摘要
针对现有昆虫飞行信息采集方法中存在的采集数据量大、采集效率低和束缚实验对象等问题,设计了一种基于视觉检测技术的自动化采集装置。该装置主要包括障碍通道模块、运动触发采集模块、飞行轨迹信息模块和自动采集控制系统。首先,根据控制功能的需求设计了间隙控制器,采用激光测距传感器和步进电机等硬件实现了闭环控制;其次,采用STM32微控制器为控制终端,并结合运动方向检测算法触发Blackfly S USB3高速相机,实现昆虫在指定运动方向的序列图像采集;然后,利用开源计算机视觉库OpenCV分析采集的序列图像,获取昆虫的飞行轨迹信息;最后,通过嵌入式控制系统协调各模块之间的通信,以达到高效、稳定获取昆虫飞行信息的设计要求。为了验证该装置的适用性和准确率,选取中华蜜蜂进行穿越间隙的实验,实验结果表明:在静态环境中,该装置能获取完整、清晰的飞行序列图像,平均准确率达到73.24%,采集性能稳定,有效提升了采集效率。通过分析蜜蜂的飞行轨迹信息,推断出其识别间隙的机制与横向运动的幅度和速度存在联系,这为深度研究蜜蜂的飞行机制提供了支持。
Aiming at the problems of large amount of data collection, low collection efficiency and binding experimental objects existing in the current insect flight information collection methods, an automatic collection device based on visual inspection technology was designed. The device mainly included obstacle channel module, motion trigger acquisition module, flight path information module and automatic acquisition control system. Firstly, gap controller was designed according to the requirements of the control function, and closed-loop control was realized by using hardware such as laser ranging sensor and stepping motor;secondly, the STM32 microcontroller was used as control terminal, and the Blackfly S USB3 high-speed camera was triggered by using motion direction detection algorithm to realize the sequence image acquisition of insects in the specified motion direction;then,using the open-source computer vision library OpenCV to analyze the collected sequence images and obtain the flight path information of insects;finally, the embedded control system coordinated the communication between modules to achieve the design requirements of efficient and stable acquisition of insect flight information. In order to verify the applicability and accuracy of the device, Apis cerana was selected for the gap crossing experiment. The experimental results showed that in a static environment, the device could obtain complete and clear flight sequence images. The average accuracy rate reached 73.24%. The acquisition performance was stable, which effectively improved the acquisition efficiency. By analyzing the flight path information of honeybee, it is inferred that the mechanism of gap recognition is related to the amplitude and speed of its lateral movement, which provides support for in-depth research on the flight mechanism of Apis cerana.
作者
陆古月
邢强
赵蔚
马磊
张小萍
王文波
LU Gu-yue;XING Qiang;ZHAO Wei;MA Lei;ZHANG Xiao-ping;WANG Wen-bo(School of Mechanical Engineering,Nantong University,Nantong 226019,China;School of Information Science and Technology,Nantong University,Nantong 226019,China;College of Mechanical and Electrical Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210016,China)
出处
《工程设计学报》
CSCD
北大核心
2022年第5期607-615,共9页
Chinese Journal of Engineering Design
基金
国家重点研发计划资助项目(2020YFB1313504)。
关键词
自动化采集装置
蜜蜂行为学
图像采集
设计
automatic collection device
honeybee behavior
image acquisition
design
作者简介
陆古月(1997-),男,江苏盐城人,硕士,从事自动化技术研究,E-mail:2009310009@stmail.ntu.edu.cn,https://orcid.org/0000-0002-7046-1748;通信联系人:邢强(1985-),男,江苏南通人,副教授,博士,从事仿生视觉等研究,E-mail:meexq@ntu.edu.cn,https://orcid.org/0000-0002-4552-1817。
