32results about How to "Realize safety supervision" patented technology

The invention discloses an integrated digital transfusion apparatus and a transfusion monitoring and management method thereof. The integrated digital transfusion apparatus is mainly structurally characterized in that a shell of devices including a control system, a display and the like and a transfusion tube clip and device shell are respectively fixed onto an integrated transfusion support and device shell. A touch display, a microprocessor, a microphone / speaker, a scanner, a wireless network connection device, a miniprinter, a transfusion tube, a Murphy dripper clamping groove, a finger peristaltic pump, a photoelectric droplet detection device, a pressure detector and an ultrasonic bubble detector are mounted on the integrated digital transfusion apparatus. Data acquired by different parts in the transfusion apparatus are used for mutual comparison of transfusion parameters. Software computation and simulation techniques are used during transfusion, and nurses can well know detailed transfusion conditions of each patient in transfusion by directly observing simulated transfusion bottles / bags and tubes and simulated transfusion apparatuses on software of the transfusion apparatuses and software on external computers, setting operation of various transfusion parameters, setting alarm thresholds of the various transfusion parameters and the like.

The invention relates to the technical field of blasting engineering priming control and discloses a digital electronic detonator large-scale priming system for engineering blasting and a method thereof. A priming network main machine is sequentially connected with a plurality of priming network auxiliary machines in series through an RS-485 serial communication cable; and each priming network auxiliary machine is connected with a digital electronic detonator group. According to operation on the priming network main machine, fault detection, order control, information inquiry and display are performed to a whole priming network. According to an order of the priming network main machine, each priming network auxiliary machine performs +24 voltage-to-digital converter (VDC) power supplying and sends an order information code to the digital electronic detonator group connected. Corresponding power supply control action-charging and priming are performed and corresponding communication information is fed back. According to the digital electronic detonator large-scale priming system for the engineering blasting and the method thereof, visible operation of large-scale complex priming can be achieved, a priming safety factor is increased, dangerous operation is reduced, environmental pollution is reduced, and good economic and social benefits are achieved and the digital electronic detonator large-scale priming system for the engineering blasting and the method thereof are suitable for being used in a high-precision millisecond blasting occasion.

The invention relates to an intelligent dangerous chemicals monitoring and management system. The intelligent dangerous chemicals monitoring and management system comprises an intelligent monitoring unit, a data processing unit, an expert unit, an emergency response unit, an emergency help unit and a software unit, wherein the intelligent monitoring unit is used for acquiring the state information and the storage environment information of the dangerous chemicals and transmitting the information to the expert unit after the information is processed through the data processing unit, not only can the abnormal information and the alarm information be included, but also any event inconsistent with original setting can be memorized and transmitted, in combination with the expert unit, setting points and an operation flow are ameliorated, the emergence response unit is used for realizing response according to the instruction information emitted by the data processing unit, and the emergence help unit is used for receiving the instruction information of the data processing unit and emitting the emergence help information to the expert unit according to the instruction information or directly controlling a safety protection device of the intelligent monitoring unit for action. The system is advantaged in that integral safety, stability and reliability of the equipment are improved, safety monitoring and control on the dangerous chemicals can be guaranteed for realization, a skill-improvement platform is further provided for operators, and a true field is further provided for experts.

The embodiment of the invention discloses a source-tracing processing method of Internet media files, a server, and a communication system. The method comprises steps that in response to an operation processing request of a user for a media file, a server acquires user information and a operation processing type carried in the operation processing request, the time of the operation processing, and the server information of the server; that the server packages the user information, the operation processing type, the time of the operation processing, and the server information of the server in order to acquire packaging information; and that the server writes the packaging information in a predetermined position in the metadata of the media file, generate a new media file, and performs corresponding operation processing on the new media file according to the operation processing request. The method may trace the sources of media files transmitted in Internet.

