Loading...
Theses
Browse
107 results
Search Results
ThesisItem Open Access Kidney stone detection from ultrasound images using masking techniques(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-12) Chaudhary, Harshita; Pandey, Binay KumarHere we are using masking techniques for stone detection that are present in the kidney. As we know Masking techniques are conspicuous approaches in contrast enhancement. For this firstly, the image is converted into grey and after that contrast of the image is enhanced. The process of contrast enhancement is done with the help of Optimum Wavelet-Based Masking (OWBM) using the Enhanced Cuckoo Search Algorith (ECSA). Afterward image segmentation and image masking have been done to detect stone from the image. The cuckoo search algorithm is used for global optimization of contrast enhancement. With the help of the Cuckoo search algorithm approximation of the coefficient has been optimized. The objective of this project is to design and implement a method to detect the presence of stone from the ultrasound image of a kidney. Here we are making are system our more intelligent.ThesisItem Open Access Multi-label classification of news titles using bidirectional long short-term memory model(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-12) Goel, Yash Kumar; Samantaray, S.D.Multi-Label Text Classification can be used when there are two or more classes as well as the information to be classified may relate to neither of the classifications or all of them at the very same time. With the rapid advancement of devices and digital telecommunications, online news has become one of the most important attributes for people's daily lives, studies, and jobs. Online news, in comparison to other conventional media, is extensive, diversified in form, and could be updated in real-time. The lack of classification makes it hard for a person to interpret or obtain data relevant to particularly preferred classifications. Text classification, as one of the key technologies of information resource organization & management, could allow users to narrow the scope of feature extraction as well as make it more convenient as well as effective to filter via massive digital information to fulfil the needs. The technique of text classification, which in the classification stage is capable of classifying instantly against several classifications on unstructured text with natural language, is used. In the proposed work, WordNet and word sense database is used to improve the efficiency of the classifier. To handle a huge amount of data the classification deep learning approach i.e. Bidirectional Long/Short-Term Memory (Bi- LSTM) is proposed. As News Titles is a short text that could lead to ambiguity in classification class and the title of the news item could be linked to a number of different sources that could lead to ambiguity in classification class, the introduction of the phrase seeks to optimize the classification method. The challenge of news classification begins with web scraping to gather real-time news Titles from news websites, which are then instantly classified using different classification methodologies and introduce the Wordnet and WordSense database for multi-label news titles classification. The acquired accuracy of (Bi-LSTM) was 97.91 per cent, which exceeded the approximate accuracy of each individual plan. This technique could be very helpful for academicians who want to investigate headlines in order to support their instruction.ThesisItem Open Access IoT based surveillance system using DNN(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-12) Rawat, Rahul; Srivastava, Ratnesh PrasadLocalization, Visibility, Proximity, Detection, Recognition has always been a challenge for surveillance system. These challenges can be felt in the industries where surveillance systems are used like armed forces, technical-agriculture and other such fields. A way to get the ease of mind would be installing a security camera. Most of the smart system available are just for the surveillance of human intervention but there is a need for a system which can be used for animals as well because with the outburst of human population and symbiotic relationship with wild animals results in life loss and damage to agriculture. There are many electrical equipment’s available for home which can do the monitoring from a remote area all at a time. In this paper we are designing to overcome these above-mentioned challenges for human and animal-based surveillance system in real time application. The system setup is done on a Raspberry pi integrated with deep-learning models which performs the classification of objects on the frames, then the classified objects is given to a face detection model for further processing. The detected face is relayed to the back-end for feature mapping with the saved log files with containing features of familiar face IDs. Four models were tested for face detection out of which the DNN model performed the best giving an accuracy of 87.88%. The system is also able to send alerts to the admin if any threat is detected