Simulation of Colour Image Processing Techniques on VHDL
Abuzar. A. Shaikh
This projects deals with color image processing technique. The different parameters of image are controlled by various mathematical functions. The processing using verilog has an advantage of speed and reconfigurability over system processing, specifically required for image filtering operation. The aim is to process image by using Threshold operation, Brightness operation and Invert operation. Fast operations and efficient simulation will be considered while designing the algorithm. The simulation is carried out by establishing a link between MATLAB and Hardware Descriptive Language (HDL).
Image classification and comparision of different Convolutional neural network srtuctures
Pooja.V.Magdum
Deep learning is a subfield of machine learning, which aims to learn a hierarchy of features from input data. Deep learning technologies are becoming the major approaches for natural signal and information processing, like image classification, speech recognition. Deep learning is a technology inspired by the functioning of human brain. Convolutional neural networks (CNN) become very popular for image classification in deep learning. In this paper, discussed about a deep learning convolutional network structures based on keras. Four different structures of CNN are compared on CPU system, with four different combinations of classifiers and activation functions.
Diabetes Detection – An Application of Machine Learning in Healthcare Sector
Girish Suhas Wangikar
Girish Suhas WangikarDiabetes is a disease which is rooted at global scale and continues to have adverse effects on people all over the world. India ranks the second in the list of countries with the most diabetic patients. India has estimated 77 million diabetes patients and behind every diabetes mellitus patient around the world one is from India. With rise of fast-food industry and stressful life style, people are at higher risk of being diagnosed with diabetes mellitus. Diabetes is not a disease but a syndrome, meaning it acts as host to other diseases ranging from heart diseases, vision impairment to sudden weight loss. Application of Machine Learning in Healthcare Industry poses as a solution to this problem. Machine learning can be used for the diagnosis of the diabetes mellitus as well as management of it in people who are already affected. Various Machine Learning algorithms can be of aid in this problem. Support Vector Machine (SVM) which is Classification algorithm is used in this project for detection of diabetes. Support Vector Machine is used as it creates a separation boundary so that feature space is divided into classes so that next point will be classified in either of the class being Diabetic and Non-Diabetic. For the chosen dataset of diabetes patients, SVM delivers an accuracy of 90% for training data and 89% for testing data.Keywords: Diabetes Mellitus, Machine Learning, Support Vector Machine.
Cleaning in Place (CIP) has been around for approximately 50 years, and is commonly used in hygiene critical industries such as food, beverage and pharmaceutical to clean a wide range of plant.CIP refers to use of mix of chemicals, vessels or pipe work without dismantling plant. The process can be one shot, where everything goes to drain, or recovery which recycles most of the liquid overall, CIP can be a very efficient way of cleaning. The CIP automation system was used in the dairy industry for cleaning process of milk tanker. This operation is done step by step, i.e first hot water process, a chemical of caustic soda process and last cold water process. The use of water in quite ample in quantity, which is not reused at all and send directly to the drainage or farmlands. For this purpose, we confirmed a new automation system to save energy and resources during cleaning work. In this system milk tanker is filled with pasteurised milk and this milk is transported from one region to another, during this period inner surface of tanker becomes oily, this oily part is removing by different processes such as hot water process, caustic soda process and cold water process. If inner surface of milk tanker is not properly clean then there is possibility of fungal growth. So avoid this problem we have develop a system.
Enhancement of active power flow capacity of a transmission line using STATCOM model.
Today the transmission line has many important considerations such as voltage drop, line losses and efficiency of transmission which is needed to be taken into consideration. Voltage drop of the transmission line is totally depending on resistance and inductance of transmission line. The overall resistance of transmission line causes the power loss. In the proposed work STATCOM has been identified as a complete compensation technique for solving above problems of the power system. STATCOM is a device which is used for the controlling purpose of VARs of the transmission line and it is a part of FACTS (Flexible AC Transmission System). STATCOM is one of the key FACTS controllers which can be either a voltage source converter or current source converter. AC output voltage is controlled such that it is right for the required reactive current flow. For any AC bus voltage, DC capacitor voltage is automatically adjusted as required to serve as a voltage source for the converter. STATCOM can be designed to act as an active filter to absorb system harmonics.
