Hand detection method and system based on depth information

Abstract

The invention discloses a hand detection method and system based on depth information. Including: obtaining the current depth image of the human body; marking the initial position coordinates of the hand in the current depth image according to the current depth image; normalizing the current depth image; According to the multi-layer convolutional neural network model, each selected area including hand detection is obtained; according to each selected area, the non-maximum value suppression operation is performed to obtain the optimal selected area; the optimal selected area is Image post-processing to obtain the final position coordinates of the hand in the current depth image. The hand detection method based on depth information of the present invention eliminates the condition that the hand must be at the forefront of the camera in the traditional depth map detection hand method, and at the same time can accurately find the position of the hand, through non-maximum suppression operations and Image post-processing can further improve the position accuracy of the palm of the hand.

Description

technical field [0001] The invention relates to the technical field of hand detection, in particular to a hand detection method based on depth information and a hand detection system based on depth information. Background technique [0002] With the continuous development of new interactive technologies and applications, especially the rapid development of AR (Augmented Reality) and VR (Virtual Reality), gesture-related technologies are regarded as the new and most representative interactive technologies of the next generation. It can realize non-contact remote human-computer interaction. In the field of smart home, smart home appliances or robots can be controlled and commanded by gestures; in the VR and AR neighborhoods, realistic practical experience can be achieved, and in the process of games and teaching, user experience can be greatly enhanced and so on. All gesture interaction techniques involve hand detection algorithms. Common hand detection algorithms are mainly...

AI Technical Summary

Problems solved by technology

Common hand detection algorithms are mainly based on detection on 2D color maps or on depth maps through traditional image processing methods to find the area where the nearest point is located. The default is the position of the hand. These methods either completely rely on pixel information and anti-interference capabilities. Poor, or there is a certain limit to the area where the hand is located

When the range of human activities is large, the detection is often not accurate enough

Images