A method and system for adjusting the camera sensitivity based on live broadcast

Abstract

The present invention relates to the field of sensitivity adjustment, and particularly to a method and system for adjusting the camera sensitivity based on live broadcast, including obtaining a sensitivity adjustment instruction during live broadcast; obtaining the current sensitivity parameter according to the sensitivity adjustment instruction; when the current sensitivity parameter is greater than a preset threshold, calculating an adjustment parameter according to a preset first formula for adjusting the sensitivity parameter in a large range; when the current sensitivity parameter is less than the preset threshold, calculating an adjustment parameter according to a preset second formula for adjusting the sensitivity parameter in a small range; and adjusting the current sensitivity based on the adjustment parameter to obtain an adjusted sensitivity. The method can adjust the sensitivity in a segmented manner according to the non-linear perception characteristics of the human eye to brightness, better conform to the visual perception of the human eye, and improve the viewing experience.

Description

Technical Field

[0001] This application relates to the technical field of sensitivity adjustment, and particularly to a method and system for adjusting the sensitivity of a camera based on live broadcast. Background Art

[0002] The adjustment of the sensitivity of a live camera refers to adjusting the sensitivity parameter of the camera device during a live broadcast to meet the shooting requirements under different lighting conditions. Sensitivity is a measure of the sensitivity of a camera device to light, usually represented by the ISO value. A lower ISO value is suitable for well-lit environments, capable of capturing more details and reducing image noise; while a higher ISO value is suitable for low-light environments, capable of increasing the sensitivity of the camera device to capture more light. In live camera shooting, sensitivity adjustment is very important because the lighting conditions may change at any time, such as entering from outdoors to indoors or changing from daytime to night. Through appropriate sensitivity adjustment, the brightness and contrast of the live broadcast image can be ensured to be appropriate, enabling the audience to clearly see the object being photographed and obtaining a good viewing experience.

[0003] In the prior art, the camera device adjusts the sensitivity in a linear manner, which means that each time the sensitivity is adjusted, the amount of light increased or decreased is equal. However, the human eye's perception of brightness is not linear. The perception of the highlight part conforms to a logarithmic model, and the perception of the dark part is close to a square root model. Therefore, more precise adjustment is required under low-light conditions, while it can be relatively rough under high-light conditions. The linear adjustment method cannot meet this non-linear perception characteristic. Secondly, some camera devices provide fixed sensitivity selection buttons, such as options like ISO 100, ISO 200, ISO 400, etc. This method limits the user's selection range and may not meet the requirements in specific scenarios. In addition, due to the lighting conditions may change at any time, switching between different fixed sensitivities may not be flexible and precise enough.

[0004] Therefore, there are deficiencies in the prior art and improvements are needed. Summary of the Invention

[0005] In order to solve one or several problems in the prior art, the main object of this application is to provide a method and system for adjusting the sensitivity of a camera based on live broadcast.

[0006] To achieve the above invention object, this application proposes a method for adjusting the sensitivity of a camera based on live broadcast, the method comprising:

[0007] Obtain a sensitivity adjustment instruction during the live broadcast;

[0008] According to the sensitivity adjustment instruction, obtain the current sensitivity parameter;

[0009] When the current sensitivity parameter is greater than a preset threshold, an adjustment parameter is calculated according to a preset first formula for adjusting the sensitivity parameter in a large range;

[0010] When the current sensitivity parameter is less than the preset threshold, an adjustment parameter is calculated according to a preset second formula for adjusting the sensitivity parameter in a small range;

[0011] Based on the adjustment parameter, the current sensitivity is adjusted to obtain an adjusted sensitivity.

[0012] Further, when the current sensitivity parameter is greater than the preset threshold, calculating an adjustment parameter according to a preset first formula includes:

[0013] When the current sensitivity parameter is greater than 1000, an adjustment parameter is calculated according to a preset first formula;

[0014] ISO

[0015] The first formula is: ΔISO = e K , where ΔISO is the adjustment parameter, ISO is the current sensitivity, e is the natural exponent, and K is the constant 100.

[0016] Further, when the current sensitivity parameter is less than the preset threshold, calculating an adjustment parameter according to a preset first formula includes:

[0017] When the current sensitivity parameter is less than or equal to 1000, an adjustment parameter is calculated according to a preset second formula;

[0018] The first formula is: ΔISO = ISO × 1000, where ΔISO is the adjustment parameter and ISO is the current sensitivity.

