本文二维码信息
二维码（扫一下试试看！）
An Improved MaskRCNN Based Paper Disease Diagnosis Algorithm
Received:May 09, 2024  
DOI：10.11980/j.issn.0254-508X.2024.12.021
Key Words:paper disease diagnosis  MaskRCNN  VOVNet  PrRoIPooling  SPANet
Fund Project:国家自然科学基金计划项目 （62073206）；西安市科技计划项目 （2020KJRC0146）。
Author NameAffiliationPostcode
TANG Wei College of Electrical and Control Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021
Shaanxi Xiwei Process Automation Engineering Co.， Ltd.， Xianyang， Shaanxi Province，712000 		712000
LIU Yingwei* College of Electrical and Control Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 710021
WANG Mengxiao Shaanxi Xiwei Process Automation Engineering Co.， Ltd.， Xianyang， Shaanxi Province，712000 712000
GENG Zhiyao College of Electrical and Control Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 710021
LIU Changchuang College of Electrical and Control Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 710021
YANG Yijun College of Electrical and Control Engineering， Shaanxi University of Science & Technology， Xi’an， Shaanxi Province， 710021 710021
Hits: 215
Download times: 74
Abstract:This paper proposed a paper disease diagnosis algorithm based on an improved MaskRCNN network. Firstly， this algorithm improved the network model by using a lightweight head backbone network VOVNet and a Precise RoIPooling （PrRoIPooling） on the basis of the original MaskRCNN network， in order to reduce the parameter usage of the original network model and improve the detection and classification speed. Secondly， a spatial pyramid attention mechanism （SPANet） was added to address the issue of low accuracy in detecting small objects in the original network model. More than 4 000 paper disease images were collected for simulation verification of the proposed algorithm. The results showed that the improved MaskRCNN model had increased average accuracy by 3 percentage points and speed by 15% compared to the original network model， which could meet the practical requirements of real-time and accuracy in paper disease diagnosis.
View Full Text  HTML  View/Add Comment  Download reader