Search results for: deep mixing method

Authors: Noureddine Mohtaram, Jeremy Patrix, Jerome Verny

Abstract:

As an enormous number of online services have been developed into web applications, security problems based on web applications are becoming more serious now. Most intrusion detection systems rely on each request to find the cyber-attack rather than on user behavior, and these systems can only protect web applications against known vulnerabilities rather than certain zero-day attacks. In order to detect new attacks, we analyze the HTTP protocols of web servers to divide them into two categories: normal attacks and malicious attacks. On the other hand, the quality of the results obtained by deep learning (DL) in various areas of big data has given an important motivation to apply it to cybersecurity. Deep learning for attack detection in cybersecurity has the potential to be a robust tool from small transformations to new attacks due to its capability to extract more high-level features. This research aims to take a new approach, deep learning to cybersecurity, to classify these two categories to eliminate attacks and protect web servers of the defense sector which encounters different web traffic compared to other sectors (such as e-commerce, web app, etc.). The result shows that by using a machine learning method, a higher accuracy rate, and a lower false alarm detection rate can be achieved.

Keywords: anomaly detection, HTTP protocol, logs, cyber attack, deep learning

Procedia PDF Downloads 177

Authors: Degale Desta, Cheng Jian

Abstract:

The most important factor in identifying a person is their face. Even identical twins have their own distinct faces. As a result, identification and face recognition are needed to tell one person from another. A face recognition system is a verification tool used to establish a person's identity using biometrics. Nowadays, face recognition is a common technique used in a variety of applications, including home security systems, criminal identification, and phone unlock systems. This system is more secure because it only requires a facial image instead of other dependencies like a key or card. Face detection and face identification are the two phases that typically make up a human recognition system.The idea behind designing and creating a face recognition system using deep learning with Azure ML Python's OpenCV is explained in this paper. Face recognition is a task that can be accomplished using deep learning, and given the accuracy of this method, it appears to be a suitable approach. To show how accurate the suggested face recognition system is, experimental results are given in 98.46% accuracy using Fast-RCNN Performance of algorithms under different training conditions.

Keywords: deep learning, face recognition, identification, fast-RCNN

Procedia PDF Downloads 91

Authors: Baha Vural Kök, Mehmet Yilmaz, Mustafa Akpolat, Cihat Sav

Abstract:

Various types of additives are used frequently in order to improve the rheological and mechanical properties of bituminous mixtures. Small devices instead of full scale machines are used for bitumen modification in the laboratory. These laboratory scale devices vary in terms of their properties such as mixing rate, mixing blade and the amount of binder. In this study, the effect of mixing rate and time during the bitumen modification processes on conventional and rheological properties of pure and crumb rubber modified binder were investigated. Penetration, softening point, rotational viscosity (RV) and dynamic shear rheometer (DSR) tests were applied to pure and CR modified bitumen. It was concluded that the penetration and softening point test did not show the efficiency of CR obtained by different mixing conditions. Besides, oxidation that occurred during the preparation processes plays a great part in the improvement effects of the modified binder.

Keywords: bitumen, crumb rubber, modification, rheological properties

Procedia PDF Downloads 278

Authors: Mohsen Solimani Babarsad, Mohammad Rastgoo, Payam Taheri

Abstract:

One of the important applications of hydraulic jets is used for discharge industrial, agricultural and urban wastewater into the rivers or other ambient water to reduce negative effects of pollutant water. Submerged jets due to turbulent condition can mix large amount of dense pollutant water with ambient flow. This study is conducted to investigate the distribution and length of the mixing zone in hydraulic jet's flow field with change in cross section shapes of nozzle. Toward this end, three shapes of cross section (square, circle and rectangular) and three saline densities current with different concentration are considered in a flume with 600 cm as long, 100 cm as high and 150 cm in width. Various discharges were used to evaluate mixing length for a wide range of densimetric Froude numbers, Frd, from 100 to 550 that is defined at the nozzle. Consequently, the circular nozzle, in comparison with other sections, has a densimetric Froude number 11% higher than square nozzle and 26% higher than rectangular nozzle.

Keywords: hydraulic jet, mixing zone, densimetric Froude number, nozzle

Procedia PDF Downloads 330

Authors: J. Alvarez Muñoz, Dominguez Lepe J. A.