with the help of a communication module.ThesisItem Open Access Experimental evaluation of stabilising potential of brick dust in clay soils(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-12) Bisht, Avani; Ajit KumarMigration of a large number of people towards cities has resulted in urbanization and construction of big engineering structures. Consequently, for accommodation of huge crowd enough suitable land is needed for engineering activities which can be fulfilled only when the soil in that area is competent enough to withstand the load impacted on it. Generally, the soils are classified as clayey and sandy type. The latter formed by physical weathering barely poses any threat, whereas clay formed by chemical weathering, is considered as a problematic soil because of its unpredictable behavioural changes. Construction on clayey soil is one of the biggest challenging tasks for engineers. Numerous studies have been done in the past to bring out a sustainable solution to this issue. The solution they have come up with is soil stabilization which is augmenting the soil properties by adding different additives in it. Many researchers have tried to modify the geotechnical characteristics of clayey soil utilizing different additives namely, lime and cement. However, the cost of such materials in recent years has risen demanding for a better alternative and cost-effective material. Many researchers after doing their critical evaluation suggested brick dust as an alternative to other expensive additives. Hence from economic perspective, brick dust is a cheaper alternative to conventional materials. India is the second largest manufacturer of brick and tons of brick waste generated each year goes in unplanned way. This waste occupies arable land creating environmental concerns. The problems could be solved by using this material as a soil stabilizer. Brick dust is a by-product obtained from brick kilns. It has greater ability to reduce the swelling potential of clayey soil (Sikha and Kumar, 2017). Brick dust is used as a substitute for sand in concrete and mortar and has almost same functions as that of sand and adds some strength and hydraulicity (Rather et al., 2019). In previous studies, utilizing brick dust as an additive is mainly done to evaluate the compaction characteristics, index properties, consistency and very limited investigations have been carried out on shear strength, consolidation and permeability characteristics of clayey soil. In view of the above, the main objective of this study is to investigate the effect of brick dust (introducing in varying percentages) on consolidation, shear strength and permeability characteristics of clayey soil. In addition, laboratory experiments like consistency limit tests, compaction test, unconfined compression test and CBR tests are also performed with an objective to reach for better results and conclusions. For the accomplishment of objectives, a thorough investigation of utilizing brick dust as additive has been planned. Firstly, brick dust was collected from Brick Kiln, located at Rudrapur, Uttarakhand. Soil was procured from the campus of G.B. Pant University of Agriculture & Technology, Pantnagar (Uttarakhand). The overall experimental programme was conducted in two phases. In the first phase, laboratory examination of soil was carried out. In the second and last phase, laboratory experiments were performed on soil mixed with different percentages of brick dust (i.e., 0%, 10%, 20%, 30%, 40% and 50%). After analysing the results obtained from various laboratory experiments, it was concluded that addition of brick dust as additive is satisfactory when mixed with clayey soil. Plasticity index was observed to decrease from 14.37% to 8.45% with increment in percentage of brick dust from 0 to 40% and beyond this a non-plastic behaviour of soil was observed. Proctor results reflect that optimum moisture content decreases from 16.0% to 10.0% while maximum dry density increases from 16.28 kN/m3 to 18.93 kN/m3 on increasing the brick dust percentage from 0 to 50%. The values of un-soaked and soaked CBR increased on addition of brick dust. The stress at failure increased up to 30% addition of brick dust and afterwards it displays a decreasing trend. The compression parameters exhibit decreasing trend on addition of brick dust. The value of cohesion decreases whereas angle of internal friction increases. The permeability of samples was increased when brick dust was mixed up to 50%. These results were compared to the results of prior studies conducted by a number of researchers, which revealed that the observations were in reasonably good accord.ThesisItem Open Access Optimal detection of leakage in water supply pipe networks(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-12) Kothari, Manjul; Mahar, P.S.Water transportation is facilitated by pipeline network and is one of the safest modes of water transportation. In terms of lost revenue and service problem, leaks in pipe networks generate considerable challenges for utilities and water users. Leak in pipelines can be caused due to corrosion