技术领域technical field
本发明涉及一种视频混合传感数据描述方法，尤其是一种面向变电站内视频与异构传感器的结构化描述方法，属于视频描述技术领域。The invention relates to a video hybrid sensor data description method, in particular to a structured description method for video and heterogeneous sensors in a substation, and belongs to the technical field of video description.
背景技术Background technique
视频信息以提供丰富和生动的信息，已经得到各行各业的广泛应用，如银行、商场、超市、交通路段、小区路口等。变电站内部为了清晰直观监测重要设备的运行状况，也都安装了视频监控，但是视频监控仅提供图像/视频信息，却无法进行预警，可以为设备故障留有视频记录，因此变电站内往往加装其他传感设备。但是由于传感设备和视频监控分属不同厂家，目前主要是将视频数据和其他传感器数据分别存储到数据库中，无法进行统一的数据传输和存储。In order to provide rich and vivid information, video information has been widely used in various industries, such as banks, shopping malls, supermarkets, traffic sections, intersections in residential areas, etc. In order to clearly and intuitively monitor the operation status of important equipment in the substation, video surveillance is also installed. However, the video surveillance only provides image/video information, but cannot provide early warning. Video records can be left for equipment failures. Therefore, other substations are often installed in the substation. sensing equipment. However, since the sensor equipment and video surveillance belong to different manufacturers, video data and other sensor data are mainly stored in the database at present, and unified data transmission and storage cannot be performed.
视频数据往往存在数据量大、时空相关度高和描述二义性等问题，难以用传统的数据模型进行描述。目前视频多采用面向对象的编码方式，并且集中在在内容和信息本身，如对视频及音频数据的压缩、编码和表达，而传感器数据单独存放在数据库中，无法进行综合的分析，需要时需要单独进行比对。Video data often has problems such as large data volume, high spatio-temporal correlation, and description ambiguity, which are difficult to describe with traditional data models. At present, video mostly adopts object-oriented encoding method, and focuses on the content and information itself, such as compression, encoding and expression of video and audio data, while sensor data is stored separately in the database, which cannot be comprehensively analyzed. Compare individually.
如何将视频信息与温度、振动等传感数据进行结合，采用结构化的描述，一直以来是一个难题。How to combine video information with sensor data such as temperature and vibration, and adopt a structured description, has always been a difficult problem.
发明内容Contents of the invention
本发明的目的是克服现有技术中存在的不足，提供一种面向变电站内视频与异构传感器的结构化描述方法，实现异构传感数据的结构化，方便综合信息的编码和传输，提高信息处理和检索的效率，使检索结果更加准确。The purpose of the present invention is to overcome the deficiencies in the prior art, provide a structured description method for video and heterogeneous sensors in substations, realize the structuring of heterogeneous sensor data, facilitate the encoding and transmission of comprehensive information, and improve The efficiency of information processing and retrieval makes retrieval results more accurate.
按照本发明提供的技术方案，一种面向变电站内视频与异构传感器的结构化描述方法，其特征是，采用分层描述，具体如下：According to the technical solution provided by the present invention, a structured description method for video and heterogeneous sensors in a substation is characterized in that a hierarchical description is used, specifically as follows:
第一层对传感信息进行采集，并对传感器数据进行预处理，提取特征值；The first layer collects sensing information, preprocesses sensor data, and extracts feature values;
第二层根据传感信息关联的时间信息，对同一时刻的视频信息进行原始图像的物理属性描述；The second layer describes the physical properties of the original image for the video information at the same time according to the time information associated with the sensing information;
第三层将图像转换成为逻辑关系，并关联传感器数据进行统一描述；The third layer converts images into logical relationships and associates sensor data for unified description;
第四层从图像、对象、以及关联传感信息表示中抽象出语义属性，以及从逻辑关系中转换的语义，获得结构化描述符。The fourth layer abstracts semantic attributes from images, objects, and associated sensory information representations, and converts semantics from logical relationships to obtain structured descriptors.
进一步的，所述第二层视频信息的物理属性描述按照视频段进行描述，抽取每帧图像的特征信息。Further, the physical attribute description of the second-layer video information is described according to video segments, and feature information of each frame of image is extracted.
进一步的，所述第二层视频信息的物理属性描述包括颜色信息、轮廓信息、纹理信息、运动信息、声音信息和能量信息。Further, the physical attribute description of the second layer video information includes color information, outline information, texture information, motion information, sound information and energy information.
进一步的，所述结构化描述符主要用于描述以下信息类型：对象描述、传感关联信息描述、关系描述、运动关系描述以及运维人员描述。Further, the structured descriptor is mainly used to describe the following types of information: object description, sensor-related information description, relationship description, motion relationship description, and operation and maintenance personnel description.