The 28 Bay Smart Pill box is connected to the new AI application Alexa device. It is voice recognition device. If patients is connected the Alexa to the smart pill box it will give all reminder to the patients according to the schedule.
2.WiFi communication:-The total system which is smart pill box is connected to the wifi module. Through this wifi module all the communication done with android application which are connected to the smart pill box.
This 28 Bay Smart Pill Box is connected to the web application. All the important data is maintained on this web. With the help of this web application if we forgot some information about the that particular patients then we can login for this patient and can take all the information about that patient. So likewise this web application is helps to the 28 Bay Smart Pill Box. So 28 Bay Smart Pill Box is having functions like communications between devices, imaging of the tablets, sensing the temperature and health between the computer and android application with human, according to the patients health treatment without any disturbing of lifestyle of the patients.
The 28 Bay Smart Pill Box is connected to the devices for communication with the patients with devices like mobile phones, any answering devices and through internet connections like web application. The all listed medicines which is gave by the doctors are kept into that date wise different slots for daily consumption of tablets. When that particular time and date is occurs at that time the machine will give the reminder through the buzz and led blinking before five minutes. The system is also give the reminder massage and calls to the patients on their resisted number. The device is very useful to avoid the double dosage of the tablets by maintain the record or log of that patients on internet web application.
The web which is used in the communication with web is pancare.panhealth.com,is a designed for the which is globally connected to overall branches of industries which include all healthcare departments. The web is maintain all the patients log and because of that patients can prevent from disease and early treatment with good quality results for patients health, in low cost.
Some companies extract edible oil from soybean. The byproduct after extracting oil is De-oiled cake (DOC). This DOC is used as poultry feed and also for making soya chunks. Moisture content in this DOC cake raises the issue of its storage. If moisture content in DOC exceeds 13.5% then it catches fungus in storage. If moisture content falls below 12 % then industry loses profit due to reduction in weight of DOC. Therefore, manual efforts are taken to maintain moisture level in DOC between 11.5% to 13%. Hence this project work aims to develop low cost indigenous moisture measurement system. To measure this moisture contents, industries mainly make use of loss on drying moisture meter method. In this method some sample is weighted, heated and weighted again. And finally, by comparing these two weights moisture content is calculated. Due to heating of the sample its properties are being affected. While using this method, the industry faced some problems like more time consumption and the test takes more than 30 minutes to complete the process when moisture content is high(>20%). Also, this system is not suitable for online application. It involves labor cost and safety becomes concern. Hence, there is a need to design a cheap system which measures moisture within low cost and continuous (online). Moisture meter based on near infrared (NIR) signal is suitable for online Applications with better accuracy. However, these systems are manufactured only in foreign countries. There price ranges in between USD $10,000 to $40,000. This cost does not include import duty. Hence this research work aims to develop low cost indigenous moisture measurement system based on NIR useful for industries as part of make in India scheme.
Abstract: In day to day life, smart and portable medical device is very essential part of every one. The Filter plays important role in signal separation and signal restoration. In ECG signal so many noises are presented. To remove this noise with the help of traditional FIR ?lter and PSO based FIR ?lter, But PSO based FIR Filter is more e?cient than traditional FIR ?lter. The dissertation work, Traditional FIR Filter is designed with partial swarm optimization algorithm for noise remove purpose and windowing function is used for R- peak detection for features calculation. The partial swarm optimization ?lter gives more e?ective result as compare to traditional FIR ?lter. The second part of this system is disease detection using support vector machine.
Project Objective: 1.Design FIR ?lter. 2. Develop Particle Swarm Optimization Filter. 3. Implement PSO ?lter on ECG signal. 4. ECG Disease Detection using SVM.