Abstract:

A suitable mixing time of the concrete, allows form a homogeneous mass, quality that leads to greater compressive strength and durability. Although there are recommendations as ASTM C94 standard these mention the time and the number of minimum and maximum speed for a ready-mix concrete of good quality, the specific behavior that would have a concrete mixed on site to variability of the mixing time is unknown. In this study was evaluated the behavior a design of mixture structural of f´c=250 kg/cm2, elaborate on site with limestone aggregate in warm sub-humid climate, subjected to different mixing times. Based on the recommendation for ready-mixed concrete ASTM C94, different times were set at 70, 90, 100, 110, 120, 140 total revolutions. A field study in which 14 works were observed where structural concrete made on site was used, allowed to set at 24 the number of revolutions to the reference mixture. For the production of concrete was used a hand feed concrete mixer with drum speed 28 RPM, the ratio w/c was 0.36 corrected, with a slump of 5-6 cm, for all mixtures. The compressive strength tests were performed at 3, 7, 14, and 28 days. The most outstanding results show increases in resistance in the mixtures of 24 to 70 revolutions between 8 and 17 percent and 70 to 90 revolutions of 3 to 8 percent. Increasing the number of revolutions at 110, 120 and 140, there was a reduction of the compressive strength of 0.5 to 8 percent. Regarding mixtures consistencies, they had a slump of 5 cm to 24, 70 and 90 rpm and less than 5 cm from 100 revolutions. Clearly, those made with more than 100 revolutions mixtures not only decrease the compressive strength but also the workability.

Keywords: compressive strength, concrete, mixing time, workability

Procedia PDF Downloads 377

Authors: Shuen-Tai Wang, Fang-An Kuo, Chau-Yi Chou, Yu-Bin Fang

Abstract:

2016 has become the year of the Artificial Intelligence explosion. AI technologies are getting more and more matured that most world well-known tech giants are making large investment to increase the capabilities in AI. Machine learning is the science of getting computers to act without being explicitly programmed, and deep learning is a subset of machine learning that uses deep neural network to train a machine to learn  features directly from data. Deep learning realizes many machine learning applications which expand the field of AI. At the present time, deep learning frameworks have been widely deployed on servers for deep learning applications in both academia and industry. In training deep neural networks, there are many standard processes or algorithms, but the performance of different frameworks might be different. In this paper we evaluate the running performance of two state-of-the-art distributed deep learning frameworks that are running training calculation in parallel over multi GPU and multi nodes in our cloud environment. We evaluate the training performance of the frameworks with ResNet-50 convolutional neural network, and we analyze what factors that result in the performance among both distributed frameworks as well. Through the experimental analysis, we identify the overheads which could be further optimized. The main contribution is that the evaluation results provide further optimization directions in both performance tuning and algorithmic design.

Keywords: artificial intelligence, machine learning, deep learning, convolutional neural networks

Procedia PDF Downloads 174

Authors: Talat Ghomashchi, Zahra Akbari, Shirin Malekpour, Marjan Alimirzaee

Abstract:

Mixing the waste water of industries with natural water has caused environmental pollution. So researcher try to obtain methods and optimum conditions for waste water treatment. One of important stage in waste water treatment is dissolved air flotation. DAF is used for the removal of suspended solids and oils from waste water. In this paper, the effect of several parameters on flotation efficiency with Cationic polyacrylamide as flocculant, was examined, namely, (a) concentration of cationic flocculants, (b) pH (c) fast mixing time, (d) fast mixing speed,(e) slow mixing time,(f) retention time and temperature. After design of experiment, in each trial turbidity of waste water was measured by spectrophotometer. Results show that contribution of pH and concentration of flocculant on flotation efficiency are 75% and 9% respectively. Cationic polyacrylamide led to a significant increase in the settling speed and effect of temperature is negligible. In the optimum condition, the outcome of the DAF unit is increased and amount of suspended solid and oil in waste water is decreased effectively.

Keywords: dissolved air flotation, oil industry, waste water, treatment

Procedia PDF Downloads 494

Authors: Hwang Ho Kim, Do Gyun Kim, Jin Young Choi, Sang Chul Park

Abstract:

