Feature value extraction method for display devices and still frame detection

Abstract

This disclosure relates to a display device and a feature value extraction method for still frame detection. The display device includes: if the target region in the current frame image is a natural region, then using the pixel stream of the target region as the input pixel stream of the current frame image, receiving the input pixel stream of the current frame image, extracting pixel feature values ​​of row unit regions in the current frame image according to row unit regions, comparing the pixel feature values ​​of the row unit regions in the current frame image with the pixel feature values ​​of the corresponding row unit regions in the adjacent previous frame image, obtaining a comparison result, and determining whether to output a still signal. It can be seen that this disclosure, by extracting pixel feature values ​​of the corresponding row unit regions in the current frame image according to row unit regions and comparing them with the pixel feature values ​​of the corresponding row unit regions in the adjacent previous frame to obtain a comparison result, can accurately reflect the difference state of corresponding regions between adjacent frames, effectively improving the accuracy of still frame detection in the display device.

Description

Technical Field

[0001] This application relates to screen display technology. More specifically, it relates to a display device and a feature value extraction method for still frame detection. Background Technology

[0002] In the field of screen display technology, still frame detection is a key function in video processing systems, especially in scenarios such as low-power display, dynamic refresh rate adjustment, and video compression.

[0003] In existing technologies, still frame detection is usually implemented using histograms or statistical features. By comparing the brightness distribution or energy statistics of two adjacent frames, although it can avoid storing the complete original pixel data of the image frame, the detection accuracy is low and it is difficult to meet the still frame detection requirements of display devices. Summary of the Invention

[0004] To solve the above-mentioned technical problems, or at least partially solve them, embodiments of this application provide a feature value extraction method for a display device and still frame detection.

[0005] In a first aspect, embodiments of this application provide a display device, including:

[0006] The controller is configured as follows:

[0007] Based on the pixel values ​​of adjacent pixels within the target region in the current frame image, determine the average intra-row gradient value of the target region, and based on the pixel values ​​of pixels within the target region, determine the pixel value variance of the target region.

[0008] Based on the average intra-row gradient value and the variance of the pixel value, the classification result of the target region is determined. If the classification result is that the target region is a natural region, the pixel stream of the target region is used as the input pixel stream of the current frame image.

[0009] Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions;

[0010] The pixel feature values ​​of the row cell region in the current frame image are compared with the pixel feature values ​​of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; the comparison result is used to determine whether to output a stationary signal.

[0011] In this embodiment, by extracting the pixel feature value of the corresponding row unit region of the current frame image according to the row unit region, and comparing it with the pixel feature value of the corresponding row unit region of the adjacent previous frame, the comparison result can accurately reflect the difference state of the corresponding region between adjacent frames, effectively improving the accuracy of still frame detection of the display device.

[0012] In some embodiments of this application, the controller is deployed with a row cache, and the controller is specifically configured as follows:

[0013] The average inter-row gradient value of the target region is determined based on the pixel values ​​of the corresponding columns of adjacent rows in the target region in the current frame image.

[0014] The classification result of the target region is determined based on the average intra-row gradient value, pixel value variance, and average inter-row gradient value.

[0015] In some embodiments of this application, the controller is specifically configured as follows:

[0016] Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell region in the current frame image according to the operation rules of the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, and the perturbation term generated by the pixel position of the current pixel to be processed.

[0017] In some embodiments of this application, the controller is deployed with a row cache, and the controller is specifically configured as follows:

[0018] Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell region in the current frame image according to the operation rules of the operation rules of the perturbation term generated by the pixel position of the current pixel and the pixel value of the current pixel to be processed, based on the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, the perturbation term generated by the pixel position of the current pixel to be processed, and the data perturbation of the preceding row cell region.

[0019] In some embodiments of this application, the controller is specifically configured as follows:

[0020] Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions using a multi-path hash combination method.

[0021] In some embodiments of this application, the controller is specifically configured as follows:

[0022] Receive the input pixel stream of the current frame image, determine the multi-path hash combination based on the independent hash channel corresponding to the color channel, and extract the pixel feature value of the row cell region in the current frame image according to the row cell region using the multi-path hash combination method.

[0023] In some embodiments of this application, the controller is specifically configured as follows:

[0024] Receive the input pixel stream of the current frame image, segment the target row cell region of the current frame image through a moving hash window of a preset size, and generate a pixel feature value sequence of the target row cell region;

[0025] The pixel feature value sequence is compared with the pixel feature value sequence of the target row unit region corresponding to the previous frame image of the current frame image to obtain the comparison result.

