-22b9b34417bc6092a744279 20220216024410558 beie:beie director@blueeyesintelligence.org WEB-FORM International Journal of Innovative Technology and Exploring Engineering IJITEE 22783075 10.35940/ijitee https://www.ijitee.org/ 02 28 2021 10 4 Department of Computer, The National University of Water and Environmental Engineering, Revine, Ukraine. Mahmoud Smaida Serhii Yaroshchak Applied Mathematics, The National University of Water and Environmental Engineering, Revine, Ukraine. Ahmed Y. Ben Sasi Department of Computer, The College of Industrial Technology, Misurata, Libya. One of the most important hyper-parameters for model training and generalization is the learning rate. Recently, many research studies have shown that optimizing the learning rate schedule is very useful for training deep neural networks to get accurate and efficient results. In this paper, different learning rate schedules using some comprehensive optimization techniques have been compared in order to measure the accuracy of a convolutional neural network CNN model to classify four ophthalmic conditions. In this work, a deep learning CNN based on Keras and TensorFlow has been deployed using Python on a database that contains 1692 images, which consists of four types of ophthalmic cases: Glaucoma, Myopia, Diabetic retinopathy, and Normal eyes. The CNN model has been trained on Google Colab. GPU with different learning rate schedules and adaptive learning algorithms. Constant learning rate, time-based decay, step-based decay, exponential decay, and adaptive learning rate optimization techniques for deep learning have been addressed. Adam adaptive learning rate method. has outperformed the other optimization techniques and achieved the best model accuracy of 92.58% for training set and 80.49% for validation datasets, respectively. 02 28 2021 211 216 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.B8259.0210421 https://www.ijitee.org/portfolio-item/B82591210220/ Department of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa. Johannes P. Angula Freddie Inambao Department of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa. In this study, a 3D Computational Fluid Dynamics (CFD) model was developed to simulate the drying process of maize ears cobs in a mixed-mode solar grain dryer. The dryer system is aimed to operate under forced convection and is integrated with a preheater to heat air prior to entering the solar collector. The 3D model was developed with great accuracy using SolidWorks software and the CFD simulation was carried out using ANSYS Fluent software. The study was aimed at analyzing and predicting temperature and airflow distribution in the mixed-mode solar dryer system. The CFD simulation was conducted at different airflow velocities varying from 0.5 m/s to 2 m/s for different temperature values of the preheater. Results from the simulation of the solar collector were satisfactory, indicating a minimum and maximum temperature of 59.7 ℃ and 70.5 ℃ at minimum and maximum drying conditions, respectively. The variation of temperature inside the drying chamber was predicted with an average maximum of 64.1 ℃ at the inlets. Results of airflow distribution in the solar collector and drying chamber indicated high turbulence and flow recirculation. This is a desirable flow combination that promotes good moisture evaporation from the maize ears during the drying process. This study proves that the use of computer software can allow one to clearly gain an understanding of the development, heat and mass transfer process, and performance of dryers used in the food drying industry. This approach can promote improvements in existing drying processes and increase food productivity. 02 28 2021 33 40 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.B8282.0210421 https://www.ijitee.org/portfolio-item/B82821210220/ Department of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa. Johannes P. Angula Freddie Inambao Department of Mechanical Engineering, University of KwaZulu-Natal, Durban, South Africa. In this study the performance of a forced convection mixed-mode solar grain dryer integrated with a preheater was evaluated. The type of grains used in the experiment were 72 freshly harvested maize cobs with a total mass of 17 kg. The experiment was conducted at various airflow speeds and preheater temperatures ranging from 0.5 m/s to 2 m/s and 30 ℃ to 40 ℃, respectively. The aim of the study was to improve the performance of an existing indirect solar dryer which was converted to a mixed-mode solar dryer. The initial thermal efficiency of the indirect solar dryer before modification was 36 %. The results from the experiment indicated a maximum thermal efficiency of 58.8 % with a corresponding drying rate of 0.0438 kg/hr. The minimum thermal efficiency for the mixed-mode solar grain dryer system was 47.7 %, with a corresponding drying rate of 0.0356 kg/hr. The fastest drying time of maize cobs was achieved in 4 hours and 34 minutes from an initial moisture content of 24.7 % wb to 12.5 % wb. The findings show a significant improvement in the dryer system’s performance. This is a clear indication that operating a solar dryer system in mixed-mode operation with forced convection and the assistance of a preheater or backup heater can significantly improve drying processes and increase food preservation. 