Search results for: deep soil mix

Authors: Atakan Aral Ormancı, Tuğçe Arslantaş, Murat Özcü

Abstract:

This study presents the design and implementation of an experimental chassis-level system for various control applications. Specifically, the height level of the chassis is controlled using proportional integrated derivative, fuzzy logic, and deep learning control methods. Real-time data obtained from height and pressure sensors installed in a 6x2 truck chassis, in combination with pulse-width modulation signal values, are utilized during the tests. A prototype pneumatic system of a 6x2 truck is added to the setup, which enables the Smart Pneumatic Actuators to function as if they were in a real-world setting. To obtain real-time signal data from height sensors, an Arduino Nano is utilized, while a Raspberry Pi processes the data using Matlab/Simulink and provides the correct output signals to control the Smart Pneumatic Actuator in the truck chassis. The objective of this research is to optimize the time it takes for the chassis to level down and up under various loads. To achieve this, proportional integrated derivative control, fuzzy logic control, and deep learning techniques are applied to the system. The results show that the deep learning method is superior in optimizing time for a non-linear system. Fuzzy logic control with a triangular membership function as the rule base achieves better outcomes than proportional integrated derivative control. Traditional proportional integrated derivative control improves the time it takes to level the chassis down and up compared to an uncontrolled system. The findings highlight the superiority of deep learning techniques in optimizing the time for a non-linear system, and the potential of fuzzy logic control. The proposed approach and the experimental results provide a valuable contribution to the field of control, automation, and systems engineering.

Keywords: automotive, chassis level control, control systems, pneumatic system control

Procedia PDF Downloads 61

Authors: John Pomerat, Aviv Segev

Abstract:

Recently, deep learning has had many theoretical breakthroughs. For deep learning to be successful in the industry, however, there need to be practical algorithms capable of handling many real-world hiccups preventing the immediate application of a learning algorithm. Although AI promises to revolutionize the healthcare industry, getting access to patient data in order to train learning algorithms has not been easy. One proposed solution to this is data- sharing. In this paper, we propose an alternative protocol, based on multi-party computation, to train deep learning models while maintaining both the privacy and security of training data. We examine three methods of training neural networks in this way: Transfer learning, average ensemble learning, and series network learning. We compare these methods to the equivalent model obtained through data-sharing across two different experiments. Additionally, we address the security concerns of this protocol. While the motivating example is healthcare, our findings regarding multi-party computation of neural network training are purely theoretical and have use-cases outside the domain of healthcare.

Keywords: neural network aggregation, multi-party computation, transfer learning, average ensemble learning

Procedia PDF Downloads 143

Authors: Nataliya Apanovich, Mark Wright

Abstract:

Uganda is among the poorest but fastest growing populations in the world. Its annual population growth of 3% puts additional stress through land fragmentation, agricultural intensification, and deforestation on already highly weathered tropical (Ferralsol) soils. All of these factors lead to decreased agricultural yields and consequently diminished food security. The central region of Uganda, Buganda Kingdom, is especially vulnerable in terms of food security as its high population density coupled with mismanagement of natural resources led to gradual loss of its soil and even changes in microclimate. These changes are negatively affecting livelihoods of smallholder farmers who comprise 80% of all population in Uganda. This research focuses on biochar for soil remediation in Masaka District, Uganda. If produced on a small scale from locally sourced materials, biochar can increase the quality of soil in a cost and time effective manner. To assess biochar potential, 151 smallholder farmers were interviewed on the types of crops grown, agricultural residues produced and their use, as well as on attitudes towards biochar use and its production on a small scale. The interviews were conducted in 7 sub-counties, 32 parishes, and 92 villages. The total farmland covered by the study was 606.2 kilometers. Additional information on the state of agricultural development and environmental degradation in the district was solicited from four local government officials via informal interviews. This project has been conducted in collaboration with the international agricultural research institution, Makerere University in Kampala, Uganda. The results of this research can have implications on the way farmers perceive the value of their agricultural residues and what they decide to do with them. The underlying objective is to help smallholders in degraded soils increase their agricultural yields through the use of biochar without diverting the already established uses of agricultural residues to a new soil management practice.

Keywords: agricultural residues, biochar, central Uganda, food security, soil erosion, soil remediation

Procedia PDF Downloads 262

Authors: Huy Hieu Pham, Houssam Salmane, Louahdi Khoudour, Alain Crouzil, Pablo Zegers, Sergio Velastin

Abstract:

We present a deep learning-based multitask framework for joint 3D human pose estimation and action recognition from color video sequences. Our approach proceeds along two stages. In the first, we run a real-time 2D pose detector to determine the precise pixel location of important key points of the body. A two-stream neural network is then designed and trained to map detected 2D keypoints into 3D poses. In the second, we deploy the Efficient Neural Architecture Search (ENAS) algorithm to find an optimal network architecture that is used for modeling the Spatio-temporal evolution of the estimated 3D poses via an image-based intermediate representation and performing action recognition. Experiments on Human3.6M, Microsoft Research Redmond (MSR) Action3D, and Stony Brook University (SBU) Kinect Interaction datasets verify the effectiveness of the proposed method on the targeted tasks. Moreover, we show that our method requires a low computational budget for training and inference.

