Monitoring and early warning method for abnormal behaviors in granary

Abstract

The invention discloses a method for monitoring and early warning abnormal behaviors in a granary. The method comprises the steps that image information in the granary is extracted in real time through a camera device; the image information is analyzed, and abnormal behaviors in the granary are judged in a one-to-one correspondence mode in combination with multiple preset analysis algorithms; in combination with a multi-level early warning mechanism, the abnormal behaviors are graded to form abnormal behavior levels; different early warning levels are formed according to different abnormal behavior levels, and an early warning is given to management personnel of the granary; wherein the abnormal behaviors comprise grain surface transaction, region invasion, environment change, pests and mildew. Through real-time monitoring and intelligent analysis and in combination with a multi-stage early warning mechanism, instant detection and quick response of various abnormal behaviors in the granary are realized, the monitoring efficiency and safety are remarkably improved, meanwhile, the workload of management personnel is relieved, effective utilization and reasonable allocation of resources are ensured, small problems are prevented from being evolved into large accidents, and the system is suitable for popularization and application. And an important reference is provided for future risk management.

Description

Technical Field

[0001] The present invention belongs to the technical field of monitoring and early warning, and particularly relates to a method for monitoring and early warning of abnormal behaviors in a granary. Background Art

[0002] In the grain storage industry, with the expansion of grain storage scale and technological progress, the requirements for the safety, quality, and management of stored grain are also constantly increasing. To ensure the quality and quantity of grain, granaries are usually equipped with a series of monitoring devices and technical means to monitor the environmental conditions and the state of grain in the granary. Existing technologies have been able to monitor environmental parameters such as temperature and humidity in the granary, and these sensors can timely feedback environmental changes to help managers understand the basic situation of the granary.

[0003] However, these technical means mainly rely on the monitoring of sensor data and often depend on regular manual inspections, making it difficult to detect and judge abnormal behaviors in the granary in a timely manner. There are still some obvious limitations and deficiencies in the actual application process. Existing monitoring means mainly rely on sensors to monitor environmental parameters, lacking comprehensive video monitoring and intelligent analysis means, and unable to comprehensively and timely detect and judge abnormal behaviors in the granary.

[0004] In addition, the method relying on manual inspections is not only time-consuming and laborious, but also inefficient when facing large granaries, making it difficult to ensure the real-time nature and accuracy of monitoring. Most existing monitoring systems require manual intervention to effectively handle abnormal behaviors and fail to fully utilize intelligent technologies to improve monitoring efficiency and management level. Moreover, for the identification of abnormal behaviors such as unauthorized personnel entering the granary, privately using grain, pest breeding, and grain mildew, existing technologies have problems such as lagging response and incomplete monitoring. Summary of the Invention

[0005] The present invention provides a method for monitoring and early warning of abnormal behaviors in a granary to solve the problems of slow response and incomplete monitoring caused by the existing granary monitoring technology mainly relying on environmental parameter monitoring and regular manual inspections.

[0006] The technical solution adopted by the present invention is as follows:

[0007] A method for monitoring and early warning of abnormal behaviors in a granary, comprising:

[0008] S101: Extract the image information in the granary in real time through a camera device;

[0009] S102: Analyze the image information and combine a variety of preset analysis algorithms to respectively judge the abnormal behaviors in the granary;

[0010] S103: Combine with a multi - level early warning mechanism to classify the abnormal behaviors and form abnormal behavior levels;

[0011] S104: According to different abnormal behavior levels, form different early warning levels and give early warnings to the management personnel of the granary;

[0012] Among them, the abnormal behaviors include grain surface movement, area intrusion, environmental change, pests and mildew.

[0013] The abnormal behavior monitoring and early warning method in the granary of the present invention further includes the following additional technical features:

[0014] Analyze the image information, combine with a variety of preset analysis algorithms to correspondently judge the abnormal behaviors in the granary, including:

[0015] According to the image information, combine with a grain surface movement recognition model to analyze the shape change of the upper surface of the grain pile;

[0016] If depressions and / or inclination changes appear on the upper surface of the grain pile, combine with the incoming and outgoing warehouse business data to determine grain surface movement.

[0017] According to the image information, combine with a grain surface movement recognition model to analyze the shape change of the upper surface of the grain pile, specifically:

[0018] Through the grain surface movement recognition model, perform grain surface segmentation processing on the image information to identify grain surface edge changes and / or internal depressions of the grain surface.

[0019] Analyze the image information, combine with a variety of preset analysis algorithms to correspondently judge the abnormal behaviors in the granary, and also include:

[0020] According to the image information, combine with a feature recognition algorithm to obtain the personnel information of the people entering the granary;

[0021] Compare the personnel information with the authorization information of the personnel authorized to enter the granary to perform area intrusion judgment.

[0022] Analyze the image information, combine with a variety of preset analysis algorithms to correspondently judge the abnormal behaviors in the granary, and also include:

[0023] Analyze the color and / or texture features in the image, combine with the temperature and / or humidity environment parameters in the granary to monitor environmental changes.

[0024] Analyze the image information, combine with a variety of preset analysis algorithms to correspondently judge the abnormal behaviors in the granary, and also include:

[0025] Extract the morphological features in the image information to identify pests and / or mildew phenomena;

[0026] When the number of the pests and / or mildew phenomena exceeds the warning threshold, it is determined as abnormal pests and mildew;

[0027] Wherein, the warning threshold is set according to the grain storage area of the granary.

[0028] The imaging device is a panoramic camera, and a plurality of the panoramic cameras are installed on the top of the granary to obtain image information of the entire area inside the granary.

[0029] Combined with a multi-level warning mechanism, the abnormal behaviors are classified to form abnormal behavior levels, specifically:

[0030] The abnormal behavior levels include level one, level two and level three.

[0031] If the abnormal behavior is regional intrusion, it is determined as a level-one abnormal behavior;

[0032] If the abnormal behaviors are grain surface movement, environmental change, pests and mildew, a first threshold and a second threshold are set according to the change rates of the grain surface movement, environmental change, pests and mildew, and the first threshold is less than the second threshold;

[0033] When the change rate is less than the first threshold, it is determined as a level-three abnormal behavior;

[0034] When the change rate is greater than the first threshold and less than the second threshold, it is determined as a level-two abnormal behavior;

[0035] When the change rate is greater than the second threshold, it is determined as a level-one abnormal behavior;

[0036] Wherein the first threshold and the second threshold are determined according to the grain storage area of the granary.

[0037] According to different abnormal behavior levels, different warning levels are formed and warnings are sent to the managers of the granary, specifically:

[0038] Acoustic and light alarms and / or SMS notifications are sent to the managers so that the managers receive abnormal behavior warnings and respectively execute disposal strategies according to the warning levels;

[0039] When the abnormal behavior level is level one, managers are dispatched for on-site verification, the abnormal behavior is confirmed, and emergency treatment measures are implemented;

[0040] When the abnormal behavior level is level two, the inspection frequency is increased, the monitoring of the abnormal area is strengthened, and local adjustment is made to the abnormal behavior area;

[0041] When the abnormal behavior level is level three, the abnormal behavior area is concerned and preventive measures are taken.

