基于倾斜矫正算法的针织物图像研究

倾斜的针织物图像在图像处理前需要对其倾斜进行矫正预处理,但由于针织物图像的特殊性,使其不能采用机织物等倾斜图像的矫正方法。针对这一难题,利用一维傅里叶变换求针织物纵密频谱和横密频谱。计算,然后根据各自频谱的峰值特性进行自动旋转倾斜矫正。给出了算法细节,同时还得到针织物的拉密度。实际研究表明,所采用的方法在针织物行业里尚属独有,运算量小,实时性高,具有很大的普适性,非常适合于在针织物的图像检测领域推广,具有广泛的实际应用价值。

Research of Knitted Fabric Image based on Slant Correction Algorithm

Tilt of the knitted fabric image in image processing is required prior to the tilt correction pretreatment. But due to the particularity of the knitted fabric image, it can not be using woven fabric tilt image correction method. To solve this problem, based on the one-dimensional FuLiye transform, knitted fabric vertical density spectrum and cross spectrum density can be calculated. Then according to their spectrum peak characteristics of automatic rotating tilt correction. In this paper the algorithm in detail is given out. According to the results of the research, this article reviews the methods used in knitting industry is unique.As its small calculating amount, high real-time performance, it has great universality. It can be very suitable for knitted fabric image detection field widely and have extensive actual application value. It has certain practicability and value of popularization.