Keywords: human action recognition, pose estimation, D-CNN, deep learning

Procedia PDF Downloads 126

Authors: Huseyn Nuraddin Qurbanov

Abstract:

The reliable food supply of the population of the republic is one of the main directions of the state's economic policy. Grain growing, which is the basis of agriculture, is important in this area. In the cultivation of cereals on the slopes, the application of equal amounts of mineral fertilizers the under the soil before sowing is a very important technological process. The low level of technical equipment in this area prevents producers from providing the country with the necessary quality cereals. Experience in the operation of modern technical means has shown that, at present, there is a need to provide an equal amount of fertilizer on the slopes to under the soil, fully meeting the agro-technical requirements. No fundamental changes have been made to the industrial machines that fertilize the under the soil, and unequal application of fertilizers under the soil on the slopes has been applied. This technological process leads to the destruction of new seedlings and reduced productivity due to intolerance to frost during the winter for the plant planted in the fall. In special climatic conditions, there is an optimal fertilization rate for each agricultural product. The application of fertilizers to the soil is one of the conditions that increase their efficiency in the field. As can be seen, the development of a new technical proposal for fertilizing and plowing the slopes in equal amounts on the slopes, improving the technological and design parameters, and taking into account the physical and mechanical properties of fertilizers is very important. Taking into account the above-mentioned issues, a combined plough was developed in our laboratory. Combined plough carries out pre-sowing technological operation in the cultivation of cereals, providing a smooth equal amount of mineral fertilizers under the soil on the slopes. Mathematical models of a smooth spreader that evenly distributes fertilizers in the field have been developed. Thus, diagrams and graphs obtained without distribution on the 8 partitions of the smooth spreader are constructed under the inclined angles of the slopes. Percentage and productivity of equal distribution in the field were noted by practical and theoretical analysis.

Keywords: combined plough, mineral fertilizer, equal sowing, fertilizer norm, grain-crops, sowing fertilizer

Procedia PDF Downloads 123

Authors: Fazıl Gökgöz, Fahrettin Filiz

Abstract:

Load forecasting has become crucial in recent years and become popular in forecasting area. Many different power forecasting models have been tried out for this purpose. Electricity load forecasting is necessary for energy policies, healthy and reliable grid systems. Effective power forecasting of renewable energy load leads the decision makers to minimize the costs of electric utilities and power plants. Forecasting tools are required that can be used to predict how much renewable energy can be utilized. The purpose of this study is to explore the effectiveness of LSTM-based neural networks for estimating renewable energy loads. In this study, we present models for predicting renewable energy loads based on deep neural networks, especially the Long Term Memory (LSTM) algorithms. Deep learning allows multiple layers of models to learn representation of data. LSTM algorithms are able to store information for long periods of time. Deep learning models have recently been used to forecast the renewable energy sources such as predicting wind and solar energy power. Historical load and weather information represent the most important variables for the inputs within the power forecasting models. The dataset contained power consumption measurements are gathered between January 2016 and December 2017 with one-hour resolution. Models use publicly available data from the Turkish Renewable Energy Resources Support Mechanism. Forecasting studies have been carried out with these data via deep neural networks approach including LSTM technique for Turkish electricity markets. 432 different models are created by changing layers cell count and dropout. The adaptive moment estimation (ADAM) algorithm is used for training as a gradient-based optimizer instead of SGD (stochastic gradient). ADAM performed better than SGD in terms of faster convergence and lower error rates. Models performance is compared according to MAE (Mean Absolute Error) and MSE (Mean Squared Error). Best five MAE results out of 432 tested models are 0.66, 0.74, 0.85 and 1.09. The forecasting performance of the proposed LSTM models gives successful results compared to literature searches.

Keywords: deep learning, long short term memory, energy, renewable energy load forecasting

Procedia PDF Downloads 247

Authors: John Gibbs, Benjamin Flanders, Dylan Pozorski, Weixuan Liu

Abstract:

Weight Painting–adjusting the influence a skeletal joint has on a given vertex in a character mesh–is an arduous and time con- suming part of the 3D animation pipeline. This process generally requires a trained technical animator and many hours of work to complete. Our skiNNer plug-in, which works within Autodesk’s Maya 3D animation software, uses Machine Learning and data pro- cessing techniques to create a deep neural network model that can accomplish the weight painting task in seconds rather than hours for bipedal quasi-humanoid character meshes. In order to create a properly trained network, a number of challenges were overcome, including curating an appropriately large data library, managing an arbitrary 3D mesh size, handling arbitrary skeletal architectures, accounting for extreme numeric values (most data points are near 0 or 1 for weight maps), and constructing an appropriate neural network model that can properly capture the high frequency alter- ation between high weight values (near 1.0) and low weight values (near 0.0). The arrived at neural network model is a cross between a traditional CNN, deep residual network, and fully dense network. The resultant network captures the unusually hard-edged features of a weight map matrix, and produces excellent results on many bipedal models.

Keywords: 3d animation, animation, character, rigging, skinning, weight painting, machine learning, artificial intelligence, neural network, deep neural network

Procedia PDF Downloads 250

Authors: Masood Varshosaz, Kamyar Hasanpour

Abstract:

In recent years, we have seen growing interest in recognizing humans in drone images for post-disaster search and rescue operations. Deep learning algorithms have shown great promise in this area, but they often require large amounts of labeled data to train the models. To keep the data acquisition cost low, augmentation techniques can be used to create additional data from existing images. There are many techniques of such that can help generate variations of an original image to improve the performance of deep learning algorithms. While data augmentation is potentially assumed to improve the accuracy and robustness of the models, it is important to ensure that the performance gains are not outweighed by the additional computational cost or complexity of implementing the techniques. To this end, it is important to evaluate the impact of data augmentation on the performance of the deep learning models. In this paper, we evaluated the most currently available 2D data augmentation techniques on a standard convolutional network which was trained for recognizing humans in drone images. The techniques include rotation, scaling, random cropping, flipping, shifting, and their combination. The results showed that the augmented models perform 1-3% better compared to a base network. However, as the augmented images only contain the human parts already visible in the original images, a new data augmentation approach is needed to include the invisible parts of the human body. Thus, we suggest a new method that employs simulated 3D human models to generate new data for training the network.

