Monitoring method based on ai gimbal-mounted searchlight, and related device

Abstract

The present application relates to the technical field of computer vision. Disclosed are a monitoring method based on an AI gimbal-mounted searchlight, and a related device. The monitoring method based on an AI gimbal-mounted searchlight comprises: acquiring a video image within a preset monitoring range in real time; detecting whether a target object presented in the video image is a risk target; and if the target object is a risk target, controlling a searchlight to illuminate the target object, and generating a warning sound until the target object cannot be detected. In the present application, when a camera device detects the risk target within a monitoring range, the searchlight is controlled to illuminate the target object, and the warning sound is generated, so as to warn and deter the risk target, thereby improving the security protection effect of a camera, and further improving the usage experience of a user.

Description

Monitoring methods and related equipment based on AI-powered PTZ searchlights Technical Field

[0001] This application relates to the field of computer vision technology, and in particular to a monitoring method and related equipment based on AI gimbal spotlights. Background Technology

[0002] Currently, with social development and technological advancements, surveillance cameras have become an important component of home and business security. They are not only used in commercial and public safety sectors but also widely applied in private residences to enhance home security. Specifically, home users install surveillance cameras primarily for theft prevention, fire prevention, and monitoring the safety of children and the elderly.

[0003] In related technologies, cameras typically send real-time images of the monitored area to the user's device, allowing the user to view the surveillance footage in real time. However, this method is ineffective at preventing unauthorized intrusion, resulting in poor security and a poor user experience.

[0004] The above content is only used to help understand the technical solution of this application and does not represent an admission that the above content is prior art.

[0005] Summary of the Invention

[0006] The main purpose of this application is to provide a monitoring method based on an AI PTZ spotlight, which aims to solve the technical problem of poor security protection effect of cameras, resulting in a poor user experience.

[0007] To achieve the above objectives, this application proposes a monitoring method based on an AI-powered gimbal spotlight, the monitoring method comprising:

[0008] Real-time acquisition of video images within a preset monitoring range;

[0009] Detect whether the target object presented in the video image is a risky target;

[0010] If the target object is a risky target, then the searchlight is controlled to illuminate the target object and generate an alarm sound until the target object can no longer be detected.

[0011] Optionally, the step of controlling the searchlight to illuminate the target object and generate an alarm sound until the target object can no longer be detected includes:

[0012] Collect the dynamic motion trajectory of the target object;

[0013] While continuously generating alarm sounds, the searchlight is controlled to illuminate the target object according to the dynamic movement trajectory until the target object can no longer be detected.

[0014] Optionally, the step of acquiring the dynamic motion trajectory of the target object includes:

[0015] The video image is converted into multiple image frames;

[0016] Based on the image frame, feature data detection is performed using a preset target tracking model to obtain a 3D detection box of the target object;

[0017] Based on the 3D detection box, target tracking is performed on the target object to obtain the target tracking result, and based on the target tracking result, the position of the target object is predicted to obtain the prediction result;

[0018] Based on the target tracking results and the prediction results, the dynamic motion trajectory of the target object is generated.

[0019] Optionally, the step of controlling the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected, includes:

[0020] If the target object is a risky target, then the first color temperature information is determined;

[0021] While generating an alarm sound, the searchlight is controlled to illuminate the target object with light corresponding to the first color temperature information until the target object can no longer be detected.

[0022] Optionally, after the step of controlling the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected, the method includes:

[0023] If the target object is not a risky target, then the second color temperature information is determined;

[0024] Turn on the lighting to generate illumination corresponding to the second color temperature information, and control the searchlight to illuminate the target object in the direction of the preset movement range of the target object, rather than illuminating the target object directly.

[0025] Optionally, the step of controlling the searchlight to illuminate the target object in the direction of a preset movement range according to the second color temperature information includes:

[0026] Collect the dynamic motion trajectory of the target object;

[0027] According to the dynamic motion trajectory, the searchlight is controlled to illuminate the target object in the direction of the preset movement range, according to the second color temperature information.

[0028] Optionally, the step of determining the second color temperature information includes:

[0029] The video image is converted into multiple image frames, and environmental features are extracted from the image frames;

[0030] Based on the environmental characteristics, color temperature adaptation analysis is performed using a preset color temperature control model to obtain second color temperature information. The color temperature control model is obtained by training a preset training model based on environmental feature samples and the color temperature information labels of the environmental feature samples.

[0031] Optionally, before the step of converting the video image into multiple image frames and extracting environmental features from the image frames, the method includes:

[0032] Obtain environmental feature samples and color temperature information labels for the environmental feature samples;

[0033] Based on the environmental feature samples and the color temperature information labels, the preset training model is iteratively trained to obtain the color temperature control model.

[0034] Optionally, the step of iteratively training a preset model to obtain a color temperature control model based on the environmental feature samples and the color temperature information labels includes:

[0035] Determine the time information of the environmental feature samples;

[0036] Based on the time information, the corresponding time weights are determined;

[0037] Based on the environmental feature samples, the color temperature information labels, and the time weights, the preset training model is iteratively trained to obtain the color temperature control model.

[0038] Optionally, the step of iteratively training a preset model to obtain a color temperature control model based on the environmental feature samples, the color temperature information labels, and the time weights includes:

[0039] Based on the environmental feature samples and the time weight, color temperature adaptation analysis is performed through a preset training model to obtain the color temperature output result.

[0040] The difference between the color temperature output result and the color temperature information label is calculated to obtain the error result;

[0041] Based on the error result, determine whether the error result meets the error standard indicated by the preset error threshold range;

[0042] If the error result does not meet the error standard indicated by the preset error threshold range, return to the step of performing color temperature adaptation analysis based on the environmental feature samples and the time weight, and obtaining the color temperature output result through the preset training model, until the error result meets the error standard indicated by the preset error threshold range, and then stop training to obtain the color temperature control model.

[0043] Furthermore, to achieve the above objectives, this application also proposes a monitoring device based on a pan-tilt searchlight, the monitoring device comprising:

[0044] The acquisition module is used to acquire video images within a preset monitoring range in real time.

[0045] The detection module is used to detect whether the target object presented in the video image is a risky target;

[0046] The control module is used to control the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected.

