition of it is a new concept in medical image processing. Moreover, segmentation of bone structures such as obturator foramen plays an essential role for clinical research in orthopedics. In this paper, we present a novel method to analyze the similarity between the substructures of the imaged region and a hand drawn template, on hip radiographs to detect obturator foramen accurately with integrated usage of Marker-controlled Watershed segmentation and Zernike moment feature descriptor. Marker-controlled Watershed segmentation is applied to seperate obturator foramen from the background effectively. Zernike moment feature descriptor is used to provide matching between binary template image and the segmented binary image for obturator foramens for final extraction. The proposed method is tested on randomly selected 100 hip radiographs. The experimental results represent that our method is able to segment obturator foramens with % 96 accuracy.Keywords: medical image analysis, segmentation of bone structures on hip radiographs, marker-controlled watershed segmentation, zernike moment feature descriptor
Procedia PDF Downloads 427338 Image Features Comparison-Based Position Estimation Method Using a Camera Sensor
Authors: Jinseon Song, Yongwan Park
Abstract:
In this paper, propose method that can user’s position that based on database is built from single camera. Previous positioning calculate distance by arrival-time of signal like GPS (Global Positioning System), RF(Radio Frequency). However, these previous method have weakness because these have large error range according to signal interference. Method for solution estimate position by camera sensor. But, signal camera is difficult to obtain relative position data and stereo camera is difficult to provide real-time position data because of a lot of image data, too. First of all, in this research we build image database at space that able to provide positioning service with single camera. Next, we judge similarity through image matching of database image and transmission image from user. Finally, we decide position of user through position of most similar database image. For verification of propose method, we experiment at real-environment like indoor and outdoor. Propose method is wide positioning range and this method can verify not only position of user but also direction.Keywords: positioning, distance, camera, features, SURF(Speed-Up Robust Features), database, estimation
Procedia PDF Downloads 342337 Investment Projects Selection Problem under Hesitant Fuzzy Environment
Authors: Irina Khutsishvili
Abstract:
In the present research, a decision support methodology for the multi-attribute group decision-making (MAGDM) problem is developed, namely for the selection of investment projects. The objective of the investment project selection problem is to choose the best project among the set of projects, seeking investment, or to rank all projects in descending order. The project selection is made considering a set of weighted attributes. To evaluate the attributes in our approach, expert assessments are used. In the proposed methodology, lingual expressions (linguistic terms) given by all experts are used as initial attribute evaluations, since they are the most natural and convenient representation of experts' evaluations. Then lingual evaluations are converted into trapezoidal fuzzy numbers, and the aggregate trapezoidal hesitant fuzzy decision matrix will be built. The case is considered when information on the attribute weights is completely unknown. The attribute weights are identified based on the De Luca and Termini information entropy concept, determined in the context of hesitant fuzzy sets. The decisions are made using the extended Technique for Order Performance by Similarity to Ideal Solution (TOPSIS) method under a hesitant fuzzy environment. Hence, a methodology is based on a trapezoidal valued hesitant fuzzy TOPSIS decision-making model with entropy weights. The ranking of alternatives is performed by the proximity of their distances to both the fuzzy positive-ideal solution (FPIS) and the fuzzy negative-ideal solution (FNIS). For this purpose, the weighted hesitant Hamming distance is used. An example of investment decision-making is shown that clearly explains the procedure of the proposed methodology.Keywords: In the present research, a decision support methodology for the multi-attribute group decision-making (MAGDM) problem is developed, namely for the selection of investment projects. The objective of the investment project selection problem is to choose the best project among the set of projects, seeking investment, or to rank all projects in descending order. The project selection is made considering a set of weighted attributes. To evaluate the attributes in our approach, expert assessments are used. In the proposed methodology, lingual expressions (linguistic terms) given by all experts are used as initial attribute evaluations since they are the most natural and convenient representation of experts' evaluations. Then lingual evaluations are converted into trapezoidal fuzzy numbers, and the aggregate trapezoidal hesitant fuzzy decision matrix will be built. The case is considered when information on the attribute weights is completely unknown. The attribute weights are identified based on the De Luca and Termini information entropy concept, determined in the context of hesitant fuzzy sets. The decisions are made using the extended Technique for Order Performance by Similarity to Ideal Solution (TOPSIS) method under a hesitant fuzzy environment. Hence, a methodology is based on a trapezoidal valued hesitant fuzzy TOPSIS decision-making model with entropy weights. The ranking of alternatives is performed by the proximity of their distances to both the fuzzy positive-ideal solution (FPIS) and the fuzzy negative-ideal solution (FNIS). For this purpose, the weighted hesitant Hamming distance is used. An example of investment decision-making is shown that clearly explains the procedure of the proposed methodology.
Procedia PDF Downloads 114336 An Accurate Brain Tumor Segmentation for High Graded Glioma Using Deep Learning
Authors: Sajeeha Ansar, Asad Ali Safi, Sheikh Ziauddin, Ahmad R. Shahid, Faraz Ahsan
Abstract:
Gliomas are most challenging and aggressive type of tumors which appear in different sizes, locations, and scattered boundaries. CNN is most efficient deep learning approach with outstanding capability of solving image analysis problems. A fully automatic deep learning based 2D-CNN model for brain tumor segmentation is presented in this paper. We used small convolution filters (3 x 3) to make architecture deeper. We increased convolutional layers for efficient learning of complex features from large dataset. We achieved better results by pushing convolutional layers up to 16 layers for HGG model. We achieved reliable and accurate results through fine-tuning among dataset and hyper-parameters. Pre-processing of this model includes generation of brain pipeline, intensity normalization, bias correction and data augmentation. We used the BRATS-2015, and Dice Similarity Coefficient (DSC) is used as performance measure for the evaluation of the proposed method. Our method achieved DSC score of 0.81 for complete, 0.79 for core, 0.80 for enhanced tumor regions. However, these results are comparable with methods already implemented 2D CNN architecture.Keywords: brain tumor segmentation, convolutional neural networks, deep learning, HGG
Procedia PDF Downloads 247