Keywords: human recognition, deep learning, drones, disaster mitigation

Procedia PDF Downloads 76

Authors: Namik Kemal Sonmez, Sahriye Sonmez, Hasan Rasit Turkkan, Hatice Tuba Selcuk

Abstract:

In this study, the spatial variation of plant and soil nutrition contents of tomato plants grown in greenhouses was investigated in Elmalı region of Antalya. For this purpose, total of 19 sampling points were determined. Coordinates of each sampling points were recorded by using a hand-held GPS device and were transferred to satellite data in GIS. Soil samples were collected from two different depths, 0-20 and 20-40 cm, and leaf were taken from different tomato greenhouses. The soil and plant samples were analyzed for N, P and K. Then, attribute tables were created with the analyses results by using GIS. Data were analyzed and semivariogram models and parameters (nugget, sill and range) of variables were determined by using GIS software. Kriged maps of variables were created by using nugget, sill and range values with geostatistical extension of ArcGIS software. Kriged maps of the N, P and K contents of plant and soil samples showed patchy or a relatively smooth distribution in the study areas. As a result, the N content of plants were sufficient approximately 66% portion of the tomato productions. It was determined that the P and K contents were sufficient of 70% and 80% portion of the areas, respectively. On the other hand, soil total K contents were generally adequate and available N and P contents were found to be highly good enough in two depths (0-20 and 20-40 cm) 90% portion of the areas.

Keywords: Elmali, nutrients, springs greenhouses, spatial variation, tomato

Procedia PDF Downloads 229

Authors: Dhanesh Sing Das, Bharat Tadikonda Venkata

Abstract:

Collapsible soils undergo significant volume reduction upon wetting under the pre-existing mechanically applied normal stress (inundation pressure). These soils exhibit a very high strength in air-dried conditions and can carry up to a considerable magnitude of normal stress without undergoing significant volume change. The soil strength is, however, lost upon saturation and results in a sudden collapse of the soil structure under the existing mechanical stress condition. The intrusion of water into the dry deposits of such soil causes ground subsidence leading to damages in the overlying buildings/structures. A study on the wetting-induced volume change behavior of collapsible soils is essential in dealing with the ground subsidence problems in various geotechnical engineering practices. The collapse of loosely compacted Kaolin soil upon wetting under various inundation pressures has been reported in recent studies. The collapse in the Kaolin soil is attributed to the alteration in the soil particle-particle association (fabric) resulting due to the changes in the various inter-particle (microscale) forces induced by the water saturation. The inundation pressure plays a significant role in the fabric evolution during the wetting process, thus controls the collapse potential of the compacted soil. A microstructural study is useful to understand the collapse mechanisms at various pore-fabric levels under different inundation pressure. Kaolin soil compacted to a dry density of 1.25 g/cc was used in this work to study the wetting-induced volume change behavior under different inundation pressures in the range of 10-1600 kPa. The compacted specimen of Kaolin soil exhibited a consistent collapse under all the studied inundation pressure. The collapse potential was observed to be increasing with an increase in the inundation pressure up to a maximum value of 13.85% under 800 kPa and then decreased to 11.7% under 1600 kPa. Microstructural analysis was carried out based on the fabric images and the pore size distributions (PSDs) obtained from FESEM analysis and mercury intrusion porosimetry (MIP), respectively. The PSDs and the soil fabric images of ‘as-compacted’ specimen and post-collapse specimen under 400 kPa were analyzed to understand the changes in the soil fabric and pores due to wetting. The pore size density curve for the post-collapse specimen was found to be on the finer side with respect to the ‘as-compacted’ specimen, indicating the reduction of the larger pores during the collapse. The inter-aggregate pores in the range of 0.1-0.5μm were identified as the major contributing pore size classes to the macroscopic volume change. Wetting under an inundation pressure results in the reduction of these pore sizes and lead to an increase in the finer pore sizes. The magnitude of inundation pressure influences the amount of reduction of these pores during the wetting process. The collapse potential was directly related to the degree of reduction in the pore volume contributed by these pore sizes.

Keywords: collapse behavior, inundation pressure, kaolin, microstructure

Procedia PDF Downloads 122

Authors: Daniel Osieko Okach, Joseph Otieno Ondier, Gerhard Rambold, John Tenhunen, Bernd Huwe, Dennis Otieno

Abstract:

Increased human activities such as grazing, logging, and agriculture alongside unpredictable rainfall patterns have been detrimental to the ecosystem service delivery, therefore compromising its productivity potential. This study aimed at simulating the impact of drought (50%) and enhanced rainfall (150%) on the future herbaceous CO2 uptake, biomass production and soil C:N dynamics in a humid savanna ecosystem influenced by livestock grazing. Rainfall pattern was predicted using manipulation experiments set up to reduce (50%) and increase (150%) ambient (100%) rainfall amounts in grazed and non-grazed plots. The impact of manipulated rainfall regime on herbaceous CO2 fluxes, biomass production and soil C:N dynamics was measured against volumetric soil water content (VWC) logged every 30 minutes using the 5TE (Decagon Devices Inc., Washington, USA) soil moisture sensors installed (at 20 cm soil depth) in every plots. Herbaceous biomass was estimated using destructive method augmented by standardized photographic imaging. CO2 fluxes were measured using the ecosystem chamber method and the gas analysed using LI-820 gas analyzer (USA). C:N ratio was calculated from the soil carbon and Nitrogen contents (analyzed using EA2400CHNS/O and EA2410 N elemental analyzers respectively) of different plots under study. The patterning of VWC was directly influenced by the rainfall amount with lower VWC observed in the grazed compared to the non-grazed plots. Rainfall variability, grazing and their interaction significantly affected changes in VWC (p < 0.05) and subsequently total biomass and CO2 fluxes. VWC had a strong influence on CO2 fluxes under 50% rainfall reduction in the grazed (r2 = 0.91; p < 0.05) and ambient rainfall in the ungrazed (r2 = 0.77; p < 0.05). The dependence of biomass on VWC across plots was enhanced under grazed (r2 = 0.78 - 0.87; p < 0.05) condition as compared to ungrazed (r2 = 0.44 - 0.85; p < 0.05). The C:N ratio was however not correlated to VWC across plots. This study provides insight on how the predicted trends in humid savanna will respond to changes influenced by rainfall variability and livestock grazing and consequently the sustainable management of such ecosystems.