进一步的，所述运维人员描述主要描述以下信息类型：人员描述、位置描述和关系描述。Further, the operation and maintenance personnel description mainly describes the following information types: personnel description, location description and relationship description.
进一步的，所述第一层传感器数据进行预处理后添加时间戳。Further, the first layer of sensor data is preprocessed and then added with a time stamp.
进一步的，所述传感信息包括温度信息、噪声信息和局放信息。Further, the sensing information includes temperature information, noise information and partial discharge information.
进一步的，所述结构化描述的前提条件为，视频监控系统和物联网系统均已经按照UTC统一时间参考体系完成时间同步。Further, the precondition of the structured description is that both the video surveillance system and the Internet of Things system have completed time synchronization according to the UTC unified time reference system.
本发明所述面向变电站内视频与异构传感器的结构化描述方法，能够将视频信息与异构传感信息进行综合关联描述，方便统一传输、存储，并提高传输的速度，也使得后期检索更加高效。The structured description method for videos and heterogeneous sensors in substations described in the present invention can comprehensively associate and describe video information and heterogeneous sensor information, facilitate unified transmission and storage, improve transmission speed, and make later retrieval easier efficient.
附图说明Description of drawings
图1为本发明所述视频与异构传感器的结构化描述的流程图。Fig. 1 is a flowchart of the structured description of the video and heterogeneous sensors of the present invention.
图2为本发明所述视频与异构传感器的结构化描述模型的示意图。Fig. 2 is a schematic diagram of a structured description model of video and heterogeneous sensors according to the present invention.
具体实施方式detailed description
下面结合具体附图对本发明作进一步说明。The present invention will be further described below in conjunction with specific drawings.
未处理的视频数据是非结构化的数据，目前所采用的信息描述方法，主要通过图像、声音和文本来表现一个完整的语义概念，仅为面向对象进行描述，无法将环境信息融入其中，已经不能满足需求，通过与传感数据结合，可以构成同一时间点的多角度信息描述，方便联合存储、传输与检索。Unprocessed video data is unstructured data. The current information description method mainly expresses a complete semantic concept through images, sounds and texts. It is only for object-oriented description and cannot integrate environmental information into it. Requirements, combined with sensor data, can form a multi-angle information description at the same time point, which is convenient for joint storage, transmission and retrieval.
本发明所述面向变电站内视频与异构传感器的结构化描述方法的前提条件，视频监控系统和物联网系统均已经按照UTC统一时间参考体系完成时间同步，采用描述符进行同一时间点的描述，从而实现视频信息与GPS、红外温度、局放等传感器信息进行关联，提高数据的交换性和数据管理的统一性，减少冗余信息，可以快速进行数据传输，和使用较少的存储空间。流程示意图如图1所示。The premise of the structured description method for video and heterogeneous sensors in substations in the present invention is that both the video monitoring system and the Internet of Things system have completed time synchronization according to the UTC unified time reference system, and use descriptors to describe the same time point. In this way, video information can be associated with GPS, infrared temperature, partial discharge and other sensor information, data exchange and data management unity can be improved, redundant information can be reduced, data transmission can be performed quickly, and less storage space is used. The flow diagram is shown in Figure 1.
如图1所示，本发明中所述结构化描述方法主要如下：As shown in Figure 1, the structured description method described in the present invention is mainly as follows:
将描述信息划分为两大类：属于视频的描述信息和属于传感器的描述信息；视频描述按照视频段进行描述，抽取每帧图像的特征信息以及语义层分析；传感器数据通过预处理，提取特征值，构建语义层分析。视频底层的信息与对象无关：主要是时间信息、格式信息、创建信息和特征信息；视频帧的关键帧主要通过颜色、纹理、运动等进行描述；结构化描述符主要用于描述以下信息类型：对象描述、传感关联信息描述、关系描述、运动关系描述以及运维人员描述。运维人员描述主要描述以下信息类型：人员描述、位置描述和关系描述。进行分层描述，提高数据的交换性和数据管理的统一性，便于通过迭代过程查询，进而可以高效地查询。The description information is divided into two categories: the description information belonging to the video and the description information belonging to the sensor; the video description is described according to the video segment, the feature information of each frame image is extracted and the semantic layer analysis; the sensor data is preprocessed to extract the feature value , to build a semantic layer analysis. The underlying information of the video has nothing to do with the object: mainly time information, format information, creation information and feature information; key frames of video frames are mainly described by color, texture, motion, etc.; structured descriptors are mainly used to describe the following types of information: Object description, sensor association information description, relationship description, motion relationship description, and operation and maintenance personnel description. Operation and maintenance personnel description mainly describes the following information types: personnel description, location description, and relationship description. Hierarchical description is carried out to improve the exchangeability of data and the unity of data management, which is convenient for query through iterative process, and then can be efficiently queried.