[0026] In some embodiments of this application, the controller is further configured to:

[0027] Receive the comparison results;

[0028] If the comparison result shows that the difference between the corresponding row cell region in the current frame image and its adjacent previous frame image is in a static state, then a static signal is output.

[0029] Secondly, embodiments of this application provide a feature value extraction method for still frame detection, applied to a display device, including:

[0030] Based on the pixel values ​​of adjacent pixels within the target region in the current frame image, determine the average intra-row gradient value of the target region, and based on the pixel values ​​of pixels within the target region, determine the pixel value variance of the target region.

[0031] Based on the average intra-row gradient value and pixel value variance, the classification result of the target region is determined. If the classification result is that the target region is a natural region, the pixel stream of the target region is used as the input pixel stream of the current frame image.

[0032] Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions;

[0033] The pixel feature values ​​of the row cell region in the current frame image are compared with the pixel feature values ​​of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; the comparison result is used to determine whether to output a stationary signal.

[0034] Thirdly, embodiments of this application provide a computer-readable storage medium, including: storing a computer program on the computer-readable storage medium, wherein when the computer program is executed by a processor, it implements the feature value extraction method for still frame detection as shown in the second aspect.

[0035] Fourthly, embodiments of this application provide a computer program product, including: when the computer program product is run on a computer, causing the computer to implement the feature value extraction method for still frame detection as shown in the second aspect. Attached Figure Description

[0036] To more clearly illustrate the implementation methods in the embodiments of this application or related technologies, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0037] Figure 1 An operational scenario between a display device and a control device according to some embodiments is illustrated;

[0038] Figure 2 A hardware configuration block diagram of a control device 100 according to some embodiments is shown;

[0039] Figure 3 A hardware configuration block diagram of a display device 200 according to some embodiments is shown;

[0040] Figure 4 One of the flowcharts for a feature value extraction method for still frame detection according to some embodiments is shown;

[0041] Figure 5 A second schematic flowchart of a feature value extraction method for still frame detection according to some embodiments is shown;

[0042] Figure 6 The third schematic diagram of a feature value extraction method for still frame detection according to some embodiments is shown;

[0043] Figure 7 The fourth schematic flowchart of a feature value extraction method for still frame detection according to some embodiments is shown;

[0044] Figure 8 The fifth flowchart illustrates a feature value extraction method for still frame detection according to some embodiments;

[0045] Figure 9 A flowchart of a feature value extraction method for still frame detection according to some embodiments is shown in diagram six.

[0046] Figure 10 The seventh flowchart illustrates a feature value extraction method for still frame detection according to some embodiments. Detailed Implementation

[0047] To make the objectives and implementation methods of this application clearer, the exemplary implementation methods of this application will be clearly and completely described below with reference to the accompanying drawings of the exemplary embodiments of this application. Obviously, the exemplary embodiments described are only some embodiments of this application, and not all embodiments.

[0048] It should be noted that the brief descriptions of terms in this application are only for the convenience of understanding the embodiments described below, and are not intended to limit the embodiments of this application. Unless otherwise stated, these terms should be understood in their ordinary and common meaning.

[0049] The terms "first," "second," "third," etc., used in the specification, claims, and accompanying drawings of this application are used to distinguish similar or related objects or entities, and do not necessarily imply a specific order or sequence, unless otherwise specified. It should be understood that such terms are interchangeable where appropriate.

[0050] The terms “comprising” and “having”, and any variations thereof, are intended to cover but not exclude inclusion, for example, a product or device that includes a range of components is not necessarily limited to all of the components that are clearly listed, but may include other components that are not clearly listed or that are inherent to such product or device.

[0051] The display device provided in this application can have various implementation forms, such as a television, a smart television, a laser projection device, a monitor, an electronic bulletin board, an electronic table, a mobile phone, a tablet computer, a laptop computer, a handheld computer, an in-vehicle electronic device, etc.

[0052] Figure 1 This is a schematic diagram illustrating an operational scenario between a display device and a control device according to an embodiment, wherein the control device includes a smart device or a control apparatus. Figure 1 As shown, the user can operate the display device 200 through the smart device 300 or the control device 100.

[0053] In some embodiments, the control device 100 may be a remote control. Communication between the remote control and the display device includes infrared protocol communication, Bluetooth protocol communication, and other short-range communication methods, controlling the display device 200 wirelessly or via wired means. Users can control the display device 200 by inputting user commands through buttons on the remote control, voice input, control panel input, etc.

