ML “Machine Learning”

Although we are still likely many decades away from anything considered close to “true AI”, many cameras and analytics systems are marketed and sold as being AI-driven or otherwise enriched through machine learning. While some of this is undoubtedly just marketing, there is truth to the notion that data analytics is making an impact in the field of video surveillance.

The ideas behind machine learning reach back into the earliest days of computer science, but only recently (within the last 5-10 years) has both the volume of data and processing power necessary become affordable enough to start making real strides. There are of course the famous industry examples like Google’s DeepMind and IBM’s Watson, but even without access to supercomputer hardware like that there are still many useful things that can be accomplished with a strong, stable connection to the cloud and a quality camera to create good data for analytics to process.

As they employ a subset of Artificial Intelligence, it would be more accurate to refer to these “smart” devices as Machine Learning, (ML) devices, that is if ever the average consumer becomes weary of the term AI being thrown around every industry from phone apps to kitchen appliances.

The way this type of technology works is through building a mathematical model using sample data, sometimes referred to as “training data”. This model is used to make predictions or decisions without the need for a programmer to directly intervene in the process. Machine learning is particularly important in the field of “Computer Vision”, the science and technology behind a computer’s processing and analysis of images and video. Because of this, ML has come to reflect a considerable portion of industry leading video surveillance and analytics systems. Historically it has been impossible to use conventional algorithms to perform such a task effectively.

Instead of trying to answer the question “Can we build machines that think?”, ML proposes that we try to “build machines that do what we (as thinking entities) do”.

“Smart” Cameras

Here we will take a look at the effect of this emerging technology and its practical application in video surveillance and analytics.

Let’s take a look at what exactly is possible using intelligent video. Smart IP cameras are able to classify the objects they see on their own. Objects entering or leaving the scene can be identified as a person, car, bike, truck, or other object. Because the camera can differentiate objects, it can be told to only alert in the case of a break-in, ignoring things like leaves in the wind or animals wondering through. New low-light cameras allow color-filtering even in scenes with very little ambient light. Even at night, color detection is possible in this way.

Alarm detection can be set to be even more specialized. Rules can be configured to look for specialized behavior patterns such as fighting, running, loitering, path following, abandoned luggage, entering an area, and more. The alarm engine in each camera coordinates with the others in a logical way to interpret this information and determine threat status. All this allows for a very robust alarm condition solution and prevents false and missed alarms.

What’s more, stored metadata enables forensic analysis at a later time. This means that retroactive searches for a specific car or person is possible even if it was not a determined item of interest until well after the event was recorded. Metadata is compact and only barely adds to the size of recordings. It is quick and easy to search through to find a specific event.

The logical next step is to continue to improve analytics for video metadata until we approach 100 percent practical use. Predictive analysis of human traffic patterns can predict shoplifting and identify potential events before they take place, and the more data that can be made useful the more accurate these types of predictions will be.

The same technology can monitor customer dwell time at different displays in a store and determine the effectiveness of in-store advertising and product locations. Analyzing customer engagement with these displays can help increase customer engagement with products and lead to increased sales and revenue. As the IoT expands, this type of technology will be more and more critical as there will be many more points of data to analyze. There is no way to fully anticipate the eventual effects this will have on our industry or the world at large.

Even though huge volumes of video data are collected every day, most statistics indicate that only 10 percent of this data is ever used. The majority of data collected loses its value very quickly after being generated. The reason for this? Our primary focus tends to be delivering the correct information in a crisis or providing it as evidence after criminal activity has taken place. This causes much data to be “wasted” in the sense that we lose out on the opportunity to perform useful analytics.

Video analytics is an increasingly powerful tool. It helps to improve usability of these vast amounts of video information. Analytics software acts as the “brain” of a surveillance system and is built into IP cameras themselves or processed in separate computing infrastructure. This creates a smarter system that “knows” what it sees and alerts guards to potential threats as soon as an alarm rule or condition is met. Analytics gives operators the chance to act faster and more efficiently with better intel.

ML Analytics is like an ever-vigilant system operator within the security system itself. Content analysis information, a form of video metadata, is stored as well. As they reduce operating costs and increase efficiency, intelligent cameras deliver a solid return on investment which can be measured in tangible results to the business or other setting in which it operates.