如图2所示，为本发明所述面向变电站内视频与异构传感器的结构化描述方法进行分层描述的模型：第一层是对如温度、噪声、局放等传感信息进行采集，并对数据进行简单的预处理，因为传感数据的变化缓慢，通过记录变化值，可以减少存储内存，并将预处理的信息添加时间戳，并统计特征，添加特征及文本注释；第二层是根据传感信息关联的时间信息，主要是对同一时刻的视频信息进行原始图像的物理属性描述，即图像客观存在的视觉能够区分的性质，如对颜色、轮廓、纹理、运动、声音、能量等物理特征进行描述；第三层是将图像转换成为逻辑关系，并关联传感数据进行统一描述，如变压器箱的轮廓信息、温度信息、噪声信息等变压器箱关联信息，断路器的轮廓信息、温度信息、SF6断路器信息等断路器的关联信息；第四层是从图像、对象、以及关联传感信息表示中抽象出语义属性，如变压器箱正常，其一般属性，以及从逻辑结构中转换的语义。As shown in Figure 2, it is a model for layered description of the structured description method for video and heterogeneous sensors in substations according to the present invention: the first layer is to collect sensing information such as temperature, noise, partial discharge, etc. And simple preprocessing of the data, because the sensor data changes slowly, by recording the change value, the storage memory can be reduced, and the preprocessed information can be added with a time stamp, and statistical features, adding features and text annotations; the second layer It is based on the time information associated with the sensing information, mainly to describe the physical properties of the original image for the video information at the same moment, that is, the visually distinguishable properties of the objective existence of the image, such as color, outline, texture, motion, sound, energy The third layer is to convert the image into a logical relationship, and associate the sensor data for a unified description, such as the outline information of the transformer box, temperature information, noise information and other related information of the transformer box, the outline information of the circuit breaker, The associated information of circuit breakers such as temperature information and SF6 circuit breaker information; the fourth layer is to abstract semantic attributes from images, objects, and associated sensor information representations, such as the normality of the transformer box, its general attributes, and conversion from the logical structure semantics.
本发明是对视频信息和传感信息进行描述的方法，现有技术视频信息描述与传感数据信息描述属于两个完全独立的系统，本发明将变电站内部的摄像机与传感器的数据进行统一的描述，便于进行统一的编码和后续传输，提高信息处理和检索的效率，使检索结果更加准确。The present invention is a method for describing video information and sensing information. In the prior art, video information description and sensing data information description belong to two completely independent systems. The present invention provides a unified description of the camera and sensor data inside the substation. , to facilitate unified coding and subsequent transmission, improve the efficiency of information processing and retrieval, and make retrieval results more accurate.
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710242861.4ACN107025292A (en) | 2017-04-14 | 2017-04-14 | The description method of video and heterogeneous sensor in towards transformer station |
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710242861.4ACN107025292A (en) | 2017-04-14 | 2017-04-14 | The description method of video and heterogeneous sensor in towards transformer station |
| Publication Number | Publication Date |
|---|---|
| CN107025292Atrue CN107025292A (en) | 2017-08-08 |
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710242861.4APendingCN107025292A (en) | 2017-04-14 | 2017-04-14 | The description method of video and heterogeneous sensor in towards transformer station |
| Country | Link |
|---|---|
| CN (1) | CN107025292A (en) |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107831409A (en)* | 2017-09-22 | 2018-03-23 | 国网山东省电力公司电力科学研究院 | The method and method for detecting abnormality of superfrequency partial discharge detection TuPu method parameter extraction |
| CN110998245A (en)* | 2017-08-09 | 2020-04-10 | 欧姆龙株式会社 | Sensor management unit, sensor device, sensor management method, and sensor management program |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102348101A (en)* | 2010-07-30 | 2012-02-08 | 深圳市先进智能技术研究所 | Examination room intelligence monitoring system and method thereof |
| CN103248703A (en)* | 2013-05-16 | 2013-08-14 | 中国农业大学 | Automatic monitoring system and method for live pig action |
| US20140082002A1 (en)* | 2012-09-20 | 2014-03-20 | Electronics And Telecommunications Research Institute | Apparatus and method for processing unstructured data event in real time |
| CN104767974A (en)* | 2015-04-09 | 2015-07-08 | 合肥三立自动化工程有限公司 | Intelligent remote monitoring terminal with video monitoring function and control method thereof |
| CN104899261A (en)* | 2015-05-20 | 2015-09-09 | 杜晓通 | Device and method for constructing structured video image information |
| CN105847752A (en)* | 2016-04-22 | 2016-08-10 | 广东中星电子有限公司 | Information encoding and decoding method, equipment and video monitoring system |
| CN106355811A (en)* | 2016-07-22 | 2017-01-25 | 河南城建学院 | Electrical fire monitoring system |
