AQI Predictor
Declining air quality index is a serious problem of the world and many International Organizations have worked to mitigate the problems arisen by Air Quality Index (AQI). Now in Pakistan there is no application or software available which can judge the AQI of our environment and help us take certain
2025-06-28 16:30:16 - Adil Khan
AQI Predictor
Project Area of Specialization Artificial IntelligenceProject SummaryDeclining air quality index is a serious problem of the world and many International Organizations have worked to mitigate the problems arisen by Air Quality Index (AQI). Now in Pakistan there is no application or software available which can judge the AQI of our environment and help us take certain measures. We have decided to build an app that can judge the AQI of local area just by a picture. It is not only frugal but also user friendly as most of the hardware devices which measure AQI user need to have prior knowledge of device to operate it and are expensive to purchase and operate.
This accomplishment will help Government to tackle the areas and alleviate the problems of bad air quality by doing proper plantation or initiating a drive for plantation. Govt. can also hinder the working of factories which emit harmful chemicals in that area until area quality reaches at normal level and take necessary measures.
Project ObjectivesAQI (Air Quality Index) is a serious problem which is inevitably increasing and with the increase of it, the effect of Global warming on environment can be seen. Currently on Government level there is no policy being implemented to mitigate the problems occurring due to worst quality of AQI.
The Government of Pakistan at this level cannot get a fair idea that which areas need its special attention regarding the AQI or which areas are on severe threat of bad air quality. Our app will help to target specifically those areas where there is need of proper plantation so that air quality can be made better there by taking proper steps.
Moreover, main objective is also that public can judge the air quality of their surroundings up to 70 percent correctly and can take proper part in making their surrounding environment clean and green.
Project Implementation MethodThe front end of the system is developed on java using android SDK. Firstly, we extract the features from our dataset, save it on our database and then we train the model using Neural Networks. From user’s device image is send on a serve. Firstly, some features are extracted from the image then we send this feature to our train model. Model will predict the value of PM2.5 and from this value we tell the AQI. By getting user’s location, we can tell the user temperature, Humidity level. Our application will tell AQI in lower cost without any sensor just by clicking an image. To get more accuracy we collect real time data and for this purpose we use sensor (PurpleAir PA-II-SD). We take image of specific area and find the PM2.5 value of that area from sensor. We will attach that sensor with the tablet we are going to use (Insignia Flex) to use the data that we are getting from the sensor to generate results and perform calculations immediately. This way we will make our whole dataset which will consist of the image and the PM 2.5. We will utilize this dataset for the training of our model which will help us achieve a higher number of accuracy.
This application will help Pakistan Meteorological Department and other supporting authorities to determine in which area of Pakistan pollution is more and if it exceeds the bearable limit. This app will help the general public to take precautions in such areas where pollution level is worse or beyond the limits.
Benefits of the ProjectBuilding an app that would be able judge and decide the Air Quality of an environment can be really helpful and beneficial for our society and possibly the world. Following are some of the benefits
- Our app will be able to determine the Air Quality index of the local area with accuracy up to 70 percent. Here it is imprudent to mention that the AQI shown by google is the AQI of whole city but not at the specific point or location. Our app will handle this flaw and can judge the AQI from a picture of person’s surrounding environment.
- We have a plan to integrate the results taken by people and send it to the national servers of EPA (Environmental Protection Agency) so they can judge which area has bad quality air.
- Govt. can take necessary steps and can start a drive of plantation based on the results generated by our app.
- The end product would be such that there won’t be any need of hardware devices to determine AQI, only the software. Hardware would only be used in the making of the project and data training.
The final product will be an android application that can run on supported systems. The application will use the current location of the user’s as they turn it on through GPS. The application will predict AQI and update on server side and will be available for other users of relevant areas.
Final Deliverable of the Project HW/SW integrated systemType of Industry IT Technologies Artificial Intelligence(AI), OthersSustainable Development Goals Climate ActionRequired Resources| Item Name | Type | No. of Units | Per Unit Cost (in Rs) | Total (in Rs) |
|---|---|---|---|---|
| Total in (Rs) | 56903 | |||
| PurpleAir PA-II-SD | Equipment | 1 | 40404 | 40404 |
| Insignia Flex 8.9 | Equipment | 1 | 16499 | 16499 |