[0042] The present invention also provides an abnormal behavior monitoring and early warning device in a granary, comprising:

[0043] A memory for storing a computer program;

[0044] A processor for implementing the steps of the abnormal behavior monitoring and early warning method in the granary when executing the computer program.

[0045] Due to the adoption of the above technical solution, the beneficial effects obtained by the present invention are as follows:

[0046] 1. In the present invention, through the imaging device, the image information inside the granary can be captured in real time, realizing all-time monitoring of the granary. This method overcomes the limitations of traditional monitoring means, ensuring that any abnormal behavior occurring at any time can be captured in time.

[0047] By carrying out real-time monitoring of the granary interior through the imaging device, it means that any abnormal behavior occurring in the granary can be captured. For example, in the event of area intrusion, the system can immediately detect the intruder and quickly notify the management staff to take action. In addition, since the video stream is transmitted in real time, the management staff can view the current state of the granary at any time through remote access to the system without waiting for regular inspection reports. This real-time monitoring ability significantly enhances the response speed to emergencies and provides a solid guarantee foundation for food security.

[0048] 2. In the present invention, by using a variety of preset analysis algorithms to intelligently analyze the image information, the abnormal behaviors occurring in the granary, such as grain surface movement, area intrusion (unauthorized personnel entering), environmental changes (environmental parameter changes), pest activities or grain mildew, etc., can be automatically identified and judged. This method improves the monitoring efficiency and reduces the need for manual intervention.

[0049] Using a variety of preset analysis algorithms to intelligently analyze the image information is not limited to one analysis algorithm, but through the integration of multiple algorithms, it can judge various abnormal situations, thereby improving the utilization rate of the images. This method not only improves the monitoring efficiency but also reduces the need for manual intervention. By integrating multiple algorithms, the system can combine the advantages of different algorithms to accurately identify and judge various abnormal situations in the granary. For example, in addition to the common image recognition algorithm for detecting grain surface movement, behavior analysis algorithms can be integrated to monitor the entry of unauthorized personnel (area intrusion), environmental monitoring algorithms for perceiving environmental parameter changes, and biometric recognition algorithms for identifying pest activities or grain mildew and other phenomena.

[0050] Through such an automated process, the system greatly reduces the workload of the staff, enabling them to focus on handling complex situations that truly require manual intervention rather than getting bogged down in daily repetitive inspections. Managers can arrange patrol tasks more targeted based on the analysis results of the system, ensuring the effective utilization of resources, and ultimately improving the overall operational efficiency and safety of the granary.

[0051] 3. In the present invention, in combination with a multi-level early warning mechanism, abnormal behaviors are classified into different levels, and corresponding early warning levels are formed accordingly. This mechanism can ensure that appropriate response measures are taken according to the severity of the abnormal behavior, preventing small problems from evolving into major accidents. The design of the multi-level early warning mechanism takes into account the risk differences that may be brought by abnormal behaviors at different levels. For example, minor environmental changes may be marked as a level-three early warning, only requiring recording and observing subsequent developments; while a severe pest invasion will be marked as a level-one early warning, and immediate action is needed to control the spread of pests.

[0052] The division of early warning levels not only helps to allocate resources reasonably but also enables managers to prioritize the handling of the most urgent situations. In addition, this mechanism also supports the dynamic adjustment of early warning levels, upgrading or downgrading the early warning in a timely manner according to the development of the situation, ensuring that the response measures always match the actual situation. This flexible early warning strategy effectively prevents the deterioration of problems caused by inappropriate responses.

[0053] 4. In the present invention, by setting up a multi-level early warning mechanism, once an abnormal behavior is identified, a corresponding early warning signal will be sent to the granary managers according to its level, prompting them to take corresponding measures to handle the problem. This greatly shortens the time from problem discovery to problem resolution and improves the overall safety of the granary. The rapid response mechanism ensures that the emergency procedure can be immediately activated after detecting an abnormality. For example, when an unauthorized person enters the granary is detected, the system can automatically trigger a level-one early warning, immediately send an alarm to the nearest security personnel or monitoring center, and at the same time lock the relevant area to prevent further intrusion. This instant feedback mechanism wins a valuable time window for handling emergency events and reduces the losses caused by delays. In addition, by continuously monitoring and recording all early warning events and their handling results, the system can also provide important reference bases for future risk management, helping granary managers continuously improve the emergency plan and continuously improve the safety management process. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings described herein are used to provide a further understanding of the present invention and form a part of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0055] Figure 1It is a schematic flow chart of the abnormal behavior monitoring and early warning method in the granary under an embodiment of the present invention. Specific embodiments

[0056] In order to more clearly illustrate the overall concept of the present invention, the following will be described in detail by way of examples in conjunction with the accompanying drawings of the specification.

[0057] In the following description, many specific details are set forth in order to provide a thorough understanding of the present invention. However, the present invention may be practiced in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited by the specific embodiments disclosed below.

[0058] As Figure 1 shown, an abnormal behavior monitoring and early warning method in a granary, the method includes:

[0059] S101: Real-time extraction of image information in the granary through a camera device.

[0060] This step aims to real-time extract image information through a camera device installed in the granary to achieve real-time monitoring of the situation in the granary. Its purpose is to provide image information for this monitoring and early warning method to judge abnormal behaviors in the granary through the analysis of image information.

[0061] In a specific embodiment, multiple high-definition cameras are installed in a large granary for real-time monitoring of the environmental conditions in the granary, including problems such as grain surface movement, area intrusion, environmental changes, pests and mildew.

[0062] Installation of camera device: Install high-definition cameras at different positions in the granary to ensure coverage of the entire granary area. For example, cameras can be installed at the four corners and the central position of the granary to obtain an omni-directional view.

[0063] Connect to the network: Connect the cameras to the network so that image information can be obtained in real time through the network. Wired or wireless methods can be used for connection to ensure the stability and security of data transmission.

[0064] Real-time image acquisition: Configure the cameras to capture images at regular intervals or continuously, and transmit the images to the server or cloud storage. For example, the cameras can be set to take a picture every minute and upload it to the server.

[0065] In this step, by deploying camera devices inside the granary and extracting image information in real time, the safety of grain storage can be significantly improved. Through the camera devices installed inside the granary, image information can be collected in real time to achieve round-the-clock monitoring of the internal conditions of the granary. This monitoring can not only help detect abnormal situations such as grain surface movement and area intrusion in a timely manner, but also enable rapid response measures to be taken in case of accidents, minimizing losses to the greatest extent. For example, if the camera captures signs of abnormal activities in the grain pile, which may be an early signal of pests, pest control measures can be taken promptly to prevent the spread of pests. In addition, real-time monitoring can also help managers detect any unauthorized entry in a timely manner, thus strengthening the security prevention of the granary.

