Flexible device display system and method based on dynamic screen morphology adaptation

Abstract

The application discloses a flexible device display system and method based on dynamic screen morphology adaptation, and relates to the technical field of flexible display. The system comprises a perception module for monitoring the multi-modal data of a dynamic screen in real time, a processing module for constructing a resolution conversion model and calculating a logical resolution, a rendering selection module for pre-rendering analysis of the dynamic screen, and a feedback module for responding to the pre-rendering signal sent by the rendering selection module, wherein the rendering selection module comprises a resolution prediction unit, a content recognition unit, a content classification unit and a dynamic reconstruction unit. The perception module realizes full-dimension real-time monitoring of multi-morphology changes such as curling and stretching. The logical resolution generation adjusts the resolution according to the curling length, informs the system to render only the visible area, thereby saving resources. Through pre-rendering, rendering resources are allocated in advance, the rendering preparation time is reduced, the application interface layout is adjusted in real time according to the screen morphology, and the user's interactive experience is improved.

Description

Technical Field

[0001] This invention relates to the field of flexible display technology, and in particular to a flexible device display system and method based on dynamic screen shape adaptation. Background Technology

[0002] With the rapid development of flexible display technology, display devices with dynamic shape change capabilities are gradually entering the market, but existing operating systems and application ecosystems cannot be well adapted to them.

[0003] This invention aims to overcome the compatibility challenges between multi-form screens and traditional operating systems and various applications through deep collaborative optimization of software and hardware, bringing users a smoother and more efficient user experience and promoting the widespread application and popularization of flexible devices. On rollable devices, when the screen changes shape (unfolding / rolling), the application interface layout cannot adjust in real time according to the screen shape, often resulting in problems such as misaligned, overlapping, and incomplete display of interface elements, making it difficult for users to use application functions normally. During screen shape switching, the display resolution changes dynamically. However, existing applications lack an effective response mechanism for dynamic resolution changes. Taking video playback as an example, when the screen shape changes, the video playback ratio may become unbalanced, resulting in image stretching and distortion, severely damaging the user's interactive experience.

[0004] Therefore, a flexible display system and method based on dynamic screen shape adaptation is needed to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a cross-platform flexible display system and method based on dynamic screen shape adaptation, so as to solve the problems mentioned in the background art.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: This flexible device display system is based on dynamic screen shape adaptation. The system includes a sensing module, a processing module, a rendering selection module, and a feedback module. The sensing module is used to monitor the multimodal data of the dynamic screen in real time when the user triggers a curling action on the flexible device, and feed it back to the processing module. The multimodal data includes physical shape data and display data. The processing module is used to construct a resolution conversion model based on the multimodal data fed back by the perception module, calculate the logical resolution, and transmit the logical resolution to the rendering selection module. The rendering selection module includes a resolution prediction unit, a content recognition unit, a content classification unit, and a dynamic reconstruction unit. The resolution prediction unit is used to predict the actual resolution of future frames; The content recognition unit is used to analyze the interface elements of the dynamic screen in real time, identify the interface elements of the dynamic screen, and assign weights to the interface elements. The content classification unit is used to calculate the priority index of interface elements based on the content recognition results and to classify the interface elements by priority. The dynamic reconstruction unit is used to pre-render elements based on the actual resolution and content classification results of future frames, and send the pre-render signal to the feedback module.

[0007] The feedback module is used to respond to the pre-rendering signal sent by the rendering selection module and pre-render the dynamic screen.

[0008] As a further preferred embodiment of the present invention, the sensing module includes a data acquisition unit and a synchronization unit: The acquisition unit monitors the physical shape data of the dynamic screen in real time through a laser rangefinder, an accelerometer, a flexible curvature sensor, and a distributed fiber optic sensor. The physical shape data includes the real-time curling length, curling speed, and curling acceleration of the dynamic screen. The unit also extracts the display data of the dynamic screen from the working log of the flexible device. The display data includes the actual resolution of the dynamic screen and the real-time pixel count distributed on the dynamic screen. The synchronization unit periodically calibrates the clock offset of the sensors using the GPS clock and PTP protocol, compensates for the reverse offset of the timestamps of the adaptation parameters using the fixed transmission delay of each sensor, and synchronizes the adaptation parameters according to the timestamps.

