Back Detection Of Human Body Based On Image And Its Application Research In The Manipulator

Aiming at the medical massage therapy,this article is to construct a detection system that can automatically track and detect the back of the human body.A prototype of a six-degree-of-freedom robotic arm is designed as a prototype for a massage physiotherapy robot.Based on the results of image detection,this robot can perform kinematics-based and image-based visual servoing.The content of this article is divided into two parts: one part is to segment the back region of the human body on the picture,detect and extract the position of human back region on the whole picture.The another part is that,based on the position of the back region of the human body,visual servo control of the movement of the end effector has been studied.Image processing is based on Open CV,an open source computer vision library,the program running platform is Visual Studio2010.In this paper,before going on the detection,the first consideration is to segment some possible human back regions in the picture.After some processing,support vector machine(SVM)classify them,and detect the human body back with its specific location in the picture.For the image dividing,focusing the use of the flood algorithm to roughly segment the image and then further highlights and separates the image by using the corresponding mathematical morphology operation.The mathematical morphology operation creates the conditions for the next step of contour extraction and cropping of the back area.Finally,we use the pre-made training documents in this paper,combined with the corresponding support vector machine algorithm,to achieve the purpose of detecting the human body back.In this paper,image-based visual servoing control(IBVS)is applied to the kinematics of massage physiotherapy robots by using the deviation between the actual feature points and the expected feature points on the human back image.Then implementing the image-based visual servoing(IBVS),the projection of the target point in the image plane of the camera is achieved as expected,so as to achieve the required motion control of the manipulator in approaching the target point.This article has achieved preliminary results in the separation and detection of human back.And for the construction of a visual servo system based on human back images for massage physiotherapy robots,a relative complete theory and method are proposed,which also provides certain conditions for the better application and practicality of massage physiotherapy robots in the future.