Mechanical Engineering Projects

KME01 - Design and Development of Electric Bicycle

Project Guide:

  • Prof. Nilesh Gaikwad

Students:

  • Mr. Sarang Kawade
  • Mr. Manish Kumar
  • Mr. Saransh Goyal

Domain:

  • eMobility

Short Description:

  • We are designing and manufacturing an electric bicycle. Our e-bicycle is more performance optimized and smart than other e-bicycles present in the market. We have developed our own customized motor controller to drive the e-bicycle. Our smart information cluster shows you all the necessary information you need, along with navigation and SOS system.

KME02 - Design, Development and Manufacturing of Power Tiller Machine

Project Guide:

  • Prof. J. D. Ganeshkar

Students:

  • Sadashiv Jeevanrao Patil
  • Varun Dilip Patil
  • Vaishnavi Kailas Shejwal
  • Tanmay Pradeep Potphode

Domain:

  • Agricultural Equipment

Short Description:

  • Weed removal machine to overcome the drawbacks in other weed removal techniques like Weed removal by hand picking, weed removal by animal operated equipment and herbicides. Our developed solution can remove weed from two rows at a time. Also, it removes weed from its roots.

KME15 - Portable Emission Detection and Continuous Monitoring Device

Project Guide:

  • Dr C. L. Ladekar

Students:

  • Umesh Rajshekhar Nandargi
  • Sakshi Ganesh Shende
  • Mansi Anil Shankhpal
  • Urvesh Vikhiat Berry

Domain:

  • Thermal and Automation

Short Description:

  • Design and fabrication of a portable device for continuous detection of vehicular pollution using gas detection sensors and microcontroller and to inform the driver and government authorities about the emissions from vehicles with the help of cloud-based storage along with an inbuilt storage of e- copies of driving documents.



KME04 - Automatic Brake Failure Indicator With Auxiliary Braking System

Project Guide:

  • Prof. Chandan R. Ingole

Students:

  • Kaustubh Salunke
  • Prashant Gayakawad
  • Ranbir Singh Saini
  • Manthan Jadhav

Domain:

Short Description:

  • Brake Failure Indicator is a device which can give early warning of brake failure while driving. The Brake Failure circuit constantly monitors the condition of the brake and gives an audio-visual indication. The brake failure condition is sensed by the sensors attached to the circuit through monitoring the brake switch.





KME05 - Design and Analysis of Compound Press Tool For Washer

Project Guide:

  • Prof. Sanjiwan K. Bhoite

Students:

  • Vighnesh Sandbhor
  • Saurabh Nagendra

Domain:

  • Manufacturing

Short Description:

  • A compound die is designed and developed for manufacturing of simple washer with an attempt to get desired accuracy and more productivity

KME06 - Low cost 3D Printer using E-waste

Project Guide:

  • Prof. Chandan R. Ingole

Students:

  • Pranav Pradip Chandugade
  • Yogesh Vasant Deshmukh
  • Yash Rajesh Nitnaware
  • Kunal Damodar Thorat

Domain:

  • Mechanical and Mechatronics

Short Description:

  • Arduino based 3D printer made using industrial E-waste which can manufacture any geometrical shape object from PLA (Polylactic acid) material.

KME07 - Design and Manufacturing of Active Rear Wing to Improve Cornering of a Car using Shape Memory Alloy Actuation System

Project Guide:

  • Prof. R. A. Gujar

Students:

  • Abhir Adiverekar
  • Amey Thonge
  • Viren Chiplunkar
  • Rahul Kulkarni

Domain:

  • Rapid Prototyping

Short Description:

  • In driving, downforce generated by the moving vehicle, has a considerable effect on the performance of the vehicle in terms of braking, cornering, and acceleration. Shape memory alloys (SMA) are characterized by the ability to recover their shape after deformation under specific conditions.
  • Our objective is to increase cornering performance of a sports car. To design and manufacture an active rear wing and a circuit for the SMA actuation system working in real-time, and individually controls both halves of the rear wing.

KME08 - Design and development of Battery Thermal Management System using Nano fluids to extend the life and range of Electric Vehicle using Li-ion Battery Pack

Project Guide:

  • Prof. S. R. Wankhade

Students:

  • Prajwal Vijay Thorat
  • Sanket Jitendra Shisode
  • Swapnil Arun Sonawane
  • Rugved Sharadrao Wankhade

Domain:

  • Sustainable Energy

Short Description:

  • Lithium ion batteries are mainly used in electric vehicles as source of Power. But these Lithium ion cells work well within a temperature range of 15 to 35 Degree Celsius. During high discharge the temperature of batteries rises and affects the working and performance of lithium ion cells and decreases range of electric vehicle. Thus an Efficient Battery Thermal Management System is incorporated using Nanofluids to reduce the maximum temperature of batteries and bring the temperature in proper working range of Lithium Ion batteries. The cooling will thus enhance the range, safety and life of lithium ion batteries.




KME09 - Design and Development of an Immersion Cooled Battery Pack for High-Performance Light Electric Vehicles

Project Guide:

  • Dr. Anindita Roy, Dr. N. R. Deore

Students:

  • Vishal Sonawane
  • Sejal Baser
  • Chaitanya Dani
  • Prajwal Dhote

Domain:

  • Lithium Ion Battery, Thermal management system

Short Description:

  • The battery packs used in Light Electric Vehicles (LEVs) and e-motorcycles are currently air cooled using natural convection or forced air convection. Air cooling is insufficient for EV battery packs and leads to reduced battery performance and life. The performance further deteriorated for Indian subcontinent due to its hot climate. The parasitic power consumed by a forced convection air cooling system is 60% more as compared to liquid cooling system. The need for an effective compact liquid cooling system is felt for thermal management of lithium ion battery packs which can be used for high performance drives and capable of handling fast charging. An in-depth literature study on national and international level reveals unavailability of direct liquid cooled application for small EV battery packs. In this project, it is aimed to develop a Battery Thermal Management System (BTMS) using Direct Liquid immersion Cooling for a 3.3 kWh battery pack. MIVOLT DFK dielectric fluid will be used for the liquid immersion cooling system.

