Human body identification and positioning method, robot and characteristic clothing

Abstract

The invention, which belongs to the technical field of machine vision, provides a human body identification and positioning method, a robot and a piece of characteristic clothing. The method comprises: a robot generates a pixel value randomly and sends the pixel value to a piece of characteristic clothing; the piece of characteristic clothing controls a display device to display a color corresponding to the pixel value; the robot obtains a first scanning image by a camera and obtains a first observation result according to the pixel value and the first scanning image; the piece of characteristic clothing obtains a second scanning image by the camera and sends the image to the robot; and the robot obtains a second observation result according to the second scanning image, determines and records a discovery relationship based on the first observation result and the second observation result, and obtains positioning information of the piece of characteristic clothing based on the discovery relationship. According to the invention, with combination of a visual angel of a robot and a visual angle of a to-be-identified human body, the path of discovering a human body by a robot is increased, so that an error caused by only using the visual angle of the robot can be eliminated.

Description

technical field [0001] The invention belongs to the technical field of machine vision, and in particular relates to a method for identifying and locating a human body, a robot and characteristic clothes. Background technique [0002] In the field of machine vision, the existing human body tracking technology mainly includes two steps: 1. Identify the target human body and lock and track the target; 2. Position the target human body. When identifying the target human body, the bone recognition technology is mainly used to identify the target human body by analyzing the skeleton information, and lock the target human body by clapping. However, the existing bone recognition technology has a large amount of calculation and poor real-time recognition. In lower versions of software, gesture recognition is needed to enable bone recognition, and the ease of use is low. And the existing technology mainly uses the robot as the main body of discovery. When the camera of the robot does...

AI Technical Summary

Problems solved by technology

However, the existing bone recognition technology has a large amount of calculation and poor real-time recognition. In the lower version of the software, gesture recognition is needed to enable bone recognition, and the ease of use is low.

And the existing technology mainly uses the robot as the main body of discovery. When the camera of the robot does not find the target object (including the target human body), it is considered that the target object is lost, which easily leads to misjudgment of loss when the target object is blocked by other objects. The discovery method is single and the probability of discovery is low

Images