Research On Intelligent Measurement Method Of Human Body Size Based On Image Ranging Algorithm

The clothing industry is a typical labor-intensive industry.Because of the mismatch between supply and demand,it leads to problems such as overcapacity and serious waste of resources.Every year,China’s clothing manufacturing industry loses more than 15 billion yuan due to inventory problems,and clothing is customized.It is a great way to solve this problem.With the development of technologies such as artificial intelligence,big data,cloud computing,and the Internet of Things,customized production has been achieved in the pattern-making,design,and logistics aspects of clothing production,but it is difficult to accurately and quickly obtain customers’ human bodies.The size of each part,the development of personalized customization is greatly hindered.Body size measurement is the most important link in personalized customization.Traditional manual measurement takes a long time and cannot meet market demand.Therefore,the intelligent measurement method has become a research hotspot.The current research is mainly divided into the method of establishing a 3D human body model from 2D photos and calculating the size of the method of directly identifying the feature points of the human body in 2D photos.Regardless of whether it is a method of converting 2D images to3 D models or directly identifying feature points,height or a single marker is currently used as the conversion scale when the pixel distance is converted to the actual distance,but there are wide-angle distortion,perspective distortion,and size when measuring the size by taking pictures.For problems such as conversion errors,if a single scale is used,it will bring about a larger local size error,and with the shooting angle and distance,it will bring greater error.In order to solve the shortcomings of the current intelligent measuring methods,this article first designed a set of elastic body suits with 357 coded patterns.The coded pattern is a three-digit circular pattern with a combination of 0-9 numbers.Based on the measuring suit designed in this paper,we have studied a whole set of intelligent measuring algorithms from code recognition and detection to size calculation.For code detection,firstly,the semantic segmentation algorithm is used to extract the circular code pattern to obtain information such as the position and size of the code,and the number that composes the code on the code is identified by the trained multi-classification network to determine the unique number of the code.After successfully identifying the encoded information,this paper proposes an image ranging algorithm based on multiple markers,and uses a convolution algorithm to calculate the actual distance corresponding to the pixel distance on the image.On the basis of the image ranging algorithm based on multiple markers,this article uses the coding pattern on the body measuring garment as the marker,and further designs the calculation method of human body size.The experiment shows that the algorithm designed in this article has high accuracy and stability.Among them,the accuracy rate is as high as 95%,and the maximum error of multiple experiments does not exceed 2cm,which meets the requirements of garment making and has strong stability.Finally,for the convenience of users,this article develops a complete intelligent measuring system based on body measuring clothes and algorithms,which realizes a complete set of functions such as photographing,ranging,storage,and display.