Search results for: fusion feature

Authors: Y. Fekir, K. Mederbal, M. A. Hammadouche, D. Anteur

Abstract:

The cultural park of the Tassili, occupying a large area of Algeria, is characterized by a rich vegetative biodiversity to be preserved and managed both in time and space. The management of a large area (case of Tassili), by its complexity, needs large amounts of data, which for the most part, are spatially localized (DEM, satellite images and socio-economic information etc.), where the use of conventional and traditional methods is quite difficult. The remote sensing, by its efficiency in environmental applications, became an indispensable solution for this kind of studies. Multispectral imaging sensors have been very useful in the last decade in very interesting applications of remote sensing. They can aid in several domains such as the de¬tection and identification of diverse surface targets, topographical details, and geological features. In this work, we try to extract vegetative areas using fusion techniques between data acquired from sensor on-board the Earth Observing 1 (EO-1) satellite and Landsat ETM+ and TM sensors. We have used images acquired over the Oasis of Djanet in the National Park of Tassili in the south of Algeria. Fusion technqiues were applied on the obtained image to extract the vegetative fraction of the different classes of land use. We compare the obtained results in vegetation end member extraction with vegetation indices calculated from both Hyperion and other multispectral sensors.

Keywords: Landsat ETM+, EO1, data fusion, vegetation, Tassili, Algeria

Procedia PDF Downloads 410

Authors: Javad Rahimipour Anaraki, Saeed Samet, Mahdi Eftekhari, Chang Wook Ahn

Abstract:

Feature selection and attribute reduction are crucial problems, and widely used techniques in the field of machine learning, data mining and pattern recognition to overcome the well-known phenomenon of the Curse of Dimensionality. This paper presents a feature selection method that efficiently carries out attribute reduction, thereby selecting the most informative features of a dataset. It consists of two components: 1) a measure for feature subset evaluation, and 2) a search strategy. For the evaluation measure, we have employed the fuzzy-rough dependency degree (FRFDD) of the lower approximation-based fuzzy-rough feature selection (L-FRFS) due to its effectiveness in feature selection. As for the search strategy, a modified version of a binary shuffled frog leaping algorithm is proposed (B-SFLA). The proposed feature selection method is obtained by hybridizing the B-SFLA with the FRDD. Nine classifiers have been employed to compare the proposed approach with several existing methods over twenty two datasets, including nine high dimensional and large ones, from the UCI repository. The experimental results demonstrate that the B-SFLA approach significantly outperforms other metaheuristic methods in terms of the number of selected features and the classification accuracy.

Keywords: binary shuffled frog leaping algorithm, feature selection, fuzzy-rough set, minimal reduct

Procedia PDF Downloads 188

Authors: Chih-Yi Ho, Tzu-Fang Chang, Chih-Chia Huang, Chia-Yen Lee

Abstract:

In dental clinics, the dentists use the periapical radiography image to assess the effectiveness of endodontic treatment of teeth with chronic apical periodontitis. Periapical radiography images are taken at different times to assess alveolar bone variation before and after the root canal treatment, and furthermore to judge whether the treatment was successful. Current clinical assessment of apical tissue recovery relies only on dentist personal experience. It is difficult to have the same standard and objective interpretations due to the dentist or radiologist personal background and knowledge. If periapical radiography images at the different time could be registered well, the endodontic treatment could be evaluated. In the image registration area, it is necessary to assign representative control points to the transformation model for good performances of registration results. However, detection of representative control points (feature points) on periapical radiography images is generally very difficult. Regardless of which traditional detection methods are practiced, sufficient feature points may not be detected due to the low-contrast characteristics of the x-ray image. Barnard detector is an algorithm for feature point detection based on grayscale value gradients, which can obtain sufficient feature points in the case of gray-scale contrast is not obvious. However, the Barnard detector would detect too many feature points, and they would be too clustered. This study uses the local extrema of clustering feature points and the suppression radius to overcome the problem, and compared different feature point detection methods. In the preliminary result, the feature points could be detected as representative control points by the proposed method.

Keywords: feature detection, Barnard detector, registration, periapical radiography image, endodontic treatment

Procedia PDF Downloads 424

Authors: Rabab M. Ramadan, Elaraby A. Elgallad

Abstract:

With extensive application, the performance of unimodal biometrics systems has to face a diversity of problems such as signal and background noise, distortion, and environment differences. Therefore, multimodal biometric systems are proposed to solve the above stated problems. This paper introduces a bimodal biometric recognition system based on the extracted features of the human palm print and iris. Palm print biometric is fairly a new evolving technology that is used to identify people by their palm features. The iris is a strong competitor together with face and fingerprints for presence in multimodal recognition systems. In this research, we introduced an algorithm to the combination of the palm and iris-extracted features using a texture-based descriptor, the Scale Invariant Feature Transform (SIFT). Since the feature sets are non-homogeneous as features of different biometric modalities are used, these features will be concatenated to form a single feature vector. Particle swarm optimization (PSO) is used as a feature selection technique to reduce the dimensionality of the feature. The proposed algorithm will be applied to the Institute of Technology of Delhi (IITD) database and its performance will be compared with various iris recognition algorithms found in the literature.

