Vision Al Blog
The importance of risk mitigation for safety and security
The use of video surveillance technology worldwide is set to achieve record numbers in 2022. Used to enhance safety and security, video surveillance is projected to be a 69.1 billion dollar market by 2026 at a CAGR of 10%.
This is driven, in part, by the industry expanding beyond the traditional approach of monitoring only critical infrastructure. A growing focus for video surveillance includes implementing next-generation technologies.
Deep learning, AI analytics, video analytics, cloud computing, and edge computing all increase the capabilities of video surveillance software offerings and use-cases such as biometric-based physical access control, real-time watchlisting, post-forensic analytics, and much more.
The history of security camera technology
Initially used by governments and the military, the first known record of the security camera’s development was in 1927 by the Russian scientist, Léon Theremin (four years after he accidentally invented the electronic musical instrument bearing his name – the theremin). Theremin assembled a simple wireless system that connected a video camera and television together. This system was eventually implemented to watch visitors coming to the Kremlin in Moscow.
This very basic type of closed-circuit television (CCTV) was taken to the next level by the Germans approximately 15 years later when CCTV cameras were installed at an air force base to monitor the launch of rockets. After 1950, video surveillance started experiencing more mainstream adoption in a variety of commercial applications such as manufacturing and medicine.
Today, the Asia Pacific (APAC) region is currently the fastest-growing region in terms of video surveillance adoption, due to heavy investments in infrastructure and smart city projects. In regions such as Europe and the United States, digital privacy concerns and the lack of universal regulatory frameworks have resulted in a slightly slower adoption rate of video surveillance technology for private and public use.
Nearly 100 years after its invention, video surveillance is now found everywhere. China currently leads the world in video surveillance; it’s estimated that there is one camera for every two people in their population. The city of Delhi, India is ranked first in the world with the highest number of CCTVs installed in public spaces.
6 ways to use video surveillance technology
Video surveillance systems are used by 68% of companies today for security and to improve operational efficiency. For the purposes of security, this seems to be a wise approach.
Below we offer a few ideas on how firms can use video surveillance technology to best serve their own specific use cases:
1. Apply and review intelligent video analytics insights
Adding a level of intelligence to your existing camera infrastructure makes your existing cameras smarter.
Many enterprises exploring video monitoring and surveillance solutions are often confused by the amount of AI or intelligence built into modern CCTV cameras. Today’s CCTV cameras are definitely smarter than their predecessors, thanks to embedded chips, which effectively add “brains” to the camera’s visuals.
While video surveillance hardware has been in use for nearly 100 years, the video surveillance software behind the camera lens itself has changed dramatically in the past five years and continues to evolve rapidly. Smart video surveillance has exploded into the market as CCTVs are now enabled with the power of AI and machine learning – producing valuable insights from video feeds. No longer are CCTVs passively gazing at scenarios. Cameras are now actively engaged in analyzing, identifying, and providing actionable insights from video feeds.
Private and public sectors are using powerful and versatile video analysis systems to enhance the security and safety of people, properties, and assets across a multitude of environments. Video analytics produce valuable insights from video feeds. Smart video surveillance systems can today detect, track, and alert people of interest (e.g., known felons, VIPs, advantage players, employees, etc.) in real-time through the use of algorithms built on convolutional neural network technology. Such systems are also capable of analyzing live video or pre-recorded video files. Data can be retrieved from different sensors, then easily navigated, processed, and represented by employing a set of powerful analytic tools and unique algorithms.
Facial recognition can often recognize faces of people entering a space, but this ability often depends on environmental factors such as the quality of the lighting, angle of the person’s face, and degree of blur or occlusion (i.e., when a person’s face is hidden behind another person) as well as the quality of the actual cameras.
Oosto uses deep learning AI to eliminate many of traditional systems’ shortcomings, by accurately capturing faces in real-world environments, even with low bandwidth CCTV cameras. The AI software instantly alerts security staff of unauthorized entries or when a person of interest (POI) is recognized and dramatically reduces false-positive rates.
The actionable insights that Oosto’s visual AI provides offer far-reaching benefits of protecting people, properties, and assets. Oosto helps retailers catch shoplifters, identify employee theft, and reduce shrinkage with real-time video surveillance and forensics which lets security teams quickly search weeks of video footage to identify suspects and streamline an investigation. The system also notifies a security team whenever a known thief is spotted by a firm’s connected cameras, allowing teams to respond immediately to a threat in real-time.
2. Keeping bad actors at bay
Video surveillance with intelligent data analytics can identify persons of interest (POIs) in real-time with live facial recognition, enabling your security team to rapidly respond to threats while protecting the privacy of bystanders.
POIs could range anywhere from VIPs whom businesses would like to offer an enhanced level of customer experience, to advantage players in casinos, known felons, shoplifters, and dangerous criminals. Knowing who is on, and within close range of, your premises at all times provides a faster, more accurate, and appropriate method of response.
Recognizing and responding to threats to safety and optimizing customer experience is what drives security and operations teams, and their success depends on accurate and fast identification of people. The human eye alone is imperfect, and most software that automates the recognition of individuals on watchlists underperforms in real-world conditions.
Oosto offers an automated watchlist alerting system that identifies persons of interest in real-time while protecting the identity of bystanders. Oosto’s neural nets are trained in the toughest conditions on low-quality images and have been battle-tested by the most demanding users and academic standards to ensure the highest accuracy even when the POI is not looking at the camera, located in a crowd, or captured in less-than-ideal conditions (e.g., poor lighting).