[0066] In addition, by deploying camera devices inside the granary and extracting image information in real time, costs can be significantly reduced and work efficiency can be improved. On the one hand, automated monitoring replaces traditional regular manual inspections, greatly reducing the input of human resources and lowering operating costs. The camera devices can monitor the conditions inside the granary continuously throughout the day, detect and record abnormal situations in a timely manner, without relying on regular manual inspections, thus reducing the burden on staff. On the other hand, real-time image information provides immediate data support for managers, enabling them to make decisions quickly, improving the response speed and efficiency of work. For example, when the images show that the grain is stacked too high or there are safety hazards, managers can immediately take measures to make adjustments to avoid potential risks. Through automated monitoring and real-time image information, not only are labor costs reduced, but work efficiency is also improved, realizing the intelligent and refined management of the granary.

[0067] S102: Analyze the image information and, in combination with a variety of preset analysis algorithms, determine the abnormal behaviors inside the granary one by one.

[0068] This step aims to analyze the image information extracted in real time by the camera device by combining a variety of preset analysis algorithms to identify and judge various abnormal behaviors inside the granary one by one. The purpose is to timely discover and handle various problems that may affect the safety of grain storage, ensuring the quality of grain and the safety of the storage environment.

[0069] In a specific embodiment, multiple high-definition cameras are installed in a large granary to monitor the environmental conditions inside the granary in real time, and a series of preset image analysis algorithms are used to analyze the image information to automatically detect and identify abnormal behaviors, including grain surface movement, area intrusion, environmental changes, pests and mildew.

[0070] Preset analysis algorithms: Define a variety of image analysis algorithms for detecting different abnormal behaviors. For example:

[0071] Grain surface movement recognition model for determining grain surface movement;

[0072] A feature recognition algorithm for identifying the personnel entering the granary;

[0073] A color and / or texture feature algorithm for identifying environmental parameter changes;

[0074] A morphological feature algorithm for identifying pests and / or mildew phenomena.

[0075] Image information analysis: Use a variety of preset analysis algorithms to analyze the collected image information frame by frame. According to the results of the algorithm analysis, judge the abnormal behaviors in the granary one by one.

[0076] In this step, by combining a variety of preset analysis algorithms, multi-dimensional analysis of the image information can be carried out, and the abnormal behaviors in the granary can be detected from multiple angles, significantly improving the accuracy and reliability of detection. Different analysis algorithms can finely identify different types of abnormal behaviors, reducing the missed details. By comprehensively applying these algorithms, not only can the conditions in the granary be comprehensively monitored from multiple dimensions, but also various potential problems can be discovered and processed in a timely manner, thus ensuring the safety and quality of grain storage.

[0077] In addition, through automatic monitoring and combined with image information for automatic recognition, the need for traditional manual recognition can be significantly reduced, greatly reducing the labor cost and improving work efficiency. This method uses a camera device to collect image information in real time and automatically identifies abnormal situations through a variety of preset analysis algorithms, replacing the traditional work mode of manually viewing images and manually identifying abnormalities. In this way, not only the dependence on human resources is reduced, the omissions and errors that may exist in manual recognition are avoided, but also the management personnel can focus more on key decisions and higher-level tasks. The automated image recognition technology can operate continuously for 24 hours, ensuring that any abnormality can be discovered and processed in a timely manner, thereby improving the overall operation efficiency and management level.

[0078] S103: Combine a multi-level early warning mechanism to classify the abnormal behaviors and form abnormal behavior levels.

[0079] The purpose of this step is to classify the abnormal behaviors in the granary by combining a multi-level early warning mechanism to form abnormal behavior levels. The purpose is to more effectively manage and respond to abnormal events of different severities, ensure that appropriate measures are taken in a timely manner, and improve the overall response efficiency and safety of the system.

[0080] In one embodiment, high-definition cameras and a variety of analysis algorithms are deployed in a large granary for real-time monitoring of the environmental conditions and abnormal behaviors in the granary. A multi-level early warning mechanism is used to classify the detected abnormal situations so as to take corresponding measures according to different levels of abnormal behaviors.

[0081] According to the abnormal behavior classification criteria, a multi-level early warning mechanism is set up, and abnormal behaviors are divided into first-level abnormal behaviors (red), second-level abnormal behaviors (yellow), and third-level abnormal behaviors (green).

[0082] Among them, the classification criteria are determined based on information such as the importance of the granary and the grain storage area in the granary. Different granaries may have different classification criteria.

[0083] When an abnormal behavior is detected, it is classified according to the multi-level early warning mechanism, and measures are taken according to the corresponding abnormal behavior level.

[0084] In this step, by combining the multi-level early warning mechanism, it is possible to ensure that the most appropriate response measures are taken in abnormal situations of different severities. The hierarchical early warning mechanism enables managers to prioritize and handle the most critical issues according to the severity of the abnormal behavior. In this way, not only the response efficiency is improved, but also the reasonable allocation of resources is ensured, thus maximizing the optimization of the management process and work efficiency while ensuring the safety of grain storage.

[0085] In addition, through the multi-level early warning mechanism, minor abnormalities can be detected and handled in a timely manner to prevent them from evolving into serious accidents. For medium abnormalities, timely reminders and suggestions for preliminary measures can prevent potential problems from escalating, thereby reducing risks and losses. This prevention-oriented strategy can not only solve problems at the budding stage, avoid small problems from accumulating into major hidden dangers, but also ensure the efficient use of resources and avoid unnecessary waste. Generally speaking, the multi-level early warning mechanism significantly reduces the risks and potential losses in the process of grain storage by detecting and preventing abnormalities in a timely manner, and ensures the safety and economy of warehousing management.

[0086] S104: According to different abnormal behavior levels, different early warning levels are formed and warnings are sent to the managers of the granary.

[0087] The purpose of this step is to form corresponding early warning levels according to different abnormal behavior levels and send warnings to the managers of the granary in a timely manner. The aim is to ensure that managers can take appropriate response measures according to the severity of the abnormal behavior, thereby improving the response efficiency and reducing risks and losses.

[0088] As a specific embodiment, warnings are sent to the managers:

[0089] For first-level abnormal behaviors: The system immediately sends emergency notifications to all relevant personnel by various means such as phone calls, text messages, and emails, and activates the emergency plan.

[0090] For second-level abnormal behaviors: The system sends alarm notifications in the form of text messages, emails, or system messages and suggests taking preliminary measures.

[0091] For third-level abnormal behaviors: The system notifies the management personnel in the form of emails or system messages, records them, and generates reports in the background to remind the management personnel to check regularly.

[0092] In this step, the multi-level early warning mechanism ensures that in the case of abnormal situations of different severities, the management personnel can quickly take the most appropriate response measures. For example, for severe abnormalities, the emergency plan is immediately activated to quickly handle the problem and prevent the situation from deteriorating further. At the same time, the hierarchical early warning mechanism enables the management personnel to prioritize and handle the most critical problems according to the severity of the abnormal behaviors, avoiding waste of resources. In this way, not only the response efficiency of the system is improved, ensuring a quick response in case of emergencies, but also the risks and losses are minimized through reasonable resource allocation, guaranteeing the safety and economy of the granary management.

