Student Magazines & Best Projects

CSE Wall-Magazine

Best-Mini Projects Videos (2018-2022 Batch)

Dataset generation using portable devices to develop dl models for eye disease detection

Computer aided diagnosis for eye diseases is receiving a lot of attention from researchers. In order to interpret the diseases directly from a model it needs to be fed using a large set of fundus images under various categories should be included. The datasets that are available are collected from the clinical environment using a high-end opthalmoscope. The scarcity of large publicly available datasets generated from the portable devices has been a bottle neck for the researchers for the successful completion of computer-aided diagnosis. This project designs a portable device that acts as a camera to capture fundus images.

Happiness Index and Mental well-being of students during pandemic using Machine Learning Algorithms

COVID-19 Pandemic Brought significant changes to our daily lives as our movements are restricted in support of efforts to contain and slow down the spread of the virus. Mental illnesses are found in people of different age groups. This study aims to collect the real time data of students from educational institutions through different survey techniques. The dataset is then pre-processed to remove any abnormalities and make it fit for training the machine learning algorithm. Based on the responses given by the students, the happiness index is measured and mental health of the individual is analyzed using machine learning techniques. Through this analysis individuals can understand their state of mind and take timely action.

Determination of happiness index of teachers during Covid

Covid pandemic indirectly impacted people’s mental health conditions, which can be overwhelming but difficult to measure. The level of mental health of a teacher has been found affected with numerous personal as well as professional demands. The proposed model detects any early symptoms of depression in the teachers by detecting the person’s mental state and their happiness level based on the vocabulary they use while answering the questions because when people write, they leave a trace of their mental state.

Color Depiction for daily sounds

The Color Depiction System would aid the elderly and disabled community to identify and react to various common events occurring in their surroundings through light emitted from the Smart Bulbs, when they are lit on activation by surrounding sounds. Some common events being opening the door when someone rings the doorbell, attending to a phone call or to respond to a person, etc.

Gesture Controlled Voice Assistant For Deaf And Elderly

A Gesture Recognition System integrated to a voice assistant, helps the elderly and disabled by offering a perpetual computing interface that captures and intercepts hand gestures. These gestures are converted into text (that can be displayed on the web interface) and speech and are processed at the back-end giving back the response in the form of text, voice and a series of gestures which are comprehensible by the end user.

Automated Interview Feedback System using NLP

This project tells the interviewees whether they are hired or not and if not in which round, they failed. It also recommends the users. This recommendation is like a feedback for the interviewee who did not get hired. This feedback tells the interviewees how they performed in their interview and in which areas they are lagging due to which they lost their opportunity so that they can focus more on it and excel in it.

Voice assistant for elderly people

The main idea of this project is to assist the elderly people in their day-to-day activities and communicate with them on the basis of their requests and functioning their requests in regional language.. A feature will be added to alexa which will receive the requests from the user in regional language and decodes the request, processes and communicates with the user in regional language. Alexa will be trained in the regional language and the request is decoded using translation techniques.

EyeX – A Vision Check Application

This system aims to make the process easier by making it easy for users to regularly keep a check on your eye health by a series of activities and tests. And based on the results and insights you can also book an appointment with your nearest Optometrist using this system and will receive information about confirmed bookings.

River network identification from Satellite Imagery using Machine Learning Algorithms

In this project the proposed model enhances and detects the complete river networks. Enhancement and feature extraction can be done using filters such as Gabor, PCA, and GLMC. Gabor filter helps in enhancing the river cross-sections and longitudinal continuity. PCA(Principal Component Analysis) reduces the dimension (columns) of a dataset. GLCM(Gray level cooccurrence matrix) is formulated to obtain statistical texture features. GLCM counts the number of pixel transitions between two pixel values.

