中国烟草学报2024，Vol.30Issue(2)：11-19,9.DOI:10.16472/j.chinatobacco.2022.T0291
基于分段拼接的卷烟侧面生成及包灰测定
Generation of cigarette side image and determination of ash integration based on segmented splicing
摘要
Abstract
[Objective]The ash integration rate is one of the main indexes for evaluating the ash packing performance of cigarettes.The objective of this paper is to accurately and completely analyze this rate under the static combustion mode.[Methods]The cigarettes were placed on a high-precision rotating platform,the video of the ash column was collected and converted into images,a series of processing methods such as image preprocessing,segmented splicing,gamma correction and histogram equalization were used to generate the side expanded image of the ash column;the cracks were extracted by using image threshold segmentation,and the ash integration rate of cigarettes was accurately calculated.This involved counting the number of pixels representing ash integration and the total number of pixels of the side expanded image,and then calculating the ratio of the two to determine the ash integration rate of the cigarettes.[Results]This method is able to generate a complete side expanded image of the ash column after the combustion of a single cigarette,and the fracture trend in the side expanded image is continuous and smooth,which achieves the accurate assessment of the ash integration rate in static combustion mode.The proposed method outperforms conventional measurement techniques.[Conclusion]The side expanded images of cigarettes generated based on segmented splicing more comprehensively reflect the cracks of the ash column,and had higher measurement accuracy for ash integration rate,which provided a high-precision method for analyzing the ash integration rate under the static combustion mode.关键词
卷烟/包灰率/分段拼接/侧面展开/图像处理Key words
cigarette/ash integration rate/segmented splicing/side expanded/image processing引用本文复制引用
张卫正,李萌,王越峰,何逸波,杨道剑,冯亚婕,张琦,柴武君,许恒誉,张伟伟..基于分段拼接的卷烟侧面生成及包灰测定[J].中国烟草学报,2024,30(2):11-19,9.基金项目
河南省科技攻关项目"基于协同代理和流形学习的大规模多模态多目标优化方法研究"(No.222102210037) （No.222102210037）
河南省高等教育教学改革研究与实践重点项目(2021SJGLX189) （2021SJGLX189）