[0047] Optionally, the control module includes:

[0048] The first acquisition module is used to acquire the dynamic motion trajectory of the target object;

[0049] The illumination module is used to continuously generate an alarm sound while controlling the searchlight to illuminate the target object according to the dynamic motion trajectory until the target object can no longer be detected.

[0050] Optionally, the first acquisition module includes:

[0051] The first conversion module is used to convert the video image into multiple image frames;

[0052] The target detection module is used to detect feature data based on the image frame using a preset target tracking model to obtain a 3D detection box of the target object.

[0053] The tracking module is used to track the target object based on the 3D detection box, obtain the target tracking result, and predict the position of the target object based on the target tracking result, thereby obtaining the prediction result.

[0054] The generation module is used to generate the dynamic motion trajectory of the target object based on the target tracking result and the prediction result.

[0055] Optionally, the control module further includes:

[0056] The determination module is used to determine the first color temperature information if the target object is a risk target;

[0057] The first illumination module is used to control the searchlight to illuminate the target object with light corresponding to the first color temperature information while generating an alarm sound, until the target object can no longer be detected.

[0058] Optionally, the monitoring device based on the AI ​​PTZ searchlight further includes:

[0059] A color temperature determination module is used to determine second color temperature information if the target object is not a risk target;

[0060] The searchlight control module is used to turn on the lighting lamp, generate illumination corresponding to the second color temperature information, and control the searchlight to illuminate the target object in the direction of the preset movement range of the target object, rather than illuminating the target object directly.

[0061] Optionally, the searchlight control module includes:

[0062] The second acquisition module is used to acquire the dynamic motion trajectory of the target object;

[0063] The second illumination module is used to control the searchlight to illuminate the target object in the direction of the preset movement range according to the dynamic movement trajectory, and to provide illumination with the corresponding second color temperature information.

[0064] Optionally, the color temperature determination module includes:

[0065] An environmental feature extraction module is used to convert the video image into multiple image frames and extract environmental features from the image frames.

[0066] The analysis module is used to perform color temperature adaptation analysis based on the environmental features and through a preset color temperature control model to obtain second color temperature information. The color temperature control model is obtained by training a preset training model based on environmental feature samples and the color temperature information labels of the environmental feature samples.

[0067] Optionally, the monitoring device based on the AI ​​PTZ searchlight further includes:

[0068] The sample acquisition module is used to acquire environmental feature samples and color temperature information labels of the environmental feature samples;

[0069] The training module is used to iteratively train a preset model to be trained based on the environmental feature samples and the color temperature information labels to obtain a color temperature control model.

[0070] Optionally, the training module includes:

[0071] A time information determination module is used to determine the time information of the environmental feature sample;

[0072] The time weight determination module is used to determine the corresponding time weight based on the time information;

[0073] The color temperature control model training module is used to iteratively train a preset model to be trained based on the environmental feature samples, the color temperature information labels, and the time weights to obtain a color temperature control model.

[0074] Optionally, the color temperature control model training module includes:

[0075] The color temperature adaptation analysis module is used to perform color temperature adaptation analysis based on the environmental feature samples and the time weight, through a preset training model, to obtain the color temperature output result.

[0076] The difference calculation module is used to calculate the difference between the color temperature output result and the color temperature information label to obtain the error result;

[0077] The judgment module is used to determine, based on the error result, whether the error result meets the error standard indicated by the preset error threshold range;

[0078] The iterative training module is used to return the step of performing color temperature adaptation analysis based on the environmental feature samples and the time weights through the preset model to be trained to obtain the color temperature output result if the error result does not meet the error standard indicated by the preset error threshold range. The training stops when the error result meets the error standard indicated by the preset error threshold range, and the color temperature control model is obtained.

[0079] In addition, to achieve the above objectives, this application also proposes a monitoring device based on a PTZ searchlight, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the monitoring method based on an AI PTZ searchlight as described above.

[0080] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the monitoring method based on the AI ​​gimbal searchlight described above.

[0081] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the monitoring method for AI-based gimbal searchlights described above.

[0082] One or more technical solutions proposed in this application have at least the following technical effects:

[0083] In contrast to related technologies where cameras typically send real-time images within a monitored area to the user for real-time viewing, this method is ineffective at preventing intrusion by strangers, resulting in poor security and a poor user experience. This application, however, acquires video images within a preset monitoring range in real time, detects whether a target object in the video image is a risky target, and if so, controls a spotlight to illuminate the target object and generates an alarm sound until the target object can no longer be detected. Essentially, when the camera device of this application detects a risky target within its monitoring range, it controls a spotlight to illuminate the target object and generates an alarm sound to alert and drive away the risky target, thereby improving the security of the camera and enhancing the user experience. Attached Figure Description

[0084] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0085] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0086] Figure 1 is a flowchart illustrating the first embodiment of the monitoring method based on an AI gimbal searchlight according to this application.

[0087] Figure 2 is a schematic diagram of the application scenario of the monitoring method based on AI gimbal searchlight in this application;

[0088] Figure 3 is a schematic diagram of color temperature division of the monitoring method based on AI gimbal searchlight in this application;

[0089] Figure 4 is a flowchart illustrating the second embodiment of the monitoring method based on AI gimbal searchlight provided in this application;

[0090] Figure 5 is a schematic diagram of the surveillance camera structure of the monitoring method based on AI gimbal searchlight in this application;

[0091] Figure 6 is a flowchart illustrating the monitoring method based on an AI gimbal spotlight according to the third embodiment of this application.

[0092] Figure 7 is a schematic diagram of the module structure of the monitoring device based on the gimbal searchlight according to an embodiment of this application;

[0093] Figure 8 is a schematic diagram of the hardware operating environment involved in the monitoring method based on AI gimbal searchlight in the embodiments of this application.

[0094] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0095] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0096] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0097] The main solution of this application embodiment is: to acquire video images within a preset monitoring range in real time; to detect whether the target object presented in the video image is a risky target; if the target object is a risky target, to control the searchlight to illuminate the target object and generate an alarm sound until the target object can no longer be detected.

