A mobile terminal remote control monitoring system and method

Abstract

The present invention relates to the field of remote monitoring technology, specifically to a mobile terminal remote control monitoring system and method, the system includes a video dynamic encoding module, a content importance analysis module, a cognitive load adjustment module and an audio optimization and alarm module. In the present invention, by dynamically adjusting the video encoding parameters, the changes in network bandwidth can be responded to in real time, significantly enhancing the stability and adaptability in data transmission, accurately capturing dynamic content and color changes by parsing video frames, instantly locating and extracting key events, thereby greatly improving the speed and accuracy of emergency response, using video stream complexity evaluation to dynamically allocate cognitive resources, adjusting resource allocation according to actual needs, ensuring monitoring quality while optimizing resource utilization efficiency, optimizing audio processing and adjusting real-time alarm strategies, ensuring clear communication of sound information and timeliness of alarms, and improving overall response capabilities and effects.

Description

Technical Field

[0001] The present invention relates to the field of remote monitoring technology, and in particular to a mobile terminal remote control monitoring system and method. Background Art

[0002] The field of remote monitoring technology covers technologies and methods for real-time monitoring and control of various devices and systems through communication networks, allowing users to be physically separated from the monitored objects and realize data collection, status detection, control command sending and other functions of the target device through the network. It is usually used in security monitoring, industrial automation, smart home, medical monitoring and other fields, involving the integrated application of multiple technologies such as data encryption, communication protocol optimization, and user interface design. With the development of mobile communications and network technologies, remote monitoring technology is becoming more efficient and common, providing a wider range of application scenarios and higher system performance.

[0003] Among them, the mobile terminal remote control monitoring system refers to a system that uses mobile terminals (such as smart phones, tablets, etc.) to implement remote monitoring and control. Its main purpose is to allow users to remotely operate and monitor home, office or industrial environments through mobile devices. Users can receive real-time data, send control commands, perform video monitoring and other operations through applications installed on mobile terminals. It is widely used in smart home systems, security monitoring, remote medical services and industrial control systems. It not only improves the convenience and flexibility of operation, but also enhances the real-time and reliability of the monitoring system.

[0004] In a dynamic network environment, existing remote monitoring technologies usually fail to effectively adapt to bandwidth fluctuations, and often have problems with unstable video quality, especially when bandwidth is limited. These problems are particularly evident. In addition, traditional monitoring systems lack real-time and flexibility when dealing with emergencies, making it difficult to quickly and accurately identify key situations, affecting the efficiency and effectiveness of monitoring. Fixed resource allocation strategies fail to adapt to the changes in the complexity of monitoring scenarios, often resulting in insufficient resources during peak periods of resource demand or wasted resources during periods of low demand. In terms of audio, the optimization of sound processing and the flexibility of the alarm system have been neglected, so that key information may not be identified and responded to in a timely manner, increasing security risks. These limitations make existing technologies insufficient in high-demand real-time monitoring scenarios, reducing user experience and system practicality. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings of the prior art and to propose a mobile terminal remote control monitoring system and method.

[0006] In order to achieve the above object, the present invention adopts the following technical solution: A mobile terminal remote control monitoring system comprises:

[0007] The video dynamic encoding module adjusts the encoding parameters of the video data based on the network bandwidth and the real-time data monitored, adjusts the bit rate and resolution to match the network changes, and obtains the encoding parameter adjustment results;

[0008] The content importance analysis module uses the encoding parameter adjustment results to perform real-time analysis of video frames, locate and extract dynamic content and significant color change images, separate foreground elements based on the location and extraction results, identify the occurrence of key actions or events, determine monitoring focus points, and obtain key content identification results;

[0009] The cognitive load adjustment module evaluates the complexity of the video stream through the key content identification results, dynamically allocates analysis resources based on the requirements of different monitoring scenarios, adjusts the monitoring video analysis depth according to the dynamic allocation results of analysis resources, optimizes the current resource configuration to match the monitoring scenarios with different complexities, and establishes a cognitive resource allocation plan;

[0010] Based on the cognitive resource allocation scheme, the audio optimization and alarm module separates key sound sources including human voice and alarm sound, enhances the clarity and prioritizes the key sound sources, adjusts the response strategy of the alarm signal in combination with real-time monitoring video analysis, and obtains a remote control alarm triggering scheme.