[0009] It should be explained that the core objective of the synchronization unit is to solve the problem of spatiotemporal data inconsistency caused by multiple sensors due to hardware differences, different data acquisition frequencies, or uneven physical location distribution. Differences in hardware sampling frequencies, signal transmission paths (such as wired / wireless transmission delays), or clock drift between different sensors can lead to millisecond-level time deviations in data collected at the same moment, directly affecting the accuracy of subsequent fusion. A high-precision hardware clock (such as a GPS clock) is designated as the master clock, and the built-in clocks of other sensors serve as slave clocks. The slave clock offsets are periodically calibrated using the IEEE 1588 Precision Time Protocol (PTP) to ensure a unified global time reference.

[0010] As a further preferred embodiment of the present invention, the processing module includes: A resolution conversion model is constructed, taking multimodal data as input and outputting logical resolution. The specific formula for the resolution conversion model is as follows: ; ; in, This indicates the deformation ratio of the dynamic screen. This indicates the real-time scroll length of the dynamic screen. This indicates the length of the scroll when the dynamic screen is fully scrolled. Indicates the logical resolution of the dynamic screen. This represents the highest resolution when the dynamic screen is fully unfolded, and k represents the characteristic coefficient of the flexible device.

[0011] It's important to explain that logical resolution acts as a bridge between the physical screen shape and software display adaptation. When the screen is rolled up, the actual display area changes, but the operating system typically relies on fixed resolution parameters. The role of logical resolution generation is to dynamically adjust these parameters, enabling the system to correctly render the UI and avoid display errors. When part of the screen is rolled up, the effective display area shrinks, and logical resolution generation adjusts the resolution according to the rollover length, instructing the system to render only the visible area, thus saving resources. Simultaneously, it maintains application compatibility, as applications still adhere to the logical resolution layout and are unaware of the physical changes.

[0012] As a further preferred embodiment of the present invention, the resolution prediction unit includes: An LSTM prediction model based on a spatiotemporal attention mechanism is constructed. The input layer receives physical morphology data and logical resolution sequences for a specified number of consecutive frames, and the output layer predicts the probability distribution of the actual resolution of future frames. The specific formula is as follows: ; in, This represents the probability matrix of the actual resolution of future frames, where M represents the total number of specified frames. This represents the time decay factor of the i-th frame. This represents the scrolling speed of the i-th frame of the dynamic screen. This represents the curl acceleration of an i-frame dynamic screen. This represents the logical resolution of the i-frame dynamic screen.

[0013] As a further preferred embodiment of the present invention, the content recognition unit includes: This method is used for real-time analysis of interface elements in dynamic screens, identifying these elements, which include video regions, text line candidate boxes, and original saliency maps. It captures the motion trajectory of dynamic images using inter-frame difference and optical flow analysis to detect video regions in the dynamic screen; obtains text line candidate boxes through text detection; and outputs the original saliency map using a lightweight DeepLabv3+ model with the screen image as input. The steps for detecting video regions include: Acquire two consecutive frames of images and calculate the pixel difference value: ; in, Indicates two consecutive frames and The difference value of the image at pixel (x,y). Representing a frame pixels, Representing a frame The pixels; When pixel difference When the difference exceeds the pixel difference threshold, the pixel (x,y) is marked as a pixel within the candidate dynamic region. The motion vectors of candidate dynamic region pixels are extracted using the Lucas-Kanade algorithm. The modulus of screening speed || The region that exceeds the specified motion threshold is marked as a video region. The text line candidate boxes are obtained by loading the EAST model with OpenCV, preprocessing the input image, performing forward propagation, and using non-maximum suppression to obtain the text line candidate boxes. The original saliency map is obtained by loading a pre-trained DeepLabv3+ model based on MobileNetV3, scaling and normalizing the image, inputting it into the model, performing forward propagation, and outputting the semantic segmentation result as the original saliency map.

[0014] As a further preferred embodiment of the present invention, the content classification unit includes: The priority index of interface elements is calculated based on the content recognition results, using the following formula: ; in, Indicates the priority index of a UI element. , , This represents the pre-configured weighting coefficients, and , This indicates the percentage of pixels in the video area on a dynamic screen. This indicates the percentage of pixels that the text line candidate boxes represent on a dynamic screen. This represents the percentage of pixels in the original saliency map on the dynamic screen, and ; Based on a priority index, interface elements are categorized into high priority, medium priority, and low priority: When a UI element has high priority, mark it to prevent it from being clipped. When an interface element is of medium priority, the interface element is marked and can be deformed to a limited extent; When a UI element has low priority, mark it and let it be handled freely.