[0098] In this embodiment, a monitoring device based on a PTZ searchlight is used as the execution subject. For ease of description, it will be referred to as "device" below.

[0099] In related technologies, cameras typically send real-time images of the monitored area to the user's device for real-time viewing. However, this method is ineffective at preventing unauthorized intrusion, resulting in poor security and a poor user experience.

[0100] This application provides a solution to alert and drive away risky targets, thereby improving the security of cameras and enhancing the user experience.

[0101] As can be seen from the above embodiments, this application acquires video images within a preset monitoring range in real time, detects whether a target object presented in the video image is a risky target, and if the target object is a risky target, controls a spotlight to illuminate the target object and generates an alarm sound until the target object can no longer be detected. It is understood that when the camera device of this application detects a risky target within the monitoring range, it controls a spotlight to illuminate the target object and generates an alarm sound to alert and drive away the risky target, thereby improving the security effect of the camera and enhancing the user experience.

[0102] It should be noted that the executing entity in this embodiment can be a computing service device with data processing, network communication, and program execution functions, such as a tablet computer, personal computer, or mobile phone, or an electronic device or terminal device capable of performing the above functions. The following description uses a monitoring device based on a PTZ searchlight as an example to illustrate this embodiment and the subsequent embodiments.

[0103] Based on this, this application provides a monitoring method based on an AI PTZ searchlight. Referring to Figure 1, Figure 1 is a flowchart of the first embodiment of the monitoring method based on an AI PTZ searchlight of this application.

[0104] In this embodiment, the monitoring method based on AI gimbal spotlight includes steps S100 to S300:

[0105] Step S100: Acquire video images within the preset monitoring range in real time;

[0106] It should be noted that the preset monitoring range refers to a pre-determined and set monitoring area, which is selected based on security considerations, monitoring needs, or other purposes. The preset monitoring range can be fixed or can be an electronic fence set by software. The video images refer to a continuous sequence of images captured by the monitoring camera, which constitute a video stream.

[0107] In a specific implementation, referring to Figure 2 for a concrete application scenario, the AI-based PTZ spotlight-based monitoring method is applied to a camera installed at the entrance of a user's residence. This device can acquire video images within a preset monitoring range in real time, instantly capturing and processing events occurring within that range without significant delay; that is, the video stream is continuous and as close as possible to the actual time of occurrence. Specifically, the above describes the monitoring function of this device, which can continuously capture and display real-time video footage within a predefined monitoring area. The device is specifically applied in fields such as security monitoring and remote monitoring to ensure timely detection and response to any abnormal activity within the monitored area.

[0108] Step S200: Detect whether the target object presented in the video image is a risky target;

[0109] It should be noted that the target object refers to a specific target in the video, specifically a person or animal within the monitoring range; the risk target refers to an unauthorized intruder or a target with suspicious behavior.

[0110] In its specific implementation, the device detects whether the target objects presented in the video image are risk targets. Specifically, the device uses computer vision and deep learning algorithms to analyze the video stream, automatically identify and locate specific targets in the video, and perform risk assessments on these targets.

[0111] Furthermore, the step of the device detecting whether a target object presented in the video image is a risky target includes:

[0112] 1. Video capture: Acquire real-time video streams via a camera;

[0113] 2. Video preprocessing: Performing noise reduction, enhancement, and decoding on the video to obtain clear and usable video images;

[0114] 3. Video segmentation: Divide the video into consecutive frames and extract relevant information;

[0115] 4. Video Feature Extraction: Using computer vision algorithms, extract key features from the video, such as people, vehicles, and scenes;

[0116] 5. Target Detection and Tracking: Using target detection algorithms, such as YOLOv8, identify and locate specific targets in the video and track them between consecutive frames. This step may involve using deep learning models, such as the YOLO series models, which can analyze video content in real time, detect anomalies, and thus make risk predictions and construct intelligent monitoring devices.

[0117] 6. Behavior Analysis: Based on the target's trajectory and actions, analyze and classify its behavior, such as human activity detection and abnormal behavior recognition;

[0118] 7. Event Recognition and Alerts: By identifying specific events in the video, such as traffic accidents and thefts, timely alerts can be issued. For example, behavioral analysis can be used to understand and analyze people's behavior and activities, sentiment analysis can be used to identify emotions and feelings, and scene analysis can be used to understand and reason about scenes, thereby enabling intelligent analysis and decision-making based on video data.

[0119] 8. Risk Assessment: Based on the analysis of detected targets and behaviors, assess whether there are any risks or threats. For example, if an unauthorized intruder or suspicious behavior is detected, the device will trigger an alarm.

[0120] Step S300: If the target object is a risky target, control the searchlight to illuminate the target object and generate an alarm sound until the target object can no longer be detected.

[0121] It should be noted that a searchlight is a high-power lighting device that can produce a strong, concentrated beam of light to illuminate distant objects or areas; the alarm sound can be a siren, a buzzer, or other preset sound patterns, used to alert that a risky target has illegally entered the designated monitoring area.

[0122] In its implementation, the device analyzes identified target objects to determine whether they pose a security risk or threat. Risk targets can be preset, such as unauthorized personnel, suspicious objects, or abnormal behavior. If the device determines that the target object is a risk target, it automatically controls a spotlight to shine on it; that is, the device uses strong light to illuminate the risk target, accompanied by an alarm sound, to deter the risk target until it disappears from the monitoring range, meaning the device can no longer detect the target object, at which point it stops shining the light and generating the alarm sound.

[0123] In its specific implementation, when the camera device of this application detects a risky target within its monitoring range, it controls the searchlight to illuminate the target object and generates an alarm sound to warn and drive away the risky target, thereby improving the security and prevention effect of the camera and enhancing the user experience.

[0124] In its specific implementation, the device controls a searchlight to illuminate the target object and generate an alarm sound until the target object can no longer be detected. This includes the following steps:

[0125] The system collects the dynamic motion trajectory of the target object; while continuously generating an alarm sound, it controls a searchlight to illuminate the target object according to the dynamic motion trajectory until the target object can no longer be detected.

[0126] In its implementation, when the device determines that a target object is a risky target, it captures images in real time through video surveillance equipment and uses computer vision technology to identify and track the target object in the images. As the target object moves, the device records its position changes within the monitored area, thereby acquiring the dynamic motion trajectory of the target object.