[0054] In some embodiments, a smart device 300 (such as a mobile terminal, tablet computer, computer, laptop computer, etc.) can also be used to control the display device 200. For example, an application running on the smart device can be used to control the display device 200.

[0055] In some embodiments, the display device may receive instructions not through the aforementioned smart devices or control devices, but through touch or gestures.

[0056] In some embodiments, the display device 200 can also be controlled in ways other than the control device 100 and the smart device 300. For example, it can be controlled by directly receiving the user's voice commands through a module configured inside the display device 200 for acquiring voice commands, or it can be controlled by receiving the user's voice commands through a voice control device set outside the display device 200.

[0057] In some embodiments, the display device 200 also communicates with the server 400. The display device 200 may communicate via a local area network (LAN), wireless local area network (WLAN), and other networks. The server 400 may provide various content and interactive features to the display device 200. The server 400 may be a cluster or multiple clusters, and may include one or more types of servers.

[0058] Figure 2 An exemplary block diagram of the configuration of the control device 100 according to an exemplary embodiment is shown. Figure 2 As shown, the control device 100 includes a controller 110, a communication interface 130, a user input / output interface 140, an external memory, and a power supply. The control device 100 can receive user input operation commands and convert the operation commands into commands that the display device 200 can recognize and respond to, thus acting as an intermediary for interaction between the user and the display device 200.

[0059] like Figure 3 The display device 200 includes at least one of the following: a tuner 210, a communicator 220, a detector 230, an external device interface 240, a controller 250, a display 260, an audio output interface 270, a user interface 280, an external memory, and a power supply.

[0060] In some embodiments, the controller includes a processor, a video processor, an audio processor, a graphics processor, RAM, ROM, and a first to an nth interface for input / output.

[0061] The display 260 includes a display screen assembly for presenting images, a driving assembly for driving image display, a component for receiving image signals from the controller output, and a user control UI interface for displaying video content, image content, menu control interface, and user control UI interface.

[0062] The display 260 can be an LCD display, an OLED display, or a projection display, and can also be a projection device and a projection screen.

[0063] The communicator 220 is a component used to communicate with external devices or servers according to various communication protocol types. For example, the communicator may include at least one of the following: a Wi-Fi module, a Bluetooth module, a wired Ethernet module, other network communication protocol chips or near-field communication protocol chips, and an infrared receiver. The display device 200 can establish the transmission and reception of control signals and data signals with the external control device 100 or the server 400 through the communicator 220.

[0064] User interface 280 can be used to receive control signals from control device 100 (such as an infrared remote control). It can also be used to directly receive user input operation commands and convert the operation commands into commands that display device 200 can recognize and respond to; in this case, it can be called a user input interface.

[0065] Detector 230 is used to collect signals from the external environment or to interact with the external environment. For example, detector 230 includes a light receiver, a sensor for collecting ambient light intensity; or, detector 230 includes an image acquisition device, such as a camera, which can be used to collect external environmental scenes, user attributes, or user interaction gestures; or, detector 230 includes a sound acquisition device, such as a microphone, for receiving external sounds.

[0066] The external device interface 240 may include, but is not limited to, one or more of the following: High Definition Multimedia Interface (HDMI), analog or high-definition component input interface (component), composite video input interface (CVBS), USB input interface (USB), RGB port, etc. It may also be a composite input / output interface formed by multiple interfaces mentioned above.

[0067] The tuner / demodulator 210 receives broadcast television signals via wired or wireless means, and demodulates audio and video signals, such as EPG data signals, from multiple wireless or wired broadcast television signals.

[0068] In some embodiments, the controller 250 and the tuner 210 may be located in different separate devices, that is, the tuner 210 may also be located in an external device of the main device where the controller 250 is located, such as an external set-top box.

[0069] The controller 250 controls the operation of the display device and responds to user operations through various software control programs stored in memory (internal or external memory). The controller 250 controls the overall operation of the display device 200. For example, in response to receiving a user command to select a UI object to display on the monitor 260, the controller 250 can perform operations related to the object selected by the user command.

[0070] In some embodiments, the controller includes at least one of a central processing unit (CPU), a video processor, an audio processor, a graphics processing unit (GPU), and random access memory (RAM), read-only memory (ROM), a first to an nth interface for input / output, a communication bus, etc.

[0071] RAM, also known as main memory, is an internal memory that directly exchanges data with the controller. It can be read and written at any time (except during refresh) and is very fast, typically serving as temporary data storage for the operating system or other running programs. Its biggest difference from ROM is data volatility; data stored in RAM is lost when power is off. RAM is used in computers and digital systems to temporarily store programs, data, and intermediate results. ROM operates in a non-destructive read-only manner, allowing only reading and not writing. Once information is written, it is fixed and will not be lost even if power is cut off; therefore, it is also called fixed-function memory.