The invention discloses a food safety unified monitoring platform and method, and relates to the technical field of food safety information regulation. The problem that an existing food safety information regulation mode is low in automation degree, poor in data timeliness and complex in inspection is solved. The food safety unified monitoring platform comprises a receiving module and a generating module. The receiving module is used for receiving food safety information input by certificate bodies, wherein the food safety information includes one or more kinds of the food raw material information, the food processing feeding information, the warehouse out-put and in-put information, the distribution flow direction information and the consumption information. The generating module is used for saving the selected food safety information and generating a food safety information record chart. The standardization and the timeliness of the food safety information, the data interaction accuracy and the data statistic automation are accordingly achieved, data can be conveniently reported and read, and the food safety unified monitoring requirements can be well met.

The invention discloses a construction personnel safety belt wearing detection method and device based on visual analysis, and the method comprises the steps: collecting an image sample of a construction personnel, forming a sample set, capturing a trunk region image of the construction personnel, and dividing the trunk region image into a safety belt wearing state and a safety belt non-wearing state; on the sample set, using a VGG training classification model to obtain a safety belt wearing classification model; acquiring a real-time video stream of a construction site monitoring camera, andextracting to obtain a key frame image; analyzing the key frame image by utilizing an OpenPose human body posture estimation algorithm, judging whether a human body key point is detected or not, andif so, positioning a trunk region of a human body; calling a safety belt wearing classification model to classify and identify the trunk area, and judging whether the constructor wears the safety beltor not. According to the invention, 24-hour all-weather nonstandard behavior detection of constructors can be realized, so that the safety supervision of a construction site is realized, and the construction efficiency and safety are improved.

The invention discloses an information security baseline learning method based on artificial intelligence. According to the method, a safety baseline training model is constructed based on the deep learning algorithm, and the finally obtained safety baseline classification model can show very strong self-learning ability and adaptive ability through training and learning, so the safety baseline classification model can be applied to different scenes in subsequent application, and the safety baseline classification model can be applied to different scenes; the information security baseline capable of evaluating the information security of the industrial Internet platform from multiple network layers and application layers is automatically generated, so dependence on manual work is eliminated, manual setting of workers is not needed, and working efficiency is improved. Moreover, as the training feature data used for constructing the security baseline classification model comes from datainvolved in the interaction process of the client and the server, the information security baseline configured based on the security baseline classification model can realize security monitoring of the client and improve the security of the server; and security monitoring of the server and the interaction process between the client and the server can be realized.

A multi-mode relay forwarding method based on data security supervision is provided. By using the relay forwarding method of the invention, firstly, the safety supervision of forwarding data can be realized on the relay satellite to avoid forwarding erroneous data to the target. Secondly, using the time-delay relay mode, the remote control and relay transmission of telemetry data can be completedwhen the ground station and spacecraft can not be visible at the same time, and can meet the needs of high-precision time telecontrol mission. At the same time, the data hosting and timing auto-control functions can be provided for spacecraft through the return and forward mode. The relay forwarding mode of the invention can realize the safety supervision of forwarding data, meet the requirementsof a plurality of tasks, improve the flexibility of the relay satellite, reduce the time cost of ground maintenance, and expand the application mode of the relay satellite.

The invention discloses a building construction site safety production real-time monitoring and early warning method based on a remote video monitoring analysis technology. The method comprises the steps: carrying out the region division of a building construction site region, and setting a three-dimensional monitoring camera and an early warning terminal in each divided sub-region to carry out remote video monitoring analysis on whether the construction waste exists in each sub-region, carrying out risk assessment on the sub-region with the construction waste, so that early warning is carried out according to the assessed risk, safety supervision of the construction waste on the building construction site is realized, the defect that the safety monitoring management direction of the building construction site is too single at present is overcome, early warning is performed on the corresponding sub-region through the monitoring management result, the personal safety of constructors in the corresponding sub-region is greatly guaranteed, further aggravated pollution of atmospheric environment pollution is avoided, and the safety supervision level of the building construction site can be improved.

