基于计算机视觉技术的天麻分选装置设计

Design of Gastrodia Selector Based on Computer Visual Technology

为了解决天麻外观品质检测问题,设计了一种在线实时检测装置,用于天麻的品质动态实时检测。为此,确定了关键部件主要结构和参数,阐述了其总体结构及工作原理。天麻通过放料部分进入滚筒,经滚筒传输至上料部分,上料部分的辊轮将其运送到翻滚槽口,经翻转后进入托盘机构,托盘机构在链条的带动下以一定速度向前输送至图像采集部分,从而获取天麻整个表面信息。图像识别系统对采集到的天麻图像进行综合分析判断,确定天麻品质和位置信息,传送给分选执行机构,对天麻进行智能分选。试验结果表明:识别模型识别效率为99.34%,步长为7.06个/s,准确率为95%,可为天麻分选装置设计生产提供理论指导。

In order to solve the problem of appearance quality detection of Gastrodia, an on-line real-time detection device was designed for Gastrodia quality dynamic and real-time detection. The main structure and parameters of the key components are determined, and the overall structure and working principle of the device are described. The main structure includes the feeding part, the feeding part, the conveying part, the computer vision recognition system, and the sorting part. Gastrodia enters the drum through the discharging part, transmits it to the upper part through the drum, and the roller of the upper part transports it to the rolling groove. After flipping, it enters the tray mechanism. The tray mechanism is driven by the chain and is transported forward at a certain speed. Image collection part, Gastrodia enters the dark box so that the image recognition system can obtain Gastrodia entire surface information. The image recognition system makes a comprehensive analysis and judgment of Gastrodia images collected, determines Gastrodia quality and location information, transmits it to the sorting execution mechanism, and performs intelligent sorting of Gastrodia. The results show that the efficiency of the model identification is 99.34 %, the step length is 7.06/S, and the accuracy of the device is 95 %. This provides theoretical guidance for the design and production of the Gastrodia separation device.