Due to increase in no. of cases of Covid, it's getting tough for hospital management and other premises to manage proper functionality of staff and patients too which results in failure of system as well as rise in cases.
Our team worked on four tracks which includes:
- Covid Analysis
- Mask Detection
- Room Occupancy
- Social Distancing
When it comes to infectious diseases data collection is difficult at the best of times. The rise of big data has provided clinicians and researchers with the systems and ability to store and work with large amounts of data, but in public health, critical surveillance systems remain slow to collect and difficult to disseminate.
Real time data acquisition and analysis of high-resolution data tends to focus on mortality and morbidity. Collecting accurate data and understanding the limitations of the data is an essential part of understanding the situation.
The data for this Project is available here 📉: https://www.mohfw.gov.in/
- Data Collection and Interpretation
- Data Cleaning
- Data Modeling and Prediction
- Data Communication
- Data Visualisation
- Infographics and Data Visualisation puts raw numbers into perspective
- Interactive and continually updated information of each state
- Continual monitoring of Cases draw from latest data
Humans have been hit by the pandemic globally, and wearing a mask has become a prerequisite. However, most of the citizens are abandoning to wear masks and being a source to spread the disease to others. Moreover, it is getting difficult for hospital management and higher officials at crowd-gathering places to keep a check on this issue. This resulted in an increase in the number of COVID-19 cases all over the world.
The current project aims to build a small model using Machine and Deep Learning with the inclusion of image detection techniques in Google Colab. It helps to edify the Face Mask Detection System, which is now seemingly gaining popularity and importance, especially at the smart hospitals for effective patient care. For safeguarding the health of citizens, the mask detector system needs to get executed at crowd-gatherings like shopping malls, warehouses, airports and many other places where foot traffic is hefty.
- Data collection
- Installation of required libraries like Tensorflow.
- Read the model
- Data Visualization where the prediction of wearing masks takes place
- Infographics and Image Classification on the basis of mask
- To check whether the person is wearing mask or not
During covid times, maintaining social distancing is necessary, whether it be rooms for patients or washrooms. But it's not possible to tell if a room is just used or not, which makes the next user or cleaner vulnerable to unhealthy circumstances.
Here, we are automating the process of checking room availability. The rooms in the facility are classified occupied or unoccupied based on passive sensor data such as temperature, humidity, light and CO2 levels. The model is based on logical regression for binary classification.
- Data Collection and Interpretation
- Data Modeling and Prediction
- Data Visualisation
- Data Classification
- Prediction of Model’s accuracy
- Effective utilisation of the available space i.e avoiding re-usage of the same room/restroom.
Social distancing is a prerequisite in the facility, however the task can’t be performed effectively by the workers, as 24/7 surveillance is not feasible.
With less staff onboard, it gets hectic for the workers to continuously keep a check on the social distancing norm being followed in the facility. So, in this track the task of checking social distancing is being automated. The system will create an alert "TO CLOSE!" if two individuals don't maintain a certain gap. Further, it could be connected to a sensor to create an alarm system in the facility.
- Installation of required libraries like CV2(Open_CV)
- Visualisations where we alert people to maintain a certain distance
- Inspects whether social distancing is followed or not