Keywords: iris recognition, particle swarm optimization, feature extraction, feature selection, palm print, the Scale Invariant Feature Transform (SIFT)

Procedia PDF Downloads 197

Authors: Dake Xiong, Ben Hankamer, Ian Ross

Abstract:

The most urgent challenge of our time is to replace the depleting resources of fossil fuels by sustainable environmentally friendly alternatives. Hydrogen is a promising CO2-neutral fuel for a more sustainable future especially when produced photo-biologically. Hydrogen can be photosynthetically produced in unicellular green alga like Chlamydomonas reinhardtii, catalysed by the inducible highly active and bidirectional [FeFe]-hydrogenase enzymes (HydA). However, evolutionary and physiological constraints severely restrict the hydrogen yield of algae for industrial scale-up, mainly due to its competition among other metabolic pathways on photosynthetic electrons. Among them, a major challenge to be resolved is the inferior competitiveness of hydrogen production (catalysed by HydA) with NADPH production (catalysed by ferredoxin-NADP+-reductase (FNR)), which is essential for cell growth and takes up ~95% of photosynthetic electrons. In this work, the in vivo hydrogen production efficiency of mutants with ferredoxin-hydrogenase (Fd*-HydA1*) fusion protein construct, where the electron donor ferredoxin (Fd*) is fused to HydA1* and expressed in the model organism C. reinhardtii was investigated. Once Fd*-HydA1* fusion gene is expressed in algal cells, the fusion enzyme is able to draw the redistributed photosynthetic electrons and use them for efficient hydrogen production. From preliminary data, mutants with Fd*-HydA1* transgene showed a ~2-fold increase in the photosynthetic hydrogen production rate compared with its parental strain, which only possesses the native HydA in vivo. Therefore, a solid method of having more efficient hydrogen production in microalgae can be achieved through the expression of the synthetic enzymes.

Keywords: Chlamydomonas reinhardtii, ferredoxin, fusion protein, hydrogen production, hydrogenase

Procedia PDF Downloads 235

Authors: Ke Ren, Huiruo Shi, Linsen Li, Baoshuai Wang, Yu Zhou

Abstract:

Since some discriminative features are required for ground moving targets classification, we propose a new feature extraction method based on micro-Doppler signature. Firstly, the time-frequency analysis of measured data indicates that the time-frequency spectrograms of the three kinds of ground moving targets, i.e., single walking person, two people walking and a moving wheeled vehicle, are discriminative. Then, a three-dimensional time-frequency feature vector is extracted from the time-frequency spectrograms to depict these differences. At last, a Support Vector Machine (SVM) classifier is trained with the proposed three-dimensional feature vector. The classification accuracy to categorize ground moving targets into the three kinds of the measured data is found to be over 96%, which demonstrates the good discriminative ability of the proposed micro-Doppler feature.

Keywords: micro-doppler, time-frequency analysis, feature extraction, radar target classification

Procedia PDF Downloads 382

Authors: Kohei Inoue, Kenji Hara, Kiichi Urahama

Abstract:

We describe a relationship between integral images and differential images. First, we derive a simple difference filter from conventional integral image. In the derivation, we show that an integral image and the corresponding differential image are related to each other by simultaneous linear equations, where the numbers of unknowns and equations are the same, and therefore, we can execute the integration and differentiation by solving the simultaneous equations. We applied the relationship to an image fusion problem, and experimentally verified the effectiveness of the proposed method.

Keywords: integral images, differential images, differential filters, image fusion

Procedia PDF Downloads 476

Authors: Huang Bai-Cheng

Abstract:

When deploying the Oriented Fast and Rotated Binary Robust Independent Elementary Feature (BRIEF) (ORB) extraction algorithm on field-programmable gate arrays (FPGA), the access of global storage for 31×31 pixel patches of the features has become the bottleneck of the system efficiency. Therefore, a feature zone strategy has been proposed. Zones are searched as features are detected. Pixels around the feature zones are extracted from global memory and distributed into patches corresponding to feature coordinates. The proposed FPGA structure is targeted on a Xilinx FPGA development board of Zynq UltraScale+ series, and multiple datasets are tested. Compared with the streaming pixel patch extraction method, the proposed architecture obtains at least two times acceleration consuming extra 3.82% Flip-Flops (FFs) and 7.78% Look-Up Tables (LUTs). Compared with the non-streaming one, the proposed architecture saves 22.3% LUT and 1.82% FF, causing a latency of only 0.2ms and a drop in frame rate for 1. Compared with the related works, the proposed strategy and hardware architecture have the superiority of keeping a balance between FPGA resources and performance.