[0011] As a further solution of the present invention, the step of obtaining the coding parameter adjustment result is:

[0012] Monitor network bandwidth and data traffic in real time, record current bandwidth and data traffic usage, and generate real-time network status records;

[0013] According to the real-time network status record, the current network capacity is evaluated for its ability to support the video data stream using the formula:

[0014] ;

[0015] Calculate and output the adjusted video stream bitrate ,in, Represents the network bandwidth monitored in real time. represents bandwidth utilization, Represents the number of current video data streams;

[0016] The adjusted video stream bit rate is used to adjust the resolution of the video encoding to match the network changes and obtain the encoding parameter adjustment result.

[0017] As a further solution of the present invention, the steps of locating and extracting the dynamic content and the significant color change picture are:

[0018] Applying the encoding parameter adjustment result to perform real-time frame analysis on the input video stream, detecting pixel changes and color fluctuations in each frame, and generating real-time video frame analysis data;

[0019] Based on the real-time video frame analysis data, the dynamic content and significant color change content in the picture are located and extracted using the formula:

[0020] ;

[0021] Calculate the average color variation of video frames , identify significant color changes and generate dynamic content positioning data, where Representative The red, green, and blue color values ​​of each pixel. is the average red, green, and blue color value of all pixels in the current frame, is the total number of pixels;

[0022] According to the dynamic content positioning data, an image recognition process is executed to extract the actual pictures of key dynamic events and significant color changes, and generate dynamic content and significant color change picture extraction results.

[0023] As a further solution of the present invention, the step of obtaining the key content identification result is:

[0024] Separating foreground elements from the background in the video image by using the dynamic content and significant color change image extraction results, comparing the color and motion data in each frame with the original background model, identifying and marking the foreground objects, and generating foreground element separation data;

[0025] Based on the foreground element separation data, key actions or events in the foreground are identified and classified, and the action type and occurrence time point are determined by analyzing the movement pattern and interaction behavior of the object to generate key action recognition results;

[0026] The key action recognition results are integrated to analyze the key monitoring points in the scene, combine the behavior information with the location information, determine the monitoring focus, and generate key content identification results.

[0027] As a further solution of the present invention, the steps of dynamically allocating analysis resources are as follows:

[0028] Based on the key content identification results, video feature extraction and analysis are performed to confirm the occurrence frequency of key events and scene transition speed, using the formula:

[0029] ;

[0030] Calculate the overall video complexity score , evaluate the video complexity and output the preliminary video complexity index, where, Representative The frequency of occurrence of key events, Representative The speed of scene switching, Indicates the number of critical events;

[0031] Based on the preliminary video complexity index, compare with historical monitoring data, identify the video segment with the largest difference from the current monitoring scene requirements, and generate a requirement difference analysis result;

[0032] According to the demand difference analysis results, the analysis resources are dynamically adjusted to match the current monitoring demand differences and obtain an analysis resource allocation plan.

[0033] As a further solution of the present invention, the steps of obtaining the cognitive resource allocation solution are:

[0034] According to the analysis resource allocation scheme, the parsing depth is adjusted for each video stream to optimize the accuracy of key content recognition, using the formula:

[0035] ;

[0036] Calculate the adjusted parsing depth , get the analytical depth adjustment result, where, is the basic analysis depth, Score the video complexity, is the complexity bound threshold, is the adjustment factor;

[0037] Using the analysis depth adjustment result, real-time monitoring of resource usage and video stream changes is performed to confirm that resource allocation dynamically corresponds to actual monitoring requirements, and to generate an optimal resource configuration plan;

[0038] According to the optimal resource configuration plan, future resource requirements and scenario changes are analyzed, resource allocation is automatically updated and monitored, and a cognitive resource allocation plan is constructed.

[0039] As a further solution of the present invention, the steps of obtaining the remote control alarm triggering solution are:

[0040] According to the cognitive resource allocation scheme, sound analysis is started, audio streams are collected from the real-time environment, segments including human voices and alarm sounds are screened, and preliminary sound source recognition results are obtained;

[0041] Based on the preliminary sound source identification results, spectrum analysis is applied to perform in-depth analysis to distinguish between human voice and alarm sound, using the formula:

[0042] ;

[0043] Calculate the quantitative index of the frequency component intensity of two sound sources , and obtain the sound source separation result, where and Represents the maximum frequency components of human voice and alarm sound respectively, It is a tiny value used to avoid division by zero error;

[0044] According to the sound source separation result, the alarm response strategy is adjusted to ensure that human voice is given priority in emergency situations and a remote control alarm triggering solution is obtained.