KME10 - Experimental Investigation of the performance of a cold storage unit integrated with phase change material

Project Guide:

  • Dr. Anindita Roy, Dr. Sonali Kale, Dr. A. B. Lingayat

Students:

  • Aditya Wavhal
  • Shamali Gawade
  • Deepankar Sengar

Domain:

  • Refrigeration, food preservation

Short Description:

  • Unavailability of reliable electricity supply and lack of cold storage in proximity to farms, poor infrastructure, and transportation facilities as well as a dearth of information are the major causes for wastage of fresh farm produce associated with price fluctuations and poverty of farmers. Distribution of cold stores to autonomously run units using local energy sources can be the next significant shift in the cold chain and food preservation market. Commercial cold stores with solar power backup and thermal storage less than 500 cubic feet are unaffordable due to huge dependence on grid electricity. The portable and movable, affordable, and low energy-consuming cold storage in the size range of less than 100 cubic feet is needed to be developed that allows farmers to have their autonomous storage. A 35 cubic feet (1 m3), mini cold storage capable of cooling about 100 kg of produce using an off-grid energy solution integrating solar-powered battery storage and phase change material (PCM)-based thermal storage is proposed

KME11 - Design and Manufacturing of Adjustable Dining Table

Project Guide:

  • Prof. Mrs. Vrushali Yogesh Bhalerao

Students:

  • Malhar Rahul Khole
  • Ritesh Shantinath Gadave
  • Aamod Vinayak Pande

Domain:

  • Manufacturing Technology

Short Description:

  • Managing space in homes is an important and challenging task to every Indian middle class. Optimum utilization of available floor area is an efficient way of addressing this problem with the use of modular, multipurpose and space saving furniture. Thus the project aims to design and develop a multi-purpose adjustable, space saving dining table for Indian middle-class homes.

KME12 - Autonomous Mobile Robot Base with Modular Extension

Project Guide:

  • Dr. Sanjay B. Matekar

Students:

  • Atharva Bhorpe
  • Aditya Jadhav
  • Nanak Arora
  • Siddharth Shrotri

Domain:

  • Robotics and Automation

Short Description:

  • Autonomous Mobile Robots (AMR) are changing the way in which many industrial and household material transportation systems work. Instead of manually driving the load in the warehouses, these advanced AMRs can perform completely without human interventions in a dynamic environment and will not require any additional cost of installing tracks that are usually required for AGVs.
  • It is planned to make a generic AMR base chassis that will have the ability to navigate from one point to another in dynamic environments and can be used to mount any additional attachments like a robotic arm, storage units, etc. as per the needs of the end-user. It can also be used for transporting materials. The AGV uses ROS for processing all the sensor data and path planning for the robot. Advanced sensors are used, like the LiDAR, Tracking Camera, Depth Camera and IMU for SLAM and Autonomous Navigation. The AMR will certainly add a lot of value to the industrial sector and automate the tiring and repetitive task of manually managing material in large warehouses.

KME13 - Development of an experimental setup to calculate specific heat of an ideal gas at constant pressure (Cp) and at constant Volume (Cv) respectively

Project Guide:

  • Prof. Ummid I. Shaikh

Students:

  • Nikhil C. Somwanshi
  • Rohit S. Thakare
  • Siddharth S. Bhosale
  • Animesh A. Wani

Domain:

  • Thermal Engineering

Short Description:

  • Our Setup include an electric resistance heating unit consisting nichrome wire as heating element which was inserted in glass aspirator bottle for heating air. For Cv, constant volume was obtained by considering bottle as closed system and pressure change with respect to temperature change was measured by precision manometer as per Gay Lussac Law considering air as an ideal gas.
  • For Cp measurement, Glass gas syringe was attached wherein air would expand keeping pressure constant and as per the Charles law change in volume with respect to temperature is measured by taking reading on syringe scale.

KME14 - Automatic Detection of Bearing Faults

Project Guide:

  • Prof. Amit Panchwadkar

Students:

  • Piyush Kishor Bhonde

Domain:

  • Condition monitoring (NVH), Machine Learning

Short Description:

  • Detection of bearing faults is one in all the foremost challenging tasks in bearing health condition monitoring, especially when the fault is at its initial stage. The defects in bearing unless detected in time may result in malfunctioning of the machinery. While performing analysis of vibration signals, it is close to impossible to separate out and focus on ‘bearing frequencies’ especially in the presence of strong masking signals from other machine components. ‘Advanced’ vibration analysis technique viz envelope analysis for localization of bearing frequencies from the masking signals generated by other machine components is used and a smart graphical user interface-based software tool has been developed for automatic detection of bearing faults. A setup consisting of vibration sensor (accelerometer), FFT analyzer, set of faulty bearings installed with rotating arrangement has been prepared for demonstration.