[0093] As a preferred embodiment of the present invention, analyze the image information, and combine a variety of preset analysis algorithms to correspondingly judge the abnormal behaviors in the granary, including:

[0094] According to the image information, combine the grain surface movement recognition model to analyze the shape change of the upper surface of the grain pile;

[0095] If there are depressions and / or inclination changes on the upper surface of the grain pile, combine the inbound and outbound business data to determine the grain surface movement.

[0096] The purpose of this step is to detect the shape change of the upper surface of the grain pile by analyzing the image information and combining the grain surface movement recognition model, and determine the grain surface movement according to the inbound and outbound business data. The purpose is to timely discover and handle abnormal situations that may lead to problems in grain storage safety, and improve the intelligent level and safety of granary management.

[0097] In a specific embodiment, high-definition cameras are deployed in a large granary to monitor the environmental conditions in the granary in real time. The image information is collected through the cameras installed in the granary, and combined with the pre-trained grain surface movement recognition model to detect the shape change of the upper surface of the grain pile.

[0098] Application of the grain surface movement recognition model: Use the pre-trained grain surface movement recognition model, which can recognize the shape changes of the upper surface of the grain pile, including but not limited to depressions, inclinations, etc.

[0099] Process and analyze the collected image information to identify abnormal changes on the grain surface.

[0100] Shape change detection: Analyze the upper surface of the grain pile in the image to detect whether there are depressions and / or inclination changes. For example, the model can identify that the contour of a certain part of the grain pile has changed, indicating that there may be a depression or inclination.

[0101] Combined with the inbound and outbound business data, determine the abnormal movement of the grain surface: Obtain the inbound and outbound business data within a recent period of time to understand whether there are normal grain in and out activities. If the abnormal movement of the grain surface occurs without inbound and outbound operations, it is determined as an abnormal situation.

[0102] In this step, through image information and model analysis, the shape changes on the upper surface of the grain pile, such as depressions and inclinations, can be detected in a timely manner, so as to discover potential problems in the early stage. Combined with the inbound and outbound business data, the system can accurately determine abnormal situations and achieve early warning. This prevention-oriented strategy can effectively prevent the further deterioration of problems and avoid greater losses by detecting and handling abnormal situations in a timely manner. In this way, not only the intelligent level of granary management is improved, but also the response efficiency and overall security of the system are significantly enhanced, ensuring the safety and economy of grain storage.

[0103] In addition, once the abnormal movement of the grain surface is detected, the emergency plan can be immediately activated to quickly handle the problem and prevent the situation from further deteriorating. Through the combined application of real-time image information and the grain surface abnormal movement recognition model, the abnormal changes on the upper surface of the grain pile, such as depressions or inclinations, can be identified in the first time, and the management personnel can be notified in time to initiate corresponding countermeasures. This rapid response mechanism can not only effectively prevent the spread of problems, but also minimize potential losses, ensuring the safety and efficiency of granary management.

[0104] Furthermore, combined with image information and model analysis, various abnormal situations on the upper surface of the grain pile can be covered to ensure that any problem will not be overlooked. Through real-time monitoring and intelligent analysis, the system can detect subtle changes on the surface of the grain pile, such as depressions and inclinations, and give early warnings in time, so as to achieve comprehensive monitoring of the grain pile condition. This technical means not only improves the accuracy and comprehensiveness of problem detection, but also ensures that potential problems can be discovered and handled in the early stage, maximizing the safety of grain storage and the efficiency of management.

[0105] As a preferred embodiment of this implementation manner, according to the image information, combined with the grain surface abnormal movement recognition model, analyze the shape changes on the upper surface of the grain pile, specifically:

[0106] Through the grain surface abnormal movement recognition model, perform grain surface segmentation processing on the image information to identify grain surface edge changes and / or internal depressions of the grain surface.

[0107] The purpose of this step is to perform grain surface segmentation processing on the image information through the grain surface abnormal movement recognition model to identify grain surface edge changes and / or internal depressions of the grain surface. The purpose is to discover abnormal situations on the surface of the grain pile in a timely manner, ensure the safety of grain storage, and improve the monitoring efficiency through automated means.

[0108] Specifically, high-definition cameras are deployed in a large granary to monitor the environmental conditions inside the granary in real time. Image information is collected through the cameras installed inside the granary, and combined with a pre-trained grain surface movement recognition model to detect the shape changes on the upper surface of the grain pile.

[0109] Grain surface segmentation processing: Use a pre-trained grain surface movement recognition model, which can segment the grain surface in the image. Through image segmentation technology, the grain surface part in the image is separated from other backgrounds for further analysis.

[0110] Edge change recognition: In the segmented image, recognize the changes in the grain surface edge. For example, if irregular changes occur in the grain surface edge, it may mean that the grain stacking is unstable or there is a risk of slipping.

[0111] Internal depression recognition: In the segmented image, recognize the depressions inside the grain surface. For example, if obvious depressions appear inside the grain surface, it may be caused by factors such as grain moisture, pests, or other reasons.

[0112] Abnormality determination and handling: If abnormal situations such as grain surface edge changes or internal depressions are recognized, the system can record these abnormalities and combine the incoming and outgoing business data to determine whether it is a normal change. If it is determined to be abnormal, the emergency plan is immediately activated, and the management staff is notified for inspection and handling.

[0113] In this step, through image segmentation technology, abnormal situations such as edge changes and internal depressions on the grain pile surface can be detected in a timely manner, thus discovering potential problems at an early stage. Image segmentation technology can accurately separate the grain surface part in the image from the background, thereby improving the ability to distinguish the details of the grain surface. This technical means can ensure that the system can accurately identify the edge changes of the grain surface, such as the irregular shapes that appear at the edge of the grain pile or the subtle depressions inside the grain surface, and then give early warnings of possible problems. For example, when the system detects abnormal changes in the grain surface edge, it can immediately notify the management staff for inspection to prevent safety accidents such as slipping or collapse caused by unstable grain stacking.

[0114] In addition, the adoption of image segmentation technology can not only improve the ability to distinguish the details of the grain surface, but also accurately distinguish the boundary between the grain surface and other objects under complex backgrounds, ensuring the accuracy of the detection results. The image segmentation technology classifies each pixel in the image through an algorithm and marks the part belonging to the grain surface, enabling the system to focus on analyzing the real situation of the grain surface. For example, in the case of a depression inside the grain pile, the system can accurately identify the location and scope of the depression and combine historical data to determine whether it is an abnormal situation. This high-resolution detection ability enables the system to maintain a keen perception of the grain surface condition in a complex warehousing environment, thereby realizing the timely discovery and handling of abnormal situations and ensuring the safety of grain storage and the efficiency of management.