Since tire production process is very complicated, company-wide management of it is very difficult, necessitating considerable amounts of capital and labors. Thus, productivity should be enhanced and maintained competitive by developing and applying effective production plans. Among major processes for tire manufacturing, consisting of mixing component preparation, building and curing, the mixing process is an essential and important step because the main component of tire, called compound, is formed at this step. Compound as a rubber synthesis with various characteristics plays its own role required for a tire as a finished product. Meanwhile, scheduling tire mixing process is similar to flexible job shop scheduling problem (FJSSP) because various kinds of compounds have their unique orders of operations, and a set of alternative machines can be used to process each operation. In addition, setup time required for different operations may differ due to alteration of additives. In other words, each operation of mixing processes requires different setup time depending on the previous one, and this kind of feature, called sequence dependent setup time (SDST), is a very important issue in traditional scheduling problems such as flexible job shop scheduling problems. However, despite of its importance, there exist few research works dealing with the tire mixing process. Thus, in this paper, we consider the scheduling problem for tire mixing process and suggest an efficient particle swarm optimization (PSO) algorithm to minimize the makespan for completing all the required jobs belonging to the process. Specifically, we design a particle encoding scheme for the considered scheduling problem, including a processing sequence for compounds and machine allocation information for each job operation, and a method for generating a tire mixing schedule from a given particle. At each iteration, the coordination and velocity of particles are updated, and the current solution is compared with new solution. This procedure is repeated until a stopping condition is satisfied. The performance of the proposed algorithm is validated through a numerical experiment by using some small-sized problem instances expressing the tire mixing process. Furthermore, we compare the solution of the proposed algorithm with it obtained by solving a mixed integer linear programming (MILP) model developed in previous research work. As for performance measure, we define an error rate which can evaluate the difference between two solutions. As a result, we show that PSO algorithm proposed in this paper outperforms MILP model with respect to the effectiveness and efficiency. As the direction for future work, we plan to consider scheduling problems in other processes such as building, curing. We can also extend our current work by considering other performance measures such as weighted makespan or processing times affected by aging or learning effects.

Keywords: compound, error rate, flexible job shop scheduling problem, makespan, particle encoding scheme, particle swarm optimization, sequence dependent setup time, tire mixing process

Procedia PDF Downloads 235

Authors: Rahma Abed, Sahbi Bahroun, Ezzeddine Zagrouba

Abstract:

Due to the huge amount of data in videos, extracting the relevant frames became a necessity and an essential step prior to performing face recognition. In this context, we propose a method for extracting keyframes from videos based on face quality and deep learning for a face recognition task. This method has two steps. We start by generating face quality scores for each face image based on the use of three face feature extractors, including Gabor, LBP, and HOG. The second step consists in training a Deep Convolutional Neural Network in a supervised manner in order to select the frames that have the best face quality. The obtained results show the effectiveness of the proposed method compared to the methods of the state of the art.

Keywords: keyframe extraction, face quality assessment, face in video recognition, convolution neural network

Procedia PDF Downloads 193

Authors: Constanza A. Barriga, Rafael Bernardi, Amokrane Berdja, Christian D. Guzman

Abstract:

Most image restoration methods in astronomy rely upon probabilistic tools that infer the best solution for a deconvolution problem. They achieve good performances when the point spread function (PSF) is spatially invariable in the image plane. However, this latter condition is not always satisfied with real optical systems. PSF angular variations cannot be evaluated directly from the observations, neither be corrected at a pixel resolution. We have developed a method for the restoration of images affected by static and anisotropic aberrations using deep neural networks that can be directly applied to sky images. The network is trained using simulated sky images corresponding to the T-80 telescope optical system, an 80 cm survey imager at Cerro Tololo (Chile), which are synthesized using a Zernike polynomial representation of the optical system. Once trained, the network can be used directly on sky images, outputting a corrected version of the image, which has a constant and known PSF across its field-of-view. The method was tested with the T-80 telescope, achieving better results than with PSF deconvolution techniques. We present the method and results on this telescope.

Keywords: aberrations, deep neural networks, image restoration, variable point spread function, wide field images

Procedia PDF Downloads 105

Authors: Revoti Prasad Bora, Nikita Katyal, Saurabh Yadav

Abstract:

Pre-processing plays an important role in various image processing applications. Most of the time due to the similar nature of images, a particular pre-processing or a set of pre-processing steps are sufficient to produce the desired results. However, in the education domain, there is a wide variety of images in various aspects like images with line-based diagrams, chemical formulas, mathematical equations, etc. Hence a single pre-processing or a set of pre-processing steps may not yield good results. Therefore, a Deep Learning based approach for dynamically selecting a relevant pre-processing technique for each image is proposed. The proposed method works as a classifier to detect hidden patterns in the images and predicts the relevant pre-processing technique needed for the image. This approach experimented for an image similarity matching problem but it can be adapted to other use cases too. Experimental results showed significant improvement in average similarity ranking with the proposed method as opposed to static pre-processing techniques.

Keywords: deep-learning, classification, pre-processing, computer vision, image processing, educational data mining

Procedia PDF Downloads 116

Authors: Bandhan Dey, Muhsina Bintoon Yiasha, Gulam Sulaman Choudhury

Abstract:

Chest disease is one of the most problematic ailments in our regular life. There are many known chest diseases out there. Diagnosing them correctly plays a vital role in the process of treatment. There are many methods available explicitly developed for different chest diseases. But the most common approach for diagnosing these diseases is through X-ray. In this paper, we proposed a multi-classification deep learning model for diagnosing COVID-19, lung cancer, pneumonia, tuberculosis, and atelectasis from chest X-rays. In the present work, we used the transfer learning method for better accuracy and fast training phase. The performance of three architectures is considered: InceptionV3, VGG-16, and VGG-19. We evaluated these deep learning architectures using public digital chest x-ray datasets with six classes (i.e., COVID-19, lung cancer, pneumonia, tuberculosis, atelectasis, and normal). The experiments are conducted on six-classification, and we found that VGG16 outperforms other proposed models with an accuracy of 95%.