[0045] A mobile terminal remote control monitoring method comprises the following steps:

[0046] S1: monitor network bandwidth and data traffic in real time, record the current bandwidth and data traffic usage, evaluate the current network capacity's support for video data streams, adjust the video stream bit rate and video encoding resolution, and obtain the encoding parameter adjustment results;

[0047] S2: Apply the encoding parameter adjustment result to perform real-time frame analysis on the input video stream, detect pixel changes and color fluctuations in each frame, extract key dynamic events and actual pictures with significant color changes, and generate dynamic content and significant color change picture extraction results;

[0048] S3: Using the dynamic content and significant color change screen extraction results, separate the foreground elements and background in the video screen, identify and classify key actions or events in the foreground, determine the action type and occurrence time, analyze the key monitoring points in the scene, and generate key content identification results;

[0049] S4: Based on the key content identification result, perform video feature extraction and analysis, evaluate video complexity, identify the video segment with the greatest difference from the current monitoring scene requirements, dynamically adjust analysis resources, and obtain an analysis resource allocation plan;

[0050] S5: According to the analysis resource allocation plan, adjust the analysis depth for each video stream, monitor resource usage and video stream changes in real time, analyze future resource requirements and scene changes, and build a cognitive resource allocation plan;

[0051] S6: According to the cognitive resource allocation scheme, start sound analysis, filter segments including human voice and alarm sound, apply spectrum analysis to perform in-depth analysis, distinguish human voice from alarm sound, adjust alarm response strategy, and obtain remote control alarm triggering scheme.

[0052] Compared with the prior art, the advantages and positive effects of the present invention are:

[0053] In the present invention, by dynamically adjusting video encoding parameters, changes in network bandwidth can be responded to in real time, significantly enhancing the stability and adaptability of data transmission, accurately capturing dynamic content and color changes by parsing video frames, instantly locating and extracting key events, thereby greatly improving the speed and accuracy of emergency response, dynamically allocating cognitive resources using video stream complexity assessment, adjusting resource configuration according to actual needs, ensuring monitoring quality while optimizing resource utilization efficiency, optimizing audio processing and adjusting real-time alarm strategies, ensuring clear communication of sound information and timeliness of alarms, and improving overall response capabilities and effects. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 is a system flow chart of the present invention;

[0055] Figure 2 A flowchart of obtaining the encoding parameter adjustment result of the present invention;

[0056] Figure 3 The flowchart of positioning and extracting dynamic content and significant color change pictures of the present invention;

[0057] Figure 4 The flowchart of obtaining the key content identification result of the present invention;

[0058] Figure 5 A flow chart of dynamic allocation of analysis resources for the present invention;

[0059] Figure 6 A flowchart for obtaining the cognitive resource allocation scheme of the present invention;

[0060] Figure 7 This is a flowchart for obtaining the remote control alarm triggering solution of the present invention. DETAILED DESCRIPTION

[0061] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0062] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0063] See also Figure 1 , a mobile terminal remote control monitoring system comprises:

[0064] The video dynamic encoding module adjusts the encoding parameters of the video data based on the network bandwidth and the real-time data monitored, adjusts the bit rate and resolution to match the network changes, and obtains the encoding parameter adjustment results;

[0065] The content importance analysis module uses the encoding parameter adjustment results to perform real-time analysis of video frames, locate and extract dynamic content and significant color change images, separate foreground elements based on the location and extraction results, identify the occurrence of key actions or events, determine the monitoring focus, and obtain key content identification results;

[0066] The cognitive load adjustment module evaluates the complexity of the video stream through the key content identification results, dynamically allocates analysis resources based on the requirements of different monitoring scenarios, adjusts the monitoring video analysis depth according to the dynamic allocation results of analysis resources, optimizes the current resource configuration to match the monitoring scenarios of different complexity, and establishes a cognitive resource allocation plan;

[0067] The audio optimization and alarm module is based on the cognitive resource allocation scheme, separates key sound sources including human voice and alarm sound, enhances the clarity and prioritizes the key sound sources, adjusts the response strategy of the alarm signal in combination with real-time monitoring video analysis, and obtains a remote control alarm triggering scheme.

[0068] The encoding parameter adjustment results include bit rate adjustment values ​​and resolution adjustment values. The key content identification results include foreground element recognition records, key action recognition records, and monitoring focus determination results. The cognitive resource allocation plan includes resource configuration optimization results, monitoring scene complexity matching results, and video analysis depth adjustment records. The remote control alarm triggering plan includes key sound source separation results, sound source priority sorting records, and alarm strategy adjustment results.

