Dynamic Face Recognition System

Product Overview

With world-leading technology in the field of deep learning, SenseNets boasts mature technology and business and much market experience in the security industry. The company takes vitalizing video value as its mission. In an era of accelerated urbanization and increasingly complicated population mobility and composition, the preventive action and management concerning personnel security in the urban public security sector are facing huge challenges. It remains difficult to obtain valuable information in an accurate and thorough manner as various restrictions are placed on artificial retrieval of information about the target objects despite the more and more prevalent application of video monitoring system. Therefore, how to collect, capture and cope with information about relevant objects in an intelligent, efficient and rapid way becomes a pressing matter for the users.

SenseNets rolled out the “dynamic face recognition system” based on computer vision and deep learning to meet the demand for applying face recognition in the public security field. The system provides a range of functions, including structuralization analysis of an array of videos, real-time face capture and surveillance, alarm handling through multi-terminal interaction, and personnel search, comparison and tracks analysis. On this basis, it effectively raises the ability to monitor special crowd, have the terrorists under surveillance and track the objects involved in a crime in the public security field, and thus plays an active role in social order and management, criminal investigation and analysis, and anti-terrorism and stability.

System Composition

  • Dynamic surveillance

  • Intelligent search

  • Personnel tracks

  • Face comparison

  • Operation and maintenance management

  • Statistical analysis

  • System configuration




Major Functions

1.Dynamic surveillance

The picture of a face captured in a real-time manner is monitored and compared with that in the surveillance database before a real-time alarm signal is generated and sent to the police for a rapid action based on multi-terminal interaction.

2.Face structuralization

A structured analysis of the face can be made for the real-time video or image to obtain information about the attributes of the person, such as gender, age, glasses and mask, which is helpful to identify the target object quickly through a direct retrieval of related attributes.

3.Tracks analysis

Searching-for-image-by-image or conditional retrieval comes as an effective tool to outline the spatio-temporal tracks of the target on the map and quickly learn the whereabouts of the suspect for the convenience of ensuing tracking and arrest.

4.Face search

A multi-dimensional retrieval method is available for assisting in confirmation of the real identity of the object on the basis of a quick search for his face.

5.Face comparison

Two pictures of face can be analyzed to help judge whether they represent the same person in light of their similarity.

Product Features

Structuralization of a multitude of real-time videos makes it possible to garner and store key information and search for and locate the target quickly when needed, thus greatly increasing the study and judgment efficiency.

Utilization of the deep learning algorithm, coupled with vast amounts of training data, is an effective solution to the problems of bad adaptability and low accuracy of recognition that exist in the traditional algorithm affected by the light, angle, scenario and other factors.

Multi-level networking application covering the platforms of province, municipality and county is supported.

The application model of interactive multi-terminals including the PC and mobile terminal offers customers access to more flexible application scenarios in a way that features full utilization of and connection between all sorts of resources and information.

It is well positioned to meet the demand for surveillance, study and judgment, management, and operation and maintenance in the practical public security business, accompanied by a deep optimization of the product functions and interactions with high efficiency, practicability and convenient operation.

Application Scenarios

Personnel information management

Dynamic personnel surveillance

Identity lookup and confirmation

Tracks collision and comparison

Real identity search

Examination on consistency between face and credentials

Evidence collection regarding the person involved in a case

Emergence plan for safeguarding stability in critical situations

Public security for large-scale activities

Face recognition-based mobile police affairs

Follow NetPosa on WeChat
Top Back Share