A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning

Abstract To realize effective data compression, reduce resource occupation, and provide a smoother video transmission experience. The security coding method of an intelligent factory detection video based on improved deep reinforcement learning is studied. First, the transmission and network model o...

Full description

Saved in:
Bibliographic Details
Main Authors: Yongmin Yang, Zhenhao Wang
Format: Article
Language:English
Published: Springer 2024-12-01
Series:Discover Applied Sciences
Subjects:
Online Access:https://doi.org/10.1007/s42452-024-06366-3
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850086000761503744
author Yongmin Yang
Zhenhao Wang
author_facet Yongmin Yang
Zhenhao Wang
author_sort Yongmin Yang
collection DOAJ
description Abstract To realize effective data compression, reduce resource occupation, and provide a smoother video transmission experience. The security coding method of an intelligent factory detection video based on improved deep reinforcement learning is studied. First, the transmission and network model of the factory inspection video are analyzed, and a conditional depth network based on Transformer is designed. The network learns the image features through a feature extractor, and processes these features through a motion estimation and compression module. Multi-layer and multi-hypothesis motion compensation and a self-encoder are used to encode features based on the predicted features. The optimal coding strategy is learned using the depth deterministic strategy gradient method, and the residual is outputted in the decoding stage to reconstruct the video frame. The experimental results show that this method can realize the security coding of intelligent inspection video in plant area, and the BDBR (Bjontegaard delta bitrate) index can be saved by 68.01%, and the BD-PSNR (Bjontegaard delta PNSR) index can be increased by 3.49 dB on average; 50 iterations minimize the inefficient strategy ratio; The maximum average return is about 152%; It has small memory occupation, low bandwidth requirements, and high real-time video transmission; High user experience rating.
format Article
id doaj-art-426503e6f85e426a9d1d889eabfd609d
institution DOAJ
issn 3004-9261
language English
publishDate 2024-12-01
publisher Springer
record_format Article
series Discover Applied Sciences
spelling doaj-art-426503e6f85e426a9d1d889eabfd609d2025-08-20T02:43:36ZengSpringerDiscover Applied Sciences3004-92612024-12-017111810.1007/s42452-024-06366-3A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learningYongmin Yang0Zhenhao Wang1Huaneng (Zhangzhou, Fujian) Energy Co., Ltd.Huaneng (Zhangzhou, Fujian) Energy Co., Ltd.Abstract To realize effective data compression, reduce resource occupation, and provide a smoother video transmission experience. The security coding method of an intelligent factory detection video based on improved deep reinforcement learning is studied. First, the transmission and network model of the factory inspection video are analyzed, and a conditional depth network based on Transformer is designed. The network learns the image features through a feature extractor, and processes these features through a motion estimation and compression module. Multi-layer and multi-hypothesis motion compensation and a self-encoder are used to encode features based on the predicted features. The optimal coding strategy is learned using the depth deterministic strategy gradient method, and the residual is outputted in the decoding stage to reconstruct the video frame. The experimental results show that this method can realize the security coding of intelligent inspection video in plant area, and the BDBR (Bjontegaard delta bitrate) index can be saved by 68.01%, and the BD-PSNR (Bjontegaard delta PNSR) index can be increased by 3.49 dB on average; 50 iterations minimize the inefficient strategy ratio; The maximum average return is about 152%; It has small memory occupation, low bandwidth requirements, and high real-time video transmission; High user experience rating.https://doi.org/10.1007/s42452-024-06366-3Deep reinforcement learningVideo codingTransformerMotion estimationMotion compensationFailure experience correction
spellingShingle Yongmin Yang
Zhenhao Wang
A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
Discover Applied Sciences
Deep reinforcement learning
Video coding
Transformer
Motion estimation
Motion compensation
Failure experience correction
title A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
title_full A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
title_fullStr A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
title_full_unstemmed A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
title_short A study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
title_sort study on secure coding of intelligent inspection video in plant areas based on improved deep reinforcement learning
topic Deep reinforcement learning
Video coding
Transformer
Motion estimation
Motion compensation
Failure experience correction
url https://doi.org/10.1007/s42452-024-06366-3
work_keys_str_mv AT yongminyang astudyonsecurecodingofintelligentinspectionvideoinplantareasbasedonimproveddeepreinforcementlearning
AT zhenhaowang astudyonsecurecodingofintelligentinspectionvideoinplantareasbasedonimproveddeepreinforcementlearning
AT yongminyang studyonsecurecodingofintelligentinspectionvideoinplantareasbasedonimproveddeepreinforcementlearning
AT zhenhaowang studyonsecurecodingofintelligentinspectionvideoinplantareasbasedonimproveddeepreinforcementlearning