[0115] As a preferred embodiment of the present invention, analyzing the image information and combining a variety of preset analysis algorithms to respectively judge the abnormal behaviors in the granary further includes:

[0116] According to the image information and combining with a feature recognition algorithm, obtain the personnel information entering the granary;

[0117] Compare the personnel information with the authorization information of the authorized personnel entering the granary to conduct a regional intrusion judgment.

[0118] This step aims to obtain the personnel information entering the granary by analyzing the image information and combining with a feature recognition algorithm, and compare it with the personnel information of the authorized personnel entering the granary, so as to conduct a regional intrusion judgment. Its purpose is to ensure that only authorized personnel can enter the granary, prevent unauthorized personnel from entering, and thus ensure the safety of the granary.

[0119] In a specific embodiment, high-definition cameras are deployed in a large granary to monitor the environmental conditions inside and outside the granary in real time. Image information is collected through the cameras installed inside and outside the granary, and combined with a pre-trained feature recognition algorithm to detect the personnel information entering the granary and compare it with the authorized personnel information.

[0120] Application of the feature recognition algorithm: Use a pre-trained feature recognition algorithm, which can recognize the human features in the image, such as facial features, body contours, etc. Process and analyze the collected image information to extract the personnel feature information entering the granary.

[0121] Obtaining personnel information: Analyze the human features in the image to obtain the personnel information entering the granary, including but not limited to facial features, identity features, etc. For example, the algorithm can recognize the facial features of the personnel entering the granary and generate corresponding feature vectors.

[0122] Authorization Information Comparison: Obtain the list of authorized personnel entering the granary and their characteristic information, which is usually stored in a database. Compare the extracted information of the personnel entering the granary with the authorized list to determine whether they are authorized personnel.

[0123] Area Intrusion Judgment: If the information of the personnel entering the granary matches the information in the authorized list, it is determined as a legal entry.

[0124] If it does not match, it is determined that an unauthorized person has entered, that is, an area intrusion has occurred.

[0125] Exception Handling: Once an unauthorized person is detected entering the granary, the system can immediately issue an alarm and notify the management staff for handling.

[0126] Record the information of all entering personnel for future auditing and review.

[0127] In this step, through the feature recognition algorithm, the system can accurately identify the information of each person entering the granary and compare it with the pre-stored authorized list in real time. Once an unauthorized person is detected attempting to enter or already in the granary, the system can immediately trigger an alarm to effectively detect illegal intrusion. This measure not only improves the overall security of the granary but also transforms the security management from passive defense to proactive prevention-based control mode, greatly reducing the risk of grain theft, contamination, or other safety accidents caused by illegal intrusion, fundamentally avoiding potential security hazards, and ensuring the integrity and security of grain reserves.

[0128] In addition, once an unauthorized person is detected entering the granary, the system can immediately activate the emergency plan to quickly handle the problem and prevent the situation from deteriorating further. Through real-time monitoring and feature recognition technology, the system can detect illegal intrusion in the first time and automatically trigger an alarm to notify relevant personnel to take action. This rapid response mechanism can not only quickly stop unauthorized entry but also effectively prevent potential security hazards from turning into actual losses or damages, ensuring that the security management of the granary always remains under control. In addition, through the automated processing flow, the system can reduce the response time, enabling the management staff to concentrate on handling emergencies and further improving the efficiency and effectiveness of emergency handling.

[0129] Secondly, by combining image information with feature recognition algorithms, all personnel entering the granary can be covered, ensuring that any unauthorized entry will not be overlooked. By collecting image information in real time through camera devices and using advanced feature recognition technologies to accurately identify and compare each entering person, the system can comprehensively monitor every entry point without leaving any blind spots. This comprehensive monitoring mechanism ensures that the information of all personnel entering the granary is accurately recorded and verified, thus preventing illegal entry by unauthorized personnel and greatly enhancing the security protection ability of the granary. In addition, continuous data analysis and algorithm optimization further improve the reliability and accuracy of the system, providing all-round security protection for grain storage.

[0130] As a preferred embodiment of the present invention, analyzing the image information and combining a variety of preset analysis algorithms to respectively judge the abnormal behaviors in the granary further includes:

[0131] Analyzing the color and / or texture features in the image and combining the temperature and / or humidity environmental parameters in the granary to monitor environmental changes.

[0132] This step aims to monitor environmental changes by analyzing the color and / or texture features in the image and combining the temperature and humidity environmental parameters in the granary. The purpose is to timely detect environmental changes in the granary, ensure the suitability of grain storage conditions, and prevent grain quality problems caused by environmental changes.

[0133] In a specific embodiment, high-definition cameras and environmental sensors are deployed in a large granary to monitor the environmental conditions in the granary in real time. Image information is collected through the cameras installed in the granary, and by combining the temperature and humidity data obtained by the environmental sensors, the color and texture features of the grain pile are analyzed to monitor environmental changes.

[0134] Image information collection and environmental parameter collection: Install camera devices at different positions in the granary to collect image information at regular intervals or continuously, and transmit the images to a server or cloud storage. Install temperature and humidity sensors in the granary to monitor environmental parameters in real time and transmit the data to the management system.

[0135] Color and texture feature analysis: Use image processing algorithms to analyze the color and texture features in the image.

[0136] For example, through color analysis, color changes on the surface of the grain can be detected, such as signs of mildew; through texture analysis, subtle changes on the surface of the grain can be detected, such as cracks or insect damage marks.

[0137] Environmental parameter monitoring: Collect temperature and humidity data provided by environmental sensors to monitor the environmental conditions inside the granary in real time. For example, monitor whether the temperature exceeds the set safe range and whether the humidity is appropriate to prevent the grain from getting damp or moldy.

[0138] Comprehensive analysis and environmental change monitoring: Combine the analysis results of color and texture features in the image, as well as temperature and humidity data, for comprehensive evaluation. If color changes or texture abnormalities are found and the environmental parameters exceed the normal range, it is determined that the environmental change is abnormal.

[0139] Abnormality determination and handling: If an abnormal environmental change is detected, the system can record these abnormalities and notify the management staff for inspection and handling. For example, if the image shows signs of mold on the grain surface and the humidity is higher than the set safe value, immediately start the ventilation or dehumidification equipment and notify the management staff to take further measures.

[0140] In this step, through image analysis technology and environmental monitoring, environmental changes inside the granary, such as abnormal changes in temperature and humidity, can be detected in a timely manner, so as to discover potential problems at an early stage. Combining the analysis of color and texture features in the image and the real-time monitoring of temperature and humidity data, the system can quickly identify color changes, texture abnormalities on the grain surface, and fluctuations in environmental parameters. This early warning mechanism enables the management staff to take measures in a timely manner to prevent grain quality problems caused by environmental changes, avoid problems such as grain getting damp, moldy, or being damaged by pests, and thus ensure the quality and storage safety of the grain. Through continuous monitoring and data analysis, the system not only improves the sensitivity to environmental changes, but also enhances the intelligent level and overall safety of granary management.

