Artificial intelligence for video surveillance utilizes computer software programs that analyze the audio and images from video surveillance cameras in order to recognize humans, vehicles, objects and events [1]. Security contractors program is the software to define restricted areas within the camera’s view (such as a fenced off area, a parking lot but not the sidewalk or public street outside the lot) and program for times of day (such as after the close of business) for the property being protected by the camera surveillance. The artificial intelligence (“A.I.”) sends an alert if it detects a trespasser breaking the “rule” set that no person is allowed in that area during that time of day.




The A.I. program functions by using machine vision. Machine vision is a series of algorithms, or mathematical procedures, which work like a flow-chart or series of questions to compare the object seen with hundreds of thousands of stored reference images of humans in different postures, angles, positions and movements. The A.I. asks itself if the observed object moves like the reference images, whether it is approximately the same size height relative to width, if it has the characteristic two arms and two legs, if it moves with similar speed, and if it is vertical instead of horizontal. Many other questions are possible, such as the degree to which the object is reflective, the degree to which it is steady or vibrating, and the smoothness with which it moves. Combining all the values from the various questions, an overall ranking is derived which gives the A.I. the probability that the object is or is not a human. If the value exceeds a limit that is set, then the alert is sent. It is characteristic of such programs that they are self-learning to a degree, learning, for example that humans or vehicles appear bigger in certain portions of the monitored image – those areas near the camera – than in other portions, those being the areas farthest from the camera.


In addition to the simple rule restricting humans or vehicles from certain areas at certain times of day, more complex rules can be set. The user of the system may wish to know if vehicles drive in one direction but not the other. Users may wish to know that there are more than a certain preset number of people within an area. The A.I. can maintain surveillance of hundreds of cameras simultaneously. Its ability to spot a trespasser in the distance or in rain or glare is superior to humans’ ability to do so.


This type of A.I. for security is known as “rule-based” because a human programmer must set rules for all of the things for which the user wishes to be alerted. This is the most prevalent form of A.I. for security. Many video surveillance camera systems today include this type of A.I. capability. The hard-drive that houses the program can either be in the cameras themselves or can be in a separate device that receives the input from the cameras.


A newer, non-rule based form of A.I. for security called “behavioral analytics” has been developed. 


This software is fully self-learning with no initial programming input by the user or security contractor. In this type of analytics, the A.I. learns what is normal behavior for people, vehicles, machines, and the environment based on its own observation of patterns of various characteristics such as size, speed, reflectivity, color, grouping, vertical or horizontal orientation and so forth. The A.I. normalizes the visual data, meaning that it classifies and tags the objects and patterns it observes, building up continuously refined definitions of what is normal or average behavior for the various observed objects. After several weeks of learning in this fashion it can recognize when things break the pattern. When it observes such anomalies, it sends an alert. For example, it is normal for cars to drive in the street. A car seen driving up onto a sidewalk would be an anomaly. If a fenced yard is normally empty at night, then a person entering that area would be an anomaly.




Screenshot 2019-05-20 at 16.55.34


In this ever-expanding era of Surveillance Data Technology (SDT), deep intelligence will become the foundation for the security industry. Technologies that “learn” will become more common and more powerful. This trend will strengthen critical security efforts in every sphere. Now DeepinView Cameras and DeepinMind NVRs will lead the way in this new world of surveillance technology by making invisible intelligence visible for users, and then putting that intelligence to good use.




  1. Traditional Intelligent Algorithm


Traditional Intelligent Algorithm ADVANTAGES OF DEEP LEARNING The number of video surveillance devices and the sheer amount of data are both rapidly increasing in their own rite, while the traditional intelligent algorithm continues to operate only on the surface. Current systems suffer from:


  1. From “Shallow” to “Deep”


The algorithmic model for deep learning has a much deeper structure than the two 3-layered structures of traditional algorithms. In deep learning, an original signal passes through layers of processing; next, it takes a partial understanding (shallow) to an overall abstraction (deep) where it can perceive an object.


  1. From “Artificial Features” to “Feature Learning”


Deep learning does not require manual intervention but relies on a computer to extract features by itself. The more features there are, the more accurate the recognition and classifications will be


Facial Detection is only the beginning for Deep Learning features. With analytics that take business solutions into the next century, and with automatic alarm accuracy above 90%, this technology is a step above and beyond anything video surveillance as we know it. Deep Learning technology filters out insignificant objects and movements in a scene that would trigger normal alarm systems. 

Vehicle data gets recorded and shapes the database to perform numerous security functions. People Counting systems give businesses an advantage in marketing and conversion efforts. Deep Learning technology solutions add value on multiple levels.


  1. Facial Detection

Facial Detection software will analyze images and determine the presence of a human face. When a face is detected, the system captures its position, size, and expression. The video stream will judge whether there is a human face. If so, the position, size, and main features will be recorded. Identifying characteristics can be obtained from this information. When compared against recorded human faces in a database, a face can be identified. Facial comparison is the process by which structured data information operates after data modeling and analysis for the human face.

  1. False Alarm Filter

The False Alarm Filter enables the system to perform secondary recognition for human body targets in behavioral detection events (line crossing detection, intrusion detection), effectively reducing the false alarms caused by shaking leaves, shadows, light variations, vehicles, small animals, etc.

  1. People Counting

The People Counting function counts people entering, exiting and passing by a specific scene, such as in a supermarket or museum where large crowds move through on foot.



Good Anti-Interference Ability Height Filtering Loitering Filtering


  1. Vehicle Structured Data


Vehicle Structured Data refers to a bivariate table formed by extracting the vehicle information, like license plate number, vehicle color, model, brand, sub-brand, etc., and is used for vehicle information retrieval.





  1. Human Body Search


The search by human body picture feature enables the system to use a provided human body picture to find matching images and information in recorded footage.





Effective protection of citizens, their property and public areas, is a concern for city authorities around the world. In Safe City Project, Deep Learning technology is adopted to identify specific personnel’s and analyze human and vehicle behaviors. This can be used for locating a fugitive at large, finding lost people, preventing potential crimes, and detecting parking violations.


Facial Recognition


Behavioral Analytic’s 


Intelligent Traffic


  • Video Vehicle Detector
  • Traffic Capture Unit
  • Traffic Incident Detection
  • ANPR
  • Parking Violation Detection