[0069] See also Figure 2 , the steps to obtain the encoding parameter adjustment results are:

[0070] Monitor network bandwidth and data traffic in real time, record current bandwidth and data traffic usage, and generate real-time network status records;

[0071] Real-time monitoring of network bandwidth and data traffic. Automatically record bandwidth utilization and data packet transmission speed per second through network monitoring, and transmit data to the central processing unit. The central processing unit uses this data for real-time analysis. By calculating the average bandwidth consumption and traffic peak per second, it generates a detailed record of the current network status, including real-time statistics of bandwidth utilization and data traffic, ensuring continuous observation and recording of network conditions. The monitoring data is subsequently used to determine whether the network can carry high-quality video transmission, which is crucial for parameter adjustment during the video encoding process.

[0072] According to the real-time network status record, the current network capacity is evaluated for its ability to support video data streams using the formula:

[0073] ;

[0074] Calculate and output the adjusted video stream bitrate ,in, Represents the network bandwidth monitored in real time. represents bandwidth utilization, Represents the number of current video data streams;

[0075] Represents the network bandwidth monitored in real time, in Mbps. The current value collected is 500Mbps. Represents bandwidth utilization. The current utilization obtained through network monitoring is 0.60; Represents the number of current video data streams. The system is transmitting 5 video streams. Substituting into the formula, we get:

[0076] ;

[0077] The results show that the new bit rate of each video stream is adjusted to 40Mbps, which helps to match the current network conditions and optimize the video transmission process.

[0078] Use the adjusted video stream bit rate to adjust the resolution of the video encoding to match the network changes and obtain the encoding parameter adjustment results;

[0079] Based on the calculated adjusted video stream bit rate, the encoding resolution of each video stream is adjusted. First, the current network load capacity and the content characteristics of each video stream are analyzed, including the complexity of dynamic scenes, color depth, and picture motion rate. For each situation, the key frame interval and compression ratio are adjusted to adapt to different network conditions and content requirements. The resolution and compression level of the video stream can be dynamically adjusted. Finally, the adjusted parameters are applied to the actual video transmission to ensure that each video stream is transmitted with the highest efficiency under the best network conditions, and the final encoding parameter adjustment results are obtained, thereby maximizing the network utilization efficiency without sacrificing video quality.

[0080] See also Figure 3 , the steps for locating and extracting dynamic content and significant color change images are:

[0081] Apply the encoding parameter adjustment results to perform real-time frame analysis on the input video stream, detect pixel changes and color fluctuations within each frame, and generate real-time video frame analysis data;

[0082] Apply the encoding parameter adjustment results and start real-time video analysis to capture the changes and color fluctuations of each pixel frame by frame, focus on motion detection and color analysis of the image, and analyze the pixel changes and color shifts in each frame. This continuous monitoring process is constantly updated and can instantly identify and record dynamic changes in video content to ensure that changes in each frame are accurately captured, providing accurate and real-time data support for subsequent image processing and event detection. Through this real-time processing, it is possible to effectively track activity changes in the video and quickly respond to any abnormal situations.

[0083] Based on real-time video frame analysis data, locate and extract dynamic content and significant color change content in the picture, using the formula:

[0084] ;

[0085] Calculate the average color variation of video frames , identify significant color changes and generate dynamic content positioning data, where Representative The red, green, and blue color values ​​of each pixel. is the average red, green, and blue color value of all pixels in the current frame, is the total number of pixels;

[0086] Consider a frame with 10,000 pixels. The color value of each pixel consists of three components: red (R), green (G), and blue (B). The average color value of the current video frame is recorded as , , , and determine the color value of a specific pixel , , , insert the value to calculate the color change of a single pixel:

[0087] ;

[0088] ;

[0089] Apply this calculation to all pixels and sum and divide by the total number of pixels (10,000) to get the average of the overall color change. . It can accurately indicate which areas or frames have experienced significant color changes, which is crucial for the detection and extraction of dynamic content. If the average color change of the entire frame is above a certain threshold, the system can mark this frame as a frame containing significant dynamic or color changes, and then perform more detailed analysis.

[0090] According to the dynamic content positioning data, the image recognition process is executed to extract the actual images of key dynamic events and significant color changes, and the dynamic content and significant color change image extraction results are generated;

[0091] Based on the dynamic content positioning data, advanced image recognition and refinement analysis are further performed. Through in-depth analysis of the extracted images, key actions or significant color changes can be identified at the detail level, such as key visual elements such as movement speed, direction changes, and color gradients. Through analysis, important events and status changes in the video can be accurately marked and classified, and a detailed visual summary of dynamic content and significant color changes can be generated, which provides important decision-making basis for security monitoring and automatic alarm systems, ensuring that the monitoring system can effectively respond and record at critical moments.