Keywords: deep learning, image classification, X-ray images, Tensorflow, Keras, chest diseases, convolutional neural networks, multi-classification

Procedia PDF Downloads 59

Authors: Gayatri Ghosh

Abstract:

In this work, a flavour theory of a neutrino mass model based on A_4 symmetry is considered to explain the phenomenology of neutrino mixing. The spontaneous symmetry breaking of A_4 symmetry in this model leads to tribimaximal mixing in the neutrino sector at a leading order. We consider the effect of Z_2 × Z_2 invariant perturbations in neutrino sector and find the allowed region of correction terms in the perturbation matrix that is consistent with 3σ ranges of the experimental values of the mixing angles. We study the entanglement of this formalism on the other phenomenological observables, such as δ_CP phase, the neutrino oscillation probability P(νµ → νe), the effective Majorana mass |mee| and |meff νe |. A Z_2 × Z_2 invariant perturbations in this model is introduced in the neutrino sector which leads to testable predictions of θ_13 and CP violation. By changing the magnitudes of perturbations in neutrino sector, one can generate viable values of δ_CP and neutrino oscillation parameters. Next we investigate the feasibility of charged lepton flavour violation in type-I seesaw models with leptonic flavour symmetries at high energy that leads to tribimaximal neutrino mixing. We consider an effective theory with an A_4 × Z_2 × Z_2 symmetry, which after spontaneous symmetry breaking at high scale which is much higher than the electroweak scale leads to charged lepton flavour violation processes once the heavy Majorana neutrino mass degeneracy is lifted either by renormalization group effects or by a soft breaking of the A_4 symmetry. In this context the implications for charged lepton flavour violation processes like µ → eγ, τ → eγ, τ → µγ are discussed.

Keywords: Z2 × Z2 invariant perturbations, CLFV, delta CP phase, tribimaximal neutrino mixing

Procedia PDF Downloads 54

Authors: Mrinalini Amritkar, Disha Patil, Swapna Kulkarni, Sukratu Barve, Suresh Gosavi

Abstract:

Micro-mixers play an important role in the lab-on-a-chip applications and micro total analysis systems to acquire the correct level of mixing for any given process. The mixing process can be classified as active or passive according to the use of external energy. Literature of microfluidics reports that most of the work is done on the models of steady laminar flow; however, the study of unsteady laminar flow is an active area of research at present. There are wide applications of this, out of which, we consider nanoparticle synthesis in micro-mixers. In this work, we have developed a model for unsteady flow to study the mixing performance of a passive micro mixer for reactants used for such synthesis. The model is developed in Finite Volume Method (FVM)-based software, OpenFOAM. The model is tested by carrying out the simulations at Re of 0.5. Mixing performance of the micro-mixer is investigated using simulated concentration values of mixed species across the width of the micro-mixer and calculating the variance across a line profile. Experimental validation is done by passing dyes through a Y shape micro-mixer fabricated using polydimethylsiloxane (PDMS) polymer and comparing variances with the simulated ones. Gold nanoparticles are later synthesized through the micro-mixer and collected at two different times leading to significantly different size distributions. These times match with the time scales over which reactant concentrations vary as obtained from simulations. Our simulations could thus be used to create design aids for passive micro-mixers used in nanoparticle synthesis.

Keywords: Lab-on-chip, LOC, micro-mixer, OpenFOAM, PDMS

Procedia PDF Downloads 135

Authors: Shun-Chang Yen, You-Lun Peng, Kuo-Ching San

Abstract:

A numerical study on a bluff-body structure with multiple holes was conducted using ANSYS Fluent computational fluid dynamics analysis. The effects of the hole number and jet inclination angles were considered under a fixed gas flow rate and nonreactive gas. The bluff body with multiple holes can transform the axial momentum into a radial and tangential momentum as well as increase the swirl number (S). The concentration distribution in the mixing of a central carbon dioxide (CO2) jet and an annular air jet was utilized to analyze the mixing efficiency. Three bluff bodies with differing hole numbers (H = 3, 6, and 12) and three jet inclination angles (θ = 45°, 60°, and 90°) were designed for analysis. The Reynolds normal stress increases with the inclination angle. The Reynolds shear stress, average turbulence intensity, and average swirl number decrease with the inclination angle. For an unsymmetrical hole configuration (i.e., H = 3), the streamline patterns exhibited an unsymmetrical flow field. The highest mixing efficiency (i.e., the lowest integral gas fraction of CO2) occurred at H = 3. Furthermore, the highest swirl number coincided with the strongest effect on the mass fraction of CO2. Therefore, an unsymmetrical hole arrangement induced a high swirl flow behind the porous disc.