Keywords: CO2 fluxes, rainfall manipulation, soil properties, sustainability

Procedia PDF Downloads 113

Authors: S. Amirhossein Tabatabaei, Abdulghader A. Aldaeef, Mohammad T. Rayhani

Abstract:

As the permafrost continues to melt in the northern regions due to global warming, a soil-water mixture is left behind with drastically lower strength; a phenomenon that directly impacts the resilience of existing structures and infrastructure systems. The frozen soil-structure interaction, which in ice-poor soils is controlled by both interface shear and ice-bonding, changes its nature into a sole frictional state. Adfreeze, the controlling mechanism in frozen soil-structure interaction, diminishes as the ground temperature approaches zero. The main purpose of this paper is to capture the altered behaviour of frozen interface with respect to rising temperature, especially near melting states. A series of pull-out tests are conducted on model piles inside a cold room to study how the strength parameters are influenced by the phase change in ice-poor soils. Steel model piles, embedded in artificially frozen cohesionless soil, are subjected to both sustained pull-out forces and constant rates of displacement to observe the creep behaviour and acquire load-deformation curves, respectively. Temperature, as the main variable of interest, is increased from a lower limit of -10°C up to the point of melting. During different stages of the temperature rise, both skin deformations and temperatures are recorded at various depths along the pile shaft. Significant reduction of pullout capacity and accelerated creep behaviour is found to be the primary consequences of rising temperature. By investigating the different pull-out capacities and deformations measured during step-wise temperature change, characteristics of the transition from frozen to unfrozen soil-structure interaction are studied.

Keywords: Adfreeze, frozen soil-structure interface, ice-poor soils, pull-out capacity, warming permafrost

Procedia PDF Downloads 95

Authors: Oustani Mabrouka, Halilat Md Tahar, Hannachi Slimane

Abstract:

This study was conducted on representative and contrasting soils of arid regions. It focuses on the compared influence of two organic fertilizers: poultry manure (PM) and bovine manure (BM) on improving the microbial functioning of non-saline (SS) and saline (SSS) soils, in particularly, the process of mineralization of nitrogen and carbon. The microbiological activity was estimated by respirometric test (CO2–C emissions) and the extraction of two forms of mineral nitrogen (NH4+-N and NO3--N). Thus, after 56 days of incubation under controlled conditions (28 degrees and 80 per cent of the field capacity), the two types of manures showed that the mineralization activity varies according to type of soil and the organic substrate itself. However, the highest cumulative quantities of CO2–C, NH4+–N and NO3-–N obtained at the end of incubation were recorded in non-saline (SS) soil treated with poultry manure with 1173.4, 4.26 and 8.40 mg/100 g of dry soil, respectively. The reductions in rates of release of CO2–C and of nitrification under saline conditions were 21 and 36, 78 %, respectively. The influence of organic substratum on the microbial density shows a stimulating effect on all microbial groups studied. The whole results show the usefulness of two types of manures for the improvement of the microbiological functioning of arid soils.

Keywords: Salinity, Organic matter, Microorganisms, Mineralization, Nitrogen, Carbon, Arid regions

Procedia PDF Downloads 260

Authors: Chahrazed Salhi, Ayoub Zeroual, Yasmina Hamitouche

Abstract:

Soil water erosion poses a significant threat to the watersheds in Algeria today. The degradation of storage capacity in large dams over the past two decades, primarily due to erosion, necessitates a comprehensive understanding of the factors that contribute to soil erosion. The Isser watershed, located in the Northwestern region of Algeria, faces additional challenges such as recurrent droughts and the presence of delicate marl and clay outcrops, which amplify its susceptibility to water erosion. This study aims to employ advanced techniques such as Geographic Information Systems (GIS) and Remote Sensing (RS), in conjunction with the Canonical Correlation Analysis (CCA) method and Soil Water Assessment Tool (SWAT) model, to predict specific erosion patterns and analyze the key factors influencing erosion in the Isser basin. To accomplish this, an array of data sources including rainfall, climatic, hydrometric, land use, soil, digital elevation, and satellite data were utilized. The application of the SWAT model to the Isser basin yielded an average annual soil loss of approximately 16 t/ha/year. Particularly high erosion rates, exceeding 12 T/ha/year, were observed in the central and southern parts of the basin, encompassing 41% of the total basin area. Through Canonical Correlation Analysis, it was determined that vegetation cover and topography exerted the most substantial influence on erosion. Consequently, the study identified significant and spatially heterogeneous erosion throughout the study area. The impact of land topography on soil loss was found to be directly proportional, while vegetation cover exhibited an inverse proportional relationship. Modeling specific erosion for the Ladrat dam sub-basin estimated a rate of around 39 T/ha/year, thus accounting for the recorded capacity loss of 17.80% compared to the bathymetric survey conducted in 2019. The findings of this research provide valuable decision-support tools for soil conservation managers, empowering them to make informed decisions regarding soil conservation measures.

Keywords: Isser watershed, RS, CCA, SWAT, vegetation cover, topography

Procedia PDF Downloads 50

Authors: Višnjica Vučetić, Mislav Anić, Jelena Bašić, Petra Sviličić, Ivana Tomašević

Abstract:

The Agroclimatic Atlas of Croatia (Atlas) for the periods 1981–2010 and 1991–2020 is monograph of six chapters in digital form. Detailed descriptions of particular agroclimatological data are given in separate chapters as follows: agroclimatic indices based on air temperature (degree days, Huglin heliothermal index), soil temperature, water balance components (precipitation, potential evapotranspiration, actual evapotranspiration, soil moisture content, runoff, recharge and soil moisture loss) and fire weather indices. The last chapter is a description of the digital methods for the spatial interpolations (R and GIS). The Atlas comprises textual description of the relevant climate characteristic, maps of the spatial distribution of climatological elements at 109 stations (26 stations for soil temperature) and tables of the 30-year mean monthly, seasonal and annual values of climatological parameters at 24 stations. The Atlas was published in 2021, on the seventieth anniversary of the agrometeorology development at the Meteorological and Hydrological Service of Croatia. It is intended to support improvement of sustainable system of agricultural production and forest protection from fire and as a rich source of information for agronomic and forestry experts, but also for the decision-making bodies to use it for the development of strategic plans.