[0015] if = 0, meaning there is no video area on the current screen. = 0, the video region contributes nothing to the priority index calculation of this interface element, and the priority index of the interface element will be determined entirely by the text line candidate boxes and the original saliency map; if = 0, meaning there are no detected text line candidate boxes on the current screen. = 0, the text line candidate box does not contribute to the calculation of the priority index of the interface element, and the priority index of the interface element will be determined entirely by the text line candidate box and the original saliency map.

[0016] As a further preferred embodiment of the present invention, the dynamic reconfiguration unit includes: The stretching direction of the dynamic screen is identified based on physical morphology data. The stretching direction includes the curling direction and the stretching direction. Pre-rendering selection for different interface elements based on priority markers: When clipping of interface elements is prohibited, only the interface elements are scaled proportionally. When the edge of the interface element reaches the edge of the maximum display area, the scaling of the interface element stops. When interface elements undergo limited deformation, apply flexible layout adjustment and non-uniform scaling to text areas within the interface elements; apply deformation compensation distortion and reverse content cropping to image areas within the interface elements; and enable fluid layout for salient images within the interface elements. Based on the actual resolution of the future frame and the pre-rendering selection results, a pre-rendering signal for the future frame is sent to the feedback module.

[0017] As a further preferred embodiment of the present invention, the specific implementation process of the feedback module includes: Based on the received pre-rendering signal, execute the pre-rendering command and store the pre-rendering result in the buffer.

[0018] A display method for a flexible device display system based on dynamic screen shape adaptation, the method comprising the following steps: Step S100: Monitor the multimodal data of the dynamic screen in real time, wherein the multimodal data includes physical shape data and display data; Step S200: Based on the multimodal data, construct a resolution conversion model and output the logical resolution; Step S300: Predict the actual resolution of future frames, analyze the interface elements of the dynamic screen in real time, identify the interface elements of the dynamic screen, and perform pre-rendering selection of the interface elements. Step S400: Perform pre-rendering on the dynamic screen based on the actual resolution of the future frame and the pre-rendering selection.

[0019] Compared with existing technologies, the beneficial effects achieved by this invention are as follows: The flexible device display system based on dynamic screen shape adaptation provided by this invention includes a perception module for real-time monitoring of multimodal data of the dynamic screen, a processing module for constructing a resolution conversion model and calculating the logical resolution, a rendering selection module for pre-rendering analysis of the dynamic screen, and a feedback module for responding to the pre-rendering signals sent by the rendering selection module. The rendering selection module includes a resolution prediction unit, a content recognition unit, a content classification unit, and a dynamic reconstruction unit. The perception module enables real-time monitoring of multi-dimensional changes such as curling and stretching. The logical resolution generation adjusts the resolution according to the curl length, notifying the system to render only the visible area, thereby saving resources. Pre-rendering allocates rendering resources in advance, reducing rendering preparation time and enabling the application interface layout to adjust in real-time according to the screen shape, thus improving the user's interactive experience. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a schematic diagram of the structure in an embodiment of the present invention; Figure 2 This is a schematic diagram of the method in an embodiment of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention are within the scope of protection of the present invention.

[0022] Example 1 like Figure 1 As shown, this is an embodiment of the present invention, which provides a flexible device display system based on dynamic screen shape adaptation. The system includes a sensing module, a processing module, a rendering selection module, and a feedback module. The sensing module is used to monitor the multimodal data of the dynamic screen in real time when the user triggers a curling action on the flexible device, and feed it back to the processing module. The multimodal data includes physical shape data and display data. The processing module is used to construct a resolution conversion model based on the multimodal data fed back by the perception module, calculate the logical resolution, and transmit the logical resolution to the rendering selection module. The rendering selection module includes a resolution prediction unit, a content recognition unit, a content classification unit, and a dynamic reconstruction unit. The resolution prediction unit is used to predict the actual resolution of future frames; The content recognition unit is used to analyze the interface elements of the dynamic screen in real time, identify the interface elements of the dynamic screen, and assign weights to the interface elements. The content classification unit is used to calculate the priority index of interface elements based on the content recognition results and to classify the interface elements by priority. The dynamic reconstruction unit is used to pre-render elements based on the actual resolution and content classification results of future frames, and send the pre-render signal to the feedback module. The feedback module is used to respond to the pre-rendering signal sent by the rendering selection module and pre-render the dynamic screen.