[0019] Further, the method further includes:

[0020] Real-time acquisition of image data of the camera during live broadcast;

[0021] Identifying the relationship between elements in the image data;

[0022] According to the recognition result between the elements, analyzing the scene mode corresponding to the elements in the image data;

[0023] According to the scene mode, feature segmentation is performed on the target features and non-target features in the scene mode;

[0024] Based on the result of feature segmentation, the sensitivity of the target feature and the sensitivity of the non-target feature are respectively adjusted.

[0025] Further, adjusting the sensitivity of the target feature and the sensitivity of the non-target feature includes:

[0026] Analyze the scene mode. When the scene mode is the portrait mode;

[0027] Obtain the brightness information of the current scene and determine whether the brightness information is within a preset brightness value range;

[0028] If the brightness information is within the preset brightness value range, reduce the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the reduced target sensitivity is lower than the sensitivity of the non-target feature;

[0029] If the brightness information is not within the preset brightness value range, increase the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the target feature is the portrait and the non-target feature is the background.

[0030] Further, adjusting the sensitivity of the target feature and the sensitivity of the non-target feature includes:

[0031] Analyze the scene mode. When the scene mode is a specific object;

[0032] Obtain the brightness information of the current scene and determine whether the brightness information is within a preset brightness value range;

[0033] If the brightness information is within the preset brightness value range, reduce the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the reduced target sensitivity is lower than the sensitivity of the non-target feature;

[0034] If the brightness information is not within the preset brightness value range, increase the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the target feature is the specific object and the non-target feature is the person and / or the background.

[0035] Further, analyzing the scene mode corresponding to the elements in the image data includes:

[0036] Continuously analyze the elements, input the elements into a real-time object detection model to obtain the behavior information of the elements;

[0037] By comparing the behavior information of the elements, determine whether the behavior information of the elements meets the conditions of a specific object;

[0038] If the appearance duration of the element meets the preset time and the movement trajectory of the element is within the preset range, it is determined that the behavior information of the element meets the conditions of a specific object;

[0039] Based on the determination result, analyze that the scene pattern corresponding to the elements in the image data is a specific object.

[0040] An embodiment of the present application also provides a live camera sensitivity adjustment system, including:

[0041] A first acquisition module, configured to acquire a sensitivity adjustment instruction during live broadcast;

[0042] A second acquisition module, configured to acquire a current sensitivity parameter according to the sensitivity adjustment instruction;

[0043] A first calculation module, configured to, when the current sensitivity parameter is greater than a preset threshold, calculate an adjustment parameter according to a preset first formula for adjusting the sensitivity parameter in a large range;

[0044] A second calculation module, configured to, when the current sensitivity parameter is less than a preset threshold, calculate an adjustment parameter according to a preset second formula for adjusting the sensitivity parameter in a small range;

[0045] An adjustment module, configured to adjust the current sensitivity based on the adjustment parameter to obtain an adjusted sensitivity.

[0046] The present application also provides a computer device, including a memory and a processor, where the memory stores a computer program, and the processor implements the steps of the method described in any one of the above when executing the computer program.

[0047] The present application also provides a computer-readable storage medium, on which a computer program is stored, and the computer program implements the steps of the method described in any one of the above when executed by a processor.

[0048] The live camera sensitivity adjustment method and system according to the embodiment of the present application can dynamically adjust according to different lighting conditions by acquiring a sensitivity adjustment instruction and a current sensitivity parameter to meet the requirements in different scenarios. This can avoid the situation of the picture being too bright or too dark and improve the viewing experience. By precisely adjusting the sensitivity parameter, the picture quality problem caused by over-adjustment or under-adjustment can be avoided. The adjusted sensitivity can make the picture clearer, brighter, and more detailed, enhancing the visual experience of the audience; the first formula is used for large-range adjustment to quickly increase or decrease the sensitivity parameter; the second formula is used for small-range fine-tuning to finely adjust the sensitivity parameter; by using different adjustment formulas, it can better meet the non-linear perception characteristics of the human eye for brightness and the adjustment needs under different lighting conditions; according to the non-linear perception characteristics of the human eye for brightness, the sensitivity is adjusted in a segmented manner, which better conforms to the visual perception of the human eye and improves the viewing experience. Description of the Drawings

[0049] Figure 1 A schematic flowchart of a method for adjusting the camera sensitivity based on live broadcast according to an embodiment of the present application;

[0050] Figure 2 A schematic flowchart of a method for adjusting the camera sensitivity based on live broadcast according to an embodiment of the present application;

[0051] Figure 3 A schematic block diagram of the structure of a system for adjusting the camera sensitivity based on live broadcast according to an embodiment of the present application;

[0052] Figure 4 A schematic block diagram of the structure of a computer device according to an embodiment of the present application.

