Volume 9 | Issue 10 | October-2021

 Abstract

In this paper a new electric braking system based on stopping time and energy regeneration is proposed for a brushless DC (BLDC) motor driven electric vehicle (EV). The new braking system is developed by integrating various regenerative methods and plugging. Aside from the present performance dimensions such as braking torque, boost ratio and maximum conversion ratio; stopping time and energy recovery for numerous methods are analysed in diverse running conditions. It is observed that the stopping time is less for plugging and increasing in the order two, three and single switch method. Besides, energy can be recovered more in single and three switch method. Based on these performances, a new braking strategy is put forward which combine all the regenerative braking methods including plugging and switch among themselves based on the depression of brake pedal. The effectiveness of the proposed method is shown in simulation results.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Electric Vehicle, Regenerative Braking , BLDC Motor

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Gaussian belief propagation algorithm (GaBP) is one of the most important distributed algorithms in signal processing and statistical learning involving Markov networks. It is known that the algorithm correctly computes marginal density functions from a high dimensional joint density function over a Markov network in a finite number of iterations when the underlying Gaussian graph is acyclic. Analysis of convergence rate is an important factor. GaBP algorithm is shown to converge faster than classical iterative methods like Jacobi method,successive over relaxation.It is more recently known that walk summability approach extends for better convergence result.Convergence rate analysis of GaBP for markov network using walk summability approach and theoretical study of convergence rate analysis using laplacian operator are considered in this work.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Gaussian belief propagation,Markov network,Convergence rate,Walk summability,Laplacian operator

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

In speech enhancement, one of the most important tasks is the removal or reduction of background noise from a noisy signal. This paper discusses about various speech enhancement techniques and a general framework proposed to estimate short-time spectral amplitudes (STSA) of speech signals in noise by joint speech detection and estimation to remove or reduce background noise, without increasing signal distortion. By combining parametric detection and estimation theories, the main idea is to take into consideration speech presence and absence in each time-frequency bin to improve the performance of Bayesian estimators. The observed signal is frequently segmented, windowed and transformed into the time-frequency domain. Then, the clean signal coeffcients are usually retrieved by applying an enhancement algorithm to the noisy observations in this domain.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

speech enhancement, parametric method, joint detection and estimation, Bayesian estimator, minimum mean square error (MMSE).

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The need for home security alarm systems nowadays is a serious demand. As the number of crimes are increasing every day, there has to be something that will keep us safe. We are all aware of the high end security systems present in the market but they are not easily available to everyone. We therefore intend to provide a solution by constructing a cost efficient electronic system that has the capability of sensing the motion of the intruders and setting off the alarm. The basic idea behind this project is that all the bodies generate some heat energy in the form of infrared which is invisible to human eyes. But, it can be detected by electronic motion sensor. The project involves the use of Arduino, motion sensor, buzzer, LCD display and a simple program. The sensor detect any motion in its permissible range and triggers the alarm. It will also send the signal to Arduino which processes the signal and set off the alarm along with detection message on display. With this system we can easily set up a security alarm in our home for unwanted intruders.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

HOME SECURITY ALARM SYSTEM USING ARDUINO

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Text recognition is a major area of experimentation under image processing domain. It is a process by which the system locates any kind of text is present and extract them from an image. The extracted text must be converted to human readable form after several processing and if required is classified them into meaningful classes based on the content present. The platform used here in discussion is MATLAB. This paper provides a detailed study on the evolution of text detection in natural images. It analyzes and also discusses the different methods to overcome existing challenges in text detection. This paper presents the different types of datasets which are used to identify text from natural images and comparative study of different text detection methods. The paper is concluded by a method to recognize and classify the multi-oriented text present in an image based on MSER and CNN.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

MSER (Maximally Stable Extremal Regions), CNN (Convolution Neural Network)

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The aim of this paper is to explore an efficient image segmentation algorithm for medical images to reduce the physicians interpretation of Computer Tomography (CT) scan images. Medical imaging techniques which are modern generate large images that are extremely difficult to analyze manually. In this paper, image preprocessing is done using adaptive median filter and contrast limited adaptive histogram equalization .For Segmentation five important methods are used and they are k-means clustering, K-median clustering ,Particle swarm optimization(PSO), inertia-weighted particle swarm optimization(IWPSO), and guaranteed convergence particle swarm optimization(GCPSO).We will verify it using matlab and will get that GCPSO will give more accuracy of about 95.89%.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Preprocessing of Lung Cancer Detection Using Image Segmentation by means of Evolutionary Algorithm

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Support Vector Machines (SVM), Melanoma, Deep learning, Dermoscopy, Benign and Malignant.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

FTSI System

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

One of the prime reasons for vehicular accidents is due to undetected potholes and road humps.. Another reason for a huge number of accidents is drunk driving. Even a small amount of alcohol in blood can lead to mid-body imbalance. Also, even after an accident occurs timely medical is not given. Due to these reasons numerous lives are lost. Through this project we will be trying to provide a solution to these problems so that there is better safety for people inside and outside the vehicle. We are trying to achieve this by using transmitter-receiver modules. The transmitter section will be placed near the obstacles on the road which will provide the alert signal. Receiver section on the car will receive this and alert the driver. Drunk driving shall be prevented by an engine-lock system. Also, new potholes can be identified using the ultrasonic sensor which will be placed on the car.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Accident Prevention, Alcohol Sensing, Engine lock, Ultrasonic Sensor, Pothole.

  License

Creative Commons Attribution 4.0 and The Open Definition

 Abstract

The major problems that are faced by the cultivators are the diseases effecting on their crops. There is no such an authenticated and globalized technique for the detection and diagnosis of plant diseases. In this paper a Mobile diagnostic system for Chili plant was developed as an example to solve this problem. Here we take Chili plant as an example. Chili is one of the widely used as well as cultivated crop not only in our home garden, but also through out our country. India is one of the major producer of Chili crop among the whole world. With the rapid development of mobile service computing, it have an increasingly important role in our daily lives. Disease detection can be more user friendly when we utilize the mobile service computing technique for this purpose. So here we build an image dataset of 4 kinds of Chili diseases along with healthy leaves and healthy fruits. They are collected from the home garden. Then realize a Mobile diagnostic system for Chili diseases by constructing a Deep convolutional neural network (D CNN). The system realized using an Android app in our Mobile device, with which users can upload images and receive diagnostic results. Here the experimental results shows that the detection accuracy of the chili diseases exceeds 90 % and we can detect and receive diagnostic results in a few minutes using this system.

Licence: creative commons attribution 4.0

  License

Creative Commons Attribution 4.0 and The Open Definition

 Keywords

Deep Convolutional Neural Network, Treatment advices, Mobile service computing, Android app

  License

Creative Commons Attribution 4.0 and The Open Definition