Measurement Of Human Body Parameters Based On Photos

With the development of computer technology and economy, people becomevery busy, time has become an important factor for the creation of wealth, so the newtrend of the shopping-online shopping has been widespread favor. The rapiddevelopment of Taobao, Jingdong, JumeiYoupin, dangdang, e-commerce shoppingwebsite, the online shopping is a new shopping way into thousands of householdsliving. Especially in the clothing, shopping demand is increasing. But online shoppingfit problems, whether for their own problems, problems such as return of trouble, andseveral problems to be become in garment e-commerce solution. Therefore,non-contact (based on photo) of human body parameter measurement technology willbe widely used in shopping mode of future, widely loved by consumers and supplierswill also. Measurement of parameters of the human body based on photos is thecomputer vision and image processing field, it is in image processing, traffic detectionsystem, public security system (public safety) in human behavior understanding,image coding, image segmentation has a wide application prospect and replace humanaccess to information, etc.. This paper focuses on clothing fitting system of edgedetection (human image segmentation) and parameters of various parts of the humanbody, acquisition, clothing fitting system and then design an online. Objective is theintegration of virtual reality technology in depth, so that the user can freely change,from the characteristics of visual experience a full range of clothing.Following are the main research direction for some of the key issues in humanbody parameter information picture acquisition and image segmentation in:1. In thephoto of information acquisition, using EXIF information acquisition mode andformat, and calculated the parameters.Based on the geometric relationship between calculated data of2.In imagesegmentation, a segmentation method based on simple background image, makehuman body and background for the formation of non-connected to image segmentation, the main processing color background color.3. Using Open CVtechnology, on the human body photographs are gray, two values, denoising,smoothing boundary operation.The main research is as follows:1, in this study, young men and women to thebody as the research object, using the method of manual measurement,photogrammetry and3D body measurement combining the measurement parametersof the human body.2, through the Open CV programming, realizes the human bodycontour extraction is the human body edge detection of the body surface parameters,data can be automatically extracted, for example: parameter arm length, shoulderangle, body side angle, hip angle, back into the corner, dorsal angle value. Features3,combined with the basic knowledge of computer graphics and anthropometricparameters in human body, image edge feature points extraction and selection, headwidth, head, neck width, neck thick, shoulder, shoulder, back, chest width thicknesswidth (find the armpits, chest, waist width) thick, thick waist, abdomen wide, thickbelly, rump width, buttocks, thigh root wide, thick thigh thick, wide, thick knee, calf,calf wide thick size value.4, the two element linear equation, establish the geometricrelations and other mathematical methods, the parameters of the manual measurementof the value of the dependent variable, the parameter extraction procedure values forparameters fitting algorithm of the model variables.5, the system uses the windowsXP operating system, Oracle as the background database, C++as the developmenttool, a three-dimensional online fitting system, realize the non-contact measurementof the body parameters based on photos, and the measurement results can be archivestorage.