首页|基于高精度机器视觉的柑橘智能分级与包装协同技术研究

基于高精度机器视觉的柑橘智能分级与包装协同技术研究

扫码查看
机器视觉是能赋予机器"甄别图像"能力的前沿尖端技术.聚焦于其在柑橘自动分级包装系统的应用研发,旨在大幅提升柑橘生产线效率,增强市场竞争力.鉴于柑橘表皮缺陷对品质和价值的直接影响,系统以缺陷大小和数量为主要分级依据.在暗箱环境获取高质量表皮图像,通过先进算法精准识别缺陷特征实现高效准确分级.包装环节中,分级结果实时反馈至机械臂控制系统,驱动其迅速抓取柑橘并放置于对应纸箱.纸箱内有限位泡沫防止碰撞摩擦,提升保护性.实验结果表明,柑橘缺陷检测精度达98.7%,精度提升1.4个百分点,分级精度达95%,包装平均速度每小时6000个.此系统提升了分级包装效率和准确性,为柑橘产业智能化升级提供技术支撑,市场前景广阔.
Research on Intelligent Grading and Packaging Collaborative Technology of Citrus Based on High Precision Machine Vision
Machine vision is a cutting-edge technology that can give machines the ability to"screen images".This paper focuses on the application of machine vision in the automatic grading and packaging system of citrus,aiming to improve the efficiency of citrus production line and enhance the competitiveness of the market.In view of the direct impact of peel defects on the quality and value of citrus,the system uses the size and number of defects as the main grading basis.High-quality skin images are obtained in a dark box environment,and advanced algorithms accurately recognize defect characteristics to achieve efficient and accurate grading.During the packing process,the grading results are fed back to the control system of the robotic arm in real time,which is then driven to quickly grab the mandarin oranges and place them in the corresponding cartons.Limit foam inside the carton prevents collision friction and enhances protection.The experimental results show that the defect detection accuracy of citrus reaches 98.7%,with an increase of 1.4 percentage points in accuracy,the grading accuracy reaches 95%,and the average packing speed is more than 6000 pieces per hour.This system improves the efficiency and accuracy of grading and packing,provides technical support for the intelligent upgrading of citrus industry,and has a broad market prospect.

computer visioncitrusgrading and packagingdefect detectionlimit foam

陈娃蕊、陈娃迪、涂志刚、熊立贵

展开 >

广东开放大学(广东理工职业学院),广东中山 528400

华南农业大学,广东 广州 510000

计算机视觉 柑橘 分级包装 缺陷检测 限位泡沫

2024

塑料包装
中轻投资有限公司

塑料包装

影响因子:0.1
ISSN:1006-9828
年,卷(期):2024.34(6)