Embedded face detection terminal

Abstract

The invention discloses an embedded face detection system. The system comprises a CMOS image sensor, an image processing main chip, an LCD display screen and a software part loaded on the image processing main chip, wherein the CMOS image sensor is used for transmitting an acquired image variable to the image processing main chip; the image processing main chip is used for changing an original image into an image suitable for being recognized by an algorithm through an image preprocessing module, scanning current image after filtering the image, outputting a position of a face in a zoomed image, expressing the position by using a coordinate of a rectangular diagonal point, determining a rectangular position of the face in the original image, drawing a circle at a corresponding position ofthe original image through a face marking module so as to obtain a final to-be-output image; and finally, the image is updated into an image slot of Qt, and the face detection result can be displayedthrough refreshing an interface of the LCD display screen. According to the system, face detection and marking can be correctly and rapidly carried out in a frame of image, so that the working efficiency is high and the performance is stable and reliable.

Description

technical field [0001] The invention relates to the technical field of face recognition, in particular to an embedded face detection terminal. Background technique [0002] Face recognition is a technology that compares the input face image with the template in the known library and determines the identity. Since the face is an important way for people to distinguish different people, it is one of the most important sources of security information. It is widely used in video surveillance, sign-in system, human-computer interaction and other fields. [0003] Face recognition includes two methods, one is a recognition method based on two-dimensional brightness images, and the other is a recognition method based on three-dimensional faces. These two methods are interfered by the brightness of the light, the position and posture of the face to be recognized, the recognition time is long, and the resolution is low, which is not conducive to the rapid development of security ...

AI Technical Summary

Problems solved by technology

These two methods are interfered by the brightness of the light, the position and posture of the face to be recognized, the recognition time is long, and the resolution is low, which is not conducive to the rapid development of security work.

Images