Conclusion: The ?lter plays important role in signal separation and signal restoration in medical devices. Traditional FIR and PSO ?lter results are compared based on parameter. The performance of PSO and FIR ?lter are evaluated considering accuracy, speci?city and sensitivity. The PSO ?lter shows the 25 percent improvement in accuracy, 10 percent improvement in speci?city and no loss in sensitivity. The support vector machine is a classi?er that is designed, implanted and tested for disease classi?cation. From all disease Myocardia-infraction is more suitable as compared to healthy control. The accuracy of myocardial Infraction 64 percent optimized as compared to healthy control, sensitivity of myocardial infarction is 30 percent optimization as compared to healthy control and speci?city of myocardial infarction is 64 percent optimized as compared to healthy control.
The adoption of the Industrial Internet of Things (IoT) are also knows as Industry 4.0. It has an effective growth in recent years due to the availability of massive computing power, innovations in the data-processing technology, and the advent of machine learning (ML) algorithms. This is prominent in the manufacturing industry where IoT-enabled smart manufacturing has supported streamlined business operations, optimized productivity, and improvements in the return on investment. In Industry 4.0, IoT is an impressive technique which allows integration of various devices in one network for data transfer without any human to machine or human to human contact. This report demonstrates the application of condition monitoring of mechanical equipment with different sensor. Labjack T7 Pro allows collecting the signals by various frequencies at single instance. Also, it can be coded for collect the data from all sensors at single time. In this study, ensemble machine learning and Ensembles of Ensemble machine learning technique is used based on fault diagnosis of mechanical component such as gears and bearings. For the experimentation purpose, available online data is used, and results are obtained for individual base classifiers, i.e. Support Vector machine (SVM), Multilayer perceptron (MLP), Decision Tree (DT), K-Nearest Neighbour (k-NN), Navie Bayes (NB) and Random Forest (RF), Adaptive Boosting (AdaBoost), Gradient Boosting (GB), Extreme Gradient Boosting (XgBoost) and Extra Tree for Ensemble machine learning and Ensembles of Ensemble machine learning techniques respectively.
Internet of Things (IOT) is an emerging technological field. The objective of this project is to connect the internet to the physical devices of the industry as sensors through the gateway. The industrial concept of IOT requires a gateway for the safe supervision of physical components. A microcontroller senses the signal from the sensors and transmits data through the gateway. The SIM module that operates on GPRS service act as a gateway to the internet. The Message Queuing Telemetry Transport (MQTT) protocol is used as a gateway protocol. MQTT increases the security of data received from the device to the internet. The result is a continuous sensor monitoring system that can be used to determine weather conditions.
Abstract: Industries, the sorting of objects is carried out manually by using human labor. Detecting a particular object and placing it in the required position is tiring work. So,this project describes the circuit Color sensor detect the color of the object and send to arduino microcontroller. The microcontroller send signal to which drives the various motors of robotic arm. By using pick and place robotic arm the classified colored object which are in an object box by picking and placing the objects in its respective pre-programmed place.
Project Objective: 1.To develop algorithm for colour sensor. 2. To interfacing colour sensor with microcontroller. 3. To provide automation for robotic arm for sorting the object. 4. To design easy and simplified operational system.
Conclusion: The speed of color sorting process by an operator is very slow as well as accuracy is less. This all problems covered by this project by using robotic arm. Robotic arm can pick high weight also gives high accuracy and less time required. Accuracy gives is 65 percent in developed system.
Moisture of product in many industries decides quality of the product. Control over the moisture is very important. There are many methods used to measure moisture like electrical, nuclear, microwave moisture measurement, etc. These methods are so expensive. Some of the industries use loss of weight on drying, which is time consuming and not suitable for continuous measurement. There is a need to develop an accurate, low cost and online moisture measurement sys- tem. Near Infrared (NIR) signals are suitable for online application and better accuracy. NIR measurement uses reflectance and absorption principle for calculating the moisture content of sample. Higher the moisture content, higher is the absorption of light. By using particular band of NIR wavelength, the moisture measurement system will be developed. These types of systems are already avail- able in the foreign countries, but they are very expensive. Hence, the aim is to develop a low cost indigenous moisture measurement system useful for industrial applications.