Electronics and Telecommunication Projects

KET01 - Vision Guided pick-and-place Robotic ARM

Project Guide:

  • Dr. S. L. Tade, Prof. Mrs. Sonali Y. Sawant

Students:

  • Shubham Panigrahi
  • Akash Biyani
  • Ashutosh Mithari

Domain:

  • Rapid Prototyping

Short Description:

  • In our approach, we have attempted to leverage the developments in Deep Learning and hardware for the Edge, in order to develop a vision-guided pick and place robotic arm. Firstly, with the help of a custom-created dataset, YOLOv4-tiny based object detection model is trained to categorize and localize the desired object. Further, a camera is used to capture the live stream of the surrounding environment, in RGB form, and is fed to the model. Any target object, if present in the surrounding, is recognized and localized within the image frame. Then the image coordinates are mapped to real-world coordinates, and based on which robotic arm trajectory and grasping of objects can be planned to perform the pick-and-place operation. We have used NVIDIA Jetson Nano to deploy our object detection model. With this, we accomplished a fully functioning model for a vision-guided robotic arm capable of pick-and-place operation on a low-end embedded device. The experimental results show the accuracy and effectiveness of our system in real-time application

KET02 - Wireless Digital Stethoscope

Project Guide:

  • Dr. Jyoti S. Kulkarni

Students:

  • Shubham Dashrath Belkhede
  • Shivprasad Somnath Bulbule

Domain:

  • Additive Manufacturing

Short Description:

  • Heart auscultation, the process of interpreting the sound produced by the heart, is a fundamental tool sophisticated in the diagnosis of cardiac disease. It serves at the most commonly employed technique in primary health care and in circumstances, where medical equipment is not available. A digital Stethoscope is able to convert an acoustic sound to electronic signal, which can be further amplified for optimal listening. These electronic signals can be further processed and transmit to a personal computer or a laptop.

KET03 - Autonomous Mobile Robot Base with Modular Extension

Project Guide:

  • Dr. M. T. Kolte

Students:

  • Piyush Sanjay Salunke
  • Faizaan Gulammustafa Sayyad
  • Swapnil Milind Inamdar

Domain:

  • Robotics

Short Description:

  • We made a generic AMR base chassis which will have the ability to navigate from one point to another in dynamic environments and can be used to mount any additional attachments like robotic arm, storage units, etc. as per the needs of the end user.

KET04 - Design and Analysis of Wearable Microstrip Patch Antenna for Breast Cancer Detection

Project Guide:

  • Prof. G. R. Rahate

Students:

  • Pratik Punekar
  • Ankush Rajput
  • Shounak Powar

Domain:

  • Health and Care

Short Description:

  • Designed Wearable Microstrip Patch antenna is useful for detection of breast cancer. Antenna is operating at 2.45GHz. Substrate material of antenna is jeans and shape of patch is rectangular.

KET05 - Fraud Detection and Prevention by Face Recognition with and without Mask for Banking Applications

Project Guide:

  • Dr. Rajani P. K.

Students:

  • Shreyas Murkute
  • Prathamesh Phalke
  • Shashank Tiwari

Domain:

  • Signal Processing and Machine Learning

Short Description:

  • Fraud in banking transactions is unauthorized and unwanted usage of an account by someone other than the owner of that account. Fraud detection involves monitoring the activities of populations of users in order to estimate, perceive or avoid objectionable behaviour, which consist of fraud, intrusion, and defaulting. Necessary prevention measures can be taken to stop this abuse and the behaviour of such fraudulent practices can be studied to minimize it and protect against similar occurrences in the future. Researchers have shown broad concern about detection and recognition of fraudsters since banks and the individual user are both suffering significant losses from fraud activities. They have proposed various solutions to counter fraudulent activity eg., multi-factor login authentication, utilizing real-time analytics, maximizing password management. However, those methods may lose effectiveness in fraud detection because fraudsters always tend to cover their tracks. It is vital that Banks are able to identify fraudulent transactions so that customers are not charged for items that they did not purchase. Such problems can be tackled with Face Recognition and its importance, along with Machine Learning, cannot be overstated. The project intends to illustrate the modelling of a data set using machine learning with Banking Fraud Detection with mask, using face recognition and machine learning algorithms. Applications include bank security, ATM, Public distribution systems.

KET06 - Fully Automatic Cup Dhoop Making Machine

Project Guide:

  • Prof. Mrs. Sharada Patil

Students:

  • Tushar Balu Aher
  • Kaushal Kishor Kulkarni
  • Akshta Jotiram Doke

Domain:

  • Automation

Short Description:

  • Automation is a technology that implements technological devices to carry out a process with minimal human intervention. It is the process of automating a system that can be accomplished using a variety of methods, including mechanical, electrical, electronic, and computer devices, as well as pneumatics systems. The main objective is to manufacture an automatic Cup Dhoop Making Machine. Traditional Machines for Making Cup Dhoops work on 3 phase supply and use hydraulic systems which require routine maintenance, such as oil changes, reservoir cleaning. These systems are expensive, heavyweight, required skilled manpower, and are harmful as well. Automation is provided in an existing machine to some extent like automatic material filling but systems are based on mechanical drives hence required technical support for any kind of maintainces. Proposed solution is based on an electro-mechanical system which uses sensors and actuators can easily overcome the drawback of the existing machine. The machine is works on single phase supply, can make up to 25 cups at a time, low cost, low maintenance, easy to use.