of pipe wall, poor fitting, abnormal pressure, etc. So, designing a leak detection technique is challenging. However, a physical measurement technique is widely used to detect leakage at the pipeline level. The length of distribution pipe network in large cities might be thousands of kilometers. As a result, physical measures can become time-consuming and costly. In this study, a nonlinear optimization model is developed for optimal detection of leak in water distribution system subjected to constraints like continuity equation, energy (head loss) equation and pressure-dependent leakage. The model applicability has been illustrated using examples of water distribution network. The simulation model has been solved in EPANET 2.2 and then, the nonlinear optimization model has been solved using LINGO 18.0 software. Leaks are assumed to be concentrated at the nodes. Different situations are analyzed for finding the leaks at different locations. The major conclusion drawn from the study shows that with the increase in number of observed heads, more precise leak location can be identified. The optimal detection of leak location are sensitive to variations in friction coefficient. This study will be helpful to water supply engineers and water managers.ThesisItem Open Access Stock price prediction using LSTM approach(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-09) Agarwal, Shweta; Singh, B.K.In today's economy, the stock market, often known as the equity market, has a significant impact. The rise or decline in the share price has a significant impact on the investor's profit. The proposed method used Long short-Term Memory (LSTM) Approach. Here I am considering multi-column LSTM model which takes more than one column to analyse and train the model and based on that it will predict the values for future days. More than one features helps the model to predict the values more accurately than providing the single feature. Here the dataset is taken from Yahoo Finance website which provides historical data to almost all of the companies listed in the stock market. The dataset is taken for a particular company PETRONET LNG from 2004 to 2018. Next 30 days values are being predicted based on that historical data. The values for 2019 is not being considered as this time was affected by corona virus and every sector of the industry was affected by this pandemic. So taking these values may provide wrong predictions as there was sudden fall and rise in the stock values during this time. I have also added 2 more features to the given historical data i.e. volatility and momentum. Volatility is basically used to capture fluctuation in the market. Momentum tells us what is the changes in the price as compare to past days. Result showa that adding these features helps model to predict more accurately.ThesisItem Open Access Predicting the Bhimtal lake water level fluctuations by using different machine learning techniques(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-11) Tripathi, Vaibhav; Kashyap, P.S.Lake water level forecasting at various time intervals using the records of past time series is an important issue in water resources planning, engineering, etc.. Variations in lake level are complex outcomes of many environmental factors, such as precipitations, direct and indirect runoffs. The future planning, management and prediction of water demand and usage should be preceded by long-term variation analysis for related parameters in order to enhance the process of developing new scenarios whether for surface-water or ground-water resources. Water level plays an important part in the community’s well-being and economic livelihoods. This study investigated the fluctuations in the water level of Bhimtal Lake in the Nainital district (India) by using different machine learning techniques. Different soft computing such as MLP based ANN, Support Vector Machine, Random Forest, Multilinear Regression and CatBoost were used to predict the daily stage. The following data required for the study spanned over 12 years (2009-2020). By using Gamma test, the best input combination of variables (rainfall and stage lagged by two days, rainfall and stage lagged by one day and present day rainfall) were determined. The performance of the calibrated models was assessed qualitatively by visual interpretation and quantitatively using statistical indicators such as coefficient of determination (R2), root mean square error (RMSE) and mean absolute error (MAE). The MLR and CatBoost models were found as the best models compared to ANN, SVR and RF models for prediction of daily stage of the study area.ThesisItem Open Access Mixed Mode I/II crack growth in dissimilar AA2024-T6 and AA7075-T6 friction stir welded joints(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-12) Tiwari, Saurabh; Gope, P.C.Aluminium alloy 7075 and 2024 are some of the widely used materials in the aerospace industry. In the present investigation, the effect of SiC nanoparticles on fatigue crack growth and mechanical properties were tested on friction stir welded specimens and compared with as-welded