The system is based on technology whose main purpose is to detect an accident and alert to the control room and nearby Hospital, so that victim can get some help. It can detect the accident and also intensity of accident without any visual contact. If this module is inserted to every vehicle then we can detect each accident. We had used GSM, GPS and Accelerometer module for this project. When accident occurs the change in the co-ordinates are calculate by using accelerometer and from co-ordinates the change in angle can be determined and if angle of rotation is greater than the expected angle then the signal will be send to GSM module and automatically a call or message will be send to user and through message, we send the live location of an accident occurred using the GPS module and the message is send without any human interaction.
The underground drainage system is an important component of urban infrastructure. It is considered to be city’s lifeline. Most management on underground drainage is manual therefore it is not efficient to have clean and working underground system also in such big cities, it is difficult for the government personnel to locate the exact manhole which is facing the problem. Therefore, it is essential to develop a system which can handle underground drainage without human intervention. Underground Drainage involves sewerage system, gas pipeline network, water pipeline, and manholes. This project describes various functions used for maintenance and monitoring of underground drainage system. It provides a system which is able to monitor the water level, atmospheric temperature, water flow and toxic gasses. If drainage system gets blocked and water overflows it can be identified by the sensor system. And that sensor sends information via the transmitter which is located in that area to the corresponding managing station.
The design of the dispenser is simple and compact. It consists of a main storage compartment which will be cylindrical with barrel. Each slot will be having a hole on the bottom. All except one of these holes will be covered by a disc connected to a servo motor. There will be an LED display and an alarm system in the dispenser. The working of the dispenser is being controlled by a Raspberry Pi. It acts as the local server and downloads or updates the patient’s dosage information from the main cloud server whenever it comes online. This feature enables the dispenser to work without an active internet connection throughout its usage. The Raspberry Pi needs to be periodically connected to the internet to update the progress.
In this system water level of tanks are monitored and controlled by using IOT (Internet Of Things) by using information present on the cloud system. To reduce wastage of water this system is useful. To fill each tank appropriately and to control overflow of water sensor is used to detect it. This system is used for hostels of college and other places where monitoring of tanks is required.
The system measures the percentage of liquid (liquid such as petrol, diesel etc.) remains in the tank. Mobile application is used to display output and give alert notification in emergency conditions. Sensor which is used to measure the liquid level which is completely developed in institute laboratory because of this cost of system is low and accuracy is more. It can implement in various applications such as diesel generator tanks, petrol pump tanks etc. It also gives information about the whether the generator is ON/OFF if system implement for diesel generator. The systems which are available on the market are costly and they are not smart. So sometime this is time consuming and it is not giving any alert message. So This “IOT based Non Conducting Liquid Level Measurement System” measure level with low cost. ESP8266 is interfaced in the project which is sending real time data on the cloud. On the mobile app, get the liquid level. It also provides emergency notifications.
As a college students and faculty member we observed that, large amount of demand charges are included in monthly electricity energy bills. As a commercial customer, pay for—two types of charges on monthly utility bill. The first is energy charge, which can calculate by multiplying total energy use for the month (measured in kilowatt hours, or kWh) by energy rate. The other is demand charge and is typically calculated by looking at the greatest amount of power needed (measured in kilowatts) during any of thousands of “demand intervals”.A demand charge/tariff is also known as a capacity charge. Where a daily charge is determined by the highest power demand (load) observed during a certain timeframe during the day in a specified period.
Smoke detector alarm plays an important role for protecting physical and financial disaster , as it indicates the risk of explosion , fire as well as different gas leakage as it can also able to detect different gas as well . And with the course of time it will get majorly used everywhere.
