Volume 2

Abstract: Face detection is an important problem in computer vision
research and applications are getting trending due to the
advancement in the file of machine learning and computer vision.
There are several algorithms and models for face recognition.
However, face detection is the first step in all implementations.
This research proposed a face detection method based on an
enhanced Multi-Task Convolution Neural Network (MTCNN) and
improves the network of MTCNN, creates a neural network model
based on MTCNN using Python, and cascades to increase the
accuracy of face location in difficult scenarios. In this research
paper, we evaluated the performance of three famous face detector
models on CPU-based machines. Experiments show that HAAR
Cascade is the fastest on CPU-based machines but in the case of
accuracy, MTCNN is better. MTCNN and DLIB based detectors
are designed for GPU-based machines.

Abstract: This research is based on the practical facts related to the human resource
department of any organization for the recruitment of personnel. As it is a
challenging and crucial aspect for any organization to select the right talent
for the right place. This paper helps in expertise finding in the different
fields of Computer Science. Employers receive a bunch of resumes upon
job openings. And the candidates are also interested are in sifting the best
among the applicants. Screening the best candidate among the pool of
resumes is a laborious task. This paper proposes an informational retrievalbased resume ranking scheme for screening and ranking the candidate’s
resumes. The primary purpose of this research study is to exploit the class
NLP techniques to perform the information retrieval task for resume
ranking based on job description similarity. In this proposed methodology,
we compared document vectors with word embedding. Experiments show
that word embedding method is more effective than the document vector.

Abstract: The best quality production and the safety of human beings is the result of moderate and required conditions which are one of the most important industrial parameters. PLC is effective, fast responsive, and has a wide bandwidth. In this paper, we propose a technique of controlling different parameters of the industry through PLC based system by connecting the temperature, pressure, and level sensors with 2A4RTD. In any case of emergency, these sensors will detect the problem and making it easy to control and monitor the situation. 2A4RTD is a combo module that is used for taking analog and resistive inputs from the source, delivering them to the PLC 20-MAR board. PLC accepts the two analog inputs with the help of the 2A4RTD module, which has reduced the additional cost. After getting the information of the input signals, we can get the ranges of temperature, pressure, and oil levels. With the help of ranges, we can estimate the solutions to control their worst levels. Some components like the cooling fan, buzzer, and air conditioner will be connected with the relay or simply connected to the PLC board, to get the issues under control. Data will be collected and monitored through HMI.

Abstract: Structural steel has become one of the most widespread building
materials from last hundred year. It is broadly used in critical infrastructure such as buildings, cylinders, and marine pipelines, etc. In this paper, natural convection heat transfer is used for cooling steel cylinders by using ANSYS Transient Thermal. A cylinder is designed to have a radius of 10 mm and a depth of 40 mm. The initial temperature of the cylinder is 120°C and the ambient temperature is 22°C. When we apply a convection coefficient of 1×10-5 W/mm 2°C, the steel cylinder starts to cool down with the passage of time. The temperature decreases from 120°C to 22.357°C in the 10000s. These results show that the heat transfer (cooling rate) is rapid at the start of simulation but gradually decreases with the increase in time. When we apply the final convection coefficient of 2.2×10-5 W/mm 2°C, the temperature decreases from 120°C to 22.755°C in 3800s. These simulations show that the cooling rate has directly related to the Convection heat coefficient.

Abstract: Hypertension is a common disease that can be caused by a number of factors. A variety of methods, including Doppler Ultrasonography, can be used to detect hypertension, which is aided by High Blood Pressure. In this paper, Fuzzy value of speed is 12 cm/s, and corresponding Mamdani’s value is 11.8 cm/s with 0.2 difference. Fuzzy value of flow rate is 14.5 mL/min, and corresponding Mamdani’s value is 14.5 mL/min with 0.0 difference. Our Fuzzy Mathematical analysis provides values of Blood Speed with 1% margin of error as well as values of Blood Flow within 0% margin of error. The fuzzy logic approach is ideal for combining human expert knowledge with pre-existing numerical data. Fuzzy logic systems use human reasoning and decision-making to generate reliable output even when precise information is unavailable. In the case of biological systems, mathematical modelling is extremely complex, but fuzzy linguistic rules have been shown to aid in the resolution of complexity.

Abstract:

Over the last few decades, technological advancements and the tendency toward miniaturization have resulted in the development of a new subject known as micro-electro-mechanical systems (MEMS). MEMS have many applications but our focus in this research was on MEMS-based thermopneumatic micro pump. It consists of a chamber, loaded with vapors or solution, a diaphragm, and a temperature control system. It has many advantages like low power consumption while maintaining high performance and accurate dose control. MATLAB fuzzy logic controller (FLC) is used for the parametric estimation of the MEMS-based thermopneumatic micropump. The effect of voltage and frequency applied on the micropump were studied on the back pressure and flow rate. The increase in voltages and frequency were decreased the back pressure and fluid flow rate. Similarly, with a decrease in voltages and frequency were increased the back pressure and fluid flow rate. The difference between the MAMDANI calculated value and the simulated value from the rule viewer is 1.61% for both the output of back pressure and flow rate.