[0141] In addition, through the analysis of color and texture features, image analysis technology can identify color changes and texture abnormalities on the grain surface, providing intuitive visual evidence to help the management staff more accurately judge the state of the grain. Combining real-time environmental parameter monitoring, the system can ensure the suitability of grain storage conditions and further improve the accuracy and reliability of monitoring. This comprehensive monitoring method can not only detect potential problems in a timely manner, but also provide a scientific basis for grain storage through continuous data analysis, ensuring that the grain is stored in the best environment, thereby improving the overall management efficiency and safety.

[0142] As a preferred embodiment of the present invention, analyze the image information, combine a variety of preset analysis algorithms, and correspond one by one to judge the abnormal behaviors inside the granary, and further include:

[0143] Extract the morphological features in the image information to identify pests and / or mildew phenomena;

[0144] When the number of the pests and / or mildew phenomena exceeds the warning threshold, it is determined that there is an abnormality of pests and mildew;

[0145] Among them, the warning threshold is set according to the grain storage area of the granary.

[0146] This step aims to identify pests and / or mildew phenomena in the granary by extracting morphological features from the image information, and determine whether there is an abnormality of pests and mildew according to the warning threshold. Its purpose is to timely discover and handle problems that may affect the safety of grain storage, and ensure that the quality of grain is not damaged.

[0147] In a specific embodiment, high-definition cameras are deployed in a large granary to monitor the environmental conditions in the granary in real time. Image information is collected through the cameras installed in the granary, and image processing algorithms are combined to identify pests and mildew phenomena.

[0148] Morphological feature extraction: Use image processing algorithms (such as edge detection, morphological operations, etc.) to extract morphological features in the image, such as the shape, size, texture of the object, etc. Pay special attention to those features that may represent pests or mildew, such as small dot-like objects (pests), patchy areas (mildew), etc.

[0149] Pests and mildew phenomena identification: Based on the extracted morphological features, use specialized identification algorithms (such as machine learning classifiers) to identify pests and mildew phenomena in the image. For example, the algorithm can identify small black dots (possibly pests) or fuzzy patches (possibly mildew) that appear in the image.

[0150] Warning threshold setting: Set the warning threshold according to the grain storage area of the granary. For example, for a larger granary, the warning threshold can be set higher, while for a smaller granary, it can be set lower. The warning threshold can be the number or density of pests or mildew phenomena.

[0151] Pests and mildew abnormality determination: If the number of identified pests and / or mildew phenomena exceeds the warning threshold, it is determined that there is an abnormality of pests and mildew. For example, if the number of pests detected in a certain image exceeds the warning threshold set according to the granary area, it is determined that there is a pest abnormality.

[0152] Abnormality handling: Once pests and mildew abnormalities are detected, the system can record these abnormalities and notify the management staff for inspection and handling. The management staff can take corresponding prevention and control measures according to the system prompts, such as spraying insecticides, ventilating and dehumidifying, etc.

[0153] In this step, through image processing technology and morphological feature recognition, pests and mildew in the granary can be detected in a timely manner, thus discovering potential problems at an early stage. Combining real-time monitoring with intelligent analysis, the system can identify signs of grain pest activities, provide intuitive visual evidence, and help managers quickly confirm the location of problems. This early warning mechanism can not only detect pests and mildew in a timely manner, but also avoid grain quality problems caused by these issues through preventive measures, prevent grain damage, and thus ensure the quality and storage safety of grain. Through continuous data analysis and automated monitoring means, the system further improves its sensitivity to environmental changes, enhances the intelligence level and overall safety of granary management.

[0154] In addition, by combining image information and morphological feature recognition, various pests and mildew in the granary can be covered, ensuring that no problem is overlooked. Through the camera device and using advanced image processing technology for morphological feature recognition, the system can comprehensively monitor every detail in the granary, detect and record any abnormal situation in a timely manner. This step not only improves the response efficiency of the system, but also ensures the accuracy and reliability of the early warning mechanism, thus continuously enhancing the intelligence level and overall safety of granary management.

[0155] As a preferred embodiment of the present invention, the camera device is a panoramic camera, and a plurality of the panoramic cameras are installed on the top of the granary to obtain image information of the entire area inside the granary.

[0156] This step aims to install a plurality of panoramic cameras on the top of the granary to obtain image information of the entire area inside the granary. Its purpose is to achieve a full range of monitoring of the internal environment of the granary, ensure that there are no monitoring blind spots, and thus improve the comprehensiveness and reliability of monitoring.

[0157] In a specific embodiment, a plurality of panoramic cameras are installed in a large granary to monitor the environmental conditions inside the granary in real time. These panoramic cameras are installed on the top of the granary to ensure coverage of the entire granary area.

[0158] Select panoramic cameras: Select panoramic cameras with wide-angle lenses to ensure that the entire area image information inside the granary can be captured. These cameras should have high resolution to ensure the clarity of the images.

[0159] Installation location planning: Plan the installation locations on the top of the granary to ensure coverage of the entire granary area. Usually, cameras will be installed at the four corners and the central position of the granary to ensure full coverage.

[0160] Image information collection: After installation, the camera periodically or continuously collects image information and transmits the images to the server or cloud storage. Each camera is responsible for monitoring a different area, but there is a certain overlap between them to ensure there are no blind spots.

[0161] In this step, by installing multiple panoramic cameras on the top of the granary and appropriately planning the overlapping areas, all-round and dead-angle-free monitoring can be achieved. The panoramic cameras can cover all areas inside the granary, ensuring there are no monitoring blind spots, thereby improving the comprehensiveness of monitoring. This multi-angle coverage method can not only capture every detail inside the granary but also form a complete view of the granary interior through image stitching technology, ensuring that any potential problems are not overlooked, thus significantly enhancing the safety and reliability of granary management.

[0162] In addition, by using panoramic cameras with high resolution, clear image information can be provided to help managers more accurately judge the state of the grain. Combining with image stitching technology, the images captured by multiple cameras can be seamlessly stitched into a complete view of the granary interior for comprehensive analysis. The combination of such high-definition images and image stitching technology not only improves the monitoring accuracy but also ensures that managers can obtain detailed and reliable visual data, thereby effectively enhancing the safety and efficiency of granary management.

[0163] As a preferred embodiment of the present invention, in combination with a multi-level early warning mechanism, the abnormal behaviors are classified to form abnormal behavior levels, specifically as follows:

[0164] The abnormal behavior levels include level one, level two, and level three.

[0165] If the abnormal behaviors are grain surface movement, environmental change, pests and mildew, according to the change rates of grain surface movement, environmental change, pests and mildew, a first threshold and a second threshold are set, and the first threshold is less than the second threshold.

[0166] When the change rate is less than the first threshold, it is determined as a level three abnormal behavior.

[0167] When the change rate is greater than the first threshold and less than the second threshold, it is determined as a level two abnormal behavior.

