Unmet Need: Automated detection and validation of Police-citizen interaction
Current events demonstrate that there is a resurgence of interest in police behavior – namely disproportionate use of force, including lethal force by police officers. This renewed interest surrounds high profile events. These incidents, and a myriad of others, focus attention on if and when biased use of force is present in an agency. It indicates that a substantial number of studies seek to understand use of force and the situational and dynamic factors associated with that force. Policing scholars have long argued that currently available data are ill-suited for understanding police use of force.
The Technology: Interaction validation using body-worn camera footage
This technology describes a new program and process for the automated detection and validation of anomalous police-Citizen interactions using body-worn camera footage. In this technology, WSU researches develop a client specific classifier using machine learning techniques and biometric measures (audio, video, and biometric) to detect anomalous police-citizen interactions, including but not limited to instances of heightened emotional activation.
• Indicative of changing stress levels or organizational changes in the overall tone of police-citizen interactions, including but not limited to increased decibel levels.
• Application Program Interface (API) provides daily, weekly, or monthly reports indicating deviation from client specific baselines.
• Better capture of the totality of a use of force incident and generates a more in-depth understanding of what police use of force looks like and the processes by which it occurs.
• Automated process and report generation.