[0023] Specifically, the sensing module includes a data acquisition unit and a synchronization unit: The acquisition unit monitors the physical shape data of the dynamic screen in real time through a laser rangefinder, an accelerometer, a flexible curvature sensor, and a distributed fiber optic sensor. The physical shape data includes the real-time curling length, curling speed, and curling acceleration of the dynamic screen. The unit also extracts the display data of the dynamic screen from the working log of the flexible device. The display data includes the actual resolution of the dynamic screen and the real-time pixel count distributed on the dynamic screen. The synchronization unit periodically calibrates the clock offset of the sensors using the GPS clock and PTP protocol, compensates for the reverse offset of the timestamps of the adaptation parameters using the fixed transmission delay of each sensor, and synchronizes the adaptation parameters according to the timestamps.

[0024] Specifically, the implementation process of the processing module includes: A resolution conversion model is constructed, taking multimodal data as input and outputting logical resolution. The specific formula for the resolution conversion model is as follows: ; ; in, This indicates the deformation ratio of the dynamic screen. This indicates the real-time scroll length of the dynamic screen. This indicates the length of the scroll when the dynamic screen is fully scrolled. Indicates the logical resolution of the dynamic screen. This represents the highest resolution when the dynamic screen is fully unfolded, and k represents the characteristic coefficient of the flexible device.

[0025] Specifically, the resolution prediction unit includes: An LSTM prediction model based on a spatiotemporal attention mechanism is constructed. The input layer receives physical morphology data and logical resolution sequences for a specified number of consecutive frames, and the output layer predicts the probability distribution of the actual resolution of future frames. The specific formula is as follows: ; in, This represents the probability matrix of the actual resolution of future frames, where M represents the total number of specified frames. This represents the time decay factor of the i-th frame. This represents the scrolling speed of the i-th frame of the dynamic screen. This represents the curl acceleration of an i-frame dynamic screen. This represents the logical resolution of the i-frame dynamic screen.

[0026] Specifically, the content recognition unit includes: This method is used for real-time analysis of interface elements in dynamic screens, identifying these elements, which include video regions, text line candidate boxes, and original saliency maps. It captures the motion trajectory of dynamic images using inter-frame difference and optical flow analysis to detect video regions in the dynamic screen; obtains text line candidate boxes through text detection; and outputs the original saliency map using a lightweight DeepLabv3+ model with the screen image as input. The steps for detecting video regions include: Acquire two consecutive frames of images and calculate the pixel difference value: ; in, Indicates two consecutive frames and The difference value of the image at pixel (x,y). Representing a frame pixels, Representing a frame The pixels; When pixel difference When the difference exceeds the pixel difference threshold, the pixel (x,y) is marked as a pixel within the candidate dynamic region. The motion vectors of candidate dynamic region pixels are extracted using the Lucas-Kanade algorithm. Filtering speed modulus Regions exceeding a specified motion threshold are marked as video regions; The text line candidate boxes are obtained by loading the EAST model with OpenCV, preprocessing the input image, performing forward propagation, and using non-maximum suppression to obtain the text line candidate boxes. The original saliency map is obtained by loading a pre-trained DeepLabv3+ model based on MobileNetV3, scaling and normalizing the image, inputting it into the model, performing forward propagation, and outputting the semantic segmentation result as the original saliency map.

[0027] Specifically, the content classification unit includes: The priority index of interface elements is calculated based on the content recognition results, using the following formula: ; in, Indicates the priority index of a UI element. , , This represents the pre-configured weighting coefficients, and , This indicates the percentage of pixels in the video area on a dynamic screen. This indicates the percentage of pixels that the text line candidate boxes represent on a dynamic screen. This represents the percentage of pixels in the original saliency map on the dynamic screen, and ; Based on a priority index, interface elements are categorized into high priority, medium priority, and low priority: When a UI element has high priority, mark it to prevent it from being clipped. When an interface element is of medium priority, the interface element is marked and can be deformed to a limited extent; When a UI element has low priority, mark it and let it be handled freely.

[0028] Specifically, the dynamic reconfiguration unit includes: The stretching direction of the dynamic screen is identified based on physical morphology data. The stretching direction includes the curling direction and the stretching direction. Pre-rendering selection for different interface elements based on priority markers: When clipping of interface elements is prohibited, only the interface elements are scaled proportionally. When the edge of the interface element reaches the edge of the maximum display area, the scaling of the interface element stops. When interface elements undergo limited deformation, apply flexible layout adjustment and non-uniform scaling to text areas within the interface elements; apply deformation compensation distortion and reverse content cropping to image areas within the interface elements; and enable fluid layout for salient images within the interface elements. Based on the actual resolution of the future frame and the pre-rendering selection results, a pre-rendering signal for the future frame is sent to the feedback module.