[0127] In its implementation, after determining the dynamic trajectory of the target object, the device triggers an alarm mechanism, emitting a continuous alarm sound. Furthermore, the device adjusts the direction of the spotlight in real time based on the collected dynamic trajectory of the target object, ensuring the spotlight beam remains constantly focused on the target object. Once the target object is no longer detected, the device automatically stops the alarm sound and the spotlight's tracking illumination, resuming normal monitoring. Understandably, the purpose of this tracking illumination is to keep the target object within the illumination range, facilitating the monitoring and removal of risky targets. This further enables alarming and repelling of risky targets, thereby improving the security and safety of the camera and enhancing the user experience.

[0128] In its specific implementation, the step of the device acquiring the dynamic motion trajectory of the target object includes:

[0129] The video image is converted into multiple image frames; based on the image frames, feature data detection is performed using a preset target tracking model to obtain a 3D detection box of the target object; based on the 3D detection box, target tracking is performed on the target object to obtain a target tracking result, and based on the target tracking result, the position of the target object is predicted to obtain a prediction result; based on the target tracking result and the prediction result, the dynamic motion trajectory of the target object is generated.

[0130] In its implementation, the device first converts the video stream into a series of image frames, where the video is composed of consecutive image frames, each representing a moment in the video. Secondly, the device analyzes each frame using a pre-defined target tracking model to detect target objects in the image and extract their feature data. This detection process involves using deep learning models, such as convolutional neural networks (CNNs), to identify and locate targets in the image.

[0131] Furthermore, the device generates a 3D detection box based on the feature data of the detected target object. The 3D detection box includes not only the target object's position (x, y coordinates) in the two-dimensional image but also its depth information (z coordinate), thus enabling a more accurate description of the target object's position in three-dimensional space. Next, the device utilizes target tracking algorithms, such as the Kalman filter or Mean-Shift algorithm, to track the target object. This involves matching and tracking the target object's movement across consecutive image frames to obtain the target object's positional information over time.

[0132] In its implementation, the device uses a prediction algorithm to predict the position of the target object in the next few frames based on the target tracking results. This prediction algorithm allows the device to prepare and adjust the camera or other sensors in advance to continuously track fast-moving targets. Finally, the device combines the target tracking results and the position prediction results to generate the dynamic motion trajectory of the target object. This trajectory can be two-dimensional (on the image plane) or three-dimensional (in actual space), depending on the requirements of the application scenario.

[0133] In practice, the above process enables the device to automatically detect, track, and predict target objects in the video. In this way, the device can monitor the behavior of target objects in real time and predict their future position and movement.

[0134] In a specific implementation, the step of controlling the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected, includes:

[0135] If the target object is a risky target, then the first color temperature information is determined; while generating an alarm sound, the searchlight is controlled to illuminate the target object with light corresponding to the first color temperature information until the target object can no longer be detected.

[0136] In its implementation, after identifying a risky target, the device determines the color temperature information corresponding to the risk situation. Referring to Figure 3, color temperature is a physical quantity describing the color of a light source. Different color temperatures can produce different lighting effects. For example, a warmer color temperature (such as 3000K) can create a comfortable and warning atmosphere, while a cooler color temperature (such as 5000K) gives people a clear and serious feeling.

[0137] In practical implementation, since the target object is a risky target, the device uses cool color temperature or flashing light to improve visibility and warning effect for the purpose of warning and driving away risky targets. High-brightness strobe lights can also be set to effectively deter suspected targets.

[0138] It should be noted that the first color temperature information and the second color temperature information are light color temperatures under different scenarios. The scenario with the first color temperature information is the scenario where the target object is a risky target, and a cool color temperature is usually set with flashing lights or high-brightness strobe lights. The scenario with the second color temperature information is the scenario where the target object is not a risky target, and a warm color temperature is usually used to welcome guests or help the target object illuminate the road in the dark.

[0139] In its implementation, this application identifies risk targets and proposes a method to control color temperature in addition to lighting conditions, thereby enhancing the security and deterrence of risk targets and improving the user experience.

[0140] In contrast to related technologies where cameras typically send real-time images within a monitored area to the user for real-time viewing, this method is ineffective at preventing intrusion by strangers, resulting in poor security and a poor user experience. This application, however, acquires video images within a preset monitoring range in real time, detects whether a target object in the video image is a risky target, and if so, controls a spotlight to illuminate the target object and generates an alarm sound until the target object can no longer be detected. Essentially, when the camera device of this application detects a risky target within its monitoring range, it controls a spotlight to illuminate the target object and generates an alarm sound to alert and drive away the risky target, thereby improving the security of the camera and enhancing the user experience.

[0141] Based on the first embodiment described above, this application also proposes another embodiment. Referring to FIG4, the monitoring method based on the AI ​​gimbal searchlight further includes:

[0142] In a specific implementation, after the step of controlling the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected, the method includes:

[0143] Step A100: If the target object is not a risky target, then determine the second color temperature information;

[0144] It should be noted that the first color temperature information and the second color temperature information are light color temperatures under different scenarios. The scenario with the first color temperature information is the scenario where the target object is a risky target, and a cool color temperature is usually set with flashing lights or high-brightness strobe lights. The scenario with the second color temperature information is the scenario where the target object is not a risky target, and a warm color temperature is usually used to welcome guests or help the target object illuminate the road in the dark.

[0145] Step A200: Turn on the lighting to generate illumination corresponding to the second color temperature information, and control the searchlight to illuminate the target object in the direction of the preset movement range of the target object with the second color temperature information, instead of illuminating the target object directly.

[0146] In the specific implementation, referring to Figure 5, the light above the camera is a wide-area illumination light, and the light below the camera is a searchlight (tracking light). After the device determines that the target object is not a risky target and obtains the second color temperature information, it turns on the illumination light to generate illumination corresponding to the second color temperature information, and controls the searchlight to illuminate the target object in the direction of the preset movement range of the target object, rather than directly illuminating the target object. Specifically, the searchlight does not directly illuminate the target object; it is used to illuminate the preset movement range of the target object. That is, the beam of the searchlight is directed towards the area where the target object may move or enter, rather than directly illuminating the target object.

