Analytics in Action: Locating a Lost Child

If you are a parent, losing sight of your child – if even for a moment – can be a scary experience. But what if your child wanders off at an amusement park or gets lost on a cruise liner during a family reunion? Scary might be an understatement. In these trying situations, you expect security and law enforcement officers to leverage everything at their disposal to return your child — safe and sound — with speed and accuracy, especially when each passing minute significantly impacts a successful outcome. In these situations, safety professionals can rely on video analytics capabilities to locate missing persons quickly and effectively.  

The Power of Advanced Video Search

When it comes to video search, leveraging capabilities, such as appearance similarity and – though more controversial – facial recognition, can make a significant impact. These enable users to pinpoint potential matches for a missing person based on an existing image or video footage of the individual. Many people don’t realize that some facial recognition technologies merely match or verify individuals by correlating biometric features against a source image alone, like digital images extracted from video or pre-defined watchlists. Often, when facial recognition is used, law enforcement or security cannot derive information about the person’s identity – they can merely locate a potential match based on an available image.

However, even when officers are running a search or investigation armed only with a description of the missing person, robust video analytics technology can still reduce time-to-target with the application of appearance similarity analytics. This solution is especially useful in situations when an image is unavailable, in locations where facial recognition is regulated or prohibited, or when camera conditions are not sufficient for accurate facial recognition.

In these 2 stories from the field, BriefCam experts were able to empower security teams to locate missing children using appearance similarity during on-site training sessions:

Missing Child on Cruise

While training the security operators of a major US cruise line, the team was notified that a child was missing. As the atmosphere immediately switched to high alert, our BriefCam implementation expert was able to support and empower them to optimize their search using video analytics. Because the parents did not have a current picture on hand, it was more expedient to use appearance similarity filters based on their description of the child. After plugging in generic identifying traits into the BriefCam Video Analytics Platform- such as child’s clothing color and gender – security officers were able to quickly narrow the search across the vessel’s 300 cameras to locate children matching the given description. After organizing the results into a reverse timeline with the click of a button, the team was able to further filter out irrelevant objects. In less than 10 minutes from the time the child was reported missing, the newly empowered security team was able to locate the missing child on one of the lower decks of the cruise liner.

Lost Child at Resort Casino

Another similar occurrence happened at a resort with an on-site casino. The parents frantically alerted security that their child was missing – without knowing the casino was in the midst of a BriefCam on-site training. The family didn’t have a picture of the child readily available, so once again, the security team had to rely on Appearance Similarity and other video analytics filters to locate the child. By first narrowing the video results to “children” our BriefCam expert was able to seamlessly guide the security operators though the steps to identify the child by filtering with the unique criteria provided by the parents.

The Advantage of Comprehensive Analytics

Facial recognition is an important technology. In certain scenarios, it is the solution that can deliver the best outcome by detecting possible matches for a unique identity in an existing image. Always, validating the accuracy of the match requires human assessment and confirmation. However, because facial recognition is not always a viable or effective option, other video analytics – such as vehicle or clothing color and appearance similarity for identifying potential matches based on non-biometric physical attributes – can be leveraged to detect, identify, and monitor the appearance of relevant objects in video.

BriefCam is committed to bringing you the most comprehensive video analytics platform in the industry. With a robust and ever-expanding feature suite, we are here to empower you to translate video into impact for your organization.