Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode

Abstract The mixed image characteristics of coal blocking and coal leakage by belt conveying are complex, and the current identification method lacks timeliness. A hybrid image detection method based on improved RetinaNet model is proposed. This method uses CCD camera and lidar to construct a hybrid...

Full description

Saved in:
Bibliographic Details
Main Authors: Qingjun Fu, Xiang Liu, Yinqiang Yan, Zhibin Guo
Format: Article
Language:English
Published: Springer 2025-05-01
Series:Discover Applied Sciences
Subjects:
Online Access:https://doi.org/10.1007/s42452-025-07068-0
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849325870320189440
author Qingjun Fu
Xiang Liu
Yinqiang Yan
Zhibin Guo
author_facet Qingjun Fu
Xiang Liu
Yinqiang Yan
Zhibin Guo
author_sort Qingjun Fu
collection DOAJ
description Abstract The mixed image characteristics of coal blocking and coal leakage by belt conveying are complex, and the current identification method lacks timeliness. A hybrid image detection method based on improved RetinaNet model is proposed. This method uses CCD camera and lidar to construct a hybrid imaging system to realize three-dimensional imaging of coal conveying belt. An improved RetinaNet model is used to train the labeled three-dimensional image data set, and features are extracted and fused through attention mechanism and feature pyramid to detect coal blockage and coal leakage, and it is deployed in production environment to realize real-time monitoring. The test results show that this method can comprehensively collect the images of coal conveying by belt, and accurately identify the detection of coal blockage and coal leakage. This method is applied to the actual production environment, which can monitor the situation of belt coal transportation in real time and accurately detect coal blockage and leakage, which is of great significance to improve the production efficiency of coal transportation, reduce economic losses and ensure production safety.
format Article
id doaj-art-1c188e73fa634f1da674cbadc8b48d0d
institution Kabale University
issn 3004-9261
language English
publishDate 2025-05-01
publisher Springer
record_format Article
series Discover Applied Sciences
spelling doaj-art-1c188e73fa634f1da674cbadc8b48d0d2025-08-20T03:48:18ZengSpringerDiscover Applied Sciences3004-92612025-05-017611610.1007/s42452-025-07068-0Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet modeQingjun Fu0Xiang Liu1Yinqiang Yan2Zhibin Guo3Guoneng Huangjinbu Power Co., Ltd.Guoneng Huangjinbu Power Co., Ltd.Changyang Technology (Beijing) Co., LtdGuoneng Huangjinbu Power Co., Ltd.Abstract The mixed image characteristics of coal blocking and coal leakage by belt conveying are complex, and the current identification method lacks timeliness. A hybrid image detection method based on improved RetinaNet model is proposed. This method uses CCD camera and lidar to construct a hybrid imaging system to realize three-dimensional imaging of coal conveying belt. An improved RetinaNet model is used to train the labeled three-dimensional image data set, and features are extracted and fused through attention mechanism and feature pyramid to detect coal blockage and coal leakage, and it is deployed in production environment to realize real-time monitoring. The test results show that this method can comprehensively collect the images of coal conveying by belt, and accurately identify the detection of coal blockage and coal leakage. This method is applied to the actual production environment, which can monitor the situation of belt coal transportation in real time and accurately detect coal blockage and leakage, which is of great significance to improve the production efficiency of coal transportation, reduce economic losses and ensure production safety.https://doi.org/10.1007/s42452-025-07068-0Improved RetinaNet modelBelt coal blockage and leakageMixed imageThree-dimensional imagingChannel characteristic diagramScale characteristic
spellingShingle Qingjun Fu
Xiang Liu
Yinqiang Yan
Zhibin Guo
Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode
Discover Applied Sciences
Improved RetinaNet model
Belt coal blockage and leakage
Mixed image
Three-dimensional imaging
Channel characteristic diagram
Scale characteristic
title Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode
title_full Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode
title_fullStr Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode
title_full_unstemmed Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode
title_short Mixed image detection method of belt coal blockage and leakage based on improved RetinaNet mode
title_sort mixed image detection method of belt coal blockage and leakage based on improved retinanet mode
topic Improved RetinaNet model
Belt coal blockage and leakage
Mixed image
Three-dimensional imaging
Channel characteristic diagram
Scale characteristic
url https://doi.org/10.1007/s42452-025-07068-0
work_keys_str_mv AT qingjunfu mixedimagedetectionmethodofbeltcoalblockageandleakagebasedonimprovedretinanetmode
AT xiangliu mixedimagedetectionmethodofbeltcoalblockageandleakagebasedonimprovedretinanetmode
AT yinqiangyan mixedimagedetectionmethodofbeltcoalblockageandleakagebasedonimprovedretinanetmode
AT zhibinguo mixedimagedetectionmethodofbeltcoalblockageandleakagebasedonimprovedretinanetmode