mercoledì, 08 set 2010
Login

Computer Vision

head_cognitive

Vision and Perception

Computer vision allows our solutions to see what a camera sees. The platform takes any visual input and then analyzes the images frame by frame. With its perception engine, it uses machine intelligence to emulate the functions of the human brain. First, the software identifies objects (any human, animal, vehicle, or inanimate object) and then classifies them, and tracks them frame by frame. our tool also improves the detection of the objects. It reduces distortions, shadows and glare, and improves resolution, which gives better clarity and recognition of objects than is currently available with existing software.

Adaptive Learning

It is in the area of adaptive learning that our solutions are most uniques and differentiated from any other existing technology. Through adaptive learning, they recognizes and anticipates behavior based on knowledge acquired over time. Behaviors are not pre-defined and hard-coded as they are in other technologies. Instead, our AIs rapidly learns what is normal for any environment and identifies and predicts abnormal, suspicious or aggressive behavior based on what it has learned. Upon recognizing such behavior, they generates responses like real-time alerts to a human operator who can take immediate action, or other local based interactions depending on project’s requirements.

The more history our AI has of a particular environment, the better informed its behaviors become. By using these techniques, our adaptive-learning solution significantly reduces false positives and greatly improves efficiency.

Cognitive-Based vs. Rules-Based

Every environment and every scene is unique. No one is able to write enough rules to cover the infinite number of possibilities for any given environment. This is why it is important to have a cognitive-based system that is able to learn what is normal for every unique environment and then recognize when there are events and/or activities that occur outside of that normal pattern. That same learning capability is also important in order to adapt to changes that may occur within any given environment over time. These two capabilities: the ability to adapt to almost any scene or environment and the ability to continue to improve upon its learning and alerting over time are the most important distinguishing factors of cognitive-based systems over rules-based video analytics systems. The benefits to businesses that adopt cognitive-based video analytics systems over rules-based systems can include everything from reduced cost due to less required coding and customization, increased effectiveness from reduced false positive alerting, and increased return on investment on the entire security infrastructure.

A Natural Evolution

Just as the information technology business has evolved to include adaptive pattern-detection solutions, so will video analytics and video management solutions. Building systems along these lines will produce the (actually under development) capability to expand the use of video analytics systems across multiple cameras connected to the same system, such as providing the ability to track behavioral patterns from one camera to another. Future capability will also allow operators or “backend” systems the ability to teach the platform through “supervised” learning activities such as teaching the system that one alert included characteristics that will be consistently of interest to other operators, and agents, no matther if humans or not.

EZD sees cognitive video analytics as the actual stage in the natural evolution of computer vision systems. Cognitive video analytics is enhancing perception and awareness and will continue to increase the scalability and effectiveness of operations over time.

Why EZD solutions

EZD is one of the fiew companies that has consolidated a method to connect the three technologies required to produce a cognitive video analytics system. Computer Vision, Video Analytics, and Machine learning are all required to produce this revolutionary product. Until now there has been no “consistend” way to connect these technologies, computer learning could not be connected because there was simply no way for it to understand input from video. This is why ALL other video analytics products adopted a “rules based” technology. The systems have to rely on a pre-programmed list of rules to base its observations on because there was no other way to do it.

The C.A.V.E. (Cognitive Analysis of Video and other inputs from EZD’) program was created to discover and invent a way to connect video (and potentially many other inputs) to machine learning. This is not to be confused with “artificial intelligence”. AI has been employed in video analytics from inception, but machine learning is an entirely separate and grossly advanced technology. Now, with the completion and deployment of our tool in facilities like private buildings, turistical sites, ports, government facilities and other high risk sites, EZD has proven that the Cognitive-based model is the only viable method for surveillance.

Cognitive-Based vs. Rules-Based
Every environment and every scene is unique. No one is able to write enough rules to cover the infinite number of possibilities for any given environment. This is why it is important to have a cognitive-based system that is able to learn what is normal for every unique environment and then recognize when there are events and/or activities that occur outside of that normal pattern. That same learning capability is also important in order to adapt to changes that may occur within any given environment over time. These two capabilities: the ability to adapt to almost any scene or environment and the ability to continue to improve upon its learning and alerting over time are the most important distinguishing factors of cognitive-based systems over rules-based video analytics systems. The benefits to businesses that adopt cognitive-based video analytics systems over rules-based systems can include everything from reduced cost due to less required coding and customization, increased effectiveness from reduced false positive alerting, and increased return on investment on the entire security infrastructure.
A Natural Evolution
Just as the information technology business has evolved to include adaptive pattern-detection solutions, so will video analytics and video management solutions. Building systems along these lines will produce the (actually under development) capability to expand the use of video analytics systems across multiple cameras connected to the same system, such as providing the ability to track behavioral patterns from one camera to another. Future capability will also allow operators or “backend” systems the ability to teach the platform through “supervised” learning activities such as teaching the system that one alert included characteristics that will be consistently of interest to other operators, and agents, no matther if humans or not.
EZD sees cognitive video analytics as the actual stage in the natural evolution of computer vision systems. Cognitive video analytics is enhancing perception and awareness and will continue to increase the scalability and effectiveness of operations over time.
Why EZD solutions
EZD is one of the fiew companies that has consolidated a method to connect the three technologies required to produce a cognitive video analytics system. Computer Vision, Video Analytics, and Machine learning are all required to produce this revolutionary product. Until now there has been no “consistend” way to connect these technologies, computer learning could not be connected because there was simply no way for it to understand input from video. This is why ALL other video analytics products adopted a “rules based” technology. The systems have to rely on a pre-programmed list of rules to base its observations on because there was no other way to do it.
EZD’s Cognitive video analytics program was created to discover and invent a way to connect video (and potentially many other inputs) to machine learning. This is not to be confused with “artificial intelligence”. AI has been employed in video analytics from inception, but machine learning is an entirely separate and grossly advanced technology. Now, with the completion and deployment of our tool in facilities like private buildings, turistical sites, ports, government facilities and other high risk sites, EZD has proven that the Cognitive-based model is the only viable method for surveillance.