The embodiment of the invention discloses an AI model supervision method and system, a server and a storage medium. The method comprises the steps of obtaining information required for development ofa to-be-supervised AI model; performing model development according to the information required for development of the AI model to obtain the AI model; and recording the AI model into a block chain. The embodiment of the invention further discloses a supervision method of the AI model. By adopting the means, supervision of key information in each life cycle of the AI model can be credible and transparent. And meanwhile, monitoring and tracing of related information of the AI model can be realized.

The invention discloses a fireproof charging operation and maintenance integrated system and an operation and maintenance method thereof. The system comprises a parking lot, a fire detection mechanism, a transmission device and a remote monitoring control platform. The parking lot is provided with a plurality of parking charging positions and at least one fire insulation processing device, and the plurality of parking charging positions are respectively used for parking a plurality of electric vehicles. The transmission device is used for sending fire information. And the remote monitoring control platform is used for receiving and displaying the fire information and transmitting the fire information to a fire emergency center. The remote monitoring control platform is further used for judging the parking charging position where a fire-happening electric vehicle is parked according to the fire information when the fire detection mechanism detects that at least one electric vehicle is in fire, and then driving a moving vehicle to move to the front side or the rear side of the parking charging position where the fire-happening electric vehicle is parked. And finally, two fire insulation roller shutters are driven to be pushed out so as to isolate the electric automobile in the fire disaster. According to the invention, the fire is prevented from further spreading, and vehicles around the fire vehicle are protected.

Provided are a video data management and control authentication method, module, device and platform, the method comprising: if there is first user equipment accessing a security control module, the security control module determining whether a current bidirectional authentication is a primary bidirectional authentication with the first user equipment; if so, the security control module performs bidirectional authentication with the first user equipment according to a locally stored initialization key, and if the first user equipment passes the bidirectional authentication, the security controlmodule imports an application authentication key from the first user equipment to the local for storage; and if the authentication is not the primary bidirectional authentication, the security control module performs bidirectional authentication with the first user equipment according to a locally stored application authentication key, and if the authentication is passed, the security control module executes a corresponding management and control operation. The security of video data management and control can be improved, planned and systematic comprehensive security supervision on the videodata is realized, and social management and credible evidence obtaining are facilitated.

The invention discloses a security monitoring intelligent management system based on a big data Internet of Things. The system comprises a track section division module, a track section image acquisition module, an obstacle recognition and analysis module, a safety database, a station statistics module, a station waiting crowd image acquisition module, a waiting area crowd statistics module, a voice prompt terminal, a station stopping parameter detection module, a station historical average pedestrian volume analysis module and a parameter processing center. The management server and the background display terminal are used for obtaining the comprehensive danger coefficient of the urban rail transit operation by fusing the rail danger coefficient, the station waiting passenger danger coefficient, the parking danger coefficient and the passenger flow danger coefficient corresponding to the urban rail transit. The method makes up the defect that the monitoring index of the current urban rail transit operation safety supervision means is too single, improves the reliability of the monitoring result, and meets the requirements of comprehensiveness and reliability of the current urban rail transit operation safety supervision.

The invention discloses a fireproof method and a fireproof controller for an electric vehicle charging pile. The charging pile comprises a fireproof base, a shell, a charging main body mechanism, a heat insulation mechanism and a heat dissipation mechanism, wherein the heat insulation mechanism comprises a fireproof box and a roller shutter motor, and a fireproof roller shutter, and the heat dissipation mechanism comprises a plurality of fans. The method comprises the following steps that whether a fire occurs outside the shell or not is detected; the real-time temperature of the shell is detected; When it is detected that a fire occurs outside the shell, the roller shutter motor is driven to rotate so as to enable the fireproof roller shutter to extend out to surround the charging main body mechanism; when it is detected that no fire occurs outside the shell, the roller shutter motor is driven to rotate so that the fireproof roller shutter can contract, and whether the real-time temperature exceeds a first preset temperature or not is judged; and when the real-time temperature exceeds the first preset temperature, the multiple fans are driven to rotate until the real-time temperature is lower than the first preset temperature. The method can prevent the charging main body mechanism from being burnt by external fire and prevent the fire from further spreading, so that the fireproof performance of the charging pile is improved, the use safety is improved, and the service life is prolonged.

