计算机与现代化 ›› 2023, Vol. 0 ›› Issue (02): 62-65.

• 人工智能 • 上一篇    下一篇

基于Faster-RCNN的自然环境下苹果识别

  

  1. (中北大学信息与通信工程学院,山西 太原 030051)
  • 出版日期:2023-04-10 发布日期:2023-04-10
  • 作者简介:石展鲲(1998—),男,山西运城人,硕士研究生,研究方向:图像处理,E-mail: szk15034593707@163.com; 杨风(1964—),女,山西万荣人,教授,研究方向:信号处理,检测技术与自动化装置,E-mail: fengyang@nuc.edu.cn; 韩建宁(1980—),男,山西太原人,教授,博士,研究方向:图形图像处理,声学超材料,E-mail: hanjn46@nuc.edu.cn; 郭鑫(1998—),男,山西晋城人,硕士研究生,研究方向:图像处理,E-mail: 937215351@qq.com; 曹尚斌(1998—),男,河北石家庄人,硕士研究生,研究方向:测试技术与仪器,E-mail: manbaout1@163.com。
  • 基金资助:
    山西省重点研发计划项目(201903D221018)

Apples Recognition in Natural Environment Based on Faster-RCNN

  1. (School of Information and Communication Engineering,North University of China,Taiyuan 030051,China)
  • Online:2023-04-10 Published:2023-04-10

摘要: 针对苹果园中存在的果实相互重叠、枝叶干扰以及复杂背景等问题,本文提出Faster-RCNN一种改进的模型。该模型通过增强Mosaic数据,使得识别小物体目标果实能力得到提升,同时,对Faster-RCNN结构中的锚框进行优化,优化后的锚框能更好地检测出距离相机较远的目标果实,以及使用Soft NMS算法对密集区域的识别效果进一步得到改进。通过对300幅未参与识别的自然环境下的苹果图像进行验证,验证结果表明:召回率为91.44%,准确率为93.35%,F1值为92.38%,每幅图像的检测可在0.2 s内完成。改进后的算法鲁棒性得到增强,能够满足在自然环境下对苹果果实的识别工作。

关键词: Faster-RCNN, Mosaic数据增强, 目标识别, Soft NMS算法

Abstract: Aiming at the problems of overlapping fruits, interference of branches and leaves, and complex backgrounds in apple orchards, the Faster-RCNN algorithm was proposed. By adding Mosaic data enhancement at the input end, the amount of data is increased and the ability to recognize small objects is enhanced. At the same time, the anchor frame in the Faster-RCNN algorithm is optimized, and the optimized anchor frame can better detect the distance. The target fruit far from the camera and the Soft NMS algorithm are used to further improve the recognition effect of dense areas. The verification results show that the recall rate is 91.44%, the accuracy rate is 93.35%, the F1 value is 92.38%, and the average detection speed per image can reach 0.2 fps. The robustness of the improved algorithm is enhanced, which can meet the recognition of apple fruits in natural environment.

Key words: Faster-RCNN, Mosaic data augmentation, target recognition, Soft NMS algorithm