[0092] See also Figure 4 , the steps to obtain the key content identification results are:

[0093] Using the results of dynamic content and significant color change image extraction, the foreground elements and background in the video image are separated, the color and motion data in each frame are compared with the original background model, the foreground objects are identified and annotated, and the foreground element separation data is generated;

[0094] Based on the previously obtained dynamic content and significant color change extraction results, the foreground and background are separated. First, the pixel changes in each frame are analyzed, especially the color and shape contrast between the dynamic object and the background. By comparing the difference between the current frame and the historical background frame pixel by pixel, the color difference and movement deviation of each pixel are determined, and a high threshold is used to filter and identify and mark the area with significant changes as the foreground, ensuring that only objects that actually move or have significant color changes are selected as the foreground, thereby improving the accuracy and response speed of monitoring.

[0095] Based on the foreground element separation data, the key actions or events in the foreground are identified and classified. By analyzing the object's motion pattern and interactive behavior, including walking, running or other activities, the action type and occurrence time are determined to generate key action recognition results;

[0096] Continue to deepen the analysis of foreground elements, determine and classify key actions or events in the image, including detailed motion capture and time series analysis, perform behavioral recognition on each marked foreground object, analyze its dynamic characteristics such as speed, direction and interactive behavior, and evaluate various behavioral patterns, from simple walking to complex interactive actions. Classification is not only based on the physical movement of the object, but also considers its interaction with the surrounding environment. Through detailed analysis, key events such as security threats or emergencies can be accurately identified and recorded, providing decisive data for subsequent processing and response.

[0097] Comprehensive key action recognition results, analyze key monitoring points in the scene, including entrances and exits or important interactive areas, combine behavior information with location information, determine monitoring focus, and generate key content identification results;

[0098] By combining the analysis results of the first two steps, the key monitoring areas and key monitoring objects can be accurately determined. Data fusion and scenario analysis are used to combine geographic and time information to determine which behaviors or events require special attention. The system assesses potential risks based on historical data of behavioral patterns and locations, automatically marks high-risk activities and key time nodes, and provides real-time alerts and activity summaries to the security monitoring center. This includes not only identifying possible security threats, but also optimizing daily monitoring, such as adjusting camera focus or increasing monitoring intensity in certain areas, thereby ensuring the effectiveness and timely response capabilities of the monitoring system and significantly improving the overall effectiveness of security management.

[0099] See also Figure 5 , the steps for dynamic allocation of analysis resources are:

[0100] Based on the key content identification results, video feature extraction and analysis are performed to confirm the frequency of key events and scene transition speed using the formula:

[0101] ;

[0102] Calculate the overall video complexity score , evaluate the video complexity and output the preliminary video complexity index, where, Representative The frequency of occurrence of key events, Representative The speed of scene switching, Indicates the number of critical events;

[0103] There are three key events in a video stream, with occurrence frequencies of 2 times / minute, 3 times / minute, and 1 time / minute, respectively. The corresponding scene transition speeds are 2 times / second, 1.5 times / second, and 1 time / second, respectively. The calculation process is as follows:

[0104] ;

[0105] ;

[0106] ;

[0107] The results show that the complexity score of the video is , indicating that the key content in the video changes frequently, the scene changes quickly, and the video stream complexity is high.

[0108] Based on the preliminary indicators of video complexity, compare with historical monitoring data to identify the video segments with the greatest difference from the current monitoring scene requirements and generate demand difference analysis results;

[0109] The preliminary indicators of video complexity are compared with historical monitoring data. The key content in the video stream is compared with historical monitoring data through cluster analysis, and the similarities and differences between the new data and the old data are evaluated (determined by calculating the Euclidean distance between the feature vector of each data point and the center point of the historical data). This method can identify video segments with significant differences, which directly affects the accuracy of resource allocation and the responsiveness of the monitoring system. Through this analysis, the demand difference analysis results are generated, which lists in detail the video segments with the largest differences from the current monitoring scene requirements, providing decision support for the next step of resource allocation.