Keywords: feature extraction, real-time, ORB, FPGA implementation

Procedia PDF Downloads 95

Authors: Sachin, Shantanu Arya, Gunjan Mehta, Md. Shamim Ansari

Abstract:

The cross-modal influence of visual information on speech perception can be illustrated by the McGurk effect which is an illusion of hearing of syllable /ta/ when a listener listens one syllable, e.g.: /pa/ while watching a synchronized video recording of syllable, /ka/. The McGurk effect is an excellent tool to investigate multisensory integration in speech perception in both normal hearing and hearing impaired populations. As the visual cue is unaffected by noise, individuals with hearing impairment rely more than normal listeners on the visual cues.However, when non congruent visual and auditory cues are processed together, audiovisual interaction seems to occur differently in normal and persons with hearing impairment. Therefore, this study aims to observe the audiovisual interaction in speech perception in Cochlear Implant users compares the same with normal hearing children. Auditory stimuli was routed through calibrated Clinical audiometer in sound field condition, and visual stimuli were presented on laptop screen placed at a distance of 1m at 0 degree azimuth. Out of 4 presentations, if 3 responses were a fusion, then McGurk effect was considered to be present. The congruent audiovisual stimuli /pa/ /pa/ and /ka/ /ka/ were perceived correctly as ‘‘pa’’ and ‘‘ka,’’ respectively by both the groups. For the non- congruent stimuli /da/ /pa/, 23 children out of 35 with normal hearing and 9 children out of 35 with cochlear implant had a fusion of sounds i.e. McGurk effect was present. For the non-congruent stimulus /pa/ /ka/, 25 children out of 35 with normal hearing and 8 children out of 35 with cochlear implant had fusion of sounds.The children who used cochlear implants for less than three years did not exhibit fusion of sound i.e. McGurk effect was absent in this group of children. To conclude, the results demonstrate that consistent fusion of visual with auditory information for speech perception is shaped by experience with bimodal spoken language during early life. When auditory experience with speech is mediated by cochlear implant, the likelihood of acquiring bimodal fusion is increased and it greatly depends on the age of implantation. All the above results strongly support the need for screening children for hearing capabilities and providing cochlear implants and aural rehabilitation as early as possible.

Keywords: cochlear implant, congruent stimuli, mcgurk effect, non-congruent stimuli

Procedia PDF Downloads 281

Authors: Raffaele A. Calogero

Abstract:

RNAseq represents an attractive methodology for the detection of functional genomic variants. RNAseq results obtained from polyA+ RNA selection protocol (POLYA) and from exonic regions capturing protocol (ACCESS) indicate that ACCESS detects 10% more coding SNV/INDELs with respect to POLYA. ACCESS requires less reads for coding SNV detection with respect to POLYA. However, if the analysis aims at identifying SNV/INDELs also in the 5’ and 3’ UTRs, POLYA is definitively the preferred method. No particular advantage comes from ACCESS or POLYA in the detection of fusion transcripts.

Keywords: fusion transcripts, INDEL, RNA-seq, WES, SNV

Procedia PDF Downloads 265

Authors: Xiaoheng Tan, Xianfang Li, Tan Guo, Yuchuan Liu, Zhijun Yang, Hongye Li, Kai Fu, Yufang Wu, Heling Gong

Abstract:

The image data collected in reality often have high dimensions, and it contains noise and redundant information. Therefore, it is necessary to extract the compact feature expression of the original perceived image. In this process, effective use of prior knowledge such as data structure distribution and sample label is the key to enhance image feature discrimination and robustness. Based on the above considerations, this paper proposes a local preserving discriminant feature learning model based on graph optimization. The model has the following characteristics: (1) Locality preserving constraint can effectively excavate and preserve the local structural relationship between data. (2) The flexibility of graph learning can be improved by constructing a new local geometric structure graph using label information and the nearest neighbor threshold. (3) The L₂,₁ norm is used to redefine LDA, and the diagonal matrix is introduced as the scale factor of LDA, and the samples are selected, which improves the robustness of feature learning. The validity and robustness of the proposed algorithm are verified by experiments in two public image datasets.

Keywords: feature extraction, graph optimization local preserving projection, linear discriminant analysis, L₂, ₁ norm

Procedia PDF Downloads 127

Authors: Tudor Barbu

Abstract:

An automatic moment-based texture segmentation approach is proposed in this paper. First, we describe the related work in this computer vision domain. Our texture feature extraction, the first part of the texture recognition process, produces a set of moment-based feature vectors. For each image pixel, a texture feature vector is computed as a sequence of area moments. Second, an automatic pixel classification approach is proposed. The feature vectors are clustered using some unsupervised classification algorithm, the optimal number of clusters being determined using a measure based on validation indexes. From the resulted pixel classes one determines easily the desired texture regions of the image.

