期刊论文

Simulation of Accurate Detection to Foreign Matter of Production for Inspection Machine in Medical Infusion

中文

计算机仿真

1006-9348

2019

36

2

330-335

10.3969/j.issn.1006-9348.2019.02.072

目基项金:国家自然科学基金项目(61401046) 国家科技支撑计划项目(2015BAF11B01) 湖南省教育厅资助科研项目(17C0046) 长沙理工大学科研创新项目(CX2016SS07).

本校为第一机构

针对250 mL及以上医药大输液生产过程中造成药液中出现毛发、漂浮物、玻璃屑等可见异物的在线实时检测难题,开发了一种基于分层联合稀疏表示的医药输液全自动智能灯检机。首先通过改进的分块主成分追踪算法对序列图像中的运动目标进行检测以排除静止背景噪声干扰,然后利用稀疏矩阵分块聚类算法得到运动目标的运动轨迹并提取运动轨迹的空间特征向量,最后为了提高传统稀疏表示分类器对异物识别的能力,提出分层联合稀疏表示分类器根据运动轨迹的空间特征向量对异物进行识别,以排除随机噪声的干扰。仿真结果证明,该系统克服了检测过程中各种噪声干扰对异物识别的影响,解决了医药大输液可见异物的在线...

Aiming at online real-time detection problem of 250 mL and above medical large infusion in the production process,which caused the emergence of hair,floating debris,glass chip and other visible foreign matter,an automatic intelligent inspection machine based on hierarchical joint sparse representation in medical infusion was proposed. Firstly, the improved block principal component tracking algorithm is used to obtain the trajectory of moving objects to eliminate static background noise interference. Then, the spatial feature vector is extracted from the motion trajectory of the moving objects...