Explore common analytics terms and phrases by scrolling through this alphabetical list or using your keyboard to find (CTRL+F) the definition you're looking for. Although this list is not all encompassing, it is meant to outline the most prominent terms and definitions one may come across when discussing cybersecurity in the physical security and surveillance industry.
Looking for a more in-depth conceptual explanation that goes beyond a simple glossary? Download the following whitepaper to help connect the dots with challenging analytics terms and concepts as they relate to artificial intelligence:
AI in video analytics whitepaper
Artificial Intelligence (AI)
A broad concept associated with machines that can solve complex tasks while demonstrating seemingly intelligent traits.
Machine Learning (ML)
A subset within AI that uses statistical learning algorithms to build systems that have the ability to automatically learn and improve during training without being explicitly programmed. This allows computers to improve their algorithms by training on existing examples instead of requiring a human to refine the algorithm.
Deep Learning (DL)
A refined version of machine learning which leverages a neural network (see definition) to combine layers of feature extraction in deep structures of rules to produce an output. These are also trained in a data-driven manner. The algorithm can automatically define what features to look for in the training data.
Neural networks
A family of algorithms that are used to recognize relationships in datasets through a process that is somewhat similar to how a human brain works. A neural network consists of a hierarchy of multiple layers of so-called "nodes" or "neurons" which are interconnected, and information is being passed along the connections, from the input layer, through the network, to the output layer. For example: an object recognition system might be fed thousands of labeled images of cars, houses, coffee cups, and so on, and it would find visual patterns in the images that consistently correlate with labels to better detect these objects in unlabeled images.
Computer vision
The discipline of making computers understand what is happening in a scene by analyzing images or videos. In order to make the data recognizable for a machine or computer, a data annotation process is necessary where the relevant objects are categorized and labeled. This is essential to AI success and is often labor-intensive by humans.
Training (learning)
When a model is fed annotated data and a training framework is used to iteratively modify and improve the model until the desired quality is reached. In other words, the model is optimized to solve a defined task.
Hardware acceleration
While you can often run a specific analytics application on several types of platforms, using dedicated hardware acceleration achieves a much higher performance when power is limited. Hardware accelerators enable power-efficient implementation of analytics applications. They can be complemented by server and cloud compute resources when suitable.
GPU (Graphics Processing Unit)
GPUs are mainly developed for graphics processing applications, but they are also used for accelerating AI on server and cloud platforms. While sometimes also used in embedded systems (edge), GPUs are not optimal for machine learning inference tasks from a power efficiency standpoint.
MLPU (Machine Learning Processing Unit)
An MLPU can accelerate inference of specific classical machine learning algorithms for solving computer vision tasks with very high power efficiency. It is designed for real-time object detection of a limited number of simultaneous object types such as humans and vehicles.
DLPU (Deep Learning Processing Unit)
Cameras with a built-in DLPU can accelerate general deep learning algorithm inference with high power efficiency, allowing for a more granular object classification.
Go back to analytics resource page