[0053] The realization, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. Detailed implementation manners

[0054] In order to make the purpose, technical solutions and advantages of the present application more clear and understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0055] Referring to Figure 1 , an embodiment of the present application provides a method for adjusting the camera sensitivity based on live broadcast, including a heat exchanger, and the method includes:

[0056] S1. Obtain a sensitivity adjustment instruction during live broadcast;

[0057] S2. Obtain the current sensitivity parameter according to the sensitivity adjustment instruction;

[0058] S3. When the current sensitivity parameter is greater than a preset threshold, calculate an adjustment parameter according to a preset first formula for adjusting the sensitivity parameter in a large range;

[0059] S4. When the current sensitivity parameter is less than a preset threshold, calculate an adjustment parameter according to a preset second formula for adjusting the sensitivity parameter in a small range;

[0060] S5. Adjust the current sensitivity based on the adjustment parameter to obtain the adjusted sensitivity.

[0061] As described in the above steps S1 - S2, it is achieved by obtaining the sensitivity adjustment instructions sent by the live camera, such as instructions transmitted through the network or set by the local control panel. These instructions can include the sensitivity adjustment value or adjustment direction (increase or decrease the sensitivity) desired by the user for subsequent corresponding processing; by obtaining the user's sensitivity adjustment instructions, dynamic adjustment of the camera sensitivity can be realized to adapt to different lighting conditions and the requirements of specific scenarios. This dynamic adjustment can improve the picture quality and viewing experience and avoid over - bright or over - dark situations. It is achieved by reading the current sensitivity parameters of the camera, such as parameters like ISO value or shutter speed. These parameters are usually stored in the chip inside the camera and can be obtained by reading the chip register or the sensor output signal. By obtaining the current sensitivity parameters, precise adjustment can be carried out in subsequent steps to meet the requirements in different scenarios. At the same time, it can avoid inaccurate adjustment caused by problems such as error accumulation or sensor offset.

[0062] As described in the above steps S3 - S4, different adjustment formulas are selected for calculation according to the size of the current sensitivity parameter and the preset threshold. Among them, the first formula is used for large - range adjustment to quickly increase or decrease the sensitivity parameter; the second formula is used for small - range fine - tuning to finely adjust the sensitivity parameter; by using different adjustment formulas, it can better meet the non - linear perception characteristics of the human eye for brightness and the adjustment needs under different lighting conditions. By adopting this segmented adjustment method, problems of over - adjustment and under - adjustment can be avoided, and the picture quality and viewing experience can be improved.

[0063] As described in the above step S5, addition or subtraction operations are performed on the current sensitivity according to the calculated adjustment parameter to obtain the adjusted sensitivity value. For example, in large - range adjustment, the current sensitivity value can be multiplied by an adjustment factor to obtain the adjusted sensitivity value; in small - range fine - tuning, a fine - tuning amount can be directly added or subtracted to obtain the fine - tuned sensitivity value. By precisely adjusting the current sensitivity based on the calculated adjustment parameter, the picture quality and viewing experience can be improved, and at the same time, inaccurate sensitivity adjustment caused by problems such as sensor noise can be avoided.

[0064] Specifically, by obtaining the sensitivity adjustment instruction and the current sensitivity parameter, dynamic adjustment can be performed according to different lighting conditions to meet the requirements in different scenarios. This can avoid the situation of the picture being too bright or too dark and improve the viewing experience. By precisely adjusting the sensitivity parameter, picture quality problems caused by over-adjustment or under-adjustment can be avoided. The adjusted sensitivity can make the picture clearer, brighter, and more detailed, enhancing the visual experience of the audience; the first formula is used for large-range adjustment to quickly increase or decrease the sensitivity parameter; the second formula is used for small-range fine-tuning to finely adjust the sensitivity parameter; by using different adjustment formulas, the non-linear perception characteristics of the human eye to brightness and the adjustment needs under different lighting conditions can be better met; according to the non-linear perception characteristics of the human eye to brightness, the sensitivity is adjusted in a segmented manner, which better conforms to the visual perception of the human eye and improves the viewing experience.