The invention discloses an electric vehicle battery safety supervision early warning system which comprises a safety supervision mechanism connected between a lithium battery and an external terminal and an early warning mechanism installed on an electric vehicle, wherein the safety supervision mechanism communicates with the early warning mechanism and a cloud early warning platform, and the cloud early warning platform communicates with a user terminal. According to the electric vehicle battery safety supervision early warning system adopting the structure, the current state of the lithium battery is monitored through the safety supervision mechanism installed on the lithium battery, whether charging or discharging can be continued or not is judged, and if the charging or discharging exceeds the safety range, connection of an external circuit is stopped immediately; therefore, lithium battery deflagration accidents caused by continuous charging or discharging are avoided, and a reason data signal for forbidding work is sent to a cloud early warning platform and an early warning mechanism on the electric vehicle, so that safety supervision of the lithium battery is realized.

The invention discloses a fireproof electric vehicle charging pile and an intelligent fireproof charging station. The charging pile comprises a fireproof base, a shell, a charging main body mechanism, a heat insulation mechanism, a fire early warning mechanism, a heat dissipation mechanism and a controller. The heat insulation mechanism comprises a fireproof box, a roller shutter motor and a fireproof roller shutter; the fire early warning mechanism comprises a temperature sensor and a flame detector; and the heat dissipation mechanism comprises a plurality of fans. The controller drives the roller shutter motor to rotate so as to enable the fireproof roller shutter to stretch out when a fire disaster occurs, drives the roller shutter motor to rotate so as to enable the fireproof roller shutter to contract when the fire disaster does not occur, and judges whether the real-time temperature exceeds the first preset temperature or not. When the real-time temperature exceeds the first preset temperature, the controller drives multiple fans to rotate until the real-time temperature is lower than the first preset temperature. The charging pile can prevent the charging main body mechanism from being burnt by external fire disaster and prevent the fire disaster from further spreading, so that the fireproof performance of the charging pile is improved, the use safety is improved, and the service life is prolonged.

The invention discloses a dangerous construction behavior identification method and device based on space-time characteristics. The dangerous construction behavior identification method comprises thesteps: A) collecting an action video sequence of a constructor, and dividing the action video sequence into a normal action and an illegal action; B) constructing an end-to-end identification network;C) extracting n frames of images in the image sequence, and inputting each frame of image into the same ResNet identification network to obtain output features corresponding to each frame of image; D) performing average pooling operation on F1 to Fn to obtain spatial features Fs of the image sequence; E) taking F1, F2,..., Fn as n time sequences, and inputting the n time sequences into an LSTM network to obtain a time feature Ft of the image sequence; and F) normalizing the Fs and the Ft to form a spatial-temporal feature F: F = Fs + Ft, and then accessing softmax to classify and identify toobtain an action type of the action video sequence. According to the invention, 24-hour all-weather illegal behavior detection of constructors can be realized, so that safety supervision of a construction site is realized, and the construction efficiency and safety are improved.

The invention provides a user safety verification method and system, and relates to the technical field of safety supervision. The method comprises the steps that human biological characteristic information and positioning information, which are collected by a user terminal, are acquired; the human biological characteristic information and the positioning information are compared with preset information; if the human biological characteristic information and / or the positioning information do not match the preset information, user abnormality information is sent to a user management terminal. According to the invention, safety verification can be carried out on a user; safety supervision of the user is realized; and safety accidents are prevented.