[0147] It should be noted that the beam of the searchlight is directed towards the area where the target object may move or enter, rather than shining directly on the target object. This is to avoid the strong light shining directly on the legitimate user's body / or face, while illuminating their possible path of movement. The light can also highlight the area near the destination (i.e., the house), allowing the legitimate user to comfortably see the path ahead, thereby improving the user experience.

[0148] In a specific implementation, the step of controlling the searchlight to illuminate the target object in the direction of a preset movement range according to the second color temperature information includes:

[0149] Collect the dynamic motion trajectory of the target object; according to the dynamic motion trajectory, control the searchlight to illuminate the target object in the direction of the preset movement range with the corresponding second color temperature information.

[0150] In its implementation, the device uses a video surveillance system to capture the motion trajectory of the target object within the monitored area. This typically involves video analytics techniques, such as target detection and tracking algorithms, to identify the target object and track its position in consecutive image frames. Based on the captured dynamic motion trajectory of the target object, the device automatically controls the direction of the spotlight to illuminate the preset range of movement the target object might move into, rather than directly illuminating the target object.

[0151] Furthermore, the searchlight not only adjusts its direction of illumination but also the color temperature of the light to meet specific monitoring needs. Specifically, the direction of the searchlight is set to the preset movement range of the target object, meaning the system predicts the possible movement path of the target object and illuminates the area in advance.

[0152] In its implementation, the above process describes an automated monitoring system that can track the movement trajectory of target objects in real time and intelligently control the direction and color temperature of the searchlights based on this information to improve monitoring efficiency and effectiveness.

[0153] In its specific implementation, the steps for determining the second color temperature information include:

[0154] The video image is converted into multiple image frames, and environmental features are extracted from the image frames. Based on the environmental features, color temperature adaptation analysis is performed through a preset color temperature control model to obtain second color temperature information. The color temperature control model is obtained by training a preset model based on environmental feature samples and the color temperature information labels of the environmental feature samples.

[0155] In its implementation, the device first converts the video stream into a series of image frames. This is achieved through video processing technology, with each frame representing a single moment in the video, providing foundational data for subsequent processing. Secondly, the device extracts environmental features from these image frames. These features may include objects in the scene, lighting conditions, color distribution, etc., which help the device understand the current environmental conditions.

[0156] Furthermore, based on the extracted environmental features, the device performs color temperature adaptation analysis using a pre-set color temperature control model. This model may be trained based on previously collected environmental feature samples and their corresponding color temperature information labels, and it can determine the most suitable color temperature according to the current environmental features. The result of the color temperature adaptation analysis is a determined "second color temperature information," which is the optimal color temperature value calculated by the device based on the current environmental features and the pre-set model.

[0157] In practice, the device can automatically adjust the color temperature of the lighting based on real-time video images and environmental characteristics, and intelligently control the direction of the spotlight to adapt to the needs of the monitoring scenario, thereby improving the user experience.

[0158] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the monitoring method of the AI ​​gimbal searchlight in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0159] Based on the first and second embodiments described above, this application also proposes another embodiment. Referring to Figure 6, the monitoring method based on the AI ​​gimbal searchlight further includes:

[0160] Prior to the step of converting the video image into multiple image frames and extracting environmental features from the image frames, the method includes:

[0161] Step B100: Obtain environmental feature samples and color temperature information labels of the environmental feature samples;

[0162] It should be noted that the device first acquires environmental feature samples. This step involves collecting image or video data under different environmental conditions using cameras, sensors, or other data acquisition devices. This data will serve as the basis for analysis and subsequent model training. Environmental feature samples may include various visual elements, such as the shape, size, color, and texture of objects, as well as the scene's lighting conditions and background noise.

[0163] Furthermore, for each collected environmental feature sample, a corresponding color temperature information label needs to be obtained. Color temperature is a physical quantity describing the color of a light source, usually expressed in Kelvin (K). Color temperature information labels can be manually labeled or automatically measured using devices such as color thermometers. These labels describe the lighting conditions of the sample at a specific color temperature.

[0164] Step B200: Based on the environmental feature samples and the color temperature information labels, iteratively train the preset training model to obtain the color temperature control model.

[0165] In practical implementation, the device first selects or designs a preset model structure. This model can accept input environmental feature samples and output the corresponding color temperature prediction. This model is untrained or "to be trained" before training begins; its parameters have not yet been adjusted to the optimal state, i.e., it is a model to be trained.

[0166] In its implementation, the device uses collected environmental feature samples and color temperature information labels to train the model. The training process typically includes the following steps:

[0167] Forward propagation: Input sample data, and obtain the predicted output through model calculation;

[0168] Loss calculation: The model's predicted output is compared with the actual color temperature information labels, and the loss function is calculated to evaluate the model's prediction accuracy;

[0169] Backpropagation: Based on the loss function, calculate the gradient of the model parameters to determine how to adjust the parameters to reduce the loss;

[0170] Parameter update: Update the model's parameters using optimization algorithms (such as gradient descent);

[0171] This process involves multiple iterations, each aimed at improving the model's performance, until the model achieves satisfactory accuracy on a given dataset.

[0172] In practice, after multiple iterations of training, the model's parameters are adjusted to an optimal state. At this point, the model can accurately predict the appropriate color temperature based on the input environmental feature samples. This trained model is called a color temperature control model, which can be used in practical applications to automatically adjust the color temperature of the light source based on real-time environmental characteristics.

[0173] Understandably, the above process describes how to use machine learning methods to train a model capable of predicting color temperature using labeled environmental feature sample data. This model can automatically adjust the color temperature of light sources under different environmental conditions to suit human eye comfort or specific application needs.

[0174] In a specific implementation, the device iteratively trains a preset training model based on the environmental feature samples and the color temperature information labels to obtain a color temperature control model, including the following steps:

[0175] Determine the time information of the environmental feature samples; determine the corresponding time weights based on the time information; iteratively train the preset training model based on the environmental feature samples, the color temperature information labels, and the time weights to obtain the color temperature control model.