[0029] Specifically, the implementation process of the feedback module includes: Based on the received pre-rendering signal, execute the pre-rendering command and store the pre-rendering result in the buffer.

[0030] Example 2 like Figure 2 As shown, this is an embodiment of the present invention, which provides a display method for a flexible device display system based on dynamic screen shape adaptation. The method includes the following steps: Step S100: Monitor the multimodal data of the dynamic screen in real time, wherein the multimodal data includes physical shape data and display data; Step S200: Based on the multimodal data, construct a resolution conversion model and output the logical resolution; Step S300: Predict the actual resolution of future frames, analyze the interface elements of the dynamic screen in real time, identify the interface elements of the dynamic screen, and perform pre-rendering selection of the interface elements. Step S400: Perform pre-rendering on the dynamic screen based on the actual resolution of the future frame and the pre-rendering selection.

[0031] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any other combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product, which includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions according to this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another.

[0032] Furthermore, the functional units in the various embodiments of this application can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. This storage medium can be a read-only memory, a disk, or an optical disk, etc.

[0033] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any person skilled in the art can easily conceive of various variations or substitutions within the technical scope disclosed in this application, and these should all 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.

Claims

1. A flexible device display system based on dynamic screen shape adaptation, characterized in that, It includes a perception module, a processing module, a rendering selection module, and a feedback module: The sensing module is used to monitor the multimodal data of the dynamic screen in real time when the user triggers a curling action on the flexible device, and feed it back to the processing module. The multimodal data includes physical shape data and display data. The physical shape data includes the real-time curling length, curling speed and curling acceleration of the dynamic screen. The display data includes the actual resolution of the dynamic screen and the real-time pixels distributed on the dynamic screen. The processing module is used to construct a resolution conversion model based on the multimodal data fed back by the perception module, calculate the logical resolution, and transmit the logical resolution to the rendering selection module. The rendering selection module includes a resolution prediction unit, a content recognition unit, a content classification unit, and a dynamic reconstruction unit. The resolution prediction unit is used to predict the actual resolution of future frames; The content recognition unit is used to analyze the interface elements of the dynamic screen in real time, identify the interface elements of the dynamic screen, and assign weights to the interface elements. The interface elements include video regions, text line candidate boxes, and original saliency maps. The content classification unit is used to calculate the priority index of interface elements based on the content recognition results and to classify the interface elements by priority. The dynamic reconstruction unit is used to pre-render elements based on the actual resolution and content classification results of future frames, and send the pre-render signal to the feedback module. The feedback module is used to respond to the pre-rendering signal sent by the rendering selection module and pre-render the dynamic screen.

2. The flexible device display system based on dynamic screen shape adaptation according to claim 1, characterized in that, The sensing module includes a data acquisition unit and a synchronization unit: The acquisition unit monitors the physical shape data of the dynamic screen in real time through a laser rangefinder, an accelerometer, a flexible curvature sensor, and a distributed fiber optic sensor; and extracts the display data of the dynamic screen through the work log of the flexible device. The synchronization unit periodically calibrates the clock offset of the sensors using the GPS clock and PTP protocol, compensates for the reverse offset of the timestamps of the adaptation parameters using the fixed transmission delay of each sensor, and synchronizes the adaptation parameters according to the timestamps.

3. The flexible device display system based on dynamic screen shape adaptation according to claim 2, characterized in that, The specific implementation process of the processing module includes: A resolution conversion model is constructed, taking multimodal data as input and outputting logical resolution. The specific formula for the resolution conversion model is as follows: ； ； in, This indicates the deformation ratio of the dynamic screen. This indicates the real-time scroll length of the dynamic screen. This indicates the length of the scroll when the dynamic screen is fully scrolled. Indicates the logical resolution of the dynamic screen. This represents the highest resolution when the dynamic screen is fully unfolded, and k represents the characteristic coefficient of the flexible device.

4. The flexible device display system based on dynamic screen shape adaptation according to claim 3, characterized in that, The resolution prediction unit includes: An LSTM prediction model based on a spatiotemporal attention mechanism is constructed. The input layer receives physical morphology data and logical resolution sequences for a specified number of consecutive frames, and the output layer predicts the probability distribution of the actual resolution of future frames. The specific formula is as follows: ； in, This represents the probability matrix of the actual resolution of future frames, where M represents the total number of specified frames. This represents the time decay factor of the i-th frame. This represents the scrolling speed of the i-th frame of the dynamic screen. This represents the curl acceleration of an i-frame dynamic screen. This represents the logical resolution of the i-frame dynamic screen.