[0110] According to the demand difference analysis results, the analysis resources are dynamically adjusted to match the current monitoring demand differences and obtain the analysis resource allocation plan;

[0111] Based on the results of the demand difference analysis, the analysis resources are adjusted, and the resources are reconfigured and optimized to ensure that each part of the resources can be effectively utilized. The availability and cost-effectiveness of various resources are taken into consideration during the implementation process to ensure that the allocation plan will not cause waste of resources while meeting the current monitoring needs. The final allocation and analysis resource plan specifies in detail how to adjust resources such as computing power, storage space, and network bandwidth according to the complexity of the video stream and monitoring needs, so as to make the resource allocation more reasonable and effectively support the operation of the monitoring system.

[0112] See also Figure 6 , the steps to obtain the cognitive resource allocation scheme are:

[0113] According to the analysis resource allocation plan, adjust the analysis depth for each video stream to optimize the accuracy of key content recognition, using the formula:

[0114] ;

[0115] Calculate the adjusted parsing depth , get the analytical depth adjustment result, where, is the basic analysis depth, Score the video complexity, is the complexity bound threshold, is the adjustment factor;

[0116] Basic analysis depth is 100 units, complexity score is 150, the complexity threshold =100, adjustment factor is 0.5. The calculation process is as follows:

[0117] ;

[0118] ;

[0119] The results show that according to the increase of video complexity score, the parsing depth should be increased to 102.03 units to adapt to the analysis needs of complex scenes.

[0120] Use the results of the deep analysis adjustment to monitor resource usage and video stream changes in real time, confirm that resource allocation dynamically corresponds to actual monitoring needs, and generate the optimal resource configuration plan;

[0121] Based on the analysis depth adjustment results, the resource usage status and video stream changes are monitored in real time, and resource configuration is updated in real time. Computing resources, storage, and bandwidth are dynamically adjusted according to the complexity of the video stream content and the real-time analysis requirements. By continuously monitoring the execution of analysis tasks and resource consumption, the performance of different resource configuration schemes is evaluated, and the resource allocation strategy is optimized in a data-driven manner. Finally, a resource configuration optimization plan is formed, which specifies in detail how to adjust resources according to monitoring needs and resource utilization efficiency to ensure that the monitoring system can operate efficiently in various scenarios, reduce resource waste and improve response speed.

[0122] According to the optimal resource allocation plan, analyze future resource requirements and scenario changes, automatically update monitoring resource allocation, and build a cognitive resource allocation plan;

[0123] Based on the optimized resource allocation results, a comprehensive cognitive resource allocation plan is established. The plan combines resource usage data, prediction models and real-time monitoring feedback to analyze future changes in resource demand and automatically adjust resource allocation based on the analysis results. The system can automatically learn and optimize resource allocation according to the complexity of the monitoring scenario. The effectiveness of the cognitive resource allocation plan is tested by simulating different scenarios to ensure that it can dynamically adapt to environmental changes in actual operation and improve the accuracy and efficiency of resource use. The final plan details how to automatically update resource configuration based on real-time data and long-term data trends to ensure the continuous and efficient operation of the monitoring system.

[0124] See also Figure 7 ,The steps to obtain the remote control alarm triggering scheme are:

[0125] According to the cognitive resource allocation plan, start sound analysis, collect audio streams from the real-time environment, filter out segments including human voices and alarm sounds, and obtain preliminary sound source identification results;

[0126] Start sound analysis, process audio streams in real-time environments, effectively separate human voices and alarm sounds from complex soundscapes, use sound features such as frequency, amplitude, and duration to distinguish different sound sources, and pre-filter audio data before processing to remove background noise and non-target sounds to ensure that only clear human voices and alarm sounds are further analyzed. The data will serve as the basic input for the next step of processing, providing key information for subsequent sound clarity enhancement and priority sorting, and can provide accurate and fast data support for sound response in emergency situations.

[0127] Based on the preliminary sound source identification results, spectrum analysis is applied to perform in-depth analysis to distinguish between human voices and alarm sounds, using the formula:

[0128] ;

[0129] Calculate the quantitative index of the frequency component intensity of two sound sources , and obtain the sound source separation result, where and Represents the maximum frequency components of human voice and alarm sound respectively, It is a tiny value used to avoid division by zero error;

[0130] Through the spectrum analysis of audio data, the maximum frequency component of human voice is obtained The maximum frequency component of the alarm sound is 500Hz. For 300Hz, set the zero-proof small constant is 0.01. Then the calculation process is:

[0131] ;

[0132] The results show that in the current audio sample, the dominant frequency component of the human voice is significantly higher than the alarm sound by about 67%, which will guide the subsequent priority sorting and sound clarity processing to give priority to the human voice and ensure that the human voice can be recognized and responded to first in an emergency.