[0176] In a specific implementation, this application further proposes that when the device collects environmental characteristic samples, it simultaneously records the time information of each sample, including the specific date and time the sample was collected, the season, and even a specific time of day (such as morning, noon, or evening).

[0177] In its implementation, the device assigns a time weight to each sample based on the sample's temporal information. This weight reflects the degree to which time factors influence the model's predictions. For example, if environmental features have a greater impact on color temperature during a certain time period, then samples from that time period may be assigned a higher time weight.

[0178] In practical implementation, the same environmental feature information has different effects at different time periods, resulting in different feature weights for the same information. Therefore, time weights are introduced to improve the accuracy of model analysis. For example, if legitimate users are monitored during the summer months of May to July, and the emitted light beam is warm-toned, it will enhance the user's perception of heat, making them feel more agitated and thus causing frustration, ultimately reducing the user experience. Therefore, in the above scenario, during the summer months of May to July, the weight of warm tones will be reduced, and the weight of cool tones will be increased. Similarly, during the winter months of September to December, the weight of cool tones will be reduced, and the weight of warm tones will be increased.

[0179] In a specific implementation, the device iteratively trains a preset training model based on the environmental feature samples, the color temperature information labels, and the time weights to obtain a color temperature control model, including the following steps:

[0180] Based on the environmental feature samples and the time weights, color temperature adaptation analysis is performed using a preset training model to obtain a color temperature output result. The difference between the color temperature output result and the color temperature information label is calculated to obtain an error result. Based on the error result, it is determined whether the error result meets the error standard indicated by a preset error threshold range. If the error result does not meet the error standard indicated by the preset error threshold range, the process returns to the step of performing color temperature adaptation analysis using the preset training model based on the environmental feature samples and the time weights to obtain a color temperature output result. Training stops when the error result meets the error standard indicated by the preset error threshold range, thus obtaining a color temperature control model.

[0181] It should be noted that, firstly, the device performs color temperature adaptation analysis based on the environmental feature samples and the time weights, using a pre-set training model to obtain a color temperature output result. Specifically, the device uses a pre-set training model, combined with environmental feature samples and corresponding time weights, to perform color temperature adaptation analysis. This process is how the model learns how to predict a suitable color temperature based on the input environmental features and time information. After the model analysis, a color temperature prediction result is output. This result is derived based on the model's current learning state and the input data.

[0182] Understandably, the device then calculates the difference between the color temperature output result and the color temperature information label to obtain the error result, that is, to verify whether the result obtained by the model during training is consistent with the known result, and to calculate the difference between the results to obtain the error result.

[0183] It should be noted that the device determines whether the error result meets the error standard indicated by the preset error threshold range based on the error result. Specifically, since there is an error between the result after model training and the actual result, the error result is allowed to be within the preset error threshold range, thereby further determining whether the error result meets the error standard indicated by the preset error threshold range.

[0184] Understandably, if the error result does not meet the error standard indicated by the preset error threshold range, it means that the model has too large an error in this training. The device returns to the step of performing color temperature adaptation analysis based on the environmental feature samples and the time weight, and obtaining the color temperature output result through the preset model to be trained. That is, iterative training is performed until the error result meets the error standard indicated by the preset error threshold range, and then training stops to obtain the color temperature control model. The above iterative process ensures that the final color temperature control model can provide accurate color temperature prediction in practical applications.

[0185] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the monitoring method of the AI ​​gimbal searchlight in this application. Any simple modifications based on this technical concept are within the protection scope of this application.

[0186] This application also provides a monitoring device based on a pan-tilt searchlight. Referring to Figure 7, the monitoring device based on a pan-tilt searchlight includes:

[0187] The acquisition module 10 is used to acquire video images within a preset monitoring range in real time;

[0188] Detection module 20 is used to detect whether the target object presented in the video image is a risky target;

[0189] The control module 30 is used to control the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected.

[0190] Optionally, the control module 30 includes:

[0191] The first acquisition module is used to acquire the dynamic motion trajectory of the target object;

[0192] The illumination module is used to continuously generate an alarm sound while controlling the searchlight to illuminate the target object according to the dynamic motion trajectory until the target object can no longer be detected.

[0193] Optionally, the first acquisition module includes:

[0194] The first conversion module is used to convert the video image into multiple image frames;

[0195] The target detection module is used to detect feature data based on the image frame using a preset target tracking model to obtain a 3D detection box of the target object.

[0196] The tracking module is used to track the target object based on the 3D detection box, obtain the target tracking result, and predict the position of the target object based on the target tracking result, thereby obtaining the prediction result.

[0197] The generation module is used to generate the dynamic motion trajectory of the target object based on the target tracking result and the prediction result.

[0198] Optionally, the control module 30 further includes:

[0199] The determination module is used to determine the first color temperature information if the target object is a risk target;

[0200] The first illumination module is used to control the searchlight to illuminate the target object with light corresponding to the first color temperature information while generating an alarm sound, until the target object can no longer be detected.

[0201] Optionally, the monitoring device based on the AI ​​PTZ searchlight further includes:

[0202] A color temperature determination module is used to determine second color temperature information if the target object is not a risk target;

[0203] The searchlight control module is used to turn on the lighting lamp, generate illumination corresponding to the second color temperature information, and control the searchlight to illuminate the target object in the direction of the preset movement range of the target object, rather than illuminating the target object directly.

[0204] Optionally, the searchlight control module includes:

[0205] The second acquisition module is used to acquire the dynamic motion trajectory of the target object;

[0206] The second illumination module is used to control the searchlight to illuminate the target object in the direction of the preset movement range according to the dynamic movement trajectory, and to provide illumination with the corresponding second color temperature information.

[0207] Optionally, the color temperature determination module includes:

[0208] An environmental feature extraction module is used to convert the video image into multiple image frames and extract environmental features from the image frames.

[0209] The analysis module is used to perform color temperature adaptation analysis based on the environmental features and through a preset color temperature control model to obtain second color temperature information. The color temperature control model is obtained by training a preset training model based on environmental feature samples and the color temperature information labels of the environmental feature samples.