KET07 - Emotion Detection and Recognition

Project Guide:

  • Dr. M. T. Kolte, Prof. P.V. Sontakke

Students:

  • Pranav Vikas Patharkar
  • Siddhant Satish Pawar
  • Pratiksha Prakash Diwanji

Domain:

  • Machine Learning

Short Description:

  • Detecting and recognizing the emotions of a person based upon the facial expressions produced by them. Usually, one has a talk with a person to know about his/her emotions or mental state. But, it seems uncomfortable for one to talk about his/her mental health. So, instead of asking a person directly to talk about it, we can design a tool which can read the emotions by using the facial expressions. This project helps to make an inference about the mental health of a person using the primary categories of emotions which are happiness, sadness, anger, surprise, fear and disgust. The proposed tool can be used at various places such as hospitals (to read a patient’s expressions), at airports (to detect suspicious facial expressions), in schools/colleges (to analyze the mood of students), etc. For example, teenagers who might be facing difficulties with their studies/social life tend to have a mental imbalance after long time of battling with their own emotions. Unfortunately, they ultimately become mentally weak and enter into a mental state generally called depression. They might take a wrong step to take their own life and end up into an unfortunate incident for the society. If the emotions of such teenagers are read using this tool in the early stages of struggling, it might help the society to help them get out of their situation.

KET08 - Smart Refrigerator System

Project Guide:

  • Dr. Mrs. S. U. Bhandari, Prof. P. V. Sontakke

Students:

  • Sanskruti Santosh More
  • Prathmesh Rajiv Satyarthi
  • Pravin Subhash Umade

Domain:

  • Internet of Things (IOT)

Short Description:

  • Modern production lines for refrigerator take advantage of automated inspection equipment that relies on cameras. As an emerging problem, refrigerator classification based on images from its front view is potentially invaluable for industrial automation of refrigerator. However, it remains an incredibly challenging task because refrigerator is commonly viewed against dense clutter in a background. In this project, we propose an automatic refrigerator image classification method which is based on a new architecture of Convolutional Neural Network (CNN). It resolves the hardships in refrigerator image classification by leveraging a data-driven mechanism and jointly optimizing both classification and similarity constraints. To our best knowledge, this is probably the first time that the deep learning architecture is applied to the field of household appliance of the refrigerator. Disclosed are various embodiments for determining whether food in a refrigerator has spoiled. An application executing on a computing device receives an image from a refrigerator of a compartment of the refrigerator. The application then analyzes the image to identify a food item in the image. Subsequently, the application analyzes the image to identify a potentially spoiled area of the food item. The application then compares the potentially spoiled area of the food item to a model of a spoiled food item. Afterwards, the application determines that the identified food item has spoiled based at least in part on a comparison of the potentially spoiled area of the identified food item to the model of the spoiled food item.

KET09 - Aerial System for Security from Rodents in Agriculture

Project Guide:

  • Prof. S. A. Patil

Students:

  • Krishna Govind Nimbalkar
  • Akshada Sanjay Patil

Domain:

  • Agriculture

Short Description:

  • In India, rodents have long been reported as having a substantial impact on crops A rise of population has immensely increased the pressure on agriculture sector. And thus, there is need to prevent agricultural produce being damaged by rodents. We propose an aerial system for security from rodents in agriculture farmland. Using Kolhapuri Chappal sound mechanisms mounted on drone to avoid rodents like snakes, rats, etc. This project aim is to prevent harms done by rodents to crops and thus providing security from rodents in agriculture.

KET10 - Analysis of EEG Signals using Machine Learning for Prediction and Detection of Stress

Project Guide:

  • Prof. A. B. Patil, Prof. Mrs. S. A. Patil

Students:

  • TUSHAR KISHOR KOTKAR
  • KAUSHIKI HARSHAL NAGPURE
  • PRATIK MILIND PHADKE

Domain:

  • Machine Learning

Short Description:

  • The working stress has really affected our lives. People around us are facing challenges due to various reasons that can be stressful and can also cause strong emotional unbalance in adults and children. All the public health actions like social distancing, to stay isolated, has lead people to be not only stressful but also anxious, which directly affects health and tends to worsening of chronic health problem. Also working under such conditions is not recommendable and henceforth in this paper we stated that using non-invasive method how we can early detect the level of stress and then the individual can take respective precautionary measure against it. Since any type of stress can be fatal thus need of stress detection is most important. There are various methods exist to detect stress like Magnetic Resonance Imaging, Electromyography, Electrocardiogram, Positron Emission Tomography which can help to quantify stress in an individual’s body. EEG are physiological features produced by brain’s electrical activity and thus we get the voltage difference when we take EEG signals form brain scalp it is a non-invasive method. Also the EEG signals are precise and reliable so it is better option to be considered for stress detection. We have considered various classifiers and compiled a detail analysis of all and among that we got the maximum accuracy for SVM. Output expected from this project is to detect stress so that further diseases can be prevented and one can lead to happy, healthy and productive life.

KET11 - Smart Number Plate Recognition System for Vehicle Using Raspberry Pi

Project Guide:

  • Dr. Mrs. Varsha Bendre

Students:

  • Sachin Dengle
  • Shreyas Patil
  • Shubham Araseed
  • Vaibhav Randale

Domain:

  • Rapid Prototyping, Additive Manufacturing, Manufacturing Technology, Machine Learning

Short Description:

  • A designing of a system which captures the image of the number plate automatically of a Shovel-Dumper combination and these details were verified using Raspberry Pi2 processor for authentication. This system captures the number plate of shovel and dumper further processing for the character recognition. Automation is the most frequently spelled term in the field of electronics. The hunger for automation brought many revolutions in the existing technologies. This paper makes use of an onboard computer, which is commonly termed as Raspberry Pi2 processor. It acts as heart of the project. This onboard computer can efficiently communicate with the output and input modules which are being used. The device which is able to perform the task is a Raspberry Pi2 processor. When any vehicle passes by the system, the image of the number plate of every vehicle is captured using camera. The image of the number plate details are fed as input to the Raspberry Pi2 processor. The Processor takes responsibility to check the authentication details of every shover and dumper. Once the details are recognized then the processor operates it detects an unauthorized image of number plate was detected. To perform this task, Raspberry Pi processor is programmed using embedded ‘Raspbian’.