Keywords: image segmentation, moment-based, texture analysis, automatic classification, validation indexes

Procedia PDF Downloads 389

Authors: Yakup Ulusu, Numan Eczacıoğlu, İsa Gökçe, Helen Waller, Jeremy H. Lakey

Abstract:

Besides having the appropriate amino acid sequence to perform the function of proteins, it is important to have correct conformation after this sequence to process. To consist of this conformation depends on the amino acid sequence at the primary structure, hydrophobic interaction, chaperones and enzymes in charge of folding etc. Misfolded proteins are not functional and tend to be aggregated. Cysteine originating disulfide cross-links make stable this conformation of functional proteins. When two of the cysteine amino acids come side by side, disulfide bond is established that forms a cystine bridge. Due to this feature cysteine plays an important role on the formation of three-dimensional structure of many proteins. There are two cysteine amino acids (C44, C69) in the Tol-A-III protein. Unlike protein disulfide bonds from within his own, any non-specific cystine bridge causes a change in the three dimensional structure of the protein. Proteins can be expressed in various host cells as directly or fusion (chimeric). As a result of overproduction of the recombinant proteins, aggregation of insoluble proteins in the host cell can occur by forming a crystal structure called inclusion body. In general fusion proteins are produced for provide affinity tags to make proteins more soluble and production of some toxic proteins via fusion protein expression system like pTolT. Proteins can be modified by using a site-directed mutagenesis. By this way, creation of non-specific disulfide crosslinks can be prevented at fusion protein expression system via the present cysteine replaced by another amino acid such as serine, glycine or etc. To do this, we need; a DNA molecule that contains the gene that encodes for the target protein, required primers for mutation to be designed according to site directed mutagenesis reaction. This study was aimed to be replaced cysteine encoding codon TGT with serine encoding codon AGT. For this sense and reverse primers designed (given below) and used site-directed mutagenesis reaction. Several new copy of the template plasmid DNA has been formed with above mentioned mutagenic primers via polymerase chain reaction (PCR). PCR product consists of both the master template DNA (wild type) and the new DNA sequences containing mutations. Dpn-l endonuclease restriction enzyme which is specific for methylated DNA and cuts them to the elimination of the master template DNA. E. coli cells obtained after transformation were incubated LB medium with antibiotic. After purification of plasmid DNA from E. coli, the presence of the mutation was determined by DNA sequence analysis. Developed this new plasmid is called PtolT-δ.

Keywords: site directed mutagenesis, Escherichia coli, pTolT, protein expression

Procedia PDF Downloads 334

Authors: Aghasadeghi MR., Bahramali G., Sadat SM., Sadeghi SA., Yousefi M., Khodaei K., Ghorbani M., Sadat Larijani M.

Abstract:

Background:The emerging COVID-19 pandemic is a serious concernfor the public health worldwide. Despite the many mutations in the virus genome, it is important to find an effective vaccine against viral mutations. Therefore, in current study, we aimed at immunoinformatic evaluation of the virus proteins immunogenicity to design a preventive vaccine candidate, which could elicit humoral and cellular immune responses as well. Methods:Three antigenic regions are included;Spike, Membrane, and Nucleocapsid amino acid sequences were obtained, and possible fusion proteins were assessed andcompared by immunogenicity, structural features, and population coverage. The best fusion protein was also evaluated for MHC-I and MHC-II T-cell epitopes and the linear and conformational B-cell epitopes. Results: Among the four predicted models, the truncated Spike protein in fusion with M and N proteins is composed of 24 highly immunogenic human MHC class I and 29 MHC class II, along with 14 B-cell linear and 61 discontinues epitopes. Also, the selected protein has high antigenicity and acceptable population coverage of 82.95% in Iran and 92.51% in Europe. Conclusion: The data indicate that the truncated Spike-M-N SARS-CoV-2form which could be potential targets of neutralizing antibodies. The protein also has the ability to stimulate humoral and cellular immunity. The in silico study provided the fusion protein as a potential preventive vaccine candidate for further in vivo evaluation.

Keywords: SARS-CoV-2, immunoinformatic, protein, vaccine

Procedia PDF Downloads 188

Authors: Sidra Naeem, Ayesha Naeem, Sahar Rahim, Nadia Nawaz Qadri

Abstract:

Citrus greening is a bacterial disease that causes considerable damage to citrus fruits worldwide. Efficient method for this disease detection must be carried out to minimize the production loss. This paper presents a pattern recognition system that comprises three stages for the detection of citrus greening from Orange leaves: segmentation, feature extraction and classification. Image segmentation is accomplished by adaptive thresholding. The feature extraction stage comprises of three visual descriptors i.e. shape, color and texture. From shape feature we have used asymmetry index, from color feature we have used histogram of Cb component from YCbCr domain and from texture feature we have used local binary pattern. Classification was done using support vector machines and k nearest neighbors. The best performances of the system is Accuracy = 88.02% and AUROC = 90.1% was achieved by automatic segmented images. Our experiments validate that: (1). Segmentation is an imperative preprocessing step for computer assisted diagnosis of citrus greening, and (2). The combination of shape, color and texture features form a complementary set towards the identification of citrus greening disease.

