Research On The Evaluation Method Ofspacesuit Modeling For Great-Nation Image

At 9:22 on June 17,2021,China’s Shenzhou 12 manned spacecraft ignited and took off and successfully entered the expected trajectory.Three astronauts were ordered to go on a expedition and will stay on the space station for three months.One of the landmark tasks is to wear a new generation of "Feitian" extravehicular spacesuit to carry out extravehicular activities.With the smooth implementation of the mission,we have carried out many positive discussions on the space mission and the new generation of "Feitian" spacesuits.Under the background of international language,the modeling design of "Feitian" series spacesuits not only has the identification needs of Chinese aesthetic identification,but also focuses more clearly on the aesthetic needs of Chinese national image.In order to meet the needs of multi-dimensional perceptual image recognition of "big country image" in space suit modeling design,a construction method of space suit modeling image evaluation model for "big country image" is proposed.The first part is the research on the construction of multi-dimensional image cognitive space of space suit.Through a large number of user surveys,this paper preliminarily selects the perceptual image set of "great country image",and obtains the representative samples of spacesuit products through cognitive experiment and cluster analysis.The semantic difference method is used to test and analyze the mapping data between the preliminary screening words of the perceptual image set of "great country image" and the representative samples of spacesuit products.The data are analyzed by principal component analysis to obtain the perceptual image cognitive space of spacesuit products about "great country image".At the same time,according to the global HIES deconstruction principle and the function constraint list of spacesuit products,the modeling feature space of spacesuit products is constructed.The multi-dimensional image cognitive space of spacesuit modeling is obtained by using semantic difference method and cognitive test.The second part is to build and test the space suit modeling image evaluation model by using BP neural network.The key elements in each group of samples are input into the neural network according to the performance of the key elements in the BP network;Based on the image cognitive value in the multi-dimensional perceptual image modeling cognitive space,the perceptual image mean of each sample under each group of image words is calculated as the output layer of BP neural network.The accuracy of the evaluation model is tested and verified by leaving one cross training method.The results show that the evaluation model of space suit modeling image is reliable.The third part is the practical application of the space suit modeling image evaluation model.The designer’s design scheme modeling is transformed into coded data according to the performance of key HIES elements,and input into the input layer of the evaluation model to obtain the image value of the design scheme output under each group of image words.Through the scheme verification,the evaluation model can effectively solve the mapping and matching problem between modeling features and multi-dimensional images,assist designers to quickly identify perceptual image targets and corresponding key modeling features,and enhance the scientificity of design scheme evaluation.