[0065] In a feasible embodiment, assume that the current sensitivity parameter range of the camera is 0 - 100, where 50 is the middle value representing normal sensitivity. By adjusting the sensitivity parameter, we hope to obtain a suitable picture brightness under different lighting conditions. Obtain the sensitivity adjustment instruction during live broadcast: Through network transmission or local control panel settings, obtain the sensitivity adjustment instruction expected by the user. For example, the user expects to adjust the sensitivity to a brighter picture. Read the current sensitivity parameter of the camera. Assume the current sensitivity parameter is 60. When the current sensitivity parameter is greater than the preset threshold (50), calculate the adjustment parameter according to the preset first formula. Assume the calculation result is 5. Adjust the current sensitivity to obtain the adjusted sensitivity: Adjusted sensitivity = Current sensitivity parameter + Adjustment parameter. The calculation result is 65. If adjusting to a darker picture, then Adjusted sensitivity = Current sensitivity parameter - Adjustment parameter. The calculation result is 55. In this example, by obtaining the user's sensitivity adjustment instruction and the current sensitivity parameter, and performing calculations and adjustments according to the preset adjustment formula, the adjusted sensitivity value is obtained. In this way, the sensitivity parameter can be dynamically adjusted according to the user's needs and changes in lighting conditions, improving the picture quality and viewing experience.

[0066] In one embodiment, when the current sensitivity parameter is greater than the preset threshold, then calculating the adjustment parameter according to the preset first formula includes:

[0067] When the current sensitivity parameter is greater than 1000, then calculate the adjustment parameter according to the preset first formula;

[0068] ISO

[0069] The first formula is: ΔISO = e K , where ΔISO is the adjustment parameter, ISO is the current sensitivity, e is the natural exponent, and K is the constant 100.

[0070] As described in the above steps, under strong light illumination, the human eye will automatically adjust the size of the pupil to reduce the amount of light entering the eye, thereby protecting the retina from damage caused by overexposure. Similarly, by adjusting the sensitivity of the camera to automatically adapt to strong light environments, the visibility of the image can be improved, and brightness distortion and detail loss caused by overexposure can be avoided. In a darker environment, the pupil of the human eye will dilate to receive more light, thereby increasing the perceived brightness of vision. Similarly, by adjusting the sensitivity of the camera, its exposure can be increased in low light environments, enhancing the visibility of the image and making details more clearly distinguishable. The human eye's perception of brightness is non-linear, that is, changes in brightness have different effects on scenes with different brightness levels. By using the first formula to calculate the sensitivity adjustment over a large range, the image brightness can be increased more quickly in brighter scenes, allowing the viewer to see details more clearly and enhancing the viewing experience. By combining the visual perception of the human eye and adjusting the sensitivity over a large range using the first formula when it is greater than the preset threshold of 1000, the visual requirements under different lighting conditions can be adapted, improving the image quality and viewing experience. This adjustment strategy can simulate the adaptive ability of the human eye, enabling the camera to better adapt to various environments and present more real and clear image information to the user.

[0071] In one embodiment, when the current sensitivity parameter is less than the preset threshold, the adjustment parameter is calculated according to the preset first formula, including:

[0072] When the current sensitivity parameter is less than or equal to 1000, the adjustment parameter is calculated according to the preset second formula;

[0073] The first formula is: ΔISO = ISO × 1000, where ΔISO is the adjustment parameter and ISO is the current sensitivity.

[0074] Refer to Figure 2 , in one embodiment, the method further includes:

[0075] S61. Real-time obtain the image data of the camera during live broadcast;

[0076] S62. Identify the relationship between elements in the image data;

[0077] S63. Analyze the scene mode corresponding to the elements in the image data according to the recognition result between the elements;

[0078] S64. Perform feature segmentation on the target features and non-target features in the scene mode according to the scene mode;

[0079] S65. Based on the result of feature segmentation, adjust the sensitivity of the target feature and the sensitivity of the non-target feature respectively.

[0080] As described in the above steps, obtain the camera image data in real time through a camera or other devices; obtain the real-time image data as the input for the subsequent steps, which is used for element recognition and scene mode analysis to achieve dynamic adjustment of the sensitivity. Using computer vision technologies, such as algorithms like object detection and image segmentation, identify and locate the elements in the image data, and determine the relationships between the elements; by identifying the elements and their relationships, important target and background information in the image can be extracted, providing a basis for scene mode analysis and feature segmentation in the subsequent steps. According to information such as the category, position, and relationship of the elements, use algorithms to analyze and judge the scene mode of the image data, such as indoor, outdoor, people, scenery, etc.; through the analysis of the scene mode, more accurate and personalized sensitivity adjustment strategies can be provided according to different scene characteristics and requirements, so that the picture maintains the best brightness and visibility in different scenes. According to the identified scene mode, use an image segmentation algorithm to segment the image data into regions of target features and non-target features; through feature segmentation, the sensitivity can be adjusted specifically for different target features and non-target features, making the target features clearer and brighter, while the non-target features relatively reduce the sensitivity to avoid overexposure or loss of details. By identifying and analyzing the elements, scene features, and target features and non-target features in the image, dynamic sensitivity adjustment for different situations is achieved. The execution of these steps can improve the image quality, enhance the visibility of the target features, and thus improve the viewing experience and user satisfaction.