Keywords: citrus greening, pattern recognition, feature extraction, classification

Procedia PDF Downloads 146

Authors: Rohit Gupta, Ravinder Agarwal

Abstract:

Surface electromyographic (sEMG) signal has the potential to identify the human activities and intention. This potential is further exploited to control the artificial limbs using the sEMG signal from residual limbs of amputees. The paper deals with the development of multichannel cost efficient sEMG signal interface for research application, along with evaluation of proposed class dependent statistical approach of the feature selection method. The sEMG signal acquisition interface was developed using ADS1298 of Texas Instruments, which is a front-end interface integrated circuit for ECG application. Further, the sEMG signal is recorded from two lower limb muscles for three locomotions namely: Plane Walk (PW), Stair Ascending (SA), Stair Descending (SD). A class dependent statistical approach is proposed for feature selection and also its performance is compared with 12 preexisting feature vectors. To make the study more extensive, performance of five different types of classifiers are compared. The outcome of the current piece of work proves the suitability of the proposed feature selection algorithm for locomotion recognition, as compared to other existing feature vectors. The SVM Classifier is found as the outperformed classifier among compared classifiers with an average recognition accuracy of 97.40%. Feature vector selection emerges as the most dominant factor affecting the classification performance as it holds 51.51% of the total variance in classification accuracy. The results demonstrate the potentials of the developed sEMG signal acquisition interface along with the proposed feature selection algorithm.

Keywords: classifiers, feature selection, locomotion, sEMG

Procedia PDF Downloads 269

Authors: Adewole A. Ayoade, Marshall R. Sweatt, John P. H. Steele, Qi Han, Khaled Al-Wahedi, Hamad Karki, William A. Yearsley

Abstract:

Understanding the behavioral characteristics of sensors is a crucial step in fusing data from several sensors of different types. This paper introduces a practical, real-time approach to integrate heterogeneous sensor data to achieve higher accuracy than would be possible from any one individual sensor in localizing a mobile robot. We use this approach in both indoor and outdoor environments and it is especially appropriate for those environments like oil and gas refineries due to their sparse and featureless nature. We have studied the individual contribution of each sensor data to the overall combined accuracy achieved from the fusion process. A Sequential Update Extended Kalman Filter(EKF) using validation gates was used to integrate GPS data, Compass data, WiFi data, Inertial Measurement Unit(IMU) data, Vehicle Velocity, and pose estimates from Fiducial marker system. Results show that the approach can enable a mobile robot to navigate autonomously in any environment using a priori information.

Keywords: inspection mobile robot, navigation, sensor fusion, sequential update extended Kalman filter

Procedia PDF Downloads 441

Authors: Daliyah S. Aljutaili, Redna A. Almutlaq, Suha A. Alharbi, Dina M. Ibrahim

Abstract:

All currencies around the world look very different from each other. For instance, the size, color, and pattern of the paper are different. With the development of modern banking services, automatic methods for paper currency recognition become important in many applications like vending machines. One of the currency recognition architecture’s phases is Feature detection and description. There are many algorithms that are used for this phase, but they still have some disadvantages. This paper proposes a feature detection algorithm, which merges the advantages given in the current SIFT and SURF algorithms, which we call, Speeded up Robust Scale-Invariant Feature Transform (SR-SIFT) algorithm. Our proposed SR-SIFT algorithm overcomes the problems of both the SIFT and SURF algorithms. The proposed algorithm aims to speed up the SIFT feature detection algorithm and keep it robust. Simulation results demonstrate that the proposed SR-SIFT algorithm decreases the average response time, especially in small and minimum number of best key points, increases the distribution of the number of best key points on the surface of the currency. Furthermore, the proposed algorithm increases the accuracy of the true best point distribution inside the currency edge than the other two algorithms.

Keywords: currency recognition, feature detection and description, SIFT algorithm, SURF algorithm, speeded up and robust features

Procedia PDF Downloads 215

Authors: Tzyy-Rong Jinn, Sheng-Kuo Hsieh, Yi-Ching Chung, Feng-Chia Hsieh

Abstract:

In this study, we attempted to develop a recombinant oleosin-based fusion expression strategy in Escherichia coli (E. coli) and coupled with the artificial oil bodies (AOB)-cyanogen bromide technology platform to produce bioactive mastoparan B (MP-B). As reported, the oleosin in AOB system plays a carrier (fusion with target protein), since oleosin possess two amphipathic regions (at the N-terminus and C-terminus), which result in the N-terminus and C-terminus of oleosin could be arranged on the surface of AOB. Thus, the target protein fused to the N-terminus or C-terminus of oleosin which also is exposed on the surface of AOB, and this process will greatly facilitate the subsequent separation and purification of target protein from AOB. In addition, oleosin, a unique structural protein of seed oil bodies, has the added advantage of helping the fused MP-B expressed in inclusion bodies, which can protect from proteolytic degradation. In this work, MP-B was fused to the C-terminus of oleosin and then was expressed in E. coli as an insoluble recombinant protein. As a consequence, we successfully developed a reliable recombinant oleosin-based fusion expression strategy in Escherichia coli and coupled with the artificial oil bodies (AOB)-cyanogen bromide technology platform to produce the small peptide, MP-B. Take together, this platform provides an insight into the production of active MP-B, which will facilitate studies and applications of this peptide in the future.

