Object identification system based on data fusion and self-learning method of object identification

Abstract

The invention discloses an object identification system based on data fusion and a self-learning method of object identification. The present invention is characterized by using an identity sensing device and a feature sensing device to respectively sense the identity data and the feature data of each object entering a monitoring area; determining a group to which each piece of feature data belongs, associating the identity data and the group to which the feature data belongs when the sensing time of the identity data is consistent with the sensing time of the feature data, and pairing the group and the identity data when judging that the association frequency of associating any group to the same identity data is greater than an association critical value. According to the invention, the identity data and the feature data of the same object can be automatically paired, so that the time and the manpower required by the manual pairing can be effectively saved.

Description

technical field [0001] The invention relates to a system and a method, in particular to an object recognition system based on data fusion and a self-learning method for object recognition. Background technique [0002] Currently, a person identification system based on data fusion technology has been proposed. The aforementioned object recognition system can use different types of sensors to sense people to obtain different types of sensing data, and perform person recognition based on different types of sensing data. [0003] see figure 1 , is a flow chart of an existing learning and identification method. For example, a person identification system may include an RFID reader (a first type of sensor) and a video camera (a second type of sensor). [0004] The person recognition system can pair the identity data and facial images of the same person by performing the learning step S10. Specifically, the administrator can operate the object recognition system to sense the R...

AI Technical Summary

Problems solved by technology

[0006] However, in the learning step of the existing learning and recognition method, the administrator must manually pair different types of sensing data (ie identity data and facial images), which makes the learning step of the existing learning and recognition method Must spend a lot of time and manpower to generate paired data

Images