Urban Space Insights Extraction using Acoustic Histogram Information
This work offers a low-cost, low-bandwidth solution for urban residential activity tracking, potentially benefiting smart city services and sustainable development.
This paper explores using low-cost analog sound sensors to detect outdoor activities and estimate rain in urban residential areas. Sound data, transmitted as histograms every 5 minutes, is processed with wavelet transformation and PCA to create robust features, which are then used for unsupervised clustering to identify activities.
Urban data mining can be identified as a highly potential area that can enhance the smart city services towards better sustainable development especially in the urban residential activity tracking. While existing human activity tracking systems have demonstrated the capability to unveil the hidden aspects of citizens' behavior, they often come with a high implementation cost and require a large communication bandwidth. In this paper, we study the implementation of low-cost analogue sound sensors to detect outdoor activities and estimate the raining period in an urban residential area. The analogue sound sensors are transmitted to the cloud every 5 minutes in histogram format, which consists of sound data sampled every 100ms (10Hz). We then use wavelet transformation (WT) and principal component analysis (PCA) to generate a more robust and consistent feature set from the histogram. After that, we performed unsupervised clustering and attempt to understand the individual characteristics of each cluster to identify outdoor residential activities. In addition, on-site validation has been conducted to show the effectiveness of our approach.