VISUAL PATTERN ANALYSIS FOR CLASSIFICATION OFTHE VISUAL BEHAVIOUR OF AN INDIVIDUAL
The scan path, sequences of fixations of the eyes on an image, provides data related to the locations and durations of the gazes that can be used to develop visual patterns to analyze the visual behavior in order to measure the concentration of the person. Our project can also be used as a screening test to diagnose disorders such as autism spectrum disorder and to observe the attention of the customer on various products on e-commerce sites.

GNITS CAMPUS ASSISTANT

GNITS Campus Assistant is similar to Google Assistant. General queries are solved with the help of Google Assistant. Similarly, details of the college can be known using GNITS Campus Assistant. This simple stand-alone web application helps less tech-savvy people or seldom visitors and students to know minor details of the college.

SYNTHESIZING REALISTIC ARMD FUNDUS IMAGES USINGGENERATIVE ADVERSARIAL NETWORKS (GAN)

Generative Adversarial Networks (GAN) is an approach to generative modelling using deep learning methods. GAN will be trained with fundus images from the Age- Related Eye Disease Study (AREDS), generating synthetic fundus images with ARMD. The performance of ARMD diagnostic DCNNs will be trained on the combination of both real and synthetic datasets. Images obtained by using GAN should appear to be realistic, and also increase the accuracy of the model. Performance of the deep learning model which uses the synthesized dataset should be close to the real images, suggesting that the dataset can be used for training humans and machines.

CLASSIFICATION OF RIVER NETWORK IN SATELLITE IMAGERY USING DEEP LEARNING TECHNIQUES

Classify the multi-spatial satellite images into river network classes and other classes using machine learning algorithms such as random forest and convolutional neural networks. Therefore, a framework constituting a three-step processing is proposed, which briefly includes image segmentation, feature extraction, and target classification. It is expected that the results to outperform other methods in producing accurate recognition results for detecting river network using multi-spatial satellite imagery.

NICE: Nervousness Indictor in Classroom Environment

NICE (Nervousness Indictor in Classroom Environment) wrist band identifies whether the student is under stress or not based on various parameters like heartrate and temperature of human body. This data can be attained by using various sensors such as Electrocardiogram (ECG), Galvanic Skin Response (GSR), etc. The product is exclusively built for students to identify the various stress inducing factors in the classroom environment on a daily basis.

AUGMENTATION OF GLAUCOMA FUNDUS IMAGES USINGGENERATIVE ADVERSARIAL NETWORK

Generative adversarial network (GAN) is an unsupervised machine learning technique which can be used to augment datasets and yield the collected images to be indistinguishable from the real-world data. Realistic images can be synthesized using the GAN-based data augmentation. This proposed model provides effective high-resolution processing of images. The data synthesized using data augmentation improves the performance of model in detecting the glaucoma disease at the earliest.

An Assistive Tool for Autistic to Teach Facial Expressions

Facial Emotion Detection is an approach towards detecting human emotions through facial expressions. This project aims to detect the emotions of autistic children from the expression of their faces. Emotions are sad, happy, anger, surprising, disgusted, fearful, neutral. To detect the emotions of autistic children, image processing and machine learning algorithms are performed.

QUICK-WITTED BEDSHEET FOR PATIENTS

The Quick-witted bedsheet continuously examines the physical movements of the patient and reports it in case of any abnormal conditions. This is done with the help of IMU sensors. The bedsheet is equipped with a device to check menstruation for bedridden women.

SHARP-WITTED BED FOR PATIENTS

The automated bed system uses sensors to continuously measure physical conditions such as movements, heart rate, and respiration rate, while the patient is bedridden. Magnetic massaging is done to remove the sores that often occur over bony prominences and other joints as a result of a prolonged lack of blood flow to the affected area, which is achieved through the pressure sensors.

OBJECT RECOGNITION AND VOICE ASSISTANT FORVISUALLY IMPAIRED

It allows the blind victims to identify and classify Real Time Based Common day-to-day objects, calculates distance to produces warnings whether he/she is very close or far away from the object and generates voice feedbacks.