Computer Engineering Projects

KCE01 - Anomaly Detection in Video Using Machine Learning

Project Guide:

  • Prof. Mrs. Archana Kadam

Students:

  • Priti Vasekar
  • Sheetal Singh
  • Samidha Bharle
  • Priti Giramkar

Domain:

  • Machine Learning

Short Description:

  • Surveillance cameras are increasingly being used in public places e.g. streets, intersections, banks, shopping malls, etc. to increase public safety. One critical task in video surveillance is detecting anomalous events such as traffic accidents, crimes or illegal activities.The goal of a practical anomaly detection system is to timely signal an activity that deviates normal patterns and identify the time window of the occurring anomaly.
  • We plan to build an application for detection of anomalous activity in surveillance videos in real time. An application which is capable to analyze the video and flag the anomalies, so that proper steps can be taken.

KCE02 - A Deep Learning based Responsive Web Platform for Cervical Cancer Detection

Project Guide:

  • Dr. Mrs. Swati Shinde

Students:

  • Tejas Morkar
  • Suyash Sonawane
  • Aditya Mahajan
  • Shubham Bodhe

Domain:

  • Healthcare, Deep Learning, Bioinformatics

Short Description:

  • The Proposed CerviTester platform makes the use of Deep Learning algorithms to detect cervical cancer and help the patients with early detection. This system aims to aid the health care workers like doctors and pathologists in speeding up the screening process. To help maximize the comfort of the patient, we have developed a completely responsive web-based platform which is aided by our robust machine learning models to help the patient get their reports from the comfort of their homes and make it easier for early detection of cervical cancer. It accepts the pap smear image or colposcopy image as the input and predicts the cervical cancer. Also it locates the abnormal portion in the image.

URLs:





KCE03 - Image Colorization of thermal images for Pedestrian detection in Surveillance of autonomous vehicles

Project Guide:

  • Dr. Mrs. Sonal Gore

Students:

  • Harish Dalal
  • Anagha Dangle
  • Jeenisha Shrungare
  • Radhika Mundada

Domain:

  • Artificial Intelligence

Short Description:

  • When driving in hazardous conditions, a person's vision may be impaired. As a result, mishaps occur. A surveillance application should assist the driver, and an enhanced vision system should be included in the solution. LIDAR and RGB cameras are currently being tested in autonomous and semi-autonomous vehicles.
  • We propose technique for colorization of thermal images and image enhancement, especially images those are captured during night time or in adverse weather conditions to detect the pedestrians. The custom Convolutional Neural Network colorizes the thermal images and those are further subject to the deblurring model for enhancement. The colorized and deblurred image is then fed into the detection model for pedestrian detection. This system is currently deployed using TensorFlow servers.





KCE04 - Edge Computing Based Portable Health Monitoring System

Project Guide:

  • Dr. Mrs. Sonal Gore

Students:

  • Aniket Shashikant Dhole
  • Mohit Sunil Gandhi
  • Shrishail Prakash Kumbhar
  • Harsh Rajesh Singhal

Domain:

  • Edge Computing, Deep Learning

Short Description:

  • A portable low-cost 3D printed device that can monitor if individuals entering the workspace are wearing masks, and whether those have an appropriate heart pulse rate and temperature. The design of IoT device is proposed for health monitoring using Microcontroller and Sensors. Tensorflow Lite Python Library and Edge Computing are utilized for optimization.

KCE05 - Booklingo (Deepfakes generation for VideoBooks)

Project Guide:

  • Dr. Mrs. Swati Shinde

Students:

  • Jayesh Shelar
  • DipakGhatole
  • Mayank Pachpande
  • Dhanashree Bhandari

Domain:

  • Videobooks, Deep fakes, Lip-sync

Short Description:

  • In today's world, most of the authors conduct their book reading sessions. They also get quite a good response from the audience and it also engages the audience more towards the story/content that the author has written. This seems a more interactive and modernized way of spreading the word through books but in an unconventional way. Now, it is not possible for the author to give book reading sessions in multiple languages and to multiple people located in the whole world.
  • Here Booklingo comes into the picture by generating realistic videos by deep fakes of the author reading the book in multiple languages. This not only solves the problem of language to language translation but also develops an engaging environment for the reader. This technique not only restricts to authors but can be used by teachers to circulate their notes in interactive Video forms and in multiple languages! This would make the teaching learning experience more fun and interactive and the specialization of the teacher can be consumed by the studentsworld-wide learning in different languages. Users only need to pass a pdf of a book/text, an author image with target language where video of that specified author will be created reading that book into specified target language.

KCE06 - A Portable Bidirectional Sign Language Translation System for Deaf and Dumb

Project Guide:

  • Dr. Mrs. K. Rajeswari, Prof. Mrs. Sushma Vispute

Students:

  • Naveen Hugar
  • Rakesh Pulli
  • Shubham Rajput
  • Rushikesh Tayade

Domain:

  • Deep Learning, Public Safety

Short Description:

  • Sign language is used by the deaf community as they are unable to hear which will lead to communication gap between deaf people and normal people because the majority of hearing people don’t understand sign language. This gap can be breached by our system Bidirectional Sign Language Translation System which is used for translating the provided signs into text and given text into related signs.

KCE07 - Automated Deep Learning based Weedicide Spraying Robot

Project Guide:

  • Prof. Mrs. Bodireddy Mahalakshmi

Students:

  • Sumedh Joshi
  • Amit Khirdekar
  • Luis Cherian
  • Ashish Kulkarni

Domain:

  • Agriculture, IoT, Deep Learning

Short Description:

  • Weed removal constitutes a significant element of the agricultural industry throughout the world. Over the past century, weed control has been a long-standing issue in the field of agriculture. Automated weed control, including weed detection and removal, has gained significant popularity in the community of precision farming over recent years.