[0081] In one embodiment, the adjustment of the sensitivity of the target feature and the sensitivity of the non-target feature includes:

[0082] Analyze the scene mode. When the scene mode is the portrait mode;

[0083] Obtain the brightness information of the current scene, and judge whether the brightness information is within the preset brightness value range;

[0084] If the brightness information is within the preset brightness value range, reduce the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the reduced target sensitivity is lower than the sensitivity of the non-target feature;

[0085] If the brightness information is not within the preset brightness value range, increase the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the target feature is the portrait and the non-target feature is the background.

[0086] As described above, the scene pattern analysis algorithm is used to determine whether the current scene is a portrait mode. Determine the scene where the sensitivity of the target feature and the non-target feature needs to be adjusted currently. Use a light sensor or other means to obtain the brightness information of the current scene, and compare it with the preset brightness value range; determine whether the current brightness is suitable for sensitivity adjustment, so as to select the corresponding adjustment strategy according to the situation. According to the judgment of the scene pattern and the brightness information, use a control algorithm to reduce the sensitivity of the target feature and the non-target feature; when the brightness is appropriate, reducing the sensitivity of the target feature can reduce the overexposure phenomenon of the portrait in the image, and at the same time reducing the sensitivity of the non-target feature can maintain the details of the background. According to the judgment of the scene pattern and the brightness information, use a control algorithm to increase the sensitivity of the target feature and reduce the sensitivity of the non-target feature; when the brightness is relatively dark, increasing the sensitivity of the target feature can improve the visibility of the portrait, and at the same time reducing the sensitivity of the non-target feature can reduce background noise. According to the judgment of the scene pattern and the brightness information, the image quality can be optimized by adjusting the sensitivity of the target feature (portrait) and the non-target feature (background), improving the visibility of the portrait, and reducing problems such as overexposure or background noise.

[0087] In one embodiment, adjusting the sensitivity of the target feature and the non-target feature includes:

[0088] Analyze the scene pattern. When the scene pattern is a specific object;

[0089] Obtain the brightness information of the current scene, and determine whether the brightness information is within the preset brightness value range;

[0090] If the brightness information is within the preset brightness value range, reduce the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the reduced target sensitivity is lower than the sensitivity of the non-target feature;

[0091] If the brightness information is not within the preset brightness value range, increase the sensitivity of the target feature and reduce the sensitivity of the non-target feature, where the target feature is a specific object and the non-target feature is a person and / or background.

[0092] As described above, the scene pattern analysis algorithm is used to determine whether the current scene is a specific object pattern; determine the scene where the sensitivity of the target feature and the non-target feature needs to be adjusted currently. The brightness information of the current scene is obtained by using a light sensor or other means and compared with the preset brightness value range; it is judged whether the current brightness is suitable for sensitivity adjustment, so as to select the corresponding adjustment strategy according to the situation. According to the judgment of the scene pattern and the brightness information, a control algorithm is used to reduce the sensitivity of the target feature and the non-target feature; when the brightness is appropriate, reducing the sensitivity of the target feature can reduce the overexposure phenomenon of specific objects in the image, and at the same time reducing the sensitivity of the non-target feature can keep the details of the background and people. According to the judgment of the scene pattern and the brightness information, a control algorithm is used to increase the sensitivity of the target feature and reduce the sensitivity of the non-target feature; when the brightness is relatively dark, increasing the sensitivity of the target feature can improve the visibility of specific objects, and at the same time reducing the sensitivity of the non-target feature can reduce the noise of the background and people. The emphasis is on specific objects.

[0093] In one embodiment, analyzing the scene pattern corresponding to the element in the image data includes:

[0094] Continuously analyze the element, and input the element into a real-time target detection model to obtain the behavior information of the element;

[0095] By comparing the behavior information of the element, judge whether the behavior information of the element meets the conditions of a specific object;

[0096] If the duration of the appearance of the element meets the preset time and the movement trajectory of the element is within the preset range, it is determined that the behavior information of the element meets the conditions of a specific object;

[0097] Based on the determined result, analyze that the scene pattern corresponding to the element in the image data is a specific object.

