Saving lives with deployment of video analytics at sea
Source: Man Overboard! — Security Today
Many of us have gone on a cruise vacation or know of a close friend or relative who has done so. Cruising is a popular vacation option enjoyed by families and individuals throughout the world. What is not so well known, though, is that man overboard events continue to be a common occurrence within the cruise industry.
Since 2005, 268 man overboard events have been reported. On average, 22 people fall off a cruise ship every year, and 86 percent of those victims do not survive, or are never found.
Man overboard events (MOB) can happen at any time during the day or night, in all types of weather, sea conditions, and from almost any location on the ship, ranging from a few tens of feet above the water, to over 180 feet. Falls from these heights can injure, render unconscious, or even kill the victim. Likewise, falls from ships operating along northern routes can expose the victim to extreme water temperatures, which can result in hypothermia and death.
Given these scenarios, one approach to increasing the survival rate of MOB events is to ensure accurate timely detection, followed by a rapid response protocol. However, few cruise ships in operation today have any form of man overboard detection system. Although they may have cameras and imaging systems to monitor activity on board the ship, these camera systems are primarily used for basic surveillance or evidential activities, and are not designed to actively notify the ship’s crew in the instance of a person falling overboard. In most cases, MOB events are reported by someone who happens to notice the fall, or by a member of the person’s party who reports the person missing. This information must all be verified by the crew before taking action, which can take a considerable amount of time.
The facts are that the typical stopping distance of a cruise ship is one mile. If it takes 15 minutes to confirm an MOB event, a ship may have moved seven miles from the original fall location and a typical man overboard detection system can report a MOB event in less than one second.
Detection Using Intelligent Video
Accurately detecting a human falling from a moving cruise ship is no an easy feat. An object falling from the top level of a large cruise ship can reach speeds of 70 mph. A MOB system must be able to detect objects moving at these speeds, while mounted on the ship itself, which is in a constant state of pitch, roll and yaw. Cruise ships are also very busy environments, with people moving about, blowing debris and a variety of normal operational activities that may involve the movement of items along the side of the ship (ladder deployments, crew boarding, water spray from decks, etc.). Although most cruises tend to target fair weather locations, they still experience all types of weather conditions and sea states. A reliable man overboard detection system must accurately detect human falls in all these situations, while avoiding false alarms and providing the crew with actionable data to react to the situation. Although a challenging problem, numerous companies currently deploy MOB solutions.
PureTech Systems, recently announced the approval of a patent addressing the use of video technology for man overboard detection. Its MOB system captures images using thermal cameras surrounding the ship’s perimeter, extending from the lowest passenger deck to the waterline. Thermal video is a logical choice for maritime situations, as it is less susceptible to harsh lighting scenarios, such as sunrise and sunset. It can also provide clear images day or night and in most weather conditions. High resolution thermal sensors can therefore deliver clear visual indications of a human target to the crew for confirmation.
The use of geospatial video further enhances the detection capabilities of this type of MOB solution. Geospatial video is the understanding of where each video pixel resides in “real” space—meaning latitude, longitude and elevation. This adds another dimension to video analysis by allowing the software to not only understand the physical location of the object, but also the real size, speed and acceleration. Therefore, although an object may be falling at the same expected speed of a human, a geospatial video solution can understand that this same object is too big or too small to be a human, and suppress the alarm.
The use of opposing cameras provides a means by which the video analytics can further confirm the event is happening on the ship itself, and is not something occurring in the background, such as on the water, in the sky or on the dock. In addition to actually being detected in both images, a detection occurring between two camera pairs must pass several additional tests for the duration of the fall; including time stamp, object size, object location, object speed, object type and fall trajectory on both cameras within the pair before being considered an alarm. This further ensures that the crew receives minimal false alarms.
System Integration and Crew Notification
In the event of a man overboard situation, accurate detection is only part of the problem. Systems must also provide a high level of situational awareness to help the ship’s crew gain a full understanding of the event as quickly as possible. MOB systems deployed with geospatially video analytics have several unique features.
Image data. Humans inherently rely on vision as one of the primary senses; the human mind will rely on its sense of sight as a means to confirm any other data received. Using video analytics for detection means data preferred for validation by a first responder— still images and looping video—is instantly available to reduce the time required to confirm the event. To further ensure rapid verification, video analytic systems highlight the potential target with graphical “bounding boxes” on both still images and full motion video clips in order to quickly communicate what event raised the alarm.
Location data. In addition to logging the GPS location of the ship at the exact time of the event, video solutions leveraging geospatial data also provide insight as to the location the event took place on the vessel itself. Including location information with each alarm allows first responders to more quickly reach the incident location and begin MOB procedures without the loss of valuable time typically required to confirm an event and its point of origin.
Industry Response and Regulations
Technology providers are not the only institutions taking actions around the need for MOB detection and the deployment of solutions. The United States government identified a need for increased measures to detect MOB situations in 2010 when it released the Cruise Vessel Security and Safety Act (CVSSA). The act applies to cruise ships carrying more 250 passengers on international voyages in which passengers embark or disembark in any US port. It calls for automatic manoverboard detection/monitoring systems within 18 months of the technology becoming available.
Since the first release of the act, several companies have stepped forward to prove their detection capabilities, prompting the current revision of the act and the development of an international standard for these man-overboard systems by the International Organization for Standardization (ISO).
Various cruise lines are also proactively testing the new technology ahead of the formal release of the ISO standards and the CVSSA requirements to ensure they understand how to most effectively incorporate this technology as an integral part of their safety systems.
While having tremendous potential to cater to the current requirements of the marine industry in detecting, recording and reporting man-overboard events in real time, this technology is directly applicable to the broader safety category of “long fall events.” In addition to other maritime industries, including ferry operators, oil platforms, commercial shipping and military vessels, these types of falls can be a safety concern to other nonmaritime industries, such as bridges.
The cruise line industry has recognized the need for man overboard detection systems and many manufacturers have invested research and development to create technology which results in accurate detection and notification, including systems based on the use of geospatial video analytics. These video-based systems monitor thermal images, process them using intelligent video algorithms and combine this data with geospatial techniques. This results in not only highly accurate detection capabilities, but also a complete solution which supplies crew members with critical event location data to reduce the delay between the time of the event and the deployment of rescue procedures.