02 28 2021 41 51 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.B8283.0210421 https://www.ijitee.org/portfolio-item/B82831210220/ Research Scholar, Dibrugarh University, Kalakata, India. Hemanta Konch North-East is a hub of many ethnic languages. This region constitutes with eight major districts; like-Assam, Arunachal Pradesh, Nagaland, Manipur, Mizoram, Tripura, Meghalaya and Sikkim. Tutsa is a minor tribe of Arunachal Pradesh. The Tutsa was migrated from the place ‘Rangkhan Sanchik’ of the South-East Asia through ‘Hakmen-Haksan’ way to Arunachal Pradesh. The Tutsa community is mainly inhabited in Tirap district and southern part of Changlang district and a few people are co-exists in Tinsukia district of Assam. The Tutsa language belongs to the Naga group of Sino-Tibetan language family. According to the Report of UNESCO, the Tutsa language is in endangered level and it included in the EGIDS Level 6B. The language has no written literature; songs, folk tales, stories are found in a colloquial form. They use Roman Script. Due to the influence of other languages it causes lack of sincerity for the use of their languages in a united form. Now-a-days the new generation is attracted for using English, Hindi and Assamese language. No study is found till now in a scientific way about the language. So, in this prospect the topic Nominal Inflection of the Tutsa Language has been selected for study. It will help to preserve the language and also help in making of dictionary, Grammar and language guide book. 02 28 2021 138 140 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8428.0210421 https://www.ijitee.org/portfolio-item/D84280210421/ Assistant professor, Department of Computer Science Ganpat University, MCA, Dr. Bhavesh Patel Machine learning techniques are used by many organizations to analyze the data and finding some meaningful hidden pattern from the data, this process is useful by an organization to take the decision making process. Various organizations used like marketing, health care, software organization and education institute etc used it in decision making. We have used machine learning techniques to enhance the performance of students. It will be ultimately used by educational institute to improve the status of educational institute. This research paper includes Naïve Bayes (NB), Logistic Regression (LR), Artificial Neural Network(ANN) and Decision Tree machine learning techniques. Performance of these models have been compared using accuracy measures parameters and ROC index. This research paper has used various parameters like academic performance and demographic information to build the model. In addition to judge the performance also used some additional parameters to measure the performance like F-measure, precision, error rate and recall. The dataset is collected using survey methodology to build the model. As a conclusion found that the Artificial Neural Network model get the best performance among all the models. 02 28 2021 1 4 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8429.0210421 https://www.ijitee.org/portfolio-item/D84290210421/ Computer Science from George Washington University, USA. Mohammad Abbadi Afaf Tareef Department of Computer Science from Mutah University, Jordan. Afnan Sarayreh Department of Computer Science from Mutah University, Jordan. The human hand has been considered a promising component for biometric-based identification and authentication systems for many decades. In this paper, hand side recognition framework is proposed based on deep learning and biometric authentication using the hashing method. The proposed approach performs in three phases: (a) hand image segmentation and enhancement by morphological filtering, automatic thresholding, and active contour deformation, (b) hand side recognition based on deep Convolutional Neural Networks (CNN), and (c) biometric authentication based on the hashing method. The proposed framework is evaluated using a very large hand dataset, which consists of 11076 hand images, including left/ right and dorsal/ palm hand images for 190 persons. Finally, the experimental results show the efficiency of the proposed framework in both dorsal-palm and left-right recognition with an average accuracy of 96.24 and 98.26, respectively, using a completely automated computer program. 02 28 2021 5 13 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8430.0210421 https://www.ijitee.org/portfolio-item/D84300210421/ Department of Electronics and Communications Eng, Kwangwoon Univ, Seoul, Korea. Hyung-Hwa Ko GilHee Choi Department of Electronics and Communications Eng, Kwangwoon Univ., Seoul, Korea. KyoungHak Lee IACF, Kwangwoon Univ, Seoul, Korea. Recently, many studies on the image completion methods make us erase obstacles and fill the hole realistically but putting a new object in its place cannot be solved with the existing Image Completion. To solve this problem, this paper proposes Image Completion which filled a new object that is created through sketch image. The proposed network use pix2pix image translation model for generating object image from sketch image. The image completion network used gated convolution to reduce the weight of meaningless pixels in the convolution process. And WGAN-GP loss is used to reduce the mode dropping. In addition, by adding a contextual attention layer in the middle of the network, image completion is performed by referring to the feature value at a distant pixel. To train the models, Places2 dataset was used as background training data for image completion and Standard Dog dataset was used as training data for pix2pix. As a result of the experiment, an image of dog is generated well by sketch image and use this image as an input of the image completion network, it can generate the realistic image as a result. 