New Swarm Intelligence Technique of Artificial Social Cockroaches for Suspicious Person Detection Using N-Gram Pixel with Visual Result Mining

Introduction And Problematic

The bio inspired is a philosophy that consists to analyse the living world in order to translate biological models of all forms (animals, plants, micro-organism, where ecosystem) to technical concepts or industrial development. The world of living is immense; it represents a large pool of ideas for researchers and companies for this reason, for each encountered problem we must observe the nature (Bouarara, 2014).

The first part of our work is to develop a new algorithm called artificial social cockroaches (ASC) inspired from the lifestyle of cockroaches, their way of communication as a decentralized system without a conductor using the odour pheromone and their antennas. The cockroach is attracted by the darker place (less luminosity), and follows the path of their congeners.

Lately, the security of human and information has become incontestably a major issue globally. Face to problems such as fighting terrorism, strengthening internal security and the rise of cybercrime; our societies are investing more and more to ensure their protection by using the technology of surveillance-video. Nowadays, the surveillance cameras are ubiquitous everywhere (streets, shops, museums, metro stations, ticket machines, shops, airports, and banks....etc.) for the purpose to detect malicious persons (thieves, criminals, terrorists). Roughly, the use of this technology has allowed a drop in crime, either in public and private places. It provides several advantages:

• •

  The resolution of criminal cases with the possibility to film at night with infrared;

• •

  Improving safety by reducing the illicit acts in the streets;

• •

  Dissuade suspicious persons by using the dummy cameras;

• •

  A source of evidence against malicious persons.

The research within this area remains in process, in order to enhance the efficiency of the cameras in term of response time and the quality of videos recorded to obtain a best identification. Merely, the use of surveillance-video in the social security, and law enforcement to identify suspicious person through facial recognition can collide with several drawbacks.

• 1.

  The inability to detect the malicious persons that disguise himself, or hide their faces with masks.

For e.g. we have a surveillance camera located in a house and there's a thief with a hidden face that enters this house for stealing, as presented in Figure 1b. The classical systems of surveillance video cannot recognize the identity of this person because they are based on facial recognition.

• 2.

  The classical surveillance-video systems are unable to warn people before the malicious act.

  • For e.g. the terrorists or wanted persons can move in the streets and airports with false identity by altering their looks. These dangerous persons cannot be distinguished by the classical surveillance video system.

• 3.

  The assault of privacy individuals.

The employment of classical surveillance-video system (based on facial recognition) in offices, streets, business establishments, and in front of the buildings does not respect the freedom of persons passing closer to the cameras.

For e.g. the use of surveillance-video system to film “reality shows” that shows us the daily life of people.

Figure 1.

Thieves with hidden face