KCE08 - AI based Workout Assistant and Fitness Guide

Project Guide:

  • Dr. Mrs. Reena Kharat

Students:

  • Pratik Dhende
  • Prathamesh Jondhalekar
  • Gourangi Taware
  • Rohit Agrawal

Domain:

  • Biomedical Applications and Soft Computing

Short Description:

  • In our work, we introduce Fitcercise, an application that detects the user's exercise pose, counts the specified exercise repetitions and provides personalized recommendations on how the user can improve their form. Pose Trainer uses the state of the art in pose estimation to detect a user’s pose, then evaluates the vector geometry of the pose through an exercise to count the repetitions and avoid physical injuries. I will also recommend a diet based on the user’s daily calorie goal.









KCE09 - Image Forgery Detection using Machine Learning

Project Guide:

  • Dr. Sudeep D. Thepade

Students:

  • Chirag Bagde
  • Sanket Bhandari
  • Rutuja Chaware
  • Krutik Lodha

Domain:

  • Security in Digital Media, Machine Learning

Short Description:

  • Machine learning based Image Forgery detection is attempted in the project
  • With the advent of digital technology ‘image tampering’ has become very common
  • Image Forgery is used mainly for fake news, election and religion polarization, increasing the political power and influence, defaming someone etc. The adverse effects of such forgery can be long lasting thus it becomes necessary to deal with such a situation so as to reduce the damage caused by such agendas.
  • Use of Machine learning algorithms can help in automated detection of forged and non forged images

KCE10 - A system for Liver Tumor Detection

Project Guide:

  • Dr. Mrs. Anuradha Thakare

Students:

  • Anjitha Nair
  • Rutuja Nemane
  • Shreya Pillai
  • Nupur Shiturkar

Domain:

  • Machine Intelligence and Optimization Algorithms

Short Description:

  • Liver cancer is found to be the sixth most common type of cancer worldwide, and is one of the major causes of death globally. Thus, it is important to diagnose liver tumors in the primitive stage itself, but it is a complex process due to some problems of CT scans like noise in the images, tilted images, low resolution and closely connected organs making it difficult to segment out different organs. The main aim of our project is to categorize an input 2-D CT image as tumorous or normal. For preprocessing the images, we are using gaussian filter to remove noise and blur the images. For image segmentation, we have compared two models that are the Unet-VGG16 and the Segnet model. Considering the fact that there is a shortage of training data in medical imaging, we have opted for the Transfer Learning (TL) technique for classification of images into tumorous or normal.

KCE11 - Management of Digital Evidence for Cyber Crime Investigation using Proxy Re-encryption Algorithm

Project Guide:

  • Dr. Mrs. Rachana Patil

Students:

  • Harshwardhan Chougule
  • Sunny Dhadiwal
  • Mehul Lokhande
  • Rohit Naikade

Domain:

  • Information and cyber Security

Short Description:

  • Current digital evidence are prone to tampering. So to avoid the tampering and to enhance the security we are thinking of integrating it with Blockchain and use the proxy re-encryption algorithm for the secure passing of delegation rights.

KCE12 - AI Based Surveillance System for Detection of Theft Activities for Security in Banks and Jewelry Stores

Project Guide:

  • Prof. Mrs. Swati Chandurkar

Students:

  • Shambhu Mohite
  • Pratik Patil
  • Tushar Patil
  • Prasad Patil

Domain:

  • AI : Machine Learning

Short Description:

The objectives of our system is to :

  • Detection of anomaly
  • Detect the mask of a person who may mean to harm others
  • Notify the ongoing situation to the related authorities
  • Create ML-based model which is capable of detecting suspicious activity
  • Implement action representation method which should be efficient to compute, effective to characterize. actions and can minimize the discrepancy between actions
  • Minimize the theft activities

KCE13 - Detection of Pulmonary Diseases from lung sounds using Multiclass Classification DL Algorithms

Project Guide:

  • Prof. Atul Pawar

Students:

  • Aditya Dawadikar
  • Anshu Srivastava
  • Neha Shelar
  • Gaurav Gaikwad

Domain:

  • Machine learning

Short Description:

  • A huge number of deaths are caused every single year due to various pulmonary diseases such as asthma, bronchitis, pneumonia, etc. Affected people suffering from such pulmonary disorders have different breathing sounds as compared to the healthy people. This includes rhonchus, crackles, wheezes, stridor and plural friction rubs which are present in breathing sounds. In order to distinguish between a healthy and an infected breathing sound various parameter checks are used.
  • To reduce this process a lot of researchers have proposed various methods using computer science technologies like ML and DL classification algorithms which help us make early diagnosis. A lot of these research use a single ML/DL algorithm for example decision tree, KNN, SVM, CNN, RNN, etc. While few have proposed that combinations of two or more such algorithm have shown better results. For example, RNN with LSTM, CNN BiLSTM, etc. Here, we propose a method where we combine feature extraction techniques with DL algorithm to achieve better results.