[0098] As described above, by continuously analyzing the elements in the image data, a real-time object detection model can be used to detect and track the elements. This model can identify different objects in the image and extract their feature and behavior information. Through the real-time object detection model, behavior information such as the position, speed, and movement trajectory of the elements can be obtained. This information can be used for further analysis of the behavior patterns of the elements. According to the behavior patterns of specific objects, corresponding conditions can be set. For example, a specific object may need to appear in the image for a certain period of time, and the movement trajectory is within a certain preset range, etc. By comparing the behavior information of the elements with these conditions, it can be determined whether the behavior of the elements meets the conditions of the specific object. If the behavior information of the elements meets the conditions of the specific object, including the duration and the movement trajectory range, etc., then it can be determined that the scene pattern corresponding to the elements in the current image data is the specific object. The advantage of judging the scene through continuous element analysis is that it can effectively deal with the situation where the prior art cannot automatically identify a specific scene, thereby achieving more efficient and accurate scene recognition and processing. Although there are already many scene recognition algorithms based on technologies such as computer vision and deep learning, their accuracy and robustness still have certain limitations. For example, in some specific scenarios, due to the influence of factors such as elements or light in the scene, traditional scene recognition algorithms may make misjudgments or omissions, resulting in the degradation of system performance. By judging the scene through continuous element analysis, the advantages of technologies such as object detection and tracking can be fully utilized, and each element in the scene can be analyzed and judged, so as to more accurately identify the scene. In addition, continuous element analysis can also effectively avoid the occurrence of misclassification and omission phenomena, and improve the accuracy and robustness of scene recognition. From the above analysis, judging the scene through continuous element analysis can make up for the deficiencies of traditional scene recognition algorithms, improve the accuracy and robustness of scene recognition, and bring more efficient and accurate scene processing effects.

[0099] In a specific example, assume that we want to determine whether a person in an image is performing a specific motion, such as rope skipping. Existing scene recognition technologies may not be able to accurately identify such a specific scene because the actions of rope skipping may be similar to other similar actions, making it difficult for traditional algorithms to distinguish; in this case, it is very useful to judge the scene through element continuous analysis. We can adopt the following steps: Object detection and tracking: Use a real-time object detection model to identify the people in the image and track their positions and movement trajectories. Action recognition: For the detected people, we can use action recognition algorithms to determine whether their current action is rope skipping. This can be achieved by training a deep learning model, inputting a series of action samples related to rope skipping, and performing action classification and recognition. Judgment conditions: Define a set of judgment conditions, such as the action of rope skipping needs to last for a certain period of time, frequency, or meet a specific action pattern. These conditions can be adjusted according to actual needs. Element continuous analysis: By analyzing the movement trajectories and action characteristics of the people, combined with the judgment conditions, continuously judge whether the people meet the conditions for rope skipping. If the conditions are met, it can be determined that the people in the current image are performing the rope skipping action. In this way, through the method of element continuous analysis, we can more accurately determine whether the people in the image are rope skipping, avoiding the possible misjudgment or missed judgment of traditional algorithms. This method can be applied to various scene recognition tasks, improving the recognition accuracy and robustness.

[0100] The live-based camera sensitivity adjustment method of the present application can dynamically adjust according to different lighting conditions by obtaining a sensitivity adjustment instruction and the current sensitivity parameter to meet the requirements in different scenarios. This can avoid the situation of the picture being too bright or too dark and improve the viewing experience. By precisely adjusting the sensitivity parameter, it is possible to avoid picture quality problems caused by over-adjustment or under-adjustment. The adjusted sensitivity can make the picture clearer, brighter, and more detailed, enhancing the visual experience of the audience; the first formula is used for large-range adjustment to quickly increase or decrease the sensitivity parameter; the second formula is used for small-range fine-tuning to finely adjust the sensitivity parameter; by using different adjustment formulas, it is possible to better meet the non-linear perception characteristics of the human eye to brightness and the adjustment needs under different lighting conditions; according to the non-linear perception characteristics of the human eye to brightness, a segmented adjustment method is adopted for sensitivity adjustment, which better conforms to the visual perception of the human eye and improves the viewing experience.

[0101] Refer to Figure 3 , an embodiment of the present application also provides a live-based camera sensitivity adjustment system, including:

[0102] The first acquisition module 1 is used to acquire the sensitivity adjustment instruction during the live broadcast;

[0103] The second acquisition module 2 is configured to acquire the current sensitivity parameter according to the sensitivity adjustment instruction;

[0104] The first calculation module 3 is configured to, when the current sensitivity parameter is greater than a preset threshold, calculate an adjustment parameter according to a preset first formula for adjusting the sensitivity parameter in a large range;

[0105] The second calculation module 4 is configured to, when the current sensitivity parameter is less than a preset threshold, calculate an adjustment parameter according to a preset second formula for adjusting the sensitivity parameter in a small range;

[0106] The adjustment module 5 is configured to adjust the current sensitivity based on the adjustment parameter to obtain an adjusted sensitivity.

