烟叶叶片结构的机器视觉检测系统

设计了一种基于机器视觉技术的烟叶叶片结构在线检测系统。介绍了该系统的组成及工作原理,包括取样、单层化、摄像和图像处理流程。重点讨论了光学检测部分的参数选择,设计了图像处理流程,详细阐述了所用算法的原理。该系统经过了物理样机实验和标准样片测试,实验结果与实际吻合度高达96%,结果令人满意。

烟叶叶片加料均匀性的快速评价方法研究

利用近红外光谱技术快速检测卷烟叶片加料均匀性。通过对切丝后叶片进行近红外检测,考察烟丝不同含水率对近红外光谱的影响,同一牌号叶组配方变化对近红外光谱的影响,结果表明烟叶含水率是近红外定量分析主要干扰因素,同一牌号叶组配方变化对近红外影响较小。收集400个具有代表性的近红外光谱,将料液假设为一种成分,运用偏最小二乘法建立料液与烟叶比例的数学模型。对所建立的数学模型进行验证,模型验证误差在5%以内,具有一定实用价值。结果表明通过近红外检测技术可以实现对卷烟叶片加料均匀性的检测。

Analysis of Growth Laws of Leaves of Flue-cured Tobacco in Various Periods

The growth laws of leaves of flue-cured tobacco in different positions in various periods were summarized. The results showed that the growth speed of new leaves of flue-cured tobacco increased constantly until their growth was the fastest, and then it reduced irreversibly. The length and width of leaves in each position increased fastest when they grew depending on photosynthates instead of or- ganic compounds from functional leaves. Therefore, it is concluded that the increases of leaf length, width and area, and dry matter, and the peak of photosynthetic rate appear successively during the process of leaves of flue-cured tobacco, and the leaf area correction coefficient of leaves of flue-cured tobacco is related to the shape of leaves, which are affected by leaf position and the variety of flue-cured tobacco. The study can provide scientific reference for further research on the growth laws of leaves of flue-cured tobacco.

近红外光谱法测定烟片加料均匀性

本文简要论述了将料液假设为一种成分,运用偏最小二乘法建立料液与烟叶比例的数学模型。通过测量实际加料后的各取样点烟片近红外光谱,计算出各样品中加料量,从而评价烟片加料的均匀性。