Keywords: bluff body with multiple holes, computational fluid dynamics, swirl-jet flow, mixing efficiency

Procedia PDF Downloads 324

Authors: Jiaqi Huang, Yuheng Wang

Abstract:

Premises selection, the task of selecting a set of axioms for proving a given conjecture, is a major bottleneck in automated theorem proving. An array of deep-learning-based methods has been established for premises selection, but a perfect performance remains challenging. Our study examines the inaccuracy of deep neural networks in premises selection. Through training network models using encoded conjecture and axiom pairs from the Mizar Mathematical Library, two potential biases are found: the network models classify more premises as necessary than unnecessary, referred to as the ‘positive bias’, and the network models perform better in proving conjectures that paired with more axioms, referred to as ‘length bias’. The ‘positive bias’ and ‘length bias’ discovered could inform the limitation of existing deep neural networks.

Keywords: automated theorem proving, premises selection, deep learning, interpreting deep learning

Procedia PDF Downloads 139

Authors: Ghayah Alsaleem

Abstract:

This research project focussed on specific compounds, whereas a literature review was completed on the broader subject of long-lasting phosphorescence. For the review and subsequent laboratory work, long lasting phosphorescence compounds were defined as materials that have an afterglow decay time greater than a few minutes. The decay time is defined as the time between the end of excitation and the moment the light intensity drops below 0.32mcd/m2. This definition is widely used in industry and in most research studies. The experimental work focused on known long-lasting phosphorescence compounds – strontium aluminate (SrAl2O4: Eu2+, Dy3+). At first, preparation was similar to literary methods. Temperature, dopant levels and mixing methods were then varied in order to expose their effects on long-lasting phosphorescence. The effect of temperature was investigated for SrAl2O4: Eu2+, Dy3+, and resulted in the discovery that 1350°C was the only temperature that the compound could be heated to in the Differential scanning calorimetry (DSC) in order to achieve any phosphorescence. However, no temperatures above 1350°C were investigated. The variation of mixing method and co-dopant level in the strontium aluminate compounds resulted in the finding that the dry mixing method using a Turbula mixer resulted in the longest afterglow. It was also found that an increase of europium inclusion, from 1mol% to 2mol% in these compounds, increased the brightest of the phosphorescence. As this increased batch was mixed using sonication, the phosphorescent time was actually reduced which produced green long-lasting phosphorescence for up to 20 minutes following 30 minutes excitation and 50 minutes when the europium content was doubled and mixed using sonication.

Keywords: long lasting, phosphorescence, excitation, europium

Procedia PDF Downloads 143

Authors: Mannal Tariq

Abstract:

Construction industry is making progress at a high pace. The trend of the world is getting more biased towards the high rise buildings. Deep beams are one of the most common elements in modern construction having small span to depth ratio. Deep beams are mostly used as transfer girders. This experimental study consists of 16 reinforced concrete (RC) deep beams. These beams were divided into two groups; A and B. Groups A and B consist of eight beams each, having 381 mm (15 in) and 457 mm (18 in) depth respectively. Each group was further subdivided into four sub groups each consisting of two identical beams. Each subgroup was comprised of solid/control beam (without opening), opening above neutral axis (NA), at NA and below NA. Except for control beams, all beams with openings were strengthened with carbon fibre reinforced polymer (CFRP) vertical strips. These eight groups differ from each other based on depth and location of openings. For testing sake, all beams have been loaded with two symmetrical point loads. All beams have been designed based on strut and tie model concept. The outcome of experimental investigation elaborates the difference in the shear behaviour of deep beams based on depth and location of circular openings variation. 457 mm (18 in) deep beam with openings above NA show the highest strength and 381 mm (15 in) deep beam with openings below NA show the least strength. CFRP sheets played a vital role in increasing the shear capacity of beams.