Automated POI alerting even allows users to identify POIs and the people they have come in contact with through both face and attribute recognition in real-time. Moreover, the same solution can be successfully applied in historical footage to identify suspects in criminal investigations.
3. Compliance – Using video surveillance for anti-money laundering
Depending on the industry, compliance-related issues may impact operations significantly if not properly planned for and mitigated.
For example, in casinos, it’s important to know your customer at all times in order to identify potential money laundering threats.
In healthcare, anytime a facility or organization stores personal health identification (PHI) (whether physical or digital), they must ensure that the data is secure and private so that only authorized personnel can access that information. For computers, this can mean replacing traditional passwords with facial recognition technology-based access and encrypted file contents. From a physical standpoint, it can entail putting privacy screens on monitors, access control on doors to sensitive files, and security cameras around a facility to document access to areas with PHI.
In Banking, the facial recognition solution that integrates clear surveillance video with transaction data and highly-accurate analytics can also help uncover identity fraud. Video can be used to identify imposters, but it can also help ensure banking staff are following protocol by always asking customers for proper identification.
At Oosto, we take data privacy seriously. We don’t store our customers’ data or even interact with it. Our responsibility to comply with laws and regulations is intended to protect our customers and the public at large.
In fact, upon the detection of a face, we have enabled the option to blur the faces of bystanders during video playback that have not opted-in to being scanned. We also avoid unnecessary data capture. Oosto enables our customers to choose whether to record the faces of individuals not enrolled in the watchlist.
Once this feature is enabled, the images of faces of non-enrolled individuals will not be displayed in the gallery and will not be saved on the server. This enables our clients to avoid unnecessary compliance risk and only record and process detections of individuals on an organization’s watchlist.
4. Improve overall workplace safety
Video surveillance isn’t just used for external audiences and monitoring. It can also be used for internal workplace safety and ensuring equipment is working correctly, while spaces are clean and free of debris or other potential dangers. You can detect changes in conditions and get alerts for potential problems. Potential slip and fall accidents at retail establishments, for example, can be alerted to security teams before they happen.
Video surveillance provides several safety benefits as opposed to in-person observations. Just a few include non-biased and accurate accident data, in-depth analysis, observations possible not in real-time, observer not exposed to risks, undisturbed work activity, and the possibility of longer observation periods. Recorded videos can help identify safe practices, the best risk management approaches, and improve training sessions.
And security cameras can help ensure other types of safety – even those of a psychological nature. CCTV cameras can record instances of employee abuse or harassment. They can also monitor visitors who come into the building by helping keep track of any suspicious activity. When criminals see a CCTV system or employees know cameras are in place, they may think twice before carrying out any illegal activity.
5. Behavior recognition: The next level of protection
Video surveillance technology can track behaviors and objects to promote health, safety, and security. Oosto and Carnegie Mellon University’s (CMU) CyLab Biometric Research Center teams are developing biometrics technology that can perform across a broad range of safety-related use cases, including object detection (e.g., weapons on school grounds) and behavioral analysis (e.g., when someone falls down). The CMU partnership will focus on early-stage research in object, body, and behavior recognition which can be particularly useful in the prevention and investigation of work-related injuries.
Video surveillance can identify anomalies in behavioral patterns, including fall detection, drowning, human behavior analysis, and the like. For instance, falls are one of the most serious health risks among senior citizens over the age of 65, adversely affecting more people than stroke and heart attacks combined. Furthermore, the U.S. Centers for Disease Control and Prevention says that deaths from falls are increasing sharply among elderly Americans. As the population ages, the toll is likely to rise.
So, when we talk about our security and of the people we love and care about, we are somewhat limited in our reliance on products that provide a false sense of security. In most cases, by the time one sits down to look at the security camera footage, the damage is already done. There are countless scenarios where people are attacked, even killed, thinking they are safe because they have a security camera. Unfortunately, help only comes when it’s too late.
6. Crime-solving with forensics applications
Oosto has proven time and again that our technology is an invaluable asset for post-event investigations. It helps law enforcement as well as public and private security agencies to achieve a radically faster time to target by enabling rapid review and search of volumes of video that would otherwise be impossible. As a result, Oosto has become the de facto standard for police departments around the globe.
Most law enforcement agencies – large or small – turn to video surveillance footage as the first line of investigation. Oosto’s ability to ingest video from disparate sources – both cameras and files, and search across them all, quickly provides clues for investigators. The ability to start searching without knowing exactly what you are looking for is a breakthrough paradigm for post-event investigations, and investigators can now afford to review video even if the chances of the subject being present are small.
Moreover, while video surveillance backed by intelligent video analytics has proven controversial, public sentiment is supportive of the technology depending on the context-specific use case. Most citizens are uncomfortable with the use of facial recognition for live, real-time mass surveillance without a specific known or immediate threat. However, the public generally supports the use of intelligent video surveillance for use by police in a criminal investigation. Most agree that the technology can be exceptionally useful to detect the presence of known criminals or terrorists in public areas and to find missing people when every moment counts.
Enhance security with AI-driven video analytics
In brief, video surveillance technology for business is a powerful tool to help protect and preserve the people, places, and assets that we need to safeguard. Vision AI-driven software is a robust and welcome addition to a security team that often provides a near-immediate return on investment in terms of efficiencies gained, security enhancement, loss prevention, and peace of mind.
To learn more about video surveillance and how biometric software solutions can be integrated into your existing technology infrastructure, please visit https://oosto.com/contact.