[0168] When the change rate is greater than the second threshold, it is determined as a level one abnormal behavior.

[0169] Wherein the first threshold and the second threshold are determined according to the grain storage area of the granary.

[0170] This step aims to determine the levels of abnormal behavior in the granary by setting different thresholds, including level one, level two, and level three. Its purpose is to standardize the criteria for judging abnormal behavior through quantitative indicators (such as the change rate), ensuring that corresponding measures are taken under different degrees of abnormality, thereby improving the accuracy of monitoring and the efficiency of management.

[0171] In a specific embodiment, high-definition cameras and environmental monitoring devices are deployed in a large granary to monitor the environmental conditions and grain status in the granary in real time. Image information is collected through panoramic cameras and other sensors installed on the top of the granary, and image processing algorithms are combined to identify phenomena such as grain surface movement, environmental changes, pests, and mildew.

[0172] Define the levels of abnormal behavior: Level three abnormal behavior: The change rate is less than the first threshold;

[0173] Level two abnormal behavior: The change rate is greater than the first threshold but less than the second threshold;

[0174] Level one abnormal behavior: The change rate is greater than the second threshold.

[0175] Set the thresholds: Determine the first threshold and the second threshold according to the grain storage area of the granary, where the first threshold is less than the second threshold. The first threshold can be set as a relatively low change rate for detecting minor abnormalities. The second threshold can be set as a relatively high change rate for detecting serious abnormalities.

[0176] Calculate the change rate: Calculate the change rates of grain surface movement, environmental changes (temperature and humidity), pests, and mildew. For example, detect the change rate of grain surface movement through image analysis algorithms.

[0177] Judge abnormal behavior: When the change rate is less than the first threshold, it is judged as level three abnormal behavior. When the change rate is greater than the first threshold and less than the second threshold, it is judged as level two abnormal behavior. When the change rate is greater than the second threshold, it is judged as level one abnormal behavior.

[0178] In this step, by setting the first threshold and the second threshold, the abnormal behavior is quantified into different levels, making the judgment of abnormal behavior more standardized and objective. This quantitative standard can not only ensure that the judgment of abnormal behavior is more accurate and consistent, but also reduce the influence of human factors, improve the scientific nature and reliability of management. At the same time, by unifying the standards, it ensures that in different situations, managers can make decisions based on the same standard, avoiding the uncertainty brought by subjective judgment, thereby improving management efficiency and the overall safety of the system.

[0179] In addition, by calculating the rate of change, the degree of abnormal behavior can be more accurately identified, enabling more effective measures to be taken. The calculation of the rate of change not only quantifies the dynamic changes in abnormal behavior but also helps managers promptly detect the development trend of potential problems, ensuring that appropriate intervention measures can be taken at an early stage. This monitoring method based on the rate of change improves the sensitivity and accuracy of anomaly detection, enabling the management system to respond to various abnormal situations more scientifically and efficiently, thereby ensuring the safety of the granary and the quality of the grain.

[0180] As a preferred embodiment under this implementation manner, if the abnormal behavior is regional intrusion, it is determined as a first-level abnormal behavior.

[0181] This step aims to, through the determination of abnormal behavior, especially when the abnormal behavior involves regional intrusion, determine it as the highest-level abnormal behavior (first-level abnormal behavior). Its purpose is to ensure that behaviors such as regional intrusion, which pose a serious threat to food security, can immediately attract attention and prompt response measures can be taken.

[0182] Specifically, high-definition cameras and environmental monitoring devices are deployed in a large granary to monitor the environmental conditions inside and outside the granary in real time. Image information is collected through the cameras installed inside and outside the granary, and image processing algorithms are combined to identify abnormal behavior.

[0183] Image information collection and feature recognition: Install high-definition cameras at the entrance and around the granary to ensure coverage of all access channels. Image information is collected regularly or continuously and transmitted to the server or cloud storage. Feature recognition algorithms are used to identify the human features in the images, such as facial features and body contours.

[0184] Personnel information acquisition and comparison: Analyze the human features in the images to obtain the information of the personnel entering the granary. The extracted personnel information is compared with the list of authorized personnel to enter the granary to determine whether they are authorized personnel.

[0185] Regional intrusion determination: If the personnel information of those entering the granary does not match the information in the authorized list, it is determined that unauthorized personnel have entered, that is, regional intrusion has occurred. Once regional intrusion is determined, it is immediately determined as a first-level abnormal behavior.

[0186] In this step, regional intrusion is regarded as the highest-level abnormal behavior, ensuring that the system can immediately activate the emergency plan, quickly handle the problem, and prevent the situation from deteriorating further. By determining regional intrusion as a first-level abnormal behavior, the system can respond immediately and take emergency measures to ensure that potential security threats are controlled in the first place. At the same time, this grading ensures that managers can prioritize dealing with such serious threats, improving management efficiency and the overall security of the system.

[0187] In addition, by setting clear criteria for the levels of abnormal behavior and directly determining regional intrusion as a first-level abnormal behavior, the determination of abnormal behavior becomes more standardized and objective. Such a quantitative standard can not only ensure more accurate and consistent determination of abnormal behavior, but also reduce the influence of human factors, improving the scientific nature and reliability of management. At the same time, by unifying the standards, it is ensured that in different situations, managers can make decisions based on the same standard, avoiding the uncertainty brought by subjective judgment, thereby improving management efficiency and the overall security of the system.

[0188] According to different levels of abnormal behavior, different warning levels are formed and warnings are sent to the managers of the granary, specifically as follows:

[0189] Acoustic and optical alarms and / or SMS notifications are sent to the managers so that the managers receive warnings of abnormal behavior and execute disposal strategies respectively according to the warning levels;

[0190] When the abnormal behavior level is the first level, managers are dispatched for on-site verification to confirm the abnormal behavior and implement emergency treatment measures;

[0191] When the abnormal behavior level is the second level, the inspection frequency is increased, the monitoring of the abnormal area is strengthened, and local adjustments are made to the abnormal behavior area;

[0192] When the abnormal behavior level is the third level, pay attention to the abnormal behavior area and take preventive measures.

[0193] This step aims to ensure that managers can receive warning information in a timely manner and take corresponding disposal strategies according to the warning levels by setting different warning levels and sending warnings to managers according to different levels of abnormal behavior. Its purpose is to improve management efficiency and ensure that various abnormal behaviors can be handled promptly and effectively, thus guaranteeing the safety of the granary.

[0194] Specifically, high-definition cameras and environmental monitoring devices are deployed in a large granary to monitor the environmental conditions inside and outside the granary in real time. Image information is collected through cameras installed inside and outside the granary, and image processing algorithms are combined to identify abnormal behavior.

[0195] Warning level setting: Different warning levels are set according to different levels of abnormal behavior. The first-level abnormal behavior corresponds to the highest warning level, the second level is the second, and the third level is the lowest.