Keywords: CFRP, deep beams, openings in deep beams, strut and tie modal, shear behaviour

Procedia PDF Downloads 272

Authors: Faouzi Ahtchi-Ali, Marcus Vitiello

Abstract:

A study was conducted in the east region of the Middle East to assess the constructability of a shoring system for a 12-meter deep excavation. Several shoring systems were considered in this study including secant concrete piling, contiguous concrete piling, and sheet-piling. The excavation was carried out in a very dense sand with the groundwater level located at 3 meters below ground surface. The study included conducting a pilot test for each shoring system listed above. The secant concrete piling included overlapping concrete piles to a depth of 16 meters. Drilling method with full steel casing was utilized to install the concrete piles. The verticality of the piles was a concern for the overlap. The contiguous concrete piling required the installation of micro-piles to seal the gap between the concrete piles. This method revealed that the gap between the piles was not fully sealed as observed by the groundwater penetration to the excavation. The sheet-piling method required pre-drilling due to the high blow count of the penetrated layer of saturated sand. This study concluded that the sheet-piling method with pre-drilling was the most cost effective and recommended a method for the shoring system.

Keywords: excavation, shoring system, middle east, Drilling method

Procedia PDF Downloads 427

Authors: Phakamani Xaba, Robert Huberts, Bilainu Oboirien

Abstract:

It has been over 200 years since margarine was discovered and manufactured using liquid oil, liquified hardstock oils and other oil phase & aqueous phase ingredients. Henry W. Bradley first used vegetable oils in liquid state and around 1871, since then; spreads have been traditionally manufactured using liquified oils. The main objective of this study was to develop a process method to produce spreads using spray dried hardstock fat powders as a structing fats in place of current liquid structuring fats. A high shear mixing system was used to condition the fat phase and the aqueous phase was prepared separately. Using a single scraped surface heat exchanger and pin stirrer, margarine was produced. The process method was developed for to produce spreads with 40%, 50% and 60% fat . The developed method was divided into three steps. In the first step, fat powders were conditioned by melting and dissolving them into liquid oils. The liquified portion of the oils were at 65 °C, whilst the spray dried fat powder was at 25 °C. The two were mixed using a mixing vessel at 900 rpm for 4 minutes. The rest of the ingredients i.e., lecithin, colorant, vitamins & flavours were added at ambient conditions to complete the fat/ oil phase. The water phase was prepared separately by mixing salt, water, preservative, acidifier in the mixing tank. Milk was also separately prepared by pasteurizing it at 79°C prior to feeding it into the aqueous phase. All the water phase contents were chilled to 8 °C. The oil phase and water phase were mixed in a tank, then fed into a single scraped surface heat exchanger. After the scraped surface heat exchanger, the emulsion was fed in a pin stirrer to work the formed crystals and produce margarine. The margarine produced using the developed process had fat levels of 40%, 50% and 60%. The margarine passed all the qualitative, stability, and taste assessments. The scores were 6/10, 7/10 & 7.5/10 for the 40%, 50% & 60% fat spreads, respectively. The success of the trials brought about differentiated knowledge on how to manufacture spreads using non micronized spray dried fat powders as hardstock. Manufacturers do not need to store structuring fats at 80-90°C and even high in winter, instead, they can adapt their processes to use fat powders which need to be stored at 25 °C. The developed process method used one scrape surface heat exchanger instead of the four to five currently used in votator based plants. The use of a single scraped surface heat exchanger translated to about 61% energy savings i.e., 23 kW per ton of product. Furthermore, it was found that the energy saved by implementing separate pasteurization was calculated to be 6.5 kW per ton of product produced.

Keywords: margarine emulsion, votator technology, margarine processing, scraped sur, fat powders

Procedia PDF Downloads 64

Authors: Amal Khalifa, Nicolas Vana Santos

Abstract:

Steganography has been known for centuries as an efficient approach for covert communication. Due to its popularity and ease of access, image steganography has attracted researchers to find secure techniques for hiding information within an innocent looking cover image. In this research, we propose a novel deep-learning approach to digital image steganography. The proposed method, DeepWaveletFusion, uses convolutional neural networks (CNN) to hide a secret image into a cover image of the same size. Two CNNs are trained back-to-back to merge the Discrete Wavelet Transform (DWT) of both colored images and eventually be able to blindly extract the hidden image. Based on two different image similarity metrics, a weighted gain function is used to guide the learning process and maximize the quality of the retrieved secret image and yet maintaining acceptable imperceptibility. Experimental results verified the high recoverability of DeepWaveletFusion which outperformed similar deep-learning-based methods.

Keywords: deep learning, steganography, image, discrete wavelet transform, fusion

Procedia PDF Downloads 42

Authors: Ruqaya H. Aljabery

Abstract:

Reinforced Concrete (RC) deep beams are one of the main critical structural elements in terms of safety since significant loads are carried in a short span. The shear capacity of these sections cannot be predicted accurately by the current design codes like ACI and EC2; thus, they must be investigated. In this research, non-linear behavior of RC deep beams strengthened in shear with Fiber Reinforced Polymer (FRP) strips, and the efficiency of FRP in terms of enhancing the shear capacity in RC deep beams are examined using Finite Element Analysis (FEA), which is conducted using the software ABAQUS. The effect of several parameters on the shear capacity of the RC deep beam are studied in this paper as well including the effect of the cross-sectional area of the FRP strip and the shear reinforcement area to the spacing ratio (As/S), and it was found that FRP enhances the shear capacity significantly and can be a substitution of steel stirrups resulting in a more economical design.