[0107] As described above, it can be understood that each component of the live broadcast-based camera sensitivity adjustment system proposed in this application can implement the functions of any one of the above-described live broadcast-based camera sensitivity adjustment methods, and the specific structure will not be elaborated.

[0108] Refer to Figure 4 , this application embodiment also provides a computer device, which may be a server, and its internal structure may be as Figure 4 shown. The computer device includes a processor, a memory, a network interface, and a database connected through a system bus. Among them, the processor of the computer design is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data such as monitoring data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, it implements a live broadcast-based camera sensitivity adjustment method.

[0109] The above processor executes the above-described live broadcast-based camera sensitivity adjustment method, including: obtaining a live broadcast sensitivity adjustment instruction; obtaining a current sensitivity parameter according to the sensitivity adjustment instruction; when the current sensitivity parameter is greater than a preset threshold, calculating an adjustment parameter according to a preset first formula for adjusting the sensitivity parameter in a large range; when the current sensitivity parameter is less than a preset threshold, calculating an adjustment parameter according to a preset second formula for adjusting the sensitivity parameter in a small range; adjusting the current sensitivity based on the adjustment parameter to obtain an adjusted sensitivity.

[0110] The above-described live broadcast-based camera sensitivity adjustment method can dynamically adjust according to different lighting conditions by obtaining a sensitivity adjustment instruction and the current sensitivity parameter, so as to meet the requirements in different scenarios. This can avoid the situation of the picture being too bright or too dark and improve the viewing experience. By precisely adjusting the sensitivity parameter, picture quality problems caused by over-adjustment or under-adjustment can be avoided. The adjusted sensitivity can make the picture clearer, brighter, and richer in details, enhancing the visual experience of the audience; the first formula is used for large-range adjustment to quickly increase or decrease the sensitivity parameter; the second formula is used for small-range fine-tuning to finely adjust the sensitivity parameter; by using different adjustment formulas, it can better meet the non-linear perception characteristics of the human eye for brightness and the adjustment needs under different lighting conditions; according to the non-linear perception characteristics of the human eye for brightness, a segmented adjustment method is adopted for sensitivity adjustment, which better conforms to the visual perception of the human eye and improves the viewing experience.

[0111] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements a live broadcast-based camera sensitivity adjustment method, including the steps of: obtaining a sensitivity adjustment instruction during a live broadcast; obtaining the current sensitivity parameter according to the sensitivity adjustment instruction; when the current sensitivity parameter is greater than a preset threshold, calculating an adjustment parameter according to a preset first formula for large-range adjustment of the sensitivity parameter; when the current sensitivity parameter is less than the preset threshold, calculating an adjustment parameter according to a preset second formula for small-range adjustment of the sensitivity parameter; and adjusting the current sensitivity based on the adjustment parameter to obtain an adjusted sensitivity.

[0112] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, storage, database, or other medium provided in the present application and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memories can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memories can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and Rambus dynamic RAM (RDRAM), etc.

[0113] It should be noted that in this document, the terms "include", "comprise", or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that includes a series of elements includes not only those elements but also other elements not expressly listed, or elements that are inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the statement "including one..." does not exclude the existence of additional identical elements in the process, apparatus, article, or method that includes such element.

[0114] The above are only the preferred embodiments of the present application, and do not limit the patent scope of the present application. Any equivalent structural or equivalent process transformation made by using the specification and drawings of the present application, or directly or indirectly applied in other related technical fields, shall be equally included in the patent protection scope of the present application.