Security Technology of South Texas is happy to offer custom access control and surveillance solutions with video analytics to the greater South Texas area, designed either turn-key and from the ground up, or integrated into an already existing series of cameras.

Please contact us through email at admin@gostst.com on our website or via phone at 210-446-4863  24/7 to schedule a consultation.

The term “Internet of Things” was coined in 1999, gaining significant traction in 2011 after a report by Gartner added it to a list of emerging technologies. As more companies worked on advancing and creating new applications for IoT, the technologies involved gained ongoing global coverage. In its inception people often referred to the IoT as “embedded internet” due to its increasingly pervasive nature and presence in many aspects of our lives. Of course with any emerging technology there are associated risks, and these threats will grow and evolve as the technology does. In this article we will look into these risks, where IoT is moving, and the ways in which organizations are mounting defenses for their networks.

IoT is an umbrella term to include all devices with IP addresses connected to the internet. There are presently five types of IoT applications.

These include:

• Consumer IoT–e.g. Light fixtures, connected thermostats and alarms, and systems such as Nest

• Commercial IoT–these applications include healthcare and transport, connected pacemakers and other medical implants or wearables, and vehicle to vehicle communication

• Industrial IoT–including network connected control systems, smart agriculture, and big data

• Infrastructure IoT–this aspect of the IoT deals with network connectivity of smart city applications such as surveillance cameras, facial recognition, and traffic analysis devices

• Military IoT–including application of IoT technologies in the military and police, to include network connected robotics and wearable biometrics for police and infantry

The technology underpinning the IoT allows users and systems to connect seamlessly to a wide array of networks and expands connectivity between physical and digital systems. With organizations and governments prioritizing this move into the cloud, the technology and protective measures must race to keep up with demand.

The number of IoT devices worldwide have been growing at a rapid pace from the late teens:

• 2018–7 billion IoT devices

• 2019–the number of devices more than triples to over 26 billion

• 127 Devices are connected to the web every second

• By 2025–more than 75 billion devices are expected to be connected

In the year 2020 it is predicted that 93 percent of enterprises will adopt IoT technology, 90 percent of cars will be web connected using IoT technology, and 3.5 billion cellular IoT connections will be installed.

According to the 2018 Open Web Application Security Project (OWASP), the most significant vulnerabilities for IoT technology include:

1. Weak, guessable, or hardcoded passwords—such as short, simple, and publicly available passwords.

2. Insecure or unneeded network services—which are installed on the device and may expose data such as sensitive and financial information to theft and eavesdropping.

3. Insecure ecosystem interfaces—external interfaces that connect to the device. The connection may compromise the device and its components.

4. Lack of secure update mechanism—such as un-encrypted data moving from outward sources towards the device, and poor security monitoring.

5. Use of insecure or outdated components—such as open-source and third-party components that weren’t scanned for vulnerabilities.

6. Insufficient privacy protection—failure to protect private information that is stored on the device and connected ecosystems.

7. Insecure data transfer and storage—such as the lack of access control and encryption during the movement of data.

8. Lack of device management—on devices deployed in production; results in poor security support.

9. Insecure default settings—the inability to fix insecure settings creates exploits in devices and systems.

10. Lack of physical hardening—creates a larger attack surface, which threat actors can leverage to take control of a device or system.

More IoT components mean a greater attack surface is exposed. The more points connected to the network, the greater the risk. Endpoint Detection and Response (EDR) tools can be employed to monitor endpoints and send alerts for critical security events. It is also important to scan devices before allowing connection to your network in order to prevent the introduction of vulnerabilities. Vulnerability scans on a regular basis help to ensure the health of the network.

It is also important to segregate network infrastructure to allow least exposure to the internet. This can be done by creating a dedicated network for IoT with limited access.

Moving into the new decade we can expect IoT devices to become more and more embedded in many aspects of our lives, both personal and professional. The technology enables a move towards digital transformation with many industries moving into the cloud. From the protection of personal devices to the defense of an entire network, it is critical that IoT security be taken seriously. With more connectivity comes increased risk of exposure. The more we entangle the physical and the digital, the more real the results of an attack or security leak become.