Keywords: Abaqus, concrete, deep beam, finite element analysis, FRP, shear strengthening, strut-and-tie

Procedia PDF Downloads 120

Authors: Baishali Deb, Hari Om Pandey, Shrilla Elangbam, Mukesh Singh, Ayon Tarafdar, A. K. S. Tomar, A. K. Pandey, Triveni Dutt

Abstract:

The goal of the present investigation was to determine the impact of social stress on behavioural characteristics, physiological responses, and haemato-biochemical indicators under various social environments in Tharparkar cattle. Serum cortisol and faecal cortisol metabolites analysis were used to determine the stress level of Tharparkar cattle. Social isolation and social mixing were the two different social circumstances used to evaluate the animals. In both the experiments i.e., social isolation and social mixing, the lying period of animals decreased significantly (p<0.05) while standing period significantly (p<0.05) increased. Frequency and duration of activities like idling, walking, exploration, oral manipulation, and elimination increased significantly (p<0.05) in Tharparkar cattle after being subjected to social isolation and social mixing. Time spent in grooming (self-grooming and allo-grooming) in respect to social isolation significantly increased during isolation and post-reunion, whereas there was a significant (p<0.05) decline in the grooming behaviour especially allo-grooming during mixing of the animals. Feeding and rumination time also decreased significantly (p<0.05) in animals during both the experiments. Physiological parameters such as respiration rate, heart rate and pulse rate increased during the treatment periods. There was no significant difference in the haematological parameters for both the experiments. There was significant (p<0.05) increase in serum cortisol and faecal cortisol metabolites (FCM) concentration in animals subjected to social stress. Therefore, it can be concluded that social stress strongly impacts the behaviour and physiological parameters of the animals, causing stress and nervousness, proving that social stress is a valid psychological stress in animals. The higher concentration of FCM in Tharparkar cattle subjected to social stress, further supported by higher serum cortisol and behaviour manifestations, suggest that FCM could be used to assess stress response as a non-invasive method.

Keywords: social stress, fecal cortisol metabolites, non-invasive, animal welfare, behaviour

Procedia PDF Downloads 60

Authors: Soofiyan Atar, Adil Shaikh, Sahil Rajpurkar, Pragnesh Bhalala, Aniket Desai, Irfan Siddavatam

Abstract:

Deep reinforcement learning (deep RL) algorithms leverage the symbolic power of complex controllers by automating it by mapping sensory inputs to low-level actions. Deep RL eliminates the complex robot dynamics with minimal engineering. Deep RL provides high-risk involvement by directly implementing it in real-world scenarios and also high sensitivity towards hyperparameters. Tuning of hyperparameters on a pneumatic quadruped robot becomes very expensive through trial-and-error learning. This paper presents an automated learning control for a pneumatic quadruped robot using sample efficient deep Q learning, enabling minimal tuning and very few trials to learn the neural network. Long training hours may degrade the pneumatic cylinder due to jerk actions originated through stochastic weights. We applied this method to the pneumatic quadruped robot, which resulted in a hopping gait. In our process, we eliminated the use of a simulator and acquired a stable gait. This approach evolves so that the resultant gait matures more sturdy towards any stochastic changes in the environment. We further show that our algorithm performed very well as compared to programmed gait using robot dynamics.

Keywords: model-based reinforcement learning, gait stability, supervised learning, pneumatic quadruped

Procedia PDF Downloads 282

Authors: Gaokai Liu

Abstract:

Automatic defect detection for nanomaterial images is widely required in industrial scenarios. Deep learning approaches are considered as the most effective solutions for the great majority of image-based tasks. In this paper, an edge guidance network for defect segmentation is proposed. First, the encoder path with multiple convolution and downsampling operations is applied to the acquisition of shared features. Then two decoder paths both are connected to the last convolution layer of the encoder and supervised by the edge and segmentation labels, respectively, to guide the whole training process. Meanwhile, the edge and encoder outputs from the same stage are concatenated to the segmentation corresponding part to further tune the segmentation result. Finally, the effectiveness of the proposed method is verified via the experiments on open nanofibrous datasets.

