基于智能视觉的透明液体可见异物检测软件设计

目前在国内的透明液体生产线上,液体通常采用人工检测。然而目前国内人工检测存在着诸如效率低、准确率低等不足,所以基于图形处理算法以及QT,我们设计了一套基于机器视觉的自动化可见异物检测智能检测软件,重点实现了软件与算法的改进优化。获取的连续图像经自适应中值滤波预处理去除图像噪声后,通过改进的序列图像二次差分算法提取出序列图中的细小的异物,然后根据通过SIFT特征度量的MeanShift目标跟踪算法,跟踪到到液体图像中的细小异物.并根据异物的运动轨迹及大小,检测到有无异物以及异物的类型,进而进一步判断透明药品质量是否合格。结果测试表明,软件的准确率达到95%以上,检测分辨率达到了44μm,基本可以替代透明液体生产线上的人工检测。

Automatic Injection Impurity Detecting software Based on Intelligent visual technology

the domestic pharmaceutical manufacturers are almost transparent using artificial method to detect the product, but the current domestic artificial drug test exist many problems such as low efficiency, low accuracy rate.So according to the national standards and requirements of 50-500ml glass bottled transparent medical production,we designed a automatic impurity detecting system based on intelligent visual technology.Compared with the low efficiency and low accuracy of domestic artificial detection, the system includes mechanical transmission module, electrical control system module and image processing module. According to the sports characteristics of the impurity and the noise, we proposed the improved second-difference algorithm to detect moving visible impurities,and used the SIFT features and MeanShift algorithm to track impurities. Finally we can judge whether the product is qualified.Experimental results showed that the accuracy rate of the system can reach 95% and the impurity to detect is above 44 microns ,Basically it can replace the original manual testing in medical production line.