specimens. Fatigue test is performed on different loading angles i.e., 0°, 30°, 45°, 60° and 90° with Pmax= 2 kN, Pmin= 0.2 kN and frequency of 5 Hz with R=0.1 and cracks length is measured using a digital camera fitted on travelling tripod and then compared them with numerically calculated values. All specimens were solution annealed at 460 °C for 2 h and then the ageing heat treatment procedure was done at 170 °C for 16 hours. The experimental data obtained was used to plot stress-strain curve, a-N and da/dN vs. Keq curve and various fatigue crack growth parameters such as Paris crack growth constants, crack opening stress intensity factor, etc. were determined. The fatigue life was found to be 37000, 36500, 33000, 30000 and 28000 cycles for friction stir welded CTS specimen at 90°, 60°, 45°, 30° and 0° respectively in as-weld condition. It was observed that there is a significant variation in the variances of crack length increment after the incorporation of SiC nanoparticles. The fatigue life after SiC incorporation was found to be 48200, 39000, 36000, 34500 and 30000 cycles at 90°, 60°, 45°, 30° and 0° respectively. The values of material constants C and m are calculated through a single Paris Curve fit. Scanning electron microscope (SEM) study of fracture surface at different locations was carried out to investigate the different modes of fracture that occurred at different loading conditions and morphology test is used to investigate the grain size distribution at different welding zones.ThesisItem Open Access Identification of groundwater potential recharge zones in Hiran River Watershed using RS and GIS(G.B. Pant University of Agriculture and Technology, Pantnagar - 263145 (Uttarakhand), 2021-10) Choudhary, Preeti; Harish ChandraGroundwater forms a substantial part of the total water resources, which provides water for irrigation, municipal supply and industrial use. Unlike mineral resources, groundwater gets replenished by the natural process of recharge. Groundwater utilization should therefore, be properly planned to achieve a balance between replenishment and extraction in order to maintain a perennial supply. In the recent years, geographic information system based studies have gained much prominence in groundwater exploration because it is rapid and will provide first-hand information on the resource for further developments. Therefore, the present study has been undertaken with objectives to analyze morphometric parameters and to identify groundwater potential recharge zones in the Hiran river watershed, Madhya Pradesh, India. The study adopts the Analytical Hierarchy Process (AHP) and Multi-Influencing factors (MIF) approaches with a combination of RS and GIS. The study addressed linear, areal and basin morphometric aspects of the watershed. The morphometric analysis of the watershed revealed that the order of Hiran river watershed was found to be eight.Out of total 7669 drainage network, 5948 were found to be of first order, 1333 were second order, 294 were third order, 65 were fourth order, 19 were fifth order, 7 were sixth order, 2 were seventh order, and 1 was eight order stream. The total length of stream was found to be longer for first order stream and found decreasing with increasing stream order. The mean bifurcation ratio obtained was 3.59 which is within standard range. Hence it was revealed that the watershed having strong structural control in drainage network. Drainage density estimated as 1.6 km/km2, which revealed that watershed, is underlain by highly permeable resistant material with vegetative cover and low relief. Areal aspects of the morphometric analysis showed that the watershed was elongated with gentle ground slope. Total eight thematic layers such as Geology, Geomorphology, Land use/ Land cover, Rainfall distribution, Soil, Slope, Lineament density and Drainage densitywere extracted from conventional and remote sensing data sources. The Spatial Analyst Tool in ArcGIS 10.4.1 was used to integrate all eight thematic layers. The weighted overlay method was used, to identify the Groundwater Potential Recharge Zones (GWPRZs) in the study area. During weighted overlay analysis, each of the thematic layers were assigned a score and weight based on the relative contribution of each of these maps to groundwater potential, which was estimated using the AHP and MIF approaches. The results from the AHP wereclassified into five categories viz., Very good (7.02%), good (36.42%), moderate (31.11%), poor (32.31%) and very poor (0.18%). Similarly in case of MIF, Very good (0.29%), good (34.130%), moderate (59.530%) and poor (6.330%). The results were validated with observed groundwater level data of observation well using ROC curve. The area under the curve for AHP and MIF was found to be 71% and 67.9%, respectively. Based on the ROC curve analysis, it was concluded that AHP approach produced relatively more accurate results than the MIF. It was also revealed from the study that accuracy of these approaches is ultimately depends on the criteria of classification, mean rating score and weight assigned to the thematic layers.