The embodiment of the invention discloses a screen information monitoring method and device, a network camera and a storage medium, and the method comprises the steps: analyzing the information content, collected in real time, displayed on a screen, and enabling an analysis result to be matched with the pre-stored standard content which is the to-be-displayed information which is approved; judgingwhether abnormal content exists in the currently displayed information or not according to a matching result; if so, controlling the infrared light supplementing lamp bead to emit an infrared remotecontrol signal to an infrared receiving controller of the screen so as to control the abnormal content displayed by the screen. According to the embodiment of the invention, the screen content is directly subjected to safety monitoring through the network camera, the manual check investment is reduced, and the screen can be directly controlled even if the network is abnormal or the video lines ofthe screen controller and the screen are damaged.

The invention discloses a construction safety judgment method based on binocular vision. The construction safety judgment method comprises the steps: collecting a sample picture of a construction site, and manually marking the position coordinates of a constructor and a crane in the sample picture; training a constructor detection and crane detection model by adopting a deep convolutional neural network; obtaining binocular camera parameters; a binocular video frame of a construction site is acquired in real time through a binocular camera; detecting the position coordinates of the constructorand the position coordinates of the crane in the binocular video frame by using the constructor detection and crane detection model; when x2 + w2 is smaller than x1 and x1 + w1 is smaller than x2, conducting stereo matching on the constructor and the crane detected in the binocular video frame, and obtaining the depth value z1 of the constructor and the depth value z2 of the crane; and when the absolute value of z1-z2 is smaller than the epsilon, considering that the constructor is located below the crane and belongs to dangerous construction behaviors. According to the invention, 24-hour all-weather dangerous behavior detection of constructors can be realized, so that safety supervision of a construction site is realized, and the construction efficiency and safety are improved.

The invention provides a normally-closed intelligent management device for a domestic gas system. The normally-closed intelligent management device comprises a gas flow meter (1), a microelectronic processor (2), a wireless signal receiver (3), a built-in electric valve (4), a gas appliance (5), electronic switch annunciators (6) and a gas pipeline (7). The normally-closed intelligent management device is characterized in that an integrated circuit board is integrated through the wireless signal receiver (3) and the microelectronic processor (2), the integrated circuit board is arranged in thegas flow meter (1), the built-in electric valve (4) is arranged in the gas flow meter (1) and is in a closed state, one end of the built-in electric valve is externally connected with a municipal gaspipe, and the other end of the built-in electric valve is connected with the gas pipeline (7) and the gas appliance (5); the electronic switch annunciators (6) are connected with a burner switching valve and an ignition switching valve which are driven by a thermocouple in a flame-out protection device of the gas appliance (5) in parallel; and the gas appliance (5) comprises a gas cooker (51) anda gas water heater (52), and the electronic switch annunciators (6) comprise a gas cooker electronic switch annunciator (61) and a gas water heater electronic switch annunciator (62).

The invention discloses a control system for a wireless detonator. The control system comprises a controller, wireless communication equipment, a server, a first client end and a second client end, wherein the controller is used for sending a first detonating request input by a blaster to the server through the wireless communication equipment; the first client end is used for sending security alert information input by an administrator to the server; and the server is used for sending the first detonating request and the security alert information to the second client end, receiving a seconddetonating request sent by the second client end, and sending the second detonating request to the first client end. The first client end is also used for sending a detonating instruction to the server, and the server is also used for sending the detonating instruction to the controller through the wireless communication equipment, so that the controller controls the detonator to be detonated according to the detonating instruction. Thus, the phenomenon of mistaken detonation can be reduced, and the security is improved.