KCE14 - Design and Development of a Web-based Tool for Idol Immersion Management and Analysis

Project Guide:

  • Dr. Mrs. Anuradha Thakare

Students:

  • Rushikesh More
  • Sagar Lahade
  • Sanjay Mahajan
  • Afrin Ali

Domain:

  • Design and Development of Tool

Short Description:

  • Every year, idols of lord Ganesh and Goddess Durga are immersed in local water bodies, such as lakes, rivers, back waters, or tanks. The material used to make and decorate idols has shifted away from biodegradable and reusable materials and toward non-biodegradable materials which contaminates these water bodies.
  • Lot of studies and research was done in the past by Govt. Agencies, NGO and individuals to identify the adverse impact to these natural water sources due to contamination. Central and State Pollution control board drafted the policies & suggestions to ensure these water sources are protected. However, there is so little done by society and individuals which is increasing the risk to ecosystem. The aim of this project is to research the studies conducted in the past, understand the Govt. guidelines, and justify the need to design and build an automated system/ portal which will assist artisans, customers and local authorities to track the idol right from its build (by artisan) to immersion (by buyer) and ensure government guidelines are adhered.

KCE15 - An automated system for rural people's complaint handling, monitoring and analysis using data analytics techniques

Project Guide:

  • Prof. Mrs. Sushma R. Vispute, Dr. Mrs. K. Rajeswari

Students:

  • Vrushali Kamble
  • Dhanashree Munot
  • Charudatta Potdar
  • Dhiraj Wakharde

Domain:

  • E- Governance, Machine Learning, Automation

Short Description:

  • In this era of digital world, the rural areas are still lacking in the usage of current technology. Their problems, queries and complaints are still done on a manual basis. The usage of handwritten letters is still their choice of expressing their problems to the authority people wherein not much of the complaints are taken care of. We intend to provide a solution in the form of a website which will act as a direct bridge between people and their leaders.

Information Technology Projects

KIT01 - Verifying Authenticity of Products based on QR Code to Avoid Counterfeiting

Project Guide:

  • Dr. Mrs. Jayashree Katti, Prof. Mrs. Priyanka Manke

Students:

  • Pranav Kulkarni
  • Kunal Patel
  • Ajinkya Tilekar
  • Abhijit Unavne
  • Sahil Hemnani

Domain:

  • Security

Short Description:

  • The market of counterfeit products is rising day by day. They are often manufactured as cheap copies of original products. Almost every company faces the threat of counterfeits because not only it affects company’s revenue but it also damages brand reputation. QR codes can be used as an effective and low cost solution that can help the industries and customers to check reliability of the product. Generating and printing a QR code on the product for identification is a simple and cheap process. The proposed system uses QR code and a hashing algorithm to maintain integrity of the product.

KIT02 - Remicare: Medicine Intake Tracker and Healthcare Assistance

Project Guide:

  • Dr. Mrs. Gulbakshee Dharmale

Students:

  • Pratiksha Shirsath
  • Abhishek Shinde
  • Vishwajeet Sawant
  • Aditi Chougule

Domain:

  • Artificial Intelligence and Machine Learning

Short Description:

  • An automated reminder mechanism is built in this Android-based application. It emphasizes the contact between doctors and patients. Patients can set a reminder to remind them when it is time to take their medicine. Multiple medications and timings, including date, time, and medicine description, can be programmed into the reminder by using image processing. Patients will be notified through a message within the system, as preferred by the patients. They have the option of looking for a doctor for assistance. In this Covid-19 pandemic situation where nurses have to remind the patients in the hospitals to take their medications, our application can be useful, alerting the patient every time of the day when he/she has to take the medicine and in what amounts. Also, all the necessary tests report and prescriptions can be saved on the cloud for later use. Patients will be provided with doctor contact information based on their availability. Also, patients will be notified of the expiry date of the medicine and the former history of the medicines can be stored for further reference. The proposed system prioritizes a good user interface and easy navigation. Image processing will be accurate and efficient with the help of powerful CNN-RNN-CTC algorithm. It also emphasizes on a secure storage of the user’s data with the help of the RSA algorithm for encryption and the gravitational search algorithm for secure cloud access. We attempted to create a Medical Reminder System that is cost-effective, time-saving, and promotes medication adherence.

KIT03 - Anti Horn System

Project Guide:

  • Prof. Mrs. Rohini Pise

Students:

  • Apurva Jagtap
  • Shubham Kadam
  • Nupur Kulkarni
  • Omkar Walhekar

Domain:

  • Internet of Things

Short Description:

  • The impact of noise pollution on the well being of human beings is severe. It not only affects the health of human beings but also has several psychological, physical and cognitive issues according to the statistics worldwide. This collectively results in negatively impacting the metropolitan cities, where noise pollution is a nuisance due to several reasons like continuous honking in no honking zones. Traditional methods have low efficiency to solve the problem. This problem can be solved efficiently using appropriate technologies like wireless communication, embedded systems etc. which can make the complete process fully automated. To curb noise pollution in No Honking Zones, an Anti Horn System is proposed. It aims at improving the overall quality of life for the people in Smart Cities by drastically reducing the honking and in turn reducing the noise pollution to a great extent in such areas.

URLs:

Civil Engineering Projects

KCI01 - Development and Optimization of Various Mixes of Concrete and Other Eco-friendly Material in 3DConcrete Printer

Project Guide:

  • Dr. D. S. Lal

Students:

  • Chinmay Poohale
  • Vaibhav Nivangune
  • Amit Hole

Domain:

  • Concrete Technology, Automation in Civil Engineering

Short Description:

  • Concrete is a construction material composed of cement, fine aggregates (sand) and coarse aggregates mixed with Water which hardens with time. But, In 3D Printing technology, Materials play key role in 3D Printing Construction. Concrete is one of the materials widely used in construction of houses. 3D printed concrete is a special type of concrete, which can be deposited through a 3D printer layer by layer without any formwork support and Vibration process. Concrete required in 3D Printing makes it Special than Normal concrete due to its Various Special Properties required in 3D printing is Maximum Compressive Strength, High Workability, Consistency, High early strength Gaining, Maximum flow ability in mixture, High Build ability after Pouring of mixture , Optimum Speed of Concrete Setting. For that, there is a need of selection of Optimum admixtures, Proportions of cement & other economical materials which need to be checked by experimentation for give better results at the time of Printing.

