毛针织服装任意衣片的自动识别

Automatic recognition of any pieces of knitted wool clothing

为提高毛针织服装CAD系统的智能性,使其能自动识别任意衣片,为后续工艺制作等工序提供计算依据,提出一种新的基于模糊理论的毛针织服装衣片自动识别算法。通过建立基于关键点的衣片数据模型,给出任意衣片与标准衣片的对比方法及基本步骤;通过对任意衣片的等比例变换,构建了基于关键点及基于轮廓点的2种任意衣片与标准衣片的隶属值公式;通过隶属度归一化处理,确定了决定任意衣片归类的总隶属度函数。编写程序对上述算法进行验证,显示该算法对任意衣片识别准确率达到95%以上,因此,该算法可较为有效地识别毛针织服装任意衣片种类。

In order to improve the intelligence of knitted wool garment CAD system to such a level that it can automatically recognize any clothing pieces and provide a computation basis for subsequent processes,this paper proposes a new automatic recognition algorithm for knitted wool garment based on fuzzy theory.Firstly,a clothing piece data model based on key points is developed,a contrast method between arbitrary and standard clothing pieces and basic steps are given.Secondly,a membership value formula of two arbitrary and standard clothing pieces based on key points and contour points according to equal proportion transformation of clothing pieces is established.Normalization processing of this membership determines the total membership function which makes the classification of clothing pieces.Some programs to verify this algorithm are compiled and the results show that the recognition accuracy of any clothing pieces reaches more than 95%,indicating that the classification of any clothing pieces can be recognized effectively by this algorithm.