| CN106485412A (en)* | 2016-09-30 | 2017-03-08 | 国网江苏省电力公司电力科学研究院 | A kind of transformer station image intelligent level of analysis multiple task distributive computational methods |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102348101A (en)* | 2010-07-30 | 2012-02-08 | 深圳市先进智能技术研究所 | Examination room intelligence monitoring system and method thereof |
| US20140082002A1 (en)* | 2012-09-20 | 2014-03-20 | Electronics And Telecommunications Research Institute | Apparatus and method for processing unstructured data event in real time |
| CN103248703A (en)* | 2013-05-16 | 2013-08-14 | 中国农业大学 | Automatic monitoring system and method for live pig action |
| CN104767974A (en)* | 2015-04-09 | 2015-07-08 | 合肥三立自动化工程有限公司 | Intelligent remote monitoring terminal with video monitoring function and control method thereof |
| CN104899261A (en)* | 2015-05-20 | 2015-09-09 | 杜晓通 | Device and method for constructing structured video image information |
| CN105847752A (en)* | 2016-04-22 | 2016-08-10 | 广东中星电子有限公司 | Information encoding and decoding method, equipment and video monitoring system |
| CN106355811A (en)* | 2016-07-22 | 2017-01-25 | 河南城建学院 | Electrical fire monitoring system |
| CN106485412A (en)* | 2016-09-30 | 2017-03-08 | 国网江苏省电力公司电力科学研究院 | A kind of transformer station image intelligent level of analysis multiple task distributive computational methods |
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110998245A (en)* | 2017-08-09 | 2020-04-10 | 欧姆龙株式会社 | Sensor management unit, sensor device, sensor management method, and sensor management program |
| CN107831409A (en)* | 2017-09-22 | 2018-03-23 | 国网山东省电力公司电力科学研究院 | The method and method for detecting abnormality of superfrequency partial discharge detection TuPu method parameter extraction |
| CN107831409B (en)* | 2017-09-22 | 2020-08-18 | 国网山东省电力公司电力科学研究院 | Ultrahigh frequency partial discharge detection map characteristic parameter extraction method and anomaly detection method |
| Publication | Publication Date | Title |
|---|---|---|
| CN102880692B (en) | A kind of monitor video semantic description towards retrieval and detection modeling method | |
| CN103023970B (en) | Method and system for storing mass data of Internet of Things (IoT) | |
| CN103106565B (en) | Real-time display method and device for grid information based on GIS and SCADA | |
| CN106981063A (en) | A kind of grid equipment state monitoring apparatus based on deep learning | |
| CN112507035A (en) | Power transmission line multi-source heterogeneous data unified standardized processing system and method | |
| CN104717468B (en) | Cluster scene intelligent monitoring method and system based on the classification of cluster track | |
| CN115424059B (en) | A Remote Sensing Land Use Classification Method Based on Pixel-Level Contrastive Learning | |
| WO2016184314A1 (en) | Device and method for establishing structured video image information | |
| CN112186901B (en) | Panoramic sensing monitoring method and system for transformer substation | |
| CN101783881A (en) | Intelligent web camera with video structural description function | |
| CN105450978A (en) | Method and device for achieving structural description in video monitoring system | |
| Miao et al. | Fire detection algorithm based on improved yolov5 | |
| CN106649652A (en) | Construction method for on-line transformer substation main equipment monitoring system based on Hadoop | |
| CN107025292A (en) | The description method of video and heterogeneous sensor in towards transformer station | |
| CN109033476A (en) | A kind of intelligent space-time data event analysis method based on event clue network | |
| CN116010652B (en) | Unstructured video data processing method and system | |
| CN105528417A (en) | Body-based community surveillance video semantic model application system and method | |
| Kieran et al. | A framework for an event driven video surveillance system | |
| Yadav et al. | Knowledge graph driven approach to represent video streams for spatiotemporal event pattern matching in complex event processing | |
| Lei et al. | Recent advances in multi-modal 3d scene understanding: A comprehensive survey and evaluation | |
| US20210014540A1 (en) | Method and system for codec of visual feature data | |
| CN106649765A (en) | Smart power grid panoramic data analysis method based on big data technology | |
| CN118069885B (en) | A dynamic video content coding retrieval method and system | |
| Zhang et al. | Research on the attribute extraction algorithm of people and vehicles based on video structure | |
| Xiaodong et al. | Artificial Intelligence in power multimodal data analysis |
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication | Application publication date:20170808 |