[0210] Optionally, the monitoring device based on the AI ​​PTZ searchlight further includes:

[0211] The sample acquisition module is used to acquire environmental feature samples and color temperature information labels of the environmental feature samples;

[0212] The training module is used to iteratively train a preset model to be trained based on the environmental feature samples and the color temperature information labels to obtain a color temperature control model.

[0213] Optionally, the training module includes:

[0214] A time information determination module is used to determine the time information of the environmental feature sample;

[0215] The time weight determination module is used to determine the corresponding time weight based on the time information;

[0216] The color temperature control model training module is used to iteratively train a preset model to be trained based on the environmental feature samples, the color temperature information labels, and the time weights to obtain a color temperature control model.

[0217] Optionally, the color temperature control model training module includes:

[0218] The color temperature adaptation analysis module is used to perform color temperature adaptation analysis based on the environmental feature samples and the time weight, through a preset training model, to obtain the color temperature output result.

[0219] The difference calculation module is used to calculate the difference between the color temperature output result and the color temperature information label to obtain the error result;

[0220] The judgment module is used to determine, based on the error result, whether the error result meets the error standard indicated by the preset error threshold range;

[0221] The iterative training module is used to return the step of performing color temperature adaptation analysis based on the environmental feature samples and the time weights through the preset model to be trained to obtain the color temperature output result if the error result does not meet the error standard indicated by the preset error threshold range. The training stops when the error result meets the error standard indicated by the preset error threshold range, and the color temperature control model is obtained.

[0222] The monitoring device based on a PTZ searchlight provided in this application employs the AI-based PTZ searchlight monitoring method described in the above embodiments, and can solve the technical problems of PTZ searchlight-based monitoring. Compared with the prior art, the beneficial effects of the PTZ searchlight-based monitoring device provided in this application are the same as those of the AI-based PTZ searchlight monitoring method provided in the above embodiments, and other technical features in the PTZ searchlight-based monitoring device are the same as those disclosed in the methods of the above embodiments, and will not be repeated here.

[0223] This application provides a monitoring device based on a PTZ searchlight. The monitoring device based on a PTZ searchlight includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the monitoring method based on an AI PTZ searchlight in the first embodiment described above.

[0224] Referring to Figure 8 below, a schematic diagram of a PTZ-based searchlight monitoring device suitable for implementing embodiments of this application is shown. The PTZ-based searchlight monitoring device in embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PADs (Portable Application Description), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. The PTZ-based searchlight monitoring device shown in Figure 8 is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0225] As shown in Figure 8, the monitoring device based on a PTZ searchlight may include a processing unit 1001 (e.g., a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to programs stored in read-only memory (ROM) 1002 or programs loaded from storage device 1003 into random access memory (RAM) 1004. RAM 1004 also stores various programs and data required for the operation of the PTZ searchlight-based monitoring device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. Input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touchscreens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the pan-tilt-zoom (PTZ) spotlight-based monitoring equipment to exchange data wirelessly or via wired communication with other devices. Although the figure shows a PTZ spotlight-based monitoring equipment with various systems, it should be understood that it is not required to implement or possess all the systems shown. More or fewer systems can be implemented or possessed alternatively.

[0226] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0227] The PTZ-based searchlight monitoring device provided in this application employs the AI-based PTZ searchlight monitoring method described in the above embodiments, which can solve the technical problems of PTZ searchlight-based monitoring. Compared with the prior art, the beneficial effects of the PTZ-based searchlight monitoring device provided in this application are the same as those of the AI-based PTZ searchlight monitoring method provided in the above embodiments, and other technical features of this PTZ-based searchlight monitoring device are the same as those disclosed in the previous embodiment method, and will not be repeated here.

[0228] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0229] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0230] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, the computer-readable program instructions being used to execute the monitoring method based on an AI gimbal searchlight in the above embodiments.

[0231] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0232] The aforementioned computer-readable storage medium may be included in a pan-tilt-zoom (PTZ) spotlight-based monitoring device; or it may exist independently and not be assembled into a PTZ spotlight-based monitoring device.

[0233] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the aforementioned one or more programs, cause the pan-tilt-zoom (PTZ) searchlight-based monitoring device to perform pan-tilt-zoom (PTZ) searchlight-based monitoring.

[0234] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0235] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0236] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0237] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described monitoring method based on an AI PTZ searchlight, thereby solving the technical problem of monitoring based on a PTZ searchlight. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the monitoring method based on an AI PTZ searchlight provided in the above embodiments, and will not be repeated here.

[0238] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the monitoring method based on an AI gimbal searchlight as described above.

[0239] The computer program product provided in this application can solve the technical problem of monitoring based on PTZ searchlights. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as those of the AI ​​PTZ searchlight monitoring method provided in the above embodiments, and will not be repeated here.

[0240] The above description is only a part of the embodiments of this application and does not limit the patent scope of this application. All equivalent structural transformations made under the technical concept of this application and using the contents of the specification and drawings of this application, or direct / indirect applications in other related technical fields, are included in the patent protection scope of this application.

