摘要
袖窿与袖子是服装结构中最复杂的部位之一,针对西服样板中肩袖造型和结构参数的非线性关系,用改进BP神经网络建立西服"袖窿袖"样板设计的模型。把"袖窿袖"的结构参数作为输入参数,视觉几何参数作为输出参数,进行神经网络训练,实现了西服样板中袖与袖窿结构到成衣肩袖造型的自动映射。通过对样本进行测试,验证了用改进BP神经网络进行西服肩袖样板设计的有效性和精确性,对西服肩袖部位的造型具有预知性,减少袖疵病的发生,并为服装的智能化生产提供理论依据。
Armhole and sleeve is one of the most complicated parts in the garment construction.Regarding the nonlinear relationship between shoulder cuff modelling and construction paramenters,an improved BP artificial neural network is presented to develop a model of armholes and sleeve of the suit.The neural network in trained using construction parameters of armhole and sleeve as input vectors and three-dimensional sizes as output.Thus,automatic mapping of the construction of armhole and sleeve along with shoulder sleeve modelling is realized.The test results show that this method is effective and accurate,and can predict the shoulder sleeve modelling of the suit,reduce defects of sleeve,and provide a theoretical basis for intelligent manufacturing garment.
出处
《纺织学报》
EI
CAS
CSCD
北大核心
2010年第8期113-116,共4页
Journal of Textile Research
基金
河南省科技攻关项目(102102210254)
关键词
BP人工神经网络
西服
结构
肩袖造型
几何参数
结构参数
BP artificial neural network
suit
pattern
model of rotator cuff
three-dimensional size
specification sizes