URLs:







KCI02 - Proposing the Design of cycle infrastructure for Nigdi Pradhikaran, Pune

Project Guide:

  • Prof. Praful Shinkar

Students:

  • Nikhil Bhanavse
  • Shubham Bhosale
  • Shivam Bondre
  • Aditya Rudrakanthwar

Domain:

  • Transportation Engineering and Planning

Short Description:

  • The five parameters used to analyze what makes a city a successful bicycling city are: planning, land use, policy, infrastructure and culture. India, being the 3rd largest consumer of bicycles in the world has also shown tremendous growth in the bicycle market since a few months. In post covid scenario a huge burden on cycling infrastructure is expected hence to have user friendly cycling infrastructure a demand-based planning is necessary. Through this project we will be focusing on cycle friendly infrastructure that includes: cycle network, junction, signage, public bicycle system.

KCI03 - Design and Development of 3D Concrete Printer

Project Guide:

  • Dr. D. S. Lal, Dr. R. N. Phursule, Mr. M. M. Narkhede

Students:

  • Kunal Patil
  • Shekhar Aditya
  • Tadge Shruti

Domain:

  • Concrete Technology, Automation in Civil Engineering

Short Description:

  • 3D printing or additive manufacturing (AM) is any of various processes for making athree-dimensional object of almost any shape from a3D model or other electronic data source primarily through additive processes in which successive layers of material are laid down under computer control. Digital fabrication methods are slowly starting to become popular in the construction industry. The elimination of formwork, less labor force, and the flexibility in the design and geometry are some of its advantages. In this method of construction,structures are built with layer-by-layer extrusion of concrete through nozzle which is attached by one arm and this whole assembly is supported by a giant frame. The major properties like extrudability, buildability, and mechanical properties are shown to be interconnected. Any discrepancy in any one of these properties can lead to a failure of the printed structure.
  • Mix required for 3D Concrete printing process needs to have characteristics like pumpability, extrudability and buildability so that requires structure can be printed without use of any type of formwork. Concrete needs to be flowable while pumping to avoid jamming of system and harden quickly enough to sustain load of subsequent layers. These characteristics are associated with behavior of concrete in fresh state and that can be controlled by including optimum dosage of suitable admixtures within mixing process.
  • In the current project, it is proposed to digitally twin the mechanism, concrete nozzle and mixing units with software to control and synchronize for effective use.

URLs:













KCI04 - Automated Layout Marking Vehicle and its Benefit to the Construction Industry

Project Guide:

  • Dr. D. S. Lal

Students:

  • Akash C. Govalkar

Domain:

  • Surveying, Automation in Civil Engineering

Short Description:

  • From past years’ progress achieved by technology in surveying is incredible. Some of the instruments like total station and scanning by laser are most popular and extensively used on construction field. Previous studies attentive on increasing the productivity of marking works by the instruments is classified into the substance technology expected on the aim to the marking works efficiency which is done by manual process and for the automated marking by using robots, workers in the construction will able to focused on building parts installing. Therefore, it will help to increase the productivity of the work.
  • The Automated Layout Marking Vehicle is an effective machine that would make the process of layout marking more efficient, faster and much easier to use. Because of the nature of the material (limestone), it is usually marked manually with bare hands with the help of rudimentary methods for precision such as the use of ropes for straight lines. Though these methods may suffice for immediate need, they may be highly inaccurate and require lots of man power and time. But, in case of ALMV this marking is done with the Automated paint dispersion device which needs only a lime paint and it demarcate the accurate marking on ground. Though there exists appliance which benefit in marking the line in case of paints however, presently there is no automation is used in the lineout marking. This process is completely automated and the robot would make a marking upon inputting the dimensions only.

URLs:











KCI05 - Effective use of Programmable Aerial Vehicle (PAV) In Construction Industry for Layout Co-Ordinates Marking

Project Guide:

  • Dr. D. S. Lal

Students:

  • Akshay M. Pudke

Domain:

  • Surveying, Automation in Civil Engineering

Short Description:

  • A building's or structure's layout depicts the design of its foundation on the ground surface, as depicted in its drawings, so that excavation may be carried out precisely where needed and the building's location and orientation can be precisely described. Layout marking is traditionally done with a peg and thread technique. Because everything was done manually, certain mistakes were discovered on the site. Manual mistakes, instrumental errors, and environmental variables such as humidity, temperature, and strong wind all contribute to this inaccuracy. Additionally, because of the pandemic scenario, labor has migrated to their home location, which has had a direct impact on labour availability on site.
  • A Programmable Aerial Vehicle (PAV) is being developed in this study to mark the layout co-ordinates of civil structures. This PAV instrument will be built using the surveying technique of selecting a station point using at least two reference points before beginning to build out the construction. The paint dispersion module, which is fitted with the vehicle in the center of the bottom, is used to indicate the layout of the building.
  • The creation of a Programmable Aerial Vehicle (PAV) will aid in the demarcation of building layouts. The PAV will assist in reducing the cost of layout activities while also performing a variety of other management responsibilities. PAV will improve the precision of dimensioning while also improving the quality of the job.
  • Due to its simplicity of deployment, cheap maintenance costs, great mobility, and capacity to hover, PAVs may be utilised in a variety of civil applications. These vehicles are used for real-time traffic monitoring, wireless coverage, remote sensing, search and rescue operations, commodities delivery, security and surveillance, precision agriculture, and civil infrastructure inspection, among other things.

URLs: