Mexico. Deep learning, typical of Artificial Intelligence (AI), helps detect anomalous behaviors in people, to prevent and attend in time to outbreaks of violent acts in sporting events, such as those presented in the city of Querétaro on March 5.
The acts of excessive aggression between the fans of the football clubs of Querétaro and Atlas, presented at the La Corregidora Stadium during day 9 of the LIGA BBVA MX, left a balance of 26 seriously injured, images that went around the world, upcoming games behind closed doors, millionaire fines and several sanctions, of which the imposition of Mikel Arriola, president of Liga MX, to credential with photograph (Fan ID) to all football bars for the month of April stands out.
These types of events, which tarnish sport and show how violence manages to permeate multiple aspects of human life, require strategies of legal control and strategic prevention. Among the latter, technology stands out as a surveillance tool to detect, attend and contain outbreaks of riots and quarrels in time.
In that sense, Mexico has the General Law of Physical Culture and Sport, which was published on June 7, 2013, with a last reform in September 2018. Chapter VI focuses on the prevention of violence in sport and establishes the promotion of solidarity attitudes, the culture of peace, legality and non-violence in any type of its manifestations. As well as the creation of the Special Commission Against Violence in Sport, and the Sports Electoral Surveillance Council (COVED). That is, within this law it is contemplated "to promote the mechanisms and actions aimed at preventing violence in sporting events and guaranteeing peaceful development in the enclosures where massive sporting events are held and for entertainment purposes and in their vicinity, as well as the safety and heritage of people, in coordination with the Public Security authorities, Private and Civil Protection corresponding".
Technology as a response
However, Dahua Technology explained that facial recognition, based on deep learning algorithms, allows control of entry and exit of people in an enclosure, as well as the capture of faces in real time, which can be added, compared and analyzed in order to generate prevention actions, such as sending alerts to a command center to activate protocols in case of emergency.
Dahua's solutions, for sports spaces, contemplate from the integration of video surveillance cameras, access control, Starlight technology to capture color images in low light conditions, video analytics that detect various attributes in visitors and, finally, the unification of all this for the generation of alerts in a matter of seconds.
Rodrigo Escamilla, Director of Business Development at Dahua Technology Mexico, explained that "applying deep learning in video surveillance makes behavioral analysis possible. By examining an activity over an extended period, the system can use deep learning to establish normal patterns of behavior of people or moving objects. This means that after analysis, abnormal behaviors such as people climbing a wall or entering a restricted area can be detected."
Thus, AI, with the help of deep learning, allows to detect aggressive behaviors and generates alerts, but to this must be added the work of security agents, law enforcement agents and others involved in the security of the various sports scenarios, which are ultimately responsible for responding to the reports of the control and surveillance rooms, containing, dismantling and prosecuting those who commit acts of vandalism and aggression.
