In this paper, we propose a method to predict the personal air quality index in an area by using the combination of the levels of the following pollutants: PM2.5, NO2, and O3, measured from the nearby weather stations of that area, and the photos of surrounding scenes taken at that area. Our approach uses the Inverse Distance Weighted (IDW) technique to estimate the missing air pollutant levels and then use regression to integrate visual features from taken photos to optimize the predicted values. After that, we can use those values to calculate the Air Quality Index (AQI). The results show that the proposed method may not improve the performance of the prediction in some cases.
The need to know the personal air pollution data is vital because it
is better to provide each individual with regional air quality data,
which seems to be more accurate than the global data measured
from far away weather stations. The problem is that the
performance of personal air quality prediction mainly interpolated from
public weather data is not good. This paper reports our solution to
tackle this challenge by finding out whether pictures of places can
improve the prediction results. To know more about this challenge
and the dataset that we will use, you can refer to the overview paper
of MediaEval 2020 - Insight for Wellbeing: Multimodal personal
health lifelog data analysis [
The experiment on using surrounding images to predict the air
quality has been conducted in several projects. For instance,
analyzing the sky images [
Because of the time limitation, we have to propose a method that
does not require an incredible training time. At first, we will use
the pure form of IDW technique [
Road surface Road River Walkway Architecture Thoroughfare
Confidence Score predicted values with an additional visual feature to produce new pollutant levels. 3.1
We use Google Cloud’s Vision API to extract information about entities in images. Each image will have a maximum of 10 labels that have the highest confidence score. For example, Table 1 shows labels of the image GH030011_005250.jpg. We create a boolean feature from those labels to define whether that location is an open space or not. It means that if an image has one of the labels in Table 2, it will be a picture of an open space area, and therefore, the __ feature has the value of 1 and vice versa. We believe that those areas usually have better air quality, so it is the reason why we use the __ attribute as a supplemental input. 3.2
The first step is to use the IDW to predict pollutant levels of PM2.5, NO2, and O3 for each hourly time frame from the known values of pollution data provided by 26 weather stations surrounding Tokyo. These predicted values will be the first input of our regression model and the second one is the __ attribute created when we extract visual features in section 3.1. We continue to fit the regression model with these two independent variables to make the prediction.
Our linear regression model has the following formula: = 1 + 2 with is the value of the pollutant level needs to be predicted, 1 is the value of the pollutant level predicted by IDW, 2 is the __ attribute, and , are the coeficients. Finding those coeficients means that the regression model will be fitted. (1)
The evaluation of PM2.5, NO2, and O3 prediction in the case of not using visual features and vice versa, provided by MediaEval task organizers are shown in Table 3, Table 4, Table 5, Table 6, Table 7, Trable 8, respectively.
In general, PM2.5, O3, and NO2 prediction results are improved, except for the case of NO2 levels of the two running courses course1test1, REFERENCES course1test2. The reason behind this could be because we did not cluster the images of each course separately. 5
We are currently investigating more advanced algorithms, such as
implementing the combination of IDW with multiple regression [