5. The flexible device display system based on dynamic screen shape adaptation according to claim 1, characterized in that, The content recognition unit includes: This tool is used for real-time analysis of interface elements on dynamic screens, identifying these elements; it captures motion trajectories of dynamic images using inter-frame difference and optical flow analysis, detecting video regions within the dynamic screen; it obtains candidate bounding boxes for text lines through text detection; and it outputs the original saliency map using the DeepLabv3+ model, taking the screen image as input. The steps for detecting video regions include: Acquire two consecutive frames of images and calculate the pixel difference value: ; in, Indicates two consecutive frames and The difference value of the image at pixel (x,y). Representing a frame pixels, Representing a frame The pixels; When pixel difference When the difference exceeds the pixel difference threshold, the pixel (x,y) is marked as a pixel within the candidate dynamic region. The motion vectors of candidate dynamic region pixels are extracted using the Lucas-Kanade algorithm. The modulus of screening speed || The region that exceeds the specified motion threshold is marked as a video region. The text line candidate boxes are obtained by loading the EAST model with OpenCV, preprocessing the input image, performing forward propagation, and using non-maximum suppression to obtain the text line candidate boxes. The original saliency map is obtained by loading a pre-trained DeepLabv3+ model based on MobileNetV3, scaling and normalizing the image, inputting it into the pre-trained DeepLabv3+ model, performing forward propagation, and outputting the semantic segmentation result as the original saliency map.

6. The flexible device display system based on dynamic screen shape adaptation according to claim 1, characterized in that, The content classification units include: The priority index of interface elements is calculated based on the content recognition results, using the following formula: ； in, Indicates the priority index of a UI element. , , This represents the pre-configured weighting coefficients, and , This indicates the percentage of pixels in the video area on a dynamic screen. This indicates the percentage of pixels that the text line candidate boxes represent on a dynamic screen. This represents the percentage of pixels in the original saliency map on the dynamic screen, and ; Based on a priority index, interface elements are categorized into high priority, medium priority, and low priority: When a UI element has high priority, mark it to prevent it from being clipped. When an interface element is of medium priority, the interface element is marked and can be deformed to a limited extent; When a UI element has low priority, mark it and let it be handled freely.

7. The flexible device display system based on dynamic screen shape adaptation according to claim 1, characterized in that, The dynamic reconfiguration unit includes: The stretching direction of the dynamic screen is identified based on physical morphology data. The stretching direction includes the curling direction and the stretching direction. Pre-rendering selection for different interface elements based on priority markers: When clipping of interface elements is prohibited, only the interface elements are scaled proportionally. When the edge of the interface element reaches the edge of the maximum display area, the scaling of the interface element stops. When interface elements undergo limited deformation, apply flexible layout adjustment and non-uniform scaling to text areas within the interface elements; apply deformation compensation distortion and reverse content cropping to image areas within the interface elements; and enable fluid layout for salient images within the interface elements. Based on the actual resolution of the future frame and the pre-rendering selection results, a pre-rendering signal for the future frame is sent to the feedback module.

8. The flexible device display system based on dynamic screen shape adaptation according to claim 1, characterized in that, The specific implementation process of the feedback module includes: Based on the received pre-rendering signal, execute the pre-rendering command and store the pre-rendering result in the buffer.

9. The display method of the flexible device display system based on dynamic screen shape adaptation according to any one of claims 1 to 8, characterized in that, The method includes the following steps: Step S100: When the user triggers a curling action on the flexible device, the multimodal data of the dynamic screen is monitored in real time. The multimodal data includes physical shape data and display data. The physical shape data includes the real-time curling length, curling speed and curling acceleration of the dynamic screen. The display data includes the actual resolution of the dynamic screen and the real-time pixels distributed on the dynamic screen. Step S200: Based on the multimodal data, construct a resolution conversion model and calculate the logical resolution; Step S300: Predict the actual resolution of future frames, analyze the interface elements of the dynamic screen in real time, identify the interface elements of the dynamic screen, and assign weights to the interface elements. The interface elements include video regions, text line candidate boxes, and original saliency maps. Based on the content recognition results, calculate the priority index of the interface elements and classify the interface elements by priority. Based on the actual resolution of future frames and the content classification results, pre-render the interface elements and issue a pre-render signal. Step S400: In response to the pre-rendering signal, perform pre-rendering on the dynamic screen.

Images