首页|基于机器视觉的卷烟包装外观质量瑕疵在线识别方法

基于机器视觉的卷烟包装外观质量瑕疵在线识别方法

扫码查看
传统卷烟外观质量瑕疵在线识别方法直接对图像瑕疵特征进行提取,未对卷烟外观图像数据进行预处理,造成传统方法识别精度低,因此提出基于机器视觉的卷烟外观质量瑕疵在线识别方法.首先,对卷烟外观图像进行数据预处理,去除背景区域,提高对比度和图像瑕疵特征提取的准确性;其次,利用提取到的瑕疵特征,设计基于机器视觉的卷烟外观质量瑕疵在线识别流程;最后,基于机器视觉实现卷烟外观质量瑕疵在线识别.通过实验结果表明:此设计方法的识别正确率为100%,具有更高的识别精度.
An online Identification Method for the Quality Defects of Cigarette Packaging Appearance Based on Machine Vision
Traditional online identification methods for the quality defects of cigarette appearance directly extract the defect features of images,without preprocessing the image data of cigarette appearance,resulting in their low identification accuracy,so this paper proposes an online identification method for the quality defects of cigarette ap-pearance based on machine vision.Firstly,it performs data preprocessing of the images of cigarette appearance,and removes background areas,in order to improve the contrast ratio and the accuracy of extracting the defect features of images.Then,it uses the extracted defect features to design an online identification process for the quality defects of cigarette appearance based on machine vision.Finally,it achieves the online identification of the quality defects of cigarette appearance based on machine vision.Experimental results show that the designed method in this paper holds an identification accuracy rate of 100%and higher identification accuracy.

Machine visionCigarette appearanceQuality defect identificationIdentification method

曾丹梦、舒娟、廖强

展开 >

重庆中烟工业有限责任公司 重庆 400060

机器视觉 卷烟外观 质量瑕疵识别 识别方法

2024

科技资讯
北京国际科技服务中心 北京合作创新国际科技服务中心

科技资讯

影响因子:0.51
ISSN:1672-3791
年,卷(期):2024.22(5)
  • 5