摘要
轮廓提取技术现已广泛应用于医疗、工业、自动化以及服装类的平面加工领域。为了解决服装轮廓提取中服装轮廓提取精度低、计算量大、速度慢等问题,基于特征自适应阈值小波收缩降噪算法(Feature Adaptive Wavelet Shrinkage, FAWS)对图像进行降噪,并且通过麻雀搜索算法(Sparrow search algorithm, SSA)对FAWS降噪算法中的参数p和q进行寻优,再经灰度化、二值化处理,完成图像预处理,最后采用Canny算法提取服装轮廓并对轮廓进行拟合。结果表明:该算法计算速度快,在处理部分细节时其像素值更接近原始图像,为服装轮廓提取提供了一种切实可行的方法。
Contour extraction technology has been widely used in medical,industrial,automation and garment plane processing.In order to solve the problems of low precision,large amount of calculation and slow speed of garment contour extraction in the current garment contour extraction,this paper denoised images based on Feature Adaptive Wavelet Shrinkage(FAWS).The parameters p and q in the FAWS denoising algorithm were optimized by the Sparrow search algorithm,then the grayscale and binarized treatment were used to complete the image preprocessing.Finally,the clothing contour was extracted with Canny algorithm and then the contour was fitted.The results show that the algorithm is fast in calculation,and its pixel value of this method is closer to the original image when processing some details.This paper provides a practical method for garment contour extraction.
作者
王超
叶川
WANG Chao;YE Chuan(Engineering Training Center of Southwest Petroleum University,Nanchong 637001,China)
出处
《北京服装学院学报（自然科学版）》
CAS
北大核心
2023年第1期16-22,共7页
Journal of Beijing Institute of Fashion Technology：Natural Science Edition
基金
南充市市校科技战略合作项目(SXQHJH058)。