Keywords: deep learning, defect detection, image segmentation, nanomaterials

Procedia PDF Downloads 115

Authors: Nitin Shravan, Sudarsun Santhiappan, B. Sivaselvan

Abstract:

Subsection identification, in the context of Electronic Health Records (EHRs), is identifying the important sections for down-stream tasks like auto-coding. In this work, we classify the text present in EHRs according to their information, using machine learning and deep learning techniques. We initially describe briefly about the problem and formulate it as a text classification problem. Then, we discuss upon the methods from the literature. We try two approaches - traditional feature extraction based machine learning methods and deep learning methods. Through experiments on a private dataset, we establish that the deep learning methods perform better than the feature extraction based Machine Learning Models.

Keywords: deep learning, machine learning, semantic clinical classification, subsection identification, text classification

Procedia PDF Downloads 182

Authors: Kazuaki Shiraishi, Narumitsu Asai, Tsukasa Kitahara, Sosuke Mieno, Takaharu Kameoka

Abstract:

The purpose of this research is to facilitate productivity advances in agriculture. To accomplish this, we developed an automatic three-dimensional (3D) recording system for growth of field crops that consists of a number of inexpensive modules: a very low-cost stereo camera, a couple of ZigBee wireless modules, a Raspberry Pi single-board computer, and a third generation (3G) wireless communication module. Our system uses an inexpensive Web stereo camera in order to keep total costs low. However, inexpensive video cameras record low-resolution images that are very noisy. Accordingly, in order to resolve these problems, we adopted a deep learning method. Based on the results of extended period of time operation test conducted without the use of an external power supply, we found that by using Super-Resolution Convolutional Neural Network method, our system could achieve a balance between the competing goals of low-cost and superior performance. Our experimental results showed the effectiveness of our system.

Keywords: 3D plant data, automatic recording, stereo camera, deep learning, image processing

Procedia PDF Downloads 249

Authors: Ahmed Krobba, Meriem Touzene, Mohamed Debeyche

Abstract:

The wireless communications system (5G) will provide more diverse applications and higher quality services for users compared to the long-term evolution 4G (LTE). 5G uses a higher carrier frequency, which suffers from information loss in 5G coverage. Most 5G users often cannot obtain high-quality communications due to transmission channel noise and channel complexity. Physical Uplink Control Channel (PUCCH-NR: Physical Uplink Control Channel New Radio) plays a crucial role in 5G NR telecommunication technology, which is mainly used to transmit link control information uplink (UCI: Uplink Control Information. This study based of evaluating the performance of channel physical uplink control PUCCH-NR under low Signal-to-Noise Ratios with various antenna numbers reception. We propose the artificial intelligence approach based on deep neural networks (Deep Learning) to estimate the PUCCH-NR channel in comparison with this approach with different conventional methods such as least-square (LS) and minimum-mean-square-error (MMSE). To evaluate the channel performance we use the block error rate (BLER) as an evaluation criterion of the communication system. The results show that the deep neural networks method gives best performance compared with MMSE and LS

Keywords: 5G network, uplink (Uplink), PUCCH channel, NR-PUCCH channel, deep learning

Procedia PDF Downloads 22

Authors: Pingping Lin, Xudong Luo, Yifan Fan

Abstract:

Sentiment analysis is a very active research topic. Every day, Facebook, Twitter, Weibo, and other social media, as well as significant e-commerce websites, generate a massive amount of comments, which can be used to analyse peoples opinions or emotions. The existing methods for sentiment analysis are based mainly on sentiment dictionaries, machine learning, and deep learning. The first two kinds of methods rely on heavily sentiment dictionaries or large amounts of labelled data. The third one overcomes these two problems. So, in this paper, we focus on the third one. Specifically, we survey various sentiment analysis methods based on convolutional neural network, recurrent neural network, long short-term memory, deep neural network, deep belief network, and memory network. We compare their futures, advantages, and disadvantages. Also, we point out the main problems of these methods, which may be worthy of careful studies in the future. Finally, we also examine the application of deep learning in multimodal sentiment analysis and aspect-level sentiment analysis.

Keywords: document analysis, deep learning, multimodal sentiment analysis, natural language processing

Procedia PDF Downloads 131

Authors: Mohammad Shameel, Peter K. K. Loh, James H. Ng

Abstract:

Cybercrime is a new category of crime which poses a different challenge for crime investigators and incident responders. Attackers can mask their identities using a suite of tools and with the help of the deep web, which makes them difficult to track down. Scouring the deep web manually takes time and is inefficient. There is a growing need for a tool to scour the deep web to obtain useful evidence or intel automatically. In this paper, we will explain the background and motivation behind the research, present a survey of existing research on related tools, describe the design of our own crime/threat intelligence hunting tool prototype, demonstrate its capability with some test cases and lastly, conclude with proposals for future enhancements.

Keywords: cybercrime, deep web, threat intelligence, web crawler

Procedia PDF Downloads 141