[0196] Warning managers: Acoustic and optical alarms and / or SMS notifications are sent to managers so that managers receive warnings of abnormal behavior. For example, an alarm is issued inside the granary through an acoustic and optical alarm, and managers are notified by SMS.

[0197] Execute disposal strategies: Different disposal strategies are executed respectively according to the warning levels:

[0198] Level 1 abnormal behavior: Dispatch management personnel for on-site verification, confirm the abnormal behavior, and implement emergency treatment measures.

[0199] Level 2 abnormal behavior: Increase the inspection frequency, strengthen the monitoring of the abnormal area, and make local adjustments to the abnormal behavior area.

[0200] Level 3 abnormal behavior: Pay attention to the abnormal behavior area and take preventive measures.

[0201] In this step, through means such as audible and visual alarms and SMS notifications, ensure that management personnel can receive early warnings of abnormal behavior in a timely manner and take prompt actions. This instant notification mechanism can not only ensure that management personnel are aware of abnormal situations in the first place but also prompt them to respond quickly and take necessary measures in a timely manner. At the same time, adopt corresponding disposal strategies according to the warning level, reasonably allocate human and technical resources, and ensure that limited resources are used in the most critical places, thereby improving management efficiency and the overall security of the system.

[0202] Once an abnormal behavior is detected, the corresponding warning mechanism can be immediately activated to quickly handle the problem and prevent the situation from deteriorating further. This rapid response mechanism can not only ensure that measures are taken in the first place to effectively control the development of abnormal situations but also reduce the workload of management personnel through an automated monitoring and warning mechanism, enabling them to focus more on decision-making and coordination work, thereby improving management efficiency and the overall security of the system.

[0203] The present invention also provides a monitoring and warning device for abnormal behavior in a granary, including:

[0204] A memory for storing computer programs;

[0205] A processor for implementing the steps of the monitoring and warning method for abnormal behavior in a granary when executing the computer program, and thus can achieve any effect of the monitoring and warning method for abnormal behavior in a granary, which will not be elaborated here.

[0206] What is not described in the present invention can be realized by adopting or referring to existing technologies.

[0207] Each embodiment in this specification is described in a progressive manner. For the same or similar parts among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments.

[0208] The above are only embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the scope of the claims of the present invention.

Claims

1. A method for monitoring and early warning of abnormal behavior in a granary, characterized in that: include: Extracting image information in the granary in real time by means of a camera device; Analyze the image information and, in combination with a plurality of preset analysis algorithms, determine abnormal behavior in the granary in a one-to-one correspondence; Combined with the multi-level early warning mechanism, the abnormal behaviors are classified to form abnormal behavior levels; Different warning levels are formed according to different levels of abnormal behavior, and warnings are given to the managers of the granary; Among them, the abnormal behaviors include grain movement, regional invasion, environmental changes, pests and mildew.

2. The method for monitoring and early warning of abnormal behavior in a granary according to claim 1 is characterized in that: Analyze the image information and combine multiple preset analysis algorithms to determine abnormal behaviors in the granary in a one-to-one correspondence, including: According to the image information, combined with the grain surface change recognition model, the shape change of the upper surface of the grain pile is analyzed; If the upper surface of the grain pile becomes concave and / or tilted, the grain surface changes can be determined in combination with the warehouse in and out business data.

3. The method for monitoring and early warning of abnormal behavior in a granary according to claim 2, characterized in that: According to the image information, combined with the grain surface change recognition model, the shape change of the upper surface of the grain pile is analyzed, specifically: The grain surface segmentation process is performed on the image information through the grain surface change recognition model to identify the grain surface edge changes and / or the grain surface internal depressions.

4. The method for monitoring and early warning of abnormal behavior in a granary according to claim 1, characterized in that: Analyzing the image information, combining a plurality of preset analysis algorithms, and judging abnormal behavior in the granary in a one-to-one correspondence, further comprising: According to the image information, combined with a feature recognition algorithm, information about people entering the granary is obtained; Compare the personnel information with the authorization information of the personnel authorized to enter the granary to make a regional intrusion judgment.

5. The method for monitoring and early warning of abnormal behavior in a granary according to claim 1, characterized in that: Analyzing the image information, combining a plurality of preset analysis algorithms, and judging abnormal behavior in the granary in a one-to-one correspondence, further comprising: The color and / or texture features in the image information are analyzed, and combined with the temperature and / or humidity environmental parameters in the granary, environmental change monitoring is performed.

6. The method for monitoring and early warning of abnormal behavior in a granary according to claim 1, characterized in that: Analyzing the image information, combining a plurality of preset analysis algorithms, and judging abnormal behavior in the granary in a one-to-one correspondence, further comprising: Extracting morphological features from the image information to identify pests and / or mildew; When the number of the pests and / or mildew phenomena exceeds the warning threshold, it is determined that the pests and mildew phenomena are abnormal; Wherein, the warning threshold is set according to the grain storage area of ​​the granary.

7. The method for monitoring and early warning of abnormal behavior in a granary according to claim 1, characterized in that: The camera device is a panoramic camera, and a plurality of the panoramic cameras are installed on the top of the granary to obtain image information of the entire area inside the granary.

8. The method for monitoring and early warning of abnormal behavior in a granary according to claim 1, characterized in that: Combined with the multi-level early warning mechanism, the abnormal behaviors are classified to form abnormal behavior levels, specifically: The abnormal behavior levels include level one, level two and level three. If the abnormal behavior is an area intrusion, it is determined to be a first-level abnormal behavior; If the abnormal behavior is grain surface movement, environmental change, pests and mildew, a first threshold and a second threshold are set according to the change rate of grain surface movement, environmental change, pests and mildew, and the first threshold is less than the second threshold; When the rate of change is less than the first threshold, it is determined to be a level 3 abnormal behavior; When the rate of change is greater than the first threshold and less than the second threshold, it is determined to be a secondary abnormal behavior; When the rate of change is greater than the second threshold, it is determined to be a first-level abnormal behavior; The first threshold and the second threshold are determined according to the grain storage area of ​​the granary.

9. The method for monitoring and early warning of abnormal behavior in a granary according to claim 8, characterized in that: Different warning levels are formed according to different abnormal behavior levels, and warnings are given to the managers of the granary, specifically: Producing sound and light alarms and / or SMS notifications to management personnel so that the management personnel receive warnings of abnormal behavior and execute handling strategies according to the warning levels; When the abnormal behavior level is level one, management personnel will be dispatched to conduct on-site inspections, confirm the abnormal behavior, and implement emergency response measures; When the abnormal behavior level is level 2, increase the inspection frequency, strengthen the monitoring of abnormal areas, and make local adjustments to the abnormal behavior areas; When the abnormal behavior level is level three, pay attention to the abnormal behavior area and take preventive measures.

10. A device for monitoring and warning abnormal behavior in a granary, characterized in that: include: Memory for storing computer programs; A processor is used to implement the steps of the abnormal behavior monitoring and early warning method in a granary as described in any one of claims 1 to 9 when executing the computer program.

Images