Oil depots, petrol stations, and unloading trucks store a large amount of refined fuel, and safety is the top priority.
In particular, the unloading process is most prone to safety hazards, the dynamic management of the tanker is difficult, the unloading process does not have real-time video, data analysis, and warning mechanisms, and it is difficult to investigate and collect evidence afterwards.
How to use intelligent means to strengthen the safety management of gas stations, ensure the safety of the internal and external areas of the gas station, and create a good and safe gas environment?
Ronglian AI Vision-Smart Eye Platform is equipped with specific algorithms in the petroleum and petrochemical industry, integrating intelligent analysis and identification, alarm push and reminding, data statistics and analysis and other business systems to integrate and control.
Intelligent analysis and recognition
Through intelligent video analysis algorithms, model the "people, things, and things" of gas stations, and automatically identify violations and behaviors.
Personnel service and off-job identification
Detection standard: Identify the uniforms of the driver/escort staff, do not leave the camera range during the unloading process, and immediately issue a violation warning once discovered.
Object detection-electrostatic clip recognition
Detection standard: Identify the staff holding the electrostatic clamp or the electrostatic clamp on the oil dump truck, otherwise the violation warning will be issued immediately.
Object detection-tripod recognition
Detection standard: The identification staff holds the tripod wood (the tripod wood is yellow) or places the tripod wood on the wheel, otherwise the violation warning will be issued immediately.
Object detection-fire extinguisher recognition
Detection standard: Identify that there is an 8kg fire extinguisher at the front and rear of the vehicle, and there is a 35kg fire extinguisher near the discharge port, and the fire extinguisher belt is unfolded, otherwise an alarm will be issued immediately.
Object detection-unloading pipe recognition
Detection standard: Recognize that the staff is approaching the ground port with the unloading pipe in hand, and then approaching the vehicle port with the other end of the unloading pipe in hand (note the order), and the process error will immediately issue a violation alarm.
Alarm push and reminder
For each gas station, there is a result push for each unloading:
1) If the process is qualified and does not alarm, the staff can view the entire process in detail.
2) If the process specification is unqualified, an alarm reminder will be automatically pushed, and the relevant staff will check and notify the shift supervisor. At the same time, you can view the specific violations in detail in the background, as well as the captured pictures of violations.
Multi-dimensional data statistical analysis capabilities
Device management: Support the management of devices connected to the camera.
Video management: Support access to video stream frame extraction, time threshold, etc. for management.
Algorithm management: Support the configuration and management of algorithm models.
Support a variety of chart types, data can be freely analyzed in detail according to the level of the dimension, and support operations such as data sorting and warning lines.
Abundant algorithms, leading deep learning platform, complete product ecology, deep industry experience and understanding, flexible service delivery methods, in the field of visual intelligence, from seeing to reading, we have more industry algorithms and scenario applications. Accelerate the intelligent upgrading of the industry.