Keywords: artificial oil bodies, Escherichia coli, Oleosin-fusion protein, Mastoparan-B

Procedia PDF Downloads 431

Authors: Heramb Somthankar, Anujoy Chakraborty

Abstract:

This paper presents a classification study on yield prediction in semiconductor production using machine learning approaches. A complicated semiconductor production process is generally monitored continuously by signals acquired from sensors and measurement sites. A monitoring system contains a variety of signals, all of which contain useful information, irrelevant information, and noise. In the case of each signal being considered a feature, "Feature Selection" is used to find the most relevant signals. The open-source UCI SECOM Dataset provides 1567 such samples, out of which 104 fail in quality assurance. Feature extraction and selection are performed on the dataset, and useful signals were considered for further study. Afterward, common machine learning algorithms were employed to predict whether the signal yields pass or fail. The most relevant algorithm is selected for prediction based on the accuracy and loss of the ML model.

Keywords: deep learning, feature extraction, feature selection, machine learning classification algorithms, semiconductor production monitoring, signal processing, time-series analysis

Procedia PDF Downloads 83

Authors: Joe Amalraj, M. Arunkumar

Abstract:

Haze is an atmospheric phenomenon that signicantly degrades the visibility of outdoor scenes. This is mainly due to the atmosphere particles that absorb and scatter the light. This paper introduces a novel single image approach that enhances the visibility of such degraded images. In this method is a fusion-based strategy that derives from two original hazy image inputs by applying a white balance and a contrast enhancing procedure. To blend effectively the information of the derived inputs to preserve the regions with good visibility, we filter their important features by computing three measures (weight maps): luminance, chromaticity, and saliency. To minimize artifacts introduced by the weight maps, our approach is designed in a multiscale fashion, using a Laplacian pyramid representation. This paper demonstrates the utility and effectiveness of a fusion-based technique for de-hazing based on a single degraded image. The method performs in a per-pixel fashion, which is straightforward to implement. The experimental results demonstrate that the method yields results comparative to and even better than the more complex state-of-the-art techniques, having the advantage of being appropriate for real-time applications.

Keywords: single image de-hazing, outdoor images, enhancing, DSP

Procedia PDF Downloads 380

Authors: Yuanchao Sang, Zhihao Gong, Longsheng Chen, Long Chen

Abstract:

Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.

Keywords: convolutional neural network, feature representation, image processing, machine modelling

Procedia PDF Downloads 114

Authors: Yasser R. Shaban

Abstract:

A practical multipurpose device for medical isotopes production is most wanted for clinical centers and researches. Unfortunately, the major supply of these radioisotopes currently comes from aging sources, and there is a great deal of uneasiness in the domestic market. There are also many cases where the cost of certain radioisotopes is too high for their introduction on a commercial scale even though the isotopes might have great benefits for society. The medical isotopes such as radiotracers PET (Positron Emission Tomography), Technetium-99 m, and Iodine-131, Lutetium-177 by is feasible to be generated by a single unit named IEMC (Inertial Electrostatic Magnetic Confinement). The IEMC fusion vessel is the upgrading unit of the Inertial Electrostatic Confinement IEC fusion vessel. Comprehensive experimental works on IEC were carried earlier with promising results. The principle of inertial electrostatic magnetic confinement IEMC fusion is based on forcing the binary fuel ions to interact in the opposite directions in ions cyclotrons orbits with different kinetic energies in order to have equal compression (forces) and with different ion cyclotron frequency ω in order to increase the rate of intersection. The IEMC features greater fusion volume than IEC by several orders of magnitude. The particles rate from the IEMC approach are projected to be 8.5 x 10¹¹ (p/s), ~ 0.2 microampere proton, for D/He-3 fusion reaction and 4.2 x 10¹² (n/s) for D/T fusion reaction. The projected values of particles yield (neutrons and protons) are suitable for medical isotope productions on-site by a single unit without any change in the fusion vessel but only the fuel gas. The PET radiotracers are usually produced on-site by medical ion accelerator whereas Technetium-99m (Tc-99m) is usually produced off-site from the irradiation facilities of nuclear power plants. Typically, hospitals receive molybdenum-99 isotope container; the isotope decays to Tc-99mwith half-life time 2.75 days. Even though the projected current from IEMC is lesser than the proton current from the medical ion accelerator but still the IEMC vessel is simpler, and reduced in components and power consumption which add a new value of populating the PET radiotracers in most clinical centers. On the other hand, the projected neutrons flux from the IEMC is lesser than the thermal neutron flux at the irradiation facilities of nuclear power plants, but in the IEMC case the productions of Technetium-99m is suggested to be at the resonance region of which the resonance integral cross section is two orders of magnitude higher than the thermal flux. Thus it can be said the net activity from both is evened. Besides, the particle accelerator cannot be considered a multipurpose particles production unless a significant change is made to the accelerator to change from neutrons mode to protons mode or vice versa. In conclusion, the projected fusion yield from IEMC is a straightforward since slightly change in the primer IEC and ion source is required.