Keywords: agrometeorology, agroclimatic indices, soil temperature, water balance components, fire weather index, meteorological and hydrological service of Croatia

Procedia PDF Downloads 109

Authors: Swapneel U. Naphade, Dushyant A. Patil, Satyabodh M. Kulkarni

Abstract:

In this work, we propose the concept and geometrical design of a soil moisture control system (SMCS) module by following the product development approach to develop an inexpensive, easy to use and quick to install product targeted towards agriculture practitioners. The module delivers water to the agricultural land efficiently by sensing the soil moisture and activating the delivery valve. We start with identifying the general needs of the potential customer. Then, based on customer needs we establish product specifications and identify important measuring quantities to evaluate our product. Keeping in mind the specifications, we develop various conceptual solutions of the product and select the best solution through concept screening and selection matrices. Then, we develop the product architecture by integrating the systems into the final product. In the end, the geometric design is done using human factors engineering concepts like heuristic analysis, task analysis, and human error reduction analysis. The result of human factors analysis reveals the remedies which should be applied while designing the geometry and software components of the product. We find that to design the best grip in terms of comfort and applied force, for a power-type grip, a grip-diameter of 35 mm is the most ideal.

Keywords: agriculture, human factors, product design, soil moisture control

Procedia PDF Downloads 160

Authors: V. Ivanov, J. Chu, V. Stabnikov

Abstract:

Low cost, sustainable, and environmentally friendly microbial cements, grouts, polysaccharides and bioplastics are useful in construction and geotechnical engineering. Construction-related biotechnologies are based on activity of different microorganisms: urease-producing, acidogenic, halophilic, alkaliphilic, denitrifying, iron- and sulphate-reducing bacteria, cyanobacteria, algae, microscopic fungi. The bio-related materials and processes can be used for the bioaggregation, soil biogrouting and bioclogging, biocementation, biodesaturation of water-satured soil, bioencapsulation of soft clay, biocoating, and biorepair of the concrete surface. Altogether with the most popular calcium- and urea based biocementation, there are possible and often are more effective such methods of ground improvement as calcium- and magnesium based biocementation, calcium phosphate strengthening of soil, calcium bicarbonate biocementation, and iron- or polysaccharide based bioclogging. The construction-related microbial biotechnologies have a lot of advantages over conventional construction materials and processes.

Keywords: ground improvement, biocementation, biogrouting, microorganisms

Procedia PDF Downloads 210

Authors: Harison Masih, Jyotsna Kiran Peter, Sundara Singh, Geetha Singh

Abstract:

The present investigation deals with diversity of dermatophytes and keratinophilic fungi from different tourist spots such as Agra Fort, Akbar tomb, It-Mat-Ud-Daulah, Mariam tomb, Radha Swami Bagh, and Taj Mahal of Agra City. These fungi are medically important which causes various infections and diseases in humans and animals. The main reservoir of these pathogens are the keratinous substances that increases due to birds and animal activities in the vicinity of monuments, where thousands (5413266) annual visitors from all over the world are visiting. The soil samples were subjected to isolate the pathogenic fungi through bait technique (buffalo skin, chicken feathers, human hair and goat tail hair). Baits were spread over the soil samples and incubated at room temperature for 30-35 days and pure culture isolates were maintained in SDA medium, stored at 4°C. Highest number of visitors were (3906453) from Taj Mahal, minimum 10785 at Mariam tomb annually, the total 271 isolates were encountered from soil samples out of these 18 genera and 38 species were found in different season. Highest incidence was 4.79% frequency shown by Chrysosporium keratinophilum while least 738% frequency occurrence by Trichophyton simii in soil samples. From the present study it was concluded that the incidence of pathogenic fungal isolates were the common in tourists soil that are etiological agents of superficial mycosis. Thus, both human and animal activity seemed to play an important role in occurrence and distribution of keratinophilic and related dermatophytes at various tourist places of Agra city.

Keywords: dermatophytic fungal diversity, bait technique, visitors at tourist spots, human and animal activities, soil samples

Procedia PDF Downloads 459

Authors: Balogun R. Ayodeji, Adefisan E. Adesanya, Adeyewa Z. Debo, E. C. Okogbue

Abstract:

Deep convective systems during active periods of the West African monsoon season have not been properly investigated over better temporal and spatial resolution in West Africa. Deep convective systems are investigated over seven climatic zones of the West African sub-region, which are; west-coast rainforest, dry rainforest, Nigeria-Cameroon rainforest, Nigeria savannah, Central African and South Sudan (CASS) Savannah, Sudano-Sahel, and Sahel, using data from Tropical Rainfall Measurement Mission (TRMM) Precipitation Feature (PF) database. The vertical structure of the convective systems indicated by the presence of at least one 40 dBZ and reaching (attaining) at least 1km in the atmosphere showed strong core (highest frequency (%)) of reflectivity values around 2 km which is below the freezing level (4-5km) for all the zones. Echoes are detected above the 15km altitude much more frequently in the rainforest and Savannah zones than the Sudano and Sahel zones during active periods in March-May (MAM), whereas during active periods in June-September (JJAS) the savannahs, Sudano and Sahel zones convections tend to reach higher altitude more frequently than the rainforest zones. The percentage frequencies of deep convection indicated that the occurrences of the systems are within the range of 2.3-2.8% during both March-May (MAM) and June-September (JJAS) active periods in the rainforest and savannah zones. On the contrary, the percentage frequencies were found to be less than 2% in the Sudano and Sahel zones, except during the active-JJAS period in the Sudano zone.