The invention belongs to the technical field of old building safety supervision and early warning systems, and particularly relates to an old building safety supervision and early warning device basedon an NB-LOT network, which comprises a wireless NB-LOT inclination angle sensor terminal, a wireless NB-LOT vibration sensor terminal, a cloud data center and a safety supervision and early warningsystem. The wireless NB-LOT inclination angle sensor terminal comprises an ARM processor, and a chip SCA103T realizes a function assembly, a memory, an NB-LOT communication module and a power supply assembly based on an MEMS. The wireless NB-LOT vibration sensor terminal comprises a detection unit MEMS micro-accelerometer, a micro-processing unit, a wireless transmission unit NB-LOT communicationmodule and a power unit. The old building safety supervision and early warning system based on the NB-LOT network has the advantages of being low in power consumption and cost, stable in network, reliable in work, real in data, high in stability, capable of achieving safety supervision of old buildings and capable of conducting early warning on the old buildings when necessary.

The embodiment of the present invention discloses a supervision method, system, server and storage medium of an AI model, including: obtaining information required for the development of the AI ​​model to be supervised; developing the model according to the information required for the development of the AI ​​model to obtain The AI ​​model; recording the AI ​​model on the block chain. The embodiment of the present invention also discloses a method for supervising an AI model, including: obtaining usage information of the AI ​​model to be supervised; and recording the usage information of the AI ​​model on a block chain. Using this method, the credible and transparent supervision of key information in each life cycle of the AI ​​model can be realized. At the same time, it can monitor and trace the information related to the AI ​​model.

The invention provides an Internet information service method and platform based on artificial intelligence, and relates to the technical field of Internet. The invention discloses an Internet information service method based on artificial intelligence. The method comprises the following steps: carrying out data acquisition and preprocessing on Internet information through a distributed data acquisition technology; inputting the collected data information into a pre-trained machine learning model so as to classify the data information, and setting corresponding service labels; evaluating the information service data added with the service label according to published file data in the Internet information service data, and storing and displaying an evaluation result. According to the method, the functions of automatic word segmentation, stop word removal and the like of information can be achieved through data collection and analysis, service information data are added into data collection, the service efficiency is improved, and after data arrangement is completed, the function of setting labels for the data is completed in order to meet subsequent requirements.

The invention relates to the technical field of blasting engineering priming control and discloses a digital electronic detonator large-scale priming system for engineering blasting and a method thereof. A priming network main machine is sequentially connected with a plurality of priming network auxiliary machines in series through an RS-485 serial communication cable; and each priming network auxiliary machine is connected with a digital electronic detonator group. According to operation on the priming network main machine, fault detection, order control, information inquiry and display are performed to a whole priming network. According to an order of the priming network main machine, each priming network auxiliary machine performs +24 voltage-to-digital converter (VDC) power supplying and sends an order information code to the digital electronic detonator group connected. Corresponding power supply control action-charging and priming are performed and corresponding communication information is fed back. According to the digital electronic detonator large-scale priming system for the engineering blasting and the method thereof, visible operation of large-scale complex priming can be achieved, a priming safety factor is increased, dangerous operation is reduced, environmental pollution is reduced, and good economic and social benefits are achieved and the digital electronic detonator large-scale priming system for the engineering blasting and the method thereof are suitable for being used in a high-precision millisecond blasting occasion.

The invention discloses an underground optical communication system based on visible light communication. The system comprises a fixed lamp and a camera which are both mounted on the side wall of a mine hole, a cleaning mechanism is arranged outside the fixed lamp; a controller is electrically connected with the fixed lamp and the photoelectric receiving plate; and the system further comprises a data server located at the ground position of the mine. The data server is electrically connected with the nearest controller; the data information communication module serves as uplink communication and transmits the uplink communication to the data server; the data server takes the control command as downlink communication and transmits the control command to the plurality of communication modules through the nearest controller to finish communication; the video monitoring and two-way communication can be achieved through an original lighting system, meanwhile, the cleaning mechanism with theexplosion-proof effect is further arranged, long-term stable operation of the system is facilitated, in addition, information can be transmitted to each individual, and therefore safety is improved,and the system is suitable for underground environment conditions.