Keywords: electrostatic versus magnetic confinement fusion vessel, ion source, medical isotopes productions, neutron activation

Procedia PDF Downloads 326

Authors: Pengchao Wei, Xiangzhong Fang

Abstract:

Noise is ubiquitous in this world. Denoising is an essential technology, especially in image semantic segmentation, where noises are generally categorized into two main types i.e. feature noise and label noise. The main focus of this paper is aiming at modeling label noise, investigating the behaviors of different types of label noise on image semantic segmentation tasks using K-Nearest-Neighbor and Convolutional Neural Network classifier. The performance without label noise and with is evaluated and illustrated in this paper. In addition to that, the influence of feature noise on the image semantic segmentation task is researched as well and a feature noise reduction method is applied to mitigate its influence in the learning procedure.

Keywords: convolutional neural network, denoising, feature noise, image semantic segmentation, k-nearest-neighbor, label noise

Procedia PDF Downloads 192

Authors: Zhou Mo, Dennis Chow

Abstract:

In the dynamic positioning system (DPS) for vessels, the reliable information transmission between each note basically relies on the wireless protocols. From the perspective of cluster-based routing protocols for wireless sensor networks, the data fusion technology based on the sleep scheduling mechanism and remaining energy in network layer is proposed, which applies the sleep scheduling mechanism to the routing protocols, considering the remaining energy of node and location information when selecting cluster-head. The problem of uneven distribution of nodes in each cluster is solved by the Equilibrium. At the same time, Classified Forwarding Mechanism as well as Redelivery Policy strategy is adopted to avoid congestion in the transmission of huge amount of data, reduce the delay in data delivery and enhance the real-time response. In this paper, a simulation test is conducted to improve the routing protocols, which turn out to reduce the energy consumption of nodes and increase the efficiency of data delivery.

Keywords: DPS for vessel, wireless sensor network, data fusion, routing protocols

Procedia PDF Downloads 497

Authors: Niyti, Aman Deep, Rajesh Kharab, Sahila Chopra, Raj. K. Gupta

Abstract:

Relatively long-lived transactinide elements (i.e., elements with atomic number Z≥104) up to Z = 108 have been produced in nuclear reactions between low Z projectiles (C to Al) and actinide targets. Cross sections have been observed to decrease steeply with increasing Z. Recently, production cross sections of several picobarns have been reported for comparatively neutron-rich nuclides of 112 through 118 produced via hot fusion reactions with 48Ca and actinide targets. Some of those heavy nuclides are reported to have lifetimes on the order of seconds or longer. The relatively high cross sections in these hot fusion reactions are not fully understood and this has renewed interest in systematic studies of heavy-ion reactions with actinide targets. The main aim of this work is to understand the dynamics hot fusion reactions 18O+ 248Cm and 22Ne+244Pu (carried out at RIKEN and TASCA respectively) using the collective clusterization technique, carried out by undertaking the decay of the compound nucleus 266Rf∗ into 4n, 5n and 6n neutron evaporation channels. Here we extend our earlier study of the excitation functions (EFs) of 266Rf∗, formed in fusion reaction 18O+248Cm, based on Dynamical Cluster-decay Model (DCM) using the pocket formula for nuclear proximity potential, to the use of other nuclear interaction potentials derived from Skyrme energy density formalism (SEDF) based on semiclassical extended Thomas Fermi (ETF) approach and also study entrance channel effects by considering the synthesis of 266Rf* in 22Ne+244Pu reaction. The Skyrme forces used are the old force SIII, and new forces GSkI and KDE0(v1). Here, the EFs for the production of 260Rf, 261Rf and 262Rf isotope via 6n, 5n and 4n decay channel from the 266Rf∗ compound nucleus are studied at Elab = 88.2 to 125 MeV, including quadrupole deformations β2i and ‘hot-optimum’ orientations θi. The calculations are made within the DCM where the neck-length ∆R is the only parameter representing the relative separation distance between two fragments and/or clusters Ai which assimilates the neck formation effects.

Keywords: entrance channel effects, fusion reactions, skyrme force, superheavy nucleus

Procedia PDF Downloads 225

Authors: Ali Alhajeri, Tarig Makki, Mosa Almutahhar, Mohammed Ahmed, Usman Ali

Abstract:

Metal powder is the raw material used for laser powder-bed fusion (LPBF) additive manufacturing (AM). There are many metal materials that can be used in LPBF. The properties of these materials are varied between each other, which can affect the building part. The objective of this paper is to do an overview of the titanium powders available in LBPF. Comparison between different literature works will lead us to study the similarities and differences between the powder properties such as size, shape, and chemical composition. Furthermore, the results of this paper will point out the significant titanium powder properties in order to clearly illustrate their effect on the build parts.