[0133] According to the sound source separation results, adjust the alarm response strategy to ensure that human voice is given priority in emergency situations and obtain remote control alarm triggering solutions;

[0134] By using the analyzed and enhanced audio data and combining it with the real-time video surveillance system, the response strategy for the alarm signal is adjusted and optimized. During this process, the system determines the priority of the response by analyzing the synchronous data of video and audio in real time, ensuring that the human voice can be recognized and responded to first at critical moments. The integrated response mechanism allows the system to dynamically adjust its alarm strategy in a variety of monitoring scenarios. By simulating different emergency situations, the system tests and verifies different alarm triggering conditions to ensure the effectiveness and accuracy of the response strategy in real scenarios. The final remote control alarm triggering solution will directly affect the overall safety operation efficiency of the monitoring system.

[0135] A mobile terminal remote control monitoring method comprises the following steps:

[0136] S1: monitor network bandwidth and data traffic in real time, record the current bandwidth and data traffic usage, evaluate the current network capacity's support for video data streams, adjust the video stream bit rate and video encoding resolution, and obtain the encoding parameter adjustment results;

[0137] S2: Apply the encoding parameter adjustment result to perform real-time frame analysis on the input video stream, detect pixel changes and color fluctuations in each frame, extract the actual pictures of key dynamic events and significant color changes, and generate dynamic content and significant color change picture extraction results;

[0138] S3: Use the dynamic content and significant color change image extraction results to separate the foreground elements and background in the video image, identify and classify the key actions or events in the foreground, determine the action type and occurrence time, analyze the key monitoring points in the scene, and generate key content identification results;

[0139] S4: Based on the key content identification results, perform video feature extraction and analysis, evaluate video complexity, identify the video segment with the greatest difference from the current monitoring scene requirements, dynamically adjust analysis resources, and obtain an analysis resource allocation plan;

[0140] S5: According to the analysis resource allocation plan, adjust the analysis depth for each video stream, monitor resource usage and video stream changes in real time, analyze future resource requirements and scenario changes, and build a cognitive resource allocation plan;

[0141] S6: According to the cognitive resource allocation plan, start sound analysis, filter the segments including human voice and alarm sound, apply spectrum analysis for in-depth analysis, distinguish human voice from alarm sound, adjust the alarm response strategy, and obtain the remote control alarm triggering plan.