Claims

1. A monitoring method based on an AI-powered PTZ searchlight, characterized in that, The monitoring method based on PTZ searchlights includes: Real-time acquisition of video images within a preset monitoring range; Detect whether the target object presented in the video image is a risky target; If the target object is a risky target, the searchlight is controlled to illuminate the target object and generate an alarm sound until the target object can no longer be detected. 2.The AI-based cloud pan-tilt spotlight-based monitoring method of claim 1, wherein, The step of controlling the searchlight to illuminate the target object and generate an alarm sound until the target object can no longer be detected includes: Collect the dynamic motion trajectory of the target object; While continuously generating alarm sounds, the searchlight is controlled to illuminate the target object according to the dynamic movement trajectory until the target object can no longer be detected. 3.The AI-based PTZ spotlight-based monitoring method of claim 2, wherein, The step of acquiring the dynamic motion trajectory of the target object includes: The video image is converted into multiple image frames; Based on the image frame, feature data detection is performed using a preset target tracking model to obtain a 3D detection box of the target object; Based on the 3D detection box, target tracking is performed on the target object to obtain the target tracking result, and based on the target tracking result, the position of the target object is predicted to obtain the prediction result; Based on the target tracking results and the prediction results, the dynamic motion trajectory of the target object is generated. 4.The AI-based cloud pan-tilt spotlight-based monitoring method of claim 1, wherein, The step of controlling the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected, includes: If the target object is a risky target, then the first color temperature information is determined; While generating an alarm sound, the searchlight is controlled to illuminate the target object with light corresponding to the first color temperature information until the target object can no longer be detected. 5.The AI-based cloud pan-tilt spotlight monitoring method of claim 1, wherein, After the step of controlling the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected, the method includes: If the target object is not a risky target, then the second color temperature information is determined; Turn on the lighting to generate illumination corresponding to the second color temperature information, and control the searchlight to illuminate the target object in the direction of the preset movement range of the target object, rather than illuminating the target object directly. 6.The AI-based PTZ spotlight-based monitoring method of claim 5, wherein, The step of controlling the searchlight to illuminate the target object in the direction of a preset movement range according to the second color temperature information includes: Collect the dynamic motion trajectory of the target object; According to the dynamic motion trajectory, the searchlight is controlled to illuminate the target object in the direction of the preset movement range, according to the second color temperature information. 7.The AI-based cloud pan-tilt spotlight-based monitoring method of claim 5, wherein, The step of determining the second color temperature information includes: The video image is converted into multiple image frames, and environmental features are extracted from the image frames; Based on the environmental characteristics, color temperature adaptation analysis is performed using a preset color temperature control model to obtain second color temperature information. The color temperature control model is obtained by training a preset training model based on environmental feature samples and the color temperature information labels of the environmental feature samples. 8.The AI-based cloud pan-tilt spotlight-based monitoring method of claim 7, wherein, Prior to the step of converting the video image into multiple image frames and extracting environmental features from the image frames, the method includes: Obtain environmental feature samples and color temperature information labels for the environmental feature samples; Based on the environmental feature samples and the color temperature information labels, the preset training model is iteratively trained to obtain the color temperature control model. 9.The AI-based pan-tilt spotlight lamp-based monitoring method of claim 8, wherein, The step of iteratively training a preset training model based on the environmental feature samples and the color temperature information labels to obtain a color temperature control model includes: Determine the time information of the environmental feature samples; Based on the time information, the corresponding time weights are determined; Based on the environmental feature samples, the color temperature information labels, and the time weights, the preset training model is iteratively trained to obtain the color temperature control model. 10.The AI-based cloud pan-tilt spotlight-based monitoring method of claim 9, wherein, The step of iteratively training a preset model to obtain a color temperature control model based on the environmental feature samples, the color temperature information labels, and the time weights includes: Based on the environmental feature samples and the time weight, color temperature adaptation analysis is performed through a preset training model to obtain the color temperature output result. The difference between the color temperature output result and the color temperature information label is calculated to obtain the error result; Based on the error result, determine whether the error result meets the error standard indicated by the preset error threshold range; If the error result does not meet the error standard indicated by the preset error threshold range, return to the step of performing color temperature adaptation analysis based on the environmental feature samples and the time weight, and obtaining the color temperature output result through the preset training model, until the error result meets the error standard indicated by the preset error threshold range, and then stop training to obtain the color temperature control model.

11. A monitoring device based on a pan-tilt searchlight, characterized by The device includes: The acquisition module is used to acquire video images within a preset monitoring range in real time. The detection module is used to detect whether the target object presented in the video image is a risky target; The control module is used to control the searchlight to illuminate the target object and generate an alarm sound if the target object is a risky target, until the target object can no longer be detected. 12.The AI-based cloud head monitoring device of claim 11, wherein The control module includes: The first acquisition module is used to acquire the dynamic motion trajectory of the target object; The illumination module is used to continuously generate an alarm sound while controlling the searchlight to illuminate the target object according to the dynamic motion trajectory until the target object can no longer be detected. 13.The AI-based cloud pan-tilt spotlight monitoring device of claim 12, wherein, The first acquisition module includes: The first conversion module is used to convert the video image into multiple image frames; The target detection module is used to detect feature data based on the image frame using a preset target tracking model to obtain a 3D detection box of the target object. The tracking module is used to track the target object based on the 3D detection box, obtain the target tracking result, and predict the position of the target object based on the target tracking result, thereby obtaining the prediction result. The generation module is used to generate the dynamic motion trajectory of the target object based on the target tracking result and the prediction result. 14.The AI-based cloud head monitoring device of claim 11, wherein The control module further includes: The determination module is used to determine the first color temperature information if the target object is a risk target; The first illumination module is used to control the searchlight to illuminate the target object with light corresponding to the first color temperature information while generating an alarm sound, until the target object can no longer be detected. 15.The AI-based cloud head monitoring device of claim 11, wherein The device further includes: A color temperature determination module is used to determine second color temperature information if the target object is not a risk target; The searchlight control module is used to turn on the lighting lamp, generate illumination corresponding to the second color temperature information, and control the searchlight to illuminate the target object in the direction of the preset movement range of the target object, rather than illuminating the target object directly. 16.The AI-based cloud dome spotlight lamp-based monitoring device according to claim 15, wherein, The searchlight control module includes: The second acquisition module is used to acquire the dynamic motion trajectory of the target object; The second illumination module is used to control the searchlight to illuminate the target object in the direction of the preset movement range according to the dynamic movement trajectory, and to illuminate it with the corresponding second color temperature information.

17. The AI-based PTZ spotlight-based monitoring device of claim 15, wherein, The color temperature determination module includes: An environmental feature extraction module is used to convert the video image into multiple image frames and extract environmental features from the image frames. The analysis module is used to perform color temperature adaptation analysis based on the environmental features and through a preset color temperature control model to obtain second color temperature information. The color temperature control model is obtained by training a preset training model based on environmental feature samples and the color temperature information labels of the environmental feature samples.

18. A monitoring device based on a pan-tilt searchlight, characterized by The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the monitoring method based on an AI PTZ searchlight as described in any one of claims 1 to 10.

19. A storage medium, characterized by The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the monitoring method based on an AI gimbal searchlight as described in any one of claims 1 to 10.

20. A computer program product, characterised in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the monitoring method based on an AI gimbal searchlight as described in any one of claims 1 to 10.

Images