Noise pollution like other pollutants is also a by- product of industrialization, urbanizations and modern civilization. Noise pollution has two sources, i.e. industrial and non- industrial. The industrial source includes the noise from various industries and big machines working at a very high speed and high noise intensity. Non- industrial source of noise includes the noise created by transport/vehicular traffic and the neighborhood noise generated by various noise pollution can also be divided in the categories, namely, natural and manmade. Noise pollution has really started to gain importance due to high population density. Loudness or sound levels are commonly measured in decibel (dB), we have some instruments which could measure the sound signals in dB, but these meters are slightly expensive .So here we are manufacturing a noise level detector which is kind of easy to manufacture and economical affordable.
The major problem identified is that there are numerous accidents occurring in the coal mines due to improper maintenance and inadequate monitoring of the mining activities. These led to numerous life losses and immeasurable recourse loss.There is no proper early detection of the uncertainty in the coal mines. Coal mining has been a very dangerous activity. The principal hazards are mine wall failures and vehicle collisions; underground mining hazards include suffocation, gas poisoning, roof collapse and gas explosions. Chronic lung diseases, such as pneumoconiosis (black lung) were once common in miners, leading to reduced life expectancy. In some mining countries black lung is still common.However, in lesser developed countries and some developing countries, many miners continue to die annually, either through direct accidents in coal mines or through adverse health consequences from working under poor conditions.
Academic Year 2019-20
SY ETC EVS Projects:
Gr. No. | Sr. No. | Roll No. | Name of Guide | Project Title |
1 | 1 | 1805015 | Prof. S.C.Bedage | Energy Monitoring System |
2 | 1801053 | |||
3 | 1805040 | |||
4 | 1805036 | |||
2 | 1 | 1805004 | Dr. J.S. Awati | Noise Level Detector |
2 | 1805024 | |||
3 | 1805026 | |||
4 | 1805054 | |||
3 | 1 | 1805032 | Prof. S. S. Patil | Waste to energy(electricity) |
2 | 1805035 | |||
3 | 1805047 | |||
4 | 1805059 | |||
4 | 1 | 1805028 | Prof. B.N. Holkar | Smart Water Bottle Reminder. |
2 | 1805030 | |||
3 | 1805033 | |||
4 | 1805046 | |||
5 | 1 | 1955002 | Dr. M.S. Kumbhar | Dry Wet Waste Separate |
2 | 1955006 | |||
3 | 1955010 | |||
4 | 1955013 | |||
7 | 1 | 1805009 | Prof. R.J Patil | Air quality monitoring system |
2 | 1805013 | |||
3 | 1805020 | |||
4 | 1805058 | |||
8 | 1 | 1805041 | Prof. B.N. Holkar | Noise Detector |
2 | 1805042 | |||
3 | 1805048 | |||
4 | 1805050 | |||
5 | 1804058 | |||
9 | 1 | 1805022 | Prof.V. S. Patil | Canny Bin |
2 | 1705004 | |||
3 | 1805062 | |||
4 | 1805056 | |||
10 | 1 | 1805001 | Prof. B.N. Holkar | LPG Gas Leakage Detection System |
2 | 1805002 | |||
3 | 1805008 | |||
4 | 1805031 | |||
11 | 1 | 1805025 | Prof. S. M. Magdum | Agricultural Soil Monitoring System |
2 | 1805037 | |||
3 | 1955015 | |||