02 28 2021 14 18 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8431.0210421 https://www.ijitee.org/portfolio-item/D84310210421/ Department of Electronics and Communication, University Institute of Technology, Rajiv Gandhi Proudyogiki Vishwavidyalaya, Bhopal, (M.P), India. Megha Deshmukh Dr. Vineeta Saxena Nigam Department of Electronics and Communication, University Institute of Technology, Rajiv Gandhi Proudyogiki Vishwavidyalaya, Bhopal, (M.P), India. Diabetic Retinopathy is a diabetic disease that directly affects the vision that causes damaged blood vessels at the back end of the eyes. It a complicated disease that cannot be recognized from normal eyes; a fundus imaging can reflect the impairments over the retina that causes partial or complete blindness that cannot be cured. It is mandatory for a routine examination that may lead to prevent from complete blindness because it can be prevented from current damaged blood vessels but it cannot be revert or treated. In the field of image processing; various diseases can be diagnosed automatically that saves humans life along with easiness for medical professionals. If a person pertains diabetes for a long time may have highest possibility for diabetic retinopathy. Here, the system has been proposed that can diagnose this disease with high level of accuracy with minimal false alarm rate. System uses Prewitt Edge Detection and Color Mapping techniques for recognizing diabetic retinopathy symptoms or damaged blood vessels from fundus imaging. Prewitt is highly sensitive for extracting impairments along with blood vessels and system is able to mask the unwanted area by using color correction tool. 02 28 2021 19 23 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8433.0210421 https://www.ijitee.org/portfolio-item/D84330210421/ Department of Electronics and Communication at Sri Jayachamarajendra College of Engineering, JSS Science and Technology University, Mysuru, Karnataka, India. Dr. U B Mahadevaswamy Meghana N Department of Electronics and Communication at Sri Jayachamarajendra College of Engineering, JSS Science and Technology University, Mysuru, Karnataka, India. Water is a very valuable and stimulating force for irrigation. The Optimum usage of water is an hourly requirement. Efficient irrigation helps to conserve water, increase plant yields, reduce fertilizer reliance, and improve the quality of crops. Various techniques are available to measure soil moisture content, both laboratory and field, including remote sensing, but the fastest and better one is with the use of soil moisture sensor electronic devices. The range of soil moisture sensors has its own benefits and drawbacks. The goal of this work is to design and develop a module for the measurement of soil moisture and temperature levels, as well as ambient temperature and humidity by using frequency concepts. The sensor is made of a corrosion resistant element and it is rugged, battery operated, low power and long range sensor using IoT. A “GND BLE Mobile Application (gSense100)” has been developed, which includes everything related to the BLE technology and soil moisture sensor, where the app uses BLE technology to transmit all sensor values to the consumer. 02 28 2021 24 28 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8438.021042 https://www.ijitee.org/portfolio-item/D84380210421/ Hydraulic Construction Institute, Hanoi, Vietnam. Vũ Ngoc Binh Do The Quynh Hydraulic Construction Institute, Hanoi, Vietnam. Peat soil is formed from river-bog sediments (abQ232) are largely distributed in Mekong river Delta provinces-Vietnam such as Kien Giang, Hau Giang, Bạc Liêu and Ca Mau. The resuls of research to improve them with many kinds of cements showed that the unconfined compressive strength of soil samples reinforced by cements had increased within 28 days, from 28 to 56 days this strength was reduced. Research for improving the soil above by cement and sodium silicate to increase the strength and stability with curing time had been conducted. The results showed that the concent of 0.5% of sodium silicate in comparison with cement mass was added to soil samples, their strength increased significantly when compared to soil samples without sodium silicate and greater than that of the soil samples reinforced by contents of 1%, 1.5% and 2% of sodium silicate in comparison with cement mass and also the concent of 0.5% of sodium silicate in comparison with cement mass added to soil sample has solved the problem of reducing soil sample strength with curing time. 02 28 2021 52 56 CC BY-NC-ND 4.0 10.35940/BEIESP.CrossMarkPolicy www.ijitee.org true 10.35940/ijitee.D8442.0210421 https://www.ijitee.org/portfolio-item/D84420210421/