PDF (1308K)
摘要
为了解决后厨安全监管效率低、现有视觉检测方法误报率高且工程落地适配性差的问题,研究设计了智慧后厨安全检测系统.系统运用计算机视觉、深度学习等技术,采用时序容错与时空关联的判定逻辑,构建了覆盖个人防护、操作规范、非法闯入三类场景的检测体系,可自动识别后厨多种违规行为并实时预警,解决了传统监管响应滞后、单帧检测误报率高的问题;同时,利用流媒体低延迟优化与分辨率自适应ROI配置技术提升系统工程部署能力,降低多硬件环境下的维护成本,解决复杂网络环境下检测滞后、部署适配性差的问题.目前,该系统已落地为可视化监管平台并投入实际运行.
Abstract
To address the issues of low efficiency in post-kitchen safety supervision, high false alarm rate of existing visual detection methods, and poor engineering applicability and adaptability, a smart post-kitchen safety detection system was designed. The system utilizes computer vision and deep learning technologies, employs timing fault tolerance and spatio-temporal correlation-based judgment logic, and constructs a detection system covering three scenarios: personal protection, operational norms, and illegal intrusion. It can automatically identify various violations in the post-kitchen and provide real-time warnings, solving the problems of delayed traditional supervision response and high false alarm rate in single-frame detection. At the same time, by optimizing low-latency streaming media and configuring resolution-adaptive ROI, the system’s engineering deployment capability is enhanced, maintenance costs are reduced in multi-hardware environments, and problems of detection lag and poor deployment adaptability in complex network environments are addressed. Currently, this system has been implemented as a visual supervision platform and put into actual operation.
关键词
Key words
[Author(id=1330192932176138410, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, orderNo=0, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1330192932239052974, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, authorId=1330192932176138410, language=EN, stringName=Lifen Gao, firstName=Lifen, middleName=null, lastName=Gao, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Guoneng Baosin Railway Group Co., Ltd. Transportation Service Branch, Erdos, Inner Mongolia, 017000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1330192932293578928, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, authorId=1330192932176138410, language=CN, stringName=高丽芬, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=国能包神铁路集团有限责任公司 运输服务分公司, 中国·内蒙古 鄂尔多斯 017000, bio={"content":"高丽芬(1980-),女,中国内蒙古鄂尔多斯人,本科,工程师,从事AI智能应用研究.
"}, bioImg=null, bioContent=高丽芬(1980-),女,中国内蒙古鄂尔多斯人,本科,工程师,从事AI智能应用研究.
, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1330192932092252325, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, xref=null, ext=[AuthorCompanyExt(id=1330192932109029542, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, companyId=1330192932092252325, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Guoneng Baosin Railway Group Co., Ltd. Transportation Service Branch, Erdos, Inner Mongolia, 017000, China), AuthorCompanyExt(id=1330192932125806759, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, companyId=1330192932092252325, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=国能包神铁路集团有限责任公司 运输服务分公司, 中国·内蒙古 鄂尔多斯 017000)])]), Author(id=1330192932343910579, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, orderNo=1, firstName=null, middleName=null, lastName=null, nameCn=null, orcid=null, stid=null, country=null, authorPic=null, dead=0, email=null, emailSecond=null, emailThird=null, correspondingAuthor=0, authorType=1, ext={EN=AuthorExt(id=1330192932411019445, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, authorId=1330192932343910579, language=EN, stringName=Zhenhua Mu, firstName=Zhenhua, middleName=null, lastName=Mu, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=Guoneng Baosin Railway Group Co., Ltd. Transportation Service Branch, Erdos, Inner Mongolia, 017000, China, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null), CN=AuthorExt(id=1330192932461351095, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, authorId=1330192932343910579, language=CN, stringName=穆振华, firstName=null, middleName=null, lastName=null, prefix=null, suffix=null, authorComment=null, nameInitials=null, affiliation=null, department=null, xref=null, address=国能包神铁路集团有限责任公司 运输服务分公司, 中国·内蒙古 鄂尔多斯 017000, bio=null, bioImg=null, bioContent=null, aboutCorrespAuthor=null)}, companyList=[AuthorCompany(id=1330192932092252325, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, xref=null, ext=[AuthorCompanyExt(id=1330192932109029542, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, companyId=1330192932092252325, language=EN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=Guoneng Baosin Railway Group Co., Ltd. Transportation Service Branch, Erdos, Inner Mongolia, 017000, China), AuthorCompanyExt(id=1330192932125806759, tenantId=1045748351789510663, journalId=1327663034936098855, articleId=1330177811802493654, companyId=1330192932092252325, language=CN, country=null, province=null, city=null, postcode=null, companyName=null, departmentName=null, remark=国能包神铁路集团有限责任公司 运输服务分公司, 中国·内蒙古 鄂尔多斯 017000)])])]
高丽芬,穆振华.
基于多维时空特征与动态时序逻辑的智慧后厨安全检测系统[J].
信息科学与工程研究, 2026, 7(8): 14-16 DOI:10.12349/iser.v7i8.11560
| [1] |
胡正平, 张乐, 殷艳华. 时空深度特征AP聚类的稀疏表示视频异常检测算法[J]. 信号处理, 2019, 35(3): 386-395.
|
| [2] |
李臣明, 吴媛媛, 漆美林, 等. 基于双曲空间图卷积的无监督视频异常检测方法[J]. 计算机应用研究, 2025, 42(7): 2234-2240.
|
| [3] |
魏明军, 周泰宇, 季展林, 张新南. 基于Mask-YOLO的复杂场景口罩佩戴检测[J]. 应用科学学报, 2022, 40(1): 93-104.
|