The Advantages Of K9 Tech Security AI Surveillance Technology
AI is the buzz word of the day. Many people had good experiences, but many also found out that some merchants use AI as a gimmick with little value. This article explains why our AI video surveillance and detection platform is an important part of our innovative property security solution and how it works to provide reliable security solution to our customers. AI video surveillance property security has become increasingly popular.
We chose the CAWAMO video surveillance analytics and management platform after reviewing and testing many other AI systems. Above is a screen from our demo site showing a detection of a person by our CAWAMO test system. CAWAMO have developed a unique AI based technology that enables us to provide a very reliable and effective solution to our clients as was demonstrated in our demo video.
The developer of any such system must have an effective way of teaching the system to be able to determine that an image is a person or an animal, for example. CAWAMO’s patented AI platform technology allows their developers to select an object and indicate that this is a person. They can do this during the development process using multiple objects until the system learns what a person is. Their AI algorithm can use these examples and learn how a specific collection of digital pixels represents a person. Once their system learns a specific object, system developer can repeat the process within the context of a larger image. For example, once the system learned to identify a facemask, the developer can teach the system how a facemask would look on various people.
Once this basic development process of a detecting an object is done, CAWAMO can further improve the detection accuracy. CAWAMO does it by showing their AI model how the detection objects look in various video clips taken by security cameras. This semi-automated process improves the detection capability to work on real-life surveillance video streams.
Therefore, CAWAMO’s AI development platform enables them to quickly develop new detection objects. The two objects that we use most often are Person Detection and Robber Detection. In the future, we may use a detection of a person holding a weapon. In some applications, CAWAMO can determine whether a vehicle that entered our customer’s property is suspicious or not. It does this by scanning its license plate or just comparing the overall vehicle image to images of vehicles it learned belong to friends or service providers. This is a great example of how AI video surveillance property security can be enhanced.
The AI based learning does not stop with the system development process. CAWAMO provides a utility to help users to improve detection. Please refer to the above picture. It has a Thumbs Up and Thumbs down icons. Once the system displays detection and a user presses on the Thumbs Up icon tells the AI algorithm that the detection was correct. When a user presses on the Thumbs Down icon, the system asks the user what the object was in real-life. For example, was the object an animal or something else. Given the user answer, the AI algorithm adjusts and learns the new rules. For example, a dog standing on its back legs could look like a person. After receiving user feedback, the system learns to avoid such an error in the future.
CAWAMO has functionality to avoid repeating alerts. This can prevent a customer or a monitoring center receiving many erroneous alerts. For example, an engine attached to a boat in a marina moving up and down on waves may look like a person. This erroneous detection can overwhelm the monitoring system with repeated alerts. CAWAMO AI can limit the alerts repeated in a pattern limiting the number of erroneous alerts received by a monitoring center.
CAWAMO has other important tools to help improve detection in real-life situations. For example, it has functionality for the system to ignore static objects as these can interfere with detection of moving objects. For example, parked cars in a parking lot can interfere with analyzing moving cars. By telling the system to ignore static objects, a user can considerably improve accuracy. This functionality adds another layer of AI video surveillance property security.
CAWAMO can perform its analytics both on the edge and on the cloud levels. The term “edge” refers to the physical location that is being protected. We can also configure how much of the work is done at what level. This is important as our platform is capable of delivering detection even when a network connection to the cloud servers are interrupted or slowed. Video clips from the security cameras can also be stored at the edge or the cloud (cloud storage can be very expensive).
The above image demonstrates another important aspect of CAWAMO’s platform functionality not directly related to AI technology — its video camera and alert management capabilities. Many competing AI analytics platforms do not include camera management functions. Security system integrators must integrate AI analytics with camera and alert management software to provide a complete solution. We liked that CAWAMO offers a rich capability in this area. This allows us to quickly on-board a new security monitoring partner or do the task in-house at a lower cost. The same capability is also available to customers who want to do their own video alert monitoring.
Leveraging CAWAMO’s AI surveillance technology and their camera and alert management platform, K9 Tech Security is capable of providing a very reliable and reasonably priced property and home security service to our customers. Of course, we do not stop with the AI technology — we leverage highly trained protection dogs to provide an immediate response and protection. This combination makes AI video surveillance property security highly effective.
Please refer to Our Technology page for more information on how we use AI video surveillance in our property security solution.