Keywords: LPBF, titanium, Ti-6Al-4V, Ti-5553, metal powder, AM

Procedia PDF Downloads 140

Authors: Zhou Mo, Dennis Chow

Abstract:

In the dynamic positioning system (DPS) for vessels, the reliable information transmission between each note basically relies on the wireless protocols. From the perspective of cluster-based routing pro-tocols for wireless sensor networks, the data fusion technology based on the sleep scheduling mechanism and remaining energy in network layer is proposed, which applies the sleep scheduling mechanism to the routing protocols, considering the remaining energy of node and location information when selecting cluster-head. The problem of uneven distribution of nodes in each cluster is solved by the Equilibrium. At the same time, Classified Forwarding Mechanism as well as Redelivery Policy strategy is adopted to avoid congestion in the transmission of huge amount of data, reduce the delay in data delivery and enhance the real-time response. In this paper, a simulation test is conducted to improve the routing protocols, which turns out to reduce the energy consumption of nodes and increase the efficiency of data delivery.

Keywords: DPS for vessel, wireless sensor network, data fusion, routing protocols

Procedia PDF Downloads 425

Authors: Juzar Vohra, Ravish Malhotra, Tim Pasang, Mana Azizi, Yuan Tao, Masami Mizutani

Abstract:

In this paper, dissimilar welding of titanium to stainless steels is reported. Laser Beam Welding (LBW) and Gas Tungsten Arc Welding (GTAW) were employed to join CPTi to SS304. The welds were examined using scanning electron microscopy (SEM) and energy dispersive X-ray spectroscopy (EDS). FeTi, Ti2Cr and Fe2Ti dendrites are formed along with beta phase titanium matrix. The hardness values of these phases are high which makes them brittle and leading to cracking along the weld pool. However, it is believed that cracking, hence, fracturing of this weld joint is largely due to the difference in thermal properties of the two alloys.

Keywords: dissimilar metals, fusion welding, intermetallics, brittle

Procedia PDF Downloads 472

Authors: Kaleem Kashif, Tayyaba Anam, Yizhi Wu

Abstract:

This paper introduces a methodology for advancing Italian speech vowels landmark detection within the distinctive feature-based speech recognition domain. Leveraging the legacy tool 'xkl' by integrating combined convolutional neural networks (CNNs) and recurrent neural networks (RNNs), the study presents a comprehensive enhancement to the 'xkl' legacy software. This integration incorporates re-assigned spectrogram methodologies, enabling meticulous acoustic analysis. Simultaneously, our proposed model, integrating combined CNNs and RNNs, demonstrates unprecedented precision and robustness in landmark detection. The augmentation of re-assigned spectrogram fusion within the 'xkl' software signifies a meticulous advancement, particularly enhancing precision related to vowel formant estimation. This augmentation catalyzes unparalleled accuracy in landmark detection, resulting in a substantial performance leap compared to conventional methods. The proposed model emerges as a state-of-the-art solution in the distinctive feature-based speech recognition systems domain. In the realm of deep learning, a synergistic integration of combined CNNs and RNNs is introduced, endowed with specialized temporal embeddings, harnessing self-attention mechanisms, and positional embeddings. The proposed model allows it to excel in capturing intricate dependencies within Italian speech vowels, rendering it highly adaptable and sophisticated in the distinctive feature domain. Furthermore, our advanced temporal modeling approach employs Bayesian temporal encoding, refining the measurement of inter-landmark intervals. Comparative analysis against state-of-the-art models reveals a substantial improvement in accuracy, highlighting the robustness and efficacy of the proposed methodology. Upon rigorous testing on a database (LaMIT) speech recorded in a silent room by four Italian native speakers, the landmark detector demonstrates exceptional performance, achieving a 95% true detection rate and a 10% false detection rate. A majority of missed landmarks were observed in proximity to reduced vowels. These promising results underscore the robust identifiability of landmarks within the speech waveform, establishing the feasibility of employing a landmark detector as a front end in a speech recognition system. The synergistic integration of re-assigned spectrogram fusion, CNNs, RNNs, and Bayesian temporal encoding not only signifies a significant advancement in Italian speech vowels landmark detection but also positions the proposed model as a leader in the field. The model offers distinct advantages, including unparalleled accuracy, adaptability, and sophistication, marking a milestone in the intersection of deep learning and distinctive feature-based speech recognition. This work contributes to the broader scientific community by presenting a methodologically rigorous framework for enhancing landmark detection accuracy in Italian speech vowels. The integration of cutting-edge techniques establishes a foundation for future advancements in speech signal processing, emphasizing the potential of the proposed model in practical applications across various domains requiring robust speech recognition systems.

Keywords: landmark detection, acoustic analysis, convolutional neural network, recurrent neural network

Procedia PDF Downloads 21