Living Body Detection Using Near-Infrared Depth and Visible Light Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional living body detection methods using a single camera have low accuracy, often mistaking non-living bodies for living ones, particularly in applications like remote account opening and social insurance where authenticity and safety are critical.
Innovation Solution
A method and apparatus utilizing two near-infrared cameras and a visible light camera to obtain multiple images, generating a depth map and using pre-trained classification models to determine if a detected human face is from a living person by analyzing depth information and RGB data, enhancing detection accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a single camera is used to acquire user pictures for living body detection, then the device complexity is low, but the detection accuracy is insufficient and non-living bodies are easily mistaken as living bodies
Solution Approach 1:
The detection system is segmented into multiple independent camera units: a first near-infrared camera, a second near-infrared camera, and a visible light camera. Each camera captures different types of information (depth data from near-infrared, color information from visible light), allowing the system to achieve high detection accuracy through multi-source data fusion while maintaining modular device structure
Solution Approach 2:
The system transitions from single-dimension (2D image) detection to multi-dimensional detection by incorporating depth information through near-infrared cameras. This adds a third dimension (depth/distance) to the traditional 2D visible light imaging, enabling the system to detect spatial relationships and distinguish living bodies from flat images or videos
2Measurement precision
If multiple cameras are used to generate depth maps for living body detection, then the detection accuracy improves, but the use of energy increases
Solution Approach 1:
The near-infrared cameras emit periodic pulsed light signals rather than continuous illumination. This periodic action reduces energy consumption while still capturing sufficient depth information through time-of-flight measurement, as the pulsed signals can be synchronized with the camera's detection cycles
Solution Approach 2:
The patent introduces light as an intermediary carrier to transmit depth information. The near-infrared light serves as a mediator between the cameras and the target object, enabling non-contact depth measurement without requiring physical sensors on the object, thus reducing overall system energy consumption
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Improves the accuracy of living body detection by utilizing depth information to differentiate between living and non-living human faces, reducing false positives and ensuring the authenticity of user information in critical applications.
Implementation Method 1
two near infrared cameras
Implementation Method 2
a third picture taken with a visible light camera
Data Source
AI summary
The present disclosure provides a living body detecting method and apparatus, a device and a storage medium. The method comprises: regarding a to-be-detected user, respectively obtaining a first picture and a second picture taken with two near infrared cameras and a third picture taken with a visible light camera; generating a depth map according to the first picture and second picture; determining whether the user is a living body according to the depth map and the third picture. The solution of the present disclosure can be applied to improve accuracy of detection results.