Keywords: active periods, convective system, frequency, reflectivity

Procedia PDF Downloads 131

Authors: Radheshyam Yadav, Ramakrishna Wusirika

Abstract:

Plant Growth Promoting Bacteria (PGPB) and Arbuscular Mycorrhizal (AM) Fungi are ubiquitous in soil and often very critical for crop yield and agriculture sustainability, and this has motivated the agricultural practices to support and promote PGPB and AM Fungi in agriculture. PGPB can be involved in a range of processes that affect Nitrogen (N) and Phosphorus (P) transformations in soil and thus influence nutrient availability and uptake to the plants. A field study with two wheat cultivars, HD-3086, and HD-2967 was performed in Malwa region, Bathinda of Punjab, India, to evaluate the effect of native and non-native PGPB alone and in combination with AM fungi as an inoculant on wheat grain yield, nutrient uptake and soil health parameters (dehydrogenase, urease, β‐glucosidase). Our results showed that despite an early insignificant increase in shoot length, plants treated with PGPB (Bacillus sp.) and AM Fungi led to a significant increase in shoot growth at maturity, aboveground biomass, nitrogen (45% - 40%) and phosphorus (40% - 34%) content in wheat grains relative to untreated control plants. Similarly, enhanced grain yield and nutrients uptake i.e. copper (27.15% - 36.25%) iron (43% - 53%) and zinc (44% - 47%) was recorded in PGPB and AM Fungi treated plants relative to untreated control. Overall, inoculation with native PGPB alone and in combination with AM Fungi provided benefits to enhance grain yield, wheat biofortification, and improved soil fertility, despite this effect varied depending on different PGPB isolates and wheat cultivars. These field study results provide evidence of the benefits of agricultural practices involving native PGPB and AM Fungi to the plants. These native strains and AM Fungi increased accumulations of copper, iron, and zinc in wheat grains, enhanced grain yield, and soil fertility.

Keywords: AM Fungi, biofortification, PGPB, soil microbial enzymes

Procedia PDF Downloads 310

Authors: Nguyen J. M., Lucas G., Brunner M., Ruan S., Antonioli D.

Abstract:

Deep learning based on convolutional neural networks (CNN) is a very powerful technique for classifying information from an image. We propose a new method, DeepNic, to transform each variable of a tabular dataset into an image where each pixel represents a set of conditions that allow the variable to make an error-free prediction. The contrast of each pixel is proportional to its prediction performance and the color of each pixel corresponds to a sub-family of NICs. NICs are probabilities that depend on the number of inputs to each neuron and the range of coefficients of the inputs. Each variable can therefore be expressed as a function of a matrix of 2 vectors corresponding to an image whose pixels express predictive capabilities. Our objective is to transform each variable of tabular data into images into an image that can be analysed by CNNs, unlike other methods which use all the variables to construct an image. We analyse the NIC information of each variable and express it as a function of the number of neurons and the range of coefficients used. The predictive value and the category of the NIC are expressed by the contrast and the color of the pixel. We have developed a pipeline to implement this technology and have successfully applied it to genomic expressions on an Affymetrix chip.

Keywords: tabular data, deep learning, perfect trees, NICS

Procedia PDF Downloads 72

Authors: N. N. Lingaraju, A. Sathish, K. N. Geetha, C. A. Srinivasamurthy, S. Bhaskar

Abstract:

Studies on spent wash utilization as a nutrient source through 'Effect of distillery spentwash application on soil properties and yield of maize (Zea may L.) and finger millet (Eleusine coracana (L.) G)' was carried out in Malavalli Taluk, Mandya District, Karnataka State, India. The study was conducted in fourteen different locations of Malavalli (12) and Maddur taluk (2) involving maize and finger millet as a test crop. The spentwash was characterized for various parameters like pH, EC, total NPK, Na, Ca, Mg, SO₄, Fe, Zn, Cu, Mn and Cl content. It was observed from the results that the pH was slightly alkaline (7.45), EC was excess (23.3 dS m⁻¹), total NPK was 0.12, 0.02, and 1.31 percent respectively, Na, Ca, Mg and SO₄ concentration was 664, 1305, 745 and 618 (mg L⁻¹) respectively, total solid content was quite high (6.7%), Fe, Zn, Cu, Mn, values were 23.5, 5.70, 3.64, 4.0 mg L⁻¹, respectively. The crops were grown by adopting different crop management practices after application of spentwash at 100 m³ ha⁻¹ to the identified farmer fields. Soil samples were drawn at three stages i.e., before sowing of crop, during crop growth stage and after harvest of the crop at 2 depths (0-30 and 30-60 cm) and analyzed for pH, EC, available K and Na parameters by adopting standard procedures. The soil analysis showed slightly acidic reaction (5.93), normal EC (0.43 dS m⁻¹), medium available potassium (267 kg ha⁻¹) before application of spentwash. Application of spentwash has enhanced pH level of soil towards neutral (6.97), EC 0.25 dS m⁻¹, available K2O to 376 kg ha⁻¹ and sodium content of 0.73 C mol (P+) kg⁻¹ during the crop growth stage. After harvest of the crops soil analysis data indicated a decrease in pH to 6.28, EC of 0.22 dS m⁻¹, available K₂O to 316 kg ha⁻¹ and Na 0.52 C mol (P⁺) kg⁻¹ compared with crop growth stage. The study showed that, there will be enhancement of potassium levels if the spentwash is applied once to dryland. The yields of both the crops were quantified and found to be in the range of 35.65 to 65.55 q ha⁻¹ and increased yield to the extent of 13.36-22.36 percent as compared to control field (11.36-22.33 q ha⁻¹) in maize crop. Also, finger millet yield was increased with the spentwash application to the extent of 14.21-20.49 percent (9.5-17.73 q ha⁻¹) higher over farmers practice (8.15-14.15 q ha⁻¹).

Keywords: distillery spentwash, finger millet, maize, waste water

Procedia PDF Downloads 330

Authors: Venkata Narayana Chejarla, Nafisa Parvez Shaik, Gopi Vara Prasad Marabathula, Deva Kumar Bejjam

Abstract:

Yoga is an advanced, well-recognized method with roots in Indian philosophy. Yoga benefits both the body and the psyche. Yoga is a regular exercise that helps people relax and sleep better while also enhancing their balance, endurance, and concentration. Yoga can be learned in a variety of settings, including at home with the aid of books and the internet as well as in yoga studios with the guidance of an instructor. Self-learning does not teach the proper yoga poses, and doing them without the right instruction could result in significant injuries. We developed "Online Yoga Asana Trainer using Deep Learning" so that people could practice yoga without a teacher. Our project is developed using Tensorflow, Movenet, and Keras models. The system makes use of data from Kaggle that includes 25 different yoga poses. The first part of the process involves applying the movement model for extracting the 17 key points of the body from the dataset, and the next part involves preprocessing, which includes building a pose classification model using neural networks. The system scores a 98.3% accuracy rate. The system is developed to work with live videos.