Claims

1. A method for adjusting camera sensitivity based on live broadcast, characterized in that: The method comprises: Get the sensitivity adjustment command during live broadcast; According to the sensitivity adjustment instruction, obtaining the current sensitivity parameter; When the current sensitivity parameter is greater than a preset threshold, an adjustment parameter is calculated according to a preset first formula to adjust the sensitivity parameter over a large range; When the current sensitivity parameter is less than a preset threshold, an adjustment parameter is calculated according to a preset second formula to adjust the sensitivity parameter in a small range; Based on the adjustment parameter, adjusting the current sensitivity to obtain an adjusted sensitivity; When the current sensitivity parameter is greater than a preset threshold, the adjustment parameter is calculated according to a preset first formula, including: when the current sensitivity parameter is greater than 1000, the adjustment parameter is calculated according to a preset first formula; the first formula is: ,in To adjust the parameters, is the current sensitivity, is the natural index, is a constant of 100; under strong light, the human eye will automatically adjust the size of the pupil to reduce the amount of light entering the eye. By adjusting the sensitivity of the camera, it can automatically adjust to the strong light environment to improve the visibility of the picture and avoid brightness distortion and detail loss caused by overexposure; in a darker environment, the pupil of the human eye will expand to receive more light, thereby increasing the brightness of visual perception; similarly, by adjusting the sensitivity of the camera, the exposure in a low-light environment is increased, the visibility of the picture is enhanced, and the details are clearer and more discernible; the human eye's perception of brightness is nonlinear, that is, changes in brightness have different effects on scenes of different brightness levels; the first formula is used to adjust the sensitivity over a large range when it is greater than the preset threshold of 1000 to adapt to the visual needs under different lighting conditions. This adjustment strategy simulates the adaptive ability of the human eye; When the current sensitivity parameter is less than a preset threshold, the adjustment parameter is calculated according to a preset second formula, including: When the current sensitivity parameter is less than or equal to 1000, the adjustment parameter is calculated according to a preset second formula; The second formula is: ,in To adjust the parameters, is the current sensitivity.

2. The method for adjusting the camera sensitivity based on live broadcast according to claim 1, characterized in that: The method further comprises: Acquire the image data of the live broadcast in real time; Identifying the relationship between elements in the image data; Analyzing the scene mode corresponding to the element in the image data according to the recognition result between the element and the element; According to the scene mode, performing feature segmentation on target features and non-target features in the scene mode; Based on the result of feature segmentation, the sensitivity of the target feature and the sensitivity of the non-target feature are adjusted respectively.

3. The method for adjusting the camera sensitivity based on live broadcast according to claim 2, characterized in that: The step of adjusting the sensitivity of the target feature and the sensitivity of the non-target feature comprises: analyzing the scene mode, when the scene mode is a portrait mode; Obtaining brightness information of the current scene, and determining whether the brightness information is within a preset brightness value range; If the brightness information is within a preset brightness value range, the sensitivity of the target feature is reduced, and the sensitivity of the non-target feature is reduced, wherein the reduced target sensitivity is lower than the sensitivity of the non-target feature; If the brightness information is not within a preset brightness value range, the sensitivity of the target feature is increased, and the sensitivity of the non-target feature is decreased, wherein the target feature is a portrait, and the non-target feature is a background.

4. The method for adjusting the camera sensitivity based on live broadcast according to claim 2, characterized in that: The step of adjusting the sensitivity of the target feature and the sensitivity of the non-target feature comprises: analyzing the scene mode, when the scene mode is a specific object; Obtaining brightness information of the current scene, and determining whether the brightness information is within a preset brightness value range; If the brightness information is within a preset brightness value range, the sensitivity of the target feature is reduced, and the sensitivity of the non-target feature is reduced, wherein the reduced target sensitivity is lower than the sensitivity of the non-target feature; If the brightness information is not within a preset brightness value range, the sensitivity of the target feature is increased and the sensitivity of the non-target feature is decreased, wherein the target feature is a specific object and the non-target feature is a person and / or background.

5. The method for adjusting the camera sensitivity based on live broadcast according to claim 4, characterized in that: The analyzing the scene mode corresponding to the element in the image data includes: Continuously analyzing the elements, and inputting the elements into a real-time target detection model to obtain behavior information of the elements; By comparing the behavior information of the element, determining whether the behavior information of the element meets the condition of the specific object; If the appearance duration of the element meets the preset time and the movement track of the element is within the preset range, it is determined that the behavior information of the element meets the condition of the specific object; Based on the determination result, the scene mode corresponding to the element in the image data is analyzed as a specific object.

6. A camera sensitivity adjustment system based on live broadcast, used in the method according to any one of claims 1 to 5, characterized in that: include: The first acquisition module is used to obtain the sensitivity adjustment instruction during live broadcast; A second acquisition module, used to acquire current sensitivity parameters according to the sensitivity adjustment instruction; A first calculation module, used for calculating an adjustment parameter according to a preset first formula when the current sensitivity parameter is greater than a preset threshold value, so as to adjust the sensitivity parameter over a large range; A second calculation module, for calculating an adjustment parameter according to a preset second formula when the current sensitivity parameter is less than a preset threshold value, so as to adjust the sensitivity parameter in a small range; The adjustment module is used to adjust the current sensitivity based on the adjustment parameter to obtain an adjusted sensitivity.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images