Monitoring Early Signs of Domestic Violence

Feasibility on Detecting Door Slamming towards Monitoring Early Signs of Domestic Violence

With the ever-increasing availability of low-cost micro- controllers and other computing devices, and advances in more lightweight machine learning techniques, it is becoming increasingly viable to make many of the everyday objects found in our homes smarter. By using low-cost microcontrollers and TinyML, we investigate the feasibility of detecting potential early warning signs of domestic violence and other anti-social behaviors within the home. We created a machine learning model to determine if a door was closed aggressively by analyzing audio data and feeding this into a convolutional neural network to classify the sample. Under test conditions, with no background noise, an accuracy of 88.89% was achieved, declining to 87.50% when assorted background noises were mixed in at a relative volume of 0.5 times that of the sample. The model is then deployed on an Arduino Nano BLE 33 Sense attached to the door, and only begins sampling once an acceleration greater than a predefined threshold acceleration is detected. The predictions made by the model can then be sent via BLE to another device, such as a smartphone of Raspberry Pi.

The COVID-19 pandemic resulted in many changes and restrictions for our daily lives, most notable being mandates to work from home where possible, as well as legal requirements to socially isolate. Between March 2020 and June 2020, police in England and Wales recorded a 7% increase in offences flagged as domestic abuse related, with the ONS noting a general increase in demand for domestic abuse victim support services (including a 65% increase in calls to the National Domestic Abuse Helpline between April and June 2020, compared to the previous quarter). Overall, the entire 12- month period between March 2020 and March 2021 saw an overall increase of 6% in domestic abuse related crimes. This follows general increases seen in previous years and could be associated with increased reporting by victims, in addition to improved recording by police forces. With this in mind, we seek to demonstrate the viability of using microcontrollers embedded within doors in the home, to assist social workers and law enforcement in the monitoring of potentially aggressive behaviors and perhaps the early signs of domestic violence in social housing.

Team

Outcomes