Keywords: yoga, deep learning, movenet, tensorflow, keras, CNN

Procedia PDF Downloads 220

Authors: Yanjun Chen, Chuanping Hu, Jie Shao, Lin Mei, Chongyang Zhang

Abstract:

Traditional image classification is based on encoding scheme (e.g. Fisher Vector, Vector of Locally Aggregated Descriptor) with low-level image features (e.g. SIFT, HoG). Compared to these low-level local features, deep convolutional features obtained at the mid-level layer of convolutional neural networks (CNN) have richer information but lack of geometric invariance. For scene classification, there are scattered objects with different size, category, layout, number and so on. It is crucial to find the distinctive objects in scene as well as their co-occurrence relationship. In this paper, we propose a method to take advantage of both deep convolutional features and the traditional encoding scheme while taking object-centric and scene-centric information into consideration. First, to exploit the object-centric and scene-centric information, two CNNs that trained on ImageNet and Places dataset separately are used as the pre-trained models to extract deep convolutional features at multiple scales. This produces dense local activations. By analyzing the performance of different CNNs at multiple scales, it is found that each CNN works better in different scale ranges. A scale-wise CNN adaption is reasonable since objects in scene are at its own specific scale. Second, a fisher kernel is applied to aggregate a global representation at each scale and then to merge into a single vector by using a post-processing method called scale-wise normalization. The essence of Fisher Vector lies on the accumulation of the first and second order differences. Hence, the scale-wise normalization followed by average pooling would balance the influence of each scale since different amount of features are extracted. Third, the Fisher vector representation based on the deep convolutional features is followed by a linear Supported Vector Machine, which is a simple yet efficient way to classify the scene categories. Experimental results show that the scale-specific feature extraction and normalization with CNNs trained on object-centric and scene-centric datasets can boost the results from 74.03% up to 79.43% on MIT Indoor67 when only two scales are used (compared to results at single scale). The result is comparable to state-of-art performance which proves that the representation can be applied to other visual recognition tasks.

Keywords: deep convolutional features, Fisher Vector, multiple scales, scale-specific normalization

Procedia PDF Downloads 313

Authors: E. D. Paul, H. Otanwa, O. F. Paul, A. J. Salifu, J. E. Toryila, C. E. Gimba

Abstract:

The concentration levels of six heavy metals (Cd, Cr, Fe, Ni, Pb, and Zn) and two mineral elements (Ca and Mg) were determined in soil samples collected from the vicinity of two auto mechanic workshops in Sabon-Gari, Kaduna state, Nigeria, using Atomic Absorption Spectrometry (AAS), in order to compare the gradation of their concentrations with distance and depth of soil from the workshop sites. At site 1, concentrations of lead, chromium, iron, and zinc were generally found to be above the World Health Organization limits, while those of Nickel and Cadmium fell within the limits. Iron had the highest concentration with a range of 176.274 ppm to 489.127 ppm at depths of 5 cm to 15 cm and a distance range of 5 m to 15 m, while the concentration of cadmium was least with a range of 0.001 ppm to 0.008 ppm at similar depth and distance ranges. In addition, there was more of calcium (11.521 ppm to 121.709 ppm), in all the samples, than magnesium (11.293 ppm to 21.635 ppm). Similar results were obtained for site II. The concentrations of all the metals analyzed showed a downward gradient with an increase in depth and distance from both workshop sites except for iron and zinc at site 2. The immediate and remote implications of these findings on the biota are discussed.

Keywords: AAS, heavy metals, mechanic workshops, soil, variation

Procedia PDF Downloads 479

Authors: Saba Ebrahimi, Saeed Ahmadian, Hedie Ashrafi

Abstract:

Selecting appropriate patients for surgery is one of the main issues in thoracic surgery (TS). Both short-term and long-term risks and benefits of surgery must be considered in the patient selection criteria. There are some limitations in the existing datasets of TS patients because of missing values of attributes and imbalanced distribution of survival classes. In this study, a novel ensemble architecture of deep learning networks is proposed based on stacking different linear and non-linear layers to deal with imbalance datasets. The categorical and numerical features are split using different layers with ability to shrink the unnecessary features. Then, after extracting the insight from the raw features, a novel biased-kernel layer is applied to reinforce the gradient of the minority class and cause the network to be trained better comparing the current methods. Finally, the performance and advantages of our proposed model over the existing models are examined for predicting patient survival after thoracic surgery using a real-life clinical data for lung cancer patients.

Keywords: deep learning, ensemble models, imbalanced classification, lung cancer, TS patient selection

Procedia PDF Downloads 122

Authors: O. Ahmed Apampa, Yinusa, A. Jimoh

Abstract:

The aim of this study was to develop an appropriate model for predicting the compaction behavior of lateritic soils and corn cob ash (CCA) stabilized lateritic soils. This was done by first adopting an equation earlier developed for fine-grained soils and subsequent adaptation by others and extending it to modified lateritic soil through the introduction of alpha and beta parameters which are polynomial functions of the CCA binder input. The polynomial equations were determined with MATLAB R2011 curve fitting tool, while the alpha and beta parameters were determined by standard linear programming techniques using the Solver function of Microsoft Excel 2010. The model so developed was a good fit with a correlation coefficient R2 value of 0.86. The paper concludes that it is possible to determine the optimum moisture content and the maximum dry density of CCA stabilized soils from the compaction test of the unmodified soil, and recommends that this procedure is extended to other binder stabilized lateritic soils to facilitate quick decision making in roadworks.