4 | 1955017 | |||
12 | 1 | 1805060 | Prof. S. M. Magdum | App to Literate Framers |
2 | 1805061 | |||
3 | 1805027 | |||
4 | 1705044 | |||
13 | 1 | 1955001 | Prof.P.P.More | Design and Development of CO Monitoring System |
2 | 1955009 | |||
3 | 1955011 | |||
4 | 1955014 | |||
14 | 1 | 1805007 | Prof.V. S. Patil | vacuum cleaner |
2 | 1805034 | |||
3 | 1805023 | |||
4 | 1805038 | |||
15 | 1 | 1805005, | Prof. R.J Patil | Automatic Flow Sensor |
2 | 1805012, | |||
3 | 1805014, | |||
4 | 1805017, | |||
5 | 1805018 | |||
16 | 1 | 1955003 | Prof. Aniket Prabhavalikar | Smoke Detector Alarm |
2 | 1955004 | |||
3 | 1955007 | |||
4 | 1955008 | |||
17 | 1 | 1955012 | Dr. M.S. Kumbhar | Sea Bin |
2 | 1955016 | |||
3 | 1955005 | |||
4 | 1801032 |
TY ETC Mini Projects:
Gr. No. | Sr. No. | Roll No. | Name of Guide | Project Title |
1 | 1 | 1705038 | Dr. M.S.Patil | Real Time Face Detection |
2 | 1705024 | |||
3 | 1705041 | |||
4 | 1705047 | |||
2 | 1 | 1855008 | Dr. M.S.Patil | Smart Real Time Drainage Monitoring System |
2 | 1855016 | |||
3 | 1855010 | |||
3 | 1 | 1855007 | Dr. M.S. Kumbhar | Heart Pulse Rate Monitoring System |
2 | 1855001 | |||
3 | 1855003 | |||
4 | 1855004 | |||
4 | 1 | 1705034 | Prof. R. T. Patil | Automatic Wheelchair Machine |
2 | 1705010 | |||
3 | 1705015 | |||
4 | 1705040 | |||
5 | 1 | 1705021 | Prof. R. T. Patil | Electricity Bill Controller |
2 | 1705023 | |||
3 | 1705025 | |||
4 | 1705027 | |||
6 | 1 | 1705019 | Prof. R. T. Patil | Power Monitoring System |
2 | 1705031 | |||
3 | 1705022 | |||
4 | 1705016 | |||
7 | 1 | 1705002 | Prof. S. S. Patil | IOT based smart water level monitoring and control |
2 | 1705005 | |||
3 | 1705014 | |||
4 | 1705017 | |||
8 | 1 | 1705008 | Prof. M.R.Jadhav | Air Pollution Detector using MSP 430 |
2 | 1705026 | |||
3 | 1705028 | |||
4 | 1705058 | |||
9 | 1 | 1705042 | Prof. B.N.Holkar | Smart Attendance System using Image Processing |
2 | 1705046 | |||
3 | 1705035 | |||
4 | 1705045 | |||
10 | 1 | 1755006 | Prof. B.N.Holkar | By using MSP430 Speaking Bus Stop Remainder |
11 | 1 | 1705003 | Prof. R. J. Patil | Real Time Noise Filter |
2 | 1705009 | |||
3 | 1705012 | |||
4 | 1705017 | |||
12 | 1 | 1705020 | Prof. R. J. Patil | Temperature Monitoring using IOT |
13 | 1 | 1705043 | Prof. V.S. Patil | Remote Area Monitoring System |
2 | 1705062 | |||
3 | 1855009 | |||
4 | 1855014 | |||
14 | 1 | 1855011 | Prof. V.S. Patil | Automatic Fish Feeding System using MSP430 |
2 | 1855012 | |||
3 | 1855006 | |||
4 | 1855015 | |||
15 | 1 | 1705050 | Prof. V.S. Patil | Coal Mine Accident Detection System. |
2 | 1705061 | |||
3 | 1705060 | |||
4 | 1855002 | |||
16 | 1 | 1705030 | Prof. Aniket Prabhavalikar | Automatic Accident Alert System |
2 | 1705032 | |||
3 | 1705053 | |||
4 | 1705054 | |||
17 | 1 | 1705013 | Prof. Aniket Prabhavalikar | Automated Medicine Dispenser |
2 | 1705001 | |||
3 | 1705018 | |||
4 | 1705006 | |||
18 | 1 | 1705033 | Prof. P.P.More | Vehicle accident detection using IOT |
2 | 1705037 | |||
3 | 1705051 | |||
4 | 1705052 | |||
19 | 1 | 1705011 | Prof. P.P.More | Deep Learning based Weeding Machine |
2 | 1705007 | |||
3 | 1705029 |