[0072] Users can input commands through a graphical user interface (GUI) displayed on the monitor 260, and the user input interface receives the user input commands through the GUI. Alternatively, users can input commands by entering specific sounds or gestures, and the user input interface receives the user input commands by recognizing the sounds or gestures through sensors.

[0073] A "user interface" is the medium through which an application or operating system interacts and exchanges information with the user. It converts information from its internal form to a form that the user can accept. A common form of user interface is the graphical user interface (GUI), which refers to a user interface related to computer operation displayed graphically. It can be an icon, window, control, or other interface element displayed on the screen of a display device. Controls can include visual interface elements such as icons, buttons, menus, tabs, text boxes, dialog boxes, status bars, navigation bars, and widgets.

[0074] In the field of screen display technology, still frame detection is a key function in video processing systems, especially in scenarios such as low-power displays, dynamic refresh rate adjustment, and video compression. Common technical approaches for still frame detection include:

[0075] Frame difference or threshold method: Static is determined by pixel-by-pixel difference and fixed threshold, which requires caching the complete data of the previous frame, resulting in high resource consumption.

[0076] Histogram-based or statistical feature-based implementation: This method compares the brightness distribution or energy statistics of two adjacent frames. Although it does not require storing frame buffer data, it has weak noise resistance and cannot achieve pixel-level consistency detection.

[0077] Based on approximate features: Extract approximate features such as the mean, variance, and texture of the image for judgment. These features are mostly used for motion recognition or scene recognition. When transferred to static detection scenes, they can only guarantee visual similarity and are difficult to meet the stringent requirement of pixel-by-pixel complete consistency. They are not suitable for high-reliability scenarios such as display verification and storage verification.

[0078] Therefore, on the one hand, it is susceptible to noise and invalid input interference. Random noise frames will appear when the system starts up, loses synchronization, or experiences channel interference. If judgment is based solely on brightness or texture stability, noisy images are easily misidentified as still frames, leading to display freezing, stuttering caused by low refresh rates, or downstream modules skipping frame updates. Therefore, it is necessary to distinguish between "noise areas" and "natural areas" before still frame detection. On the other hand, it lacks adaptability and scalability. Parameters are heavily dependent on resolution, noise level, channel characteristics, and timing. Migrating between different platforms requires repeated debugging, and its adaptability to frame rate, grayscale, and line cycle is poor, making it difficult to simultaneously consider power consumption, latency, and accuracy. Overall, the accuracy of still frame detection is low, making it difficult to meet the still frame detection requirements of display devices.

[0079] To address the aforementioned technical problems, this application provides a display device, including: a controller configured to: determine an average intra-row gradient value of a target region based on the pixel values ​​of adjacent pixels in a target region within a current frame image; determine a pixel value variance of the target region based on the pixel values ​​of pixels in the target region; determine a classification result of the target region based on the average intra-row gradient value and the pixel value variance; if the classification result indicates that the target region is a natural region, then use the pixel stream of the target region as the input pixel stream of the current frame image; receive the input pixel stream of the current frame image; extract pixel feature values ​​of row unit regions in the current frame image according to row unit regions; compare the pixel feature values ​​of row unit regions in the current frame image with the pixel feature values ​​of corresponding row unit regions in the adjacent previous frame image of the current frame image to obtain a comparison result; wherein the comparison result is used to determine whether to output a stationary signal.

[0080] In this embodiment of the application, the controller can be a timing controller (TCON).

[0081] The input pixel stream can be a continuous stream of pixel data input sequentially in the current frame image. Pixel feature values ​​can be feature parameters generated based on the inherent attributes and related information of pixels, used to characterize the image information of the corresponding row unit region in the current frame image. The row unit region can be the region corresponding to the pixel feature values ​​generated pixel-by-pixel in a row-unit-level granularity. The row unit level includes single-row level or m-row level, where m is a positive integer greater than or equal to 2. Pixel-by-pixel streaming means processing each pixel sequentially according to the pixel input order. For example, when the row unit level is single-row level, the row unit region is the image region covered by a single horizontal row in the current frame image. All pixels within this region participate in the processing pixel by pixel in the input order, ultimately generating pixel feature values ​​characterizing the image information of that single row, which serve as the basic unit for difference comparison with the corresponding single row region of the adjacent previous frame. When the row unit level is multi-row level, the row unit region is the image region covered by multiple horizontal rows in the current frame image.