Keywords: compaction, corn cob ash, lateritic soil, stabilization

Procedia PDF Downloads 513

Authors: O. U. Nwosu, C. O. Osuagwu, N. Nnawugwu, C. T. Amanze

Abstract:

Phytoremediation is a green and sustainable approach to decontaminate and restore contaminated sites while maintaining the biological activity and physical structure of soils. A pot experiment was conducted for a period of 70 days to evaluate the remediation potentials of Vetiveira zizanioides, Heianthus annuus L., Zea mays, and Glycine max in concurrent removal of anthracene and Pb in co-contaminated soil. Sandy loam soils were polluted with Pb chloride salt and anthracene at three different levels (50mg/kg of Pb, 100mg/kg of Pb, and 100mg/kg of Pb+100mg/kg of anthracene) and laid out in a completely randomized design with three replicates. Shoot dry matter weight was significantly reduced (p≤0.05) in comparison to control treatments by 33%, 32%, 40%, and 6.7% when exposed to 100mg kg⁻¹ of Pb, respectively in G.max, H.annuus, Z.mays, and vetiver. There was 42%, 41%, 48%, and 7.1% growth inhibition of shoot dry matter weight of G.max, H.annuus, Z.mays, and vetiver relative to control treatments when 100 mg Pb kg⁻¹ was mixed with 100 mgkg⁻¹ anthracene. Root and shoot metal concentration in G.max, H.annuus, Z.mays, and vetiver increased with increasing concentration of Pb. Translocation factor (TF < 1) obtained for G.max, Z.mays, and vetiver suggests that these plant species predominantly retain Pb in the root portion, while the TF value (TF≥1) obtained for H.annuus suggests that it predominantly retains Pb in the shoot portion. The extractable anthracene decreased significantly (p ≤ 0.05) in soil planted with G.max, H.annuus, Z.mays, and vetiver, as well as in pots without plants. This accounted for 53% to 71% of anthracene dissipation in planted soil and 40% dissipation in unplanted soil. This result suggested that the plant species used are a promising candidate for phytoremediation.

Keywords: phytoremediation, heavy metals, polyaromatic hydrocarbon, co-contaminated soil

Procedia PDF Downloads 102

Authors: I. Hosni, L. Bennaceur Farah, N. Saber, R Bennaceur

Abstract:

Soil moisture content is a key variable in many environmental sciences. Even though it represents a small proportion of the liquid freshwater on Earth, it modulates interactions between the land surface and the atmosphere, thereby influencing climate and weather. Accurate modeling of the above processes depends on the ability to provide a proper spatial characterization of soil moisture. The measurement of soil moisture content allows assessment of soil water resources in the field of hydrology and agronomy. The second parameter in interaction with the radar signal is the geometric structure of the soil. Most traditional electromagnetic models consider natural surfaces as single scale zero mean stationary Gaussian random processes. Roughness behavior is characterized by statistical parameters like the Root Mean Square (RMS) height and the correlation length. Then, the main problem is that the agreement between experimental measurements and theoretical values is usually poor due to the large variability of the correlation function, and as a consequence, backscattering models have often failed to predict correctly backscattering. In this study, surfaces are considered as band-limited fractal random processes corresponding to a superposition of a finite number of one-dimensional Gaussian process each one having a spatial scale. Multiscale roughness is characterized by two parameters, the first one is proportional to the RMS height, and the other one is related to the fractal dimension. Soil moisture is related to the complex dielectric constant. This multiscale description has been adapted to two-dimensional profiles using the bi-dimensional wavelet transform and the Mallat algorithm to describe more correctly natural surfaces. We characterize the soil surfaces and sub-surfaces by a three layers geo-electrical model. The upper layer is described by its dielectric constant, thickness, a multiscale bi-dimensional surface roughness model by using the wavelet transform and the Mallat algorithm, and volume scattering parameters. The lower layer is divided into three fictive layers separated by an assumed plane interface. These three layers were modeled by an effective medium characterized by an apparent effective dielectric constant taking into account the presence of air pockets in the soil. We have adopted the 2D multiscale three layers small perturbations model including, firstly air pockets in the soil sub-structure, and then a vegetable canopy in the soil surface structure, that is to simulate the radar backscattering. A sensitivity analysis of backscattering coefficient dependence on multiscale roughness and new soil moisture has been performed. Later, we proposed to change the dielectric constant of the multilayer medium because it takes into account the different moisture values of each layer in the soil. A sensitivity analysis of the backscattering coefficient, including the air pockets in the volume structure with respect to the multiscale roughness parameters and the apparent dielectric constant, was carried out. Finally, we proposed to study the behavior of the backscattering coefficient of the radar on a soil having a vegetable layer in its surface structure.

Keywords: multiscale, bidimensional, wavelets, backscattering, multilayer, SPM, air pockets

Procedia PDF Downloads 106

Authors: Somia Lakehal Ayat, Ibtissem Samai, Srara Lakehal Ayat, Chaima Dahmani

Abstract:

The aim of this work relates to the physicochemical diversity and the characterization of the different types of soils of the edough massif (North East of Algeria) and to the evaluation and characterization of the existing organic matter as well as to the evolution. and the dynamics of the latter, also on its influence on changes in the physical properties of soils. In order to know the soil properties of seraidi forest, we established a stratified sampling plan. The results obtained show that we are in the presence of a great diversity of soils, such as neutral to alkaline, whose adsorbent complex is sufficiently saturated. Also, the presence of limestone offers the soil a fairly significant buffering capacity. In our study region, the texture of the soils is varied between clayey and silty, where it offers medium porosity, there is a strong accumulation of organic matter, therefore soils rich in organic matter.The fractionation of the organic matter of the soils allowed to obtain a very high rate of humification. The soil characteristics of the edough massif (North East of Algeria) are controlled by the contribution of organic matter, which presents a dynamic and an important evolution and which varies with the climatic conditions and the nature and the type of plant formation, and these the latter have a capital and important role in the rate of mineralization of organic matter.

Keywords: organic matter, soil, foresty, diversity, mineralization

Procedia PDF Downloads 66