[0142] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A mobile terminal remote control monitoring system, characterized in that: The system comprises: The video dynamic encoding module adjusts the encoding parameters of the video data based on the network bandwidth and the real-time data monitored, adjusts the bit rate and resolution to match the network changes, and obtains the encoding parameter adjustment results; The content importance analysis module uses the encoding parameter adjustment results to perform real-time analysis of video frames, locate and extract dynamic content and significant color change images, separate foreground elements based on the location and extraction results, identify the occurrence of key actions or events, determine monitoring focus points, and obtain key content identification results; The cognitive load adjustment module evaluates the complexity of the video stream through the key content identification results, dynamically allocates analysis resources based on the requirements of different monitoring scenarios, adjusts the monitoring video analysis depth according to the dynamic allocation results of analysis resources, optimizes the current resource configuration to match the monitoring scenarios with different complexities, and establishes a cognitive resource allocation plan; The audio optimization and alarm module separates key sound sources including human voice and alarm sound based on the cognitive resource allocation scheme, enhances the clarity and prioritizes the key sound sources, adjusts the response strategy of the alarm signal in combination with real-time monitoring video analysis, and obtains a remote control alarm triggering scheme; The steps for obtaining the encoding parameter adjustment result are: Monitor network bandwidth and data traffic in real time, record current bandwidth and data traffic usage, and generate real-time network status records; According to the real-time network status record, the current network capacity is evaluated for its ability to support the video data stream using the formula: Calculate and output the adjusted video stream bitrate ,in, Represents the network bandwidth monitored in real time. represents bandwidth utilization, Represents the number of current video data streams; Using the adjusted video stream bit rate, adjusting the resolution of the video encoding to match the network change, and obtaining the encoding parameter adjustment result; The steps of locating and extracting the dynamic content and the significant color change pictures are as follows: Applying the encoding parameter adjustment result to perform real-time frame analysis on the input video stream, detecting pixel changes and color fluctuations in each frame, and generating real-time video frame analysis data; Based on the real-time video frame analysis data, the dynamic content and significant color change content in the picture are located and extracted using the formula: Calculate the average color variation of video frames , identify significant color changes and generate dynamic content positioning data, where Representative The red, green, and blue color values ​​of each pixel. is the average red, green, and blue color value of all pixels in the current frame, is the total number of pixels; According to the dynamic content positioning data, an image recognition process is executed to extract the actual pictures of key dynamic events and significant color changes, and generate dynamic content and significant color change picture extraction results; The steps for obtaining the key content identification result are: Separating foreground elements from the background in the video image by using the dynamic content and significant color change image extraction results, comparing the color and motion data in each frame with the original background model, identifying and marking foreground objects, and generating foreground element separation data; Based on the foreground element separation data, key actions or events in the foreground are identified and classified, and the action type and occurrence time point are determined by analyzing the movement pattern and interaction behavior of the object to generate key action recognition results; Combining the key action recognition results, analyzing the key monitoring points in the scene, combining the behavior information with the location information, determining the monitoring focus, and generating key content identification results; The steps of dynamic allocation of analysis resources are as follows: Based on the key content identification results, video feature extraction and analysis are performed to confirm the occurrence frequency of key events and scene transition speed, using the formula: Calculate the overall video complexity score , evaluate the video complexity and output the preliminary video complexity index, where, Representative The frequency of occurrence of key events, Representative The speed of scene switching, Indicates the number of critical events; Based on the preliminary video complexity index, compare with historical monitoring data, identify the video segment with the largest difference from the current monitoring scene requirements, and generate a requirement difference analysis result; According to the demand difference analysis results, the analysis resources are dynamically adjusted to match the current monitoring demand differences and obtain an analysis resource allocation plan; The steps for obtaining the cognitive resource allocation scheme are: According to the analysis resource allocation scheme, the parsing depth is adjusted for each video stream to optimize the accuracy of key content recognition, using the formula: Calculate the adjusted parsing depth , get the analytical depth adjustment result, where, is the basic parsing depth, Score the video complexity, is the complexity bound threshold, is the adjustment factor; Using the analysis depth adjustment result, real-time monitoring of resource usage and video stream changes is performed to confirm that resource allocation dynamically corresponds to actual monitoring requirements, and to generate an optimal resource configuration plan; According to the optimal resource allocation plan, analyze future resource requirements and scenario changes, automatically update monitoring resource allocation, and build a cognitive resource allocation plan; The steps for obtaining the remote control alarm triggering scheme are: According to the cognitive resource allocation scheme, sound analysis is started, audio streams are collected from the real-time environment, segments including human voices and alarm sounds are screened, and preliminary sound source recognition results are obtained; Based on the preliminary sound source identification results, spectrum analysis is applied to perform in-depth analysis to distinguish between human voice and alarm sound, using the formula: Calculate the quantitative index of the frequency component intensity of two sound sources , and obtain the sound source separation result, where and Represents the maximum frequency components of human voice and alarm sound respectively, is a tiny value used to avoid division by zero errors; According to the sound source separation result, the alarm response strategy is adjusted to ensure that human voice is given priority in emergency situations and a remote control alarm triggering solution is obtained.

2. A mobile terminal remote control monitoring method, characterized in that: According to claim 1, the mobile terminal remote control monitoring system is implemented, comprising the following steps: Monitor network bandwidth and data traffic in real time, record current bandwidth and data traffic usage, evaluate the current network capacity's support for video data streams, adjust video stream bit rate and video encoding resolution, and obtain encoding parameter adjustment results; Apply the encoding parameter adjustment result to perform real-time frame analysis on the input video stream, detect pixel changes and color fluctuations in each frame, extract key dynamic events and actual pictures with significant color changes, and generate dynamic content and significant color change picture extraction results; Using the dynamic content and significant color change screen extraction results, separate the foreground elements and background in the video screen, identify and classify key actions or events in the foreground, determine the action type and occurrence time, analyze the key monitoring points in the scene, and generate key content identification results; Based on the key content identification results, perform video feature extraction and analysis, evaluate video complexity, identify the video segment with the greatest difference from the current monitoring scene requirements, dynamically adjust analysis resources, and obtain an analysis resource allocation plan; According to the analysis resource allocation plan, the analysis depth is adjusted for each video stream, resource usage and video stream changes are monitored in real time, future resource requirements and scene changes are analyzed, and a cognitive resource allocation plan is constructed; According to the cognitive resource allocation scheme, sound analysis is started, segments including human voices and alarm sounds are screened, spectrum analysis is applied for in-depth analysis, human voices and alarm sounds are distinguished, alarm response strategies are adjusted, and remote control alarm triggering schemes are obtained.

Images