[0082] In this embodiment, if the row unit level is a single row level, the controller can quickly locate the changed rows by outputting the row pixel feature values ​​of the current frame image line by line and comparing them with the same row pixel feature values ​​of the adjacent previous frame; if the row unit level is an m-row level, the controller can detect changes in the partitioned area by outputting the pixel feature values ​​of the m rows of the current frame image and comparing them with the same m rows of the adjacent previous frame.

[0083] In this embodiment, after receiving the input pixel stream of the current frame image in pixel order, the controller processes each pixel sequentially according to the input order, using row unit level as the generation granularity. This generates pixel feature values ​​that characterize the image information of the row unit region in the current frame image. Then, the controller compares the pixel feature values ​​of the row unit region in the current frame image with the pre-stored pixel feature values ​​of the corresponding row unit region in the adjacent previous frame image to obtain a comparison result. This comparison result reflects the image differences between the current frame and the adjacent previous frame at the same position in the row unit region. Therefore, by generating pixel feature values ​​of the corresponding row unit region of the current frame image pixel by pixel at the row unit level and comparing them with the pixel feature values ​​of the corresponding row unit region in the adjacent previous frame, the controller can accurately reflect the differences between corresponding regions in adjacent frames, effectively improving the accuracy of still frame detection in the display device.

[0084] In some embodiments of this application, the controller is specifically configured to: receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell region in the current frame image according to the operation rules of the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, and the perturbation term generated by the pixel position of the current pixel to be processed.

[0085] The hash value of the current hash register can be the value currently stored in the register used to store intermediate results of streaming hash operations. It is the basis for updating pixel feature values ​​pixel by pixel and is dynamically updated iteratively with the processing of each pixel. The shift amount of the hash value can be the number of bits shifted when performing a left circular shift operation on the hash value of the current hash register. The pixel value of the current pixel to be processed can be the pixel value of the currently processed pixel in the input pixel stream in sequence. The perturbation term can be a nonlinear constant calculated based on the coordinate position of the current pixel to be processed in the image. The operation rule can be a way of combining the above hash value, shift amount, pixel value, and perturbation term according to a preset logic, such as using left circular shift, XOR, addition, etc. to achieve pixel-by-pixel iterative update of the hash value.

[0086] For example, in this embodiment of the application, a streaming hash update method is used to process the input pixel stream of the current frame image, and the corresponding hash update formula is as follows:

[0087] .

[0088] in, After the current pixel is processed, the hash register is updated with a new hash value. The hash value of the current hash register, such as the hash value currently stored in the 32-bit hash register. This is the pixel value of the current pixel to be processed, i.e., the input pixel. This is a left circular shift of r bits, which is the amount of shift in the hash value of the current hash register; The perturbation term generated for the pixel position of the current pixel to be processed, i.e., based on the pixel position of the current pixel to be processed. The generated perturbation constant, specifically, It can be a nonlinear term, for example This avoids generating predictable repetitive patterns in inputs with structural rules, such as grid image patches.

[0089] In this embodiment, after receiving the input pixel stream of the current frame image, the controller uses the row unit level as the generation granularity of pixel feature values. During the generation process, the controller uses the hash value stored in the current hash register as the basic operation data, combined with the preset hash value shift amount, the pixel value of the current pixel to be processed, and the perturbation term generated by the pixel position of the current pixel to be processed. According to the preset operation rules corresponding to the above parameters, the controller performs streaming processing on each pixel in sequence, iteratively updates the operation results pixel by pixel, and generates the pixel feature values ​​corresponding to the row unit region in the current frame image.

[0090] In some embodiments of this application, the controller is deployed with a row buffer. Specifically, the controller is configured to: receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row unit regions in the current frame image according to the row unit regions based on the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, the perturbation term generated by the pixel position of the current pixel to be processed, and the operation rules of the data perturbation of the preceding row unit regions.

[0091] The row buffer is an optional data caching function deployed in the controller, used to store pixel data from the previous row region (e.g., the row before the currently being processed) in the current frame image. The previous row region can be a row region in the current frame image that has already completed pixel feature value generation before the currently being processed row region, following the screen's line-by-line scanning order; for example, it could be the row before or several rows before the currently being processed row region. Data perturbation of the previous row region refers to perturbation data generated through preset operations, such as a left circular shift operation, based on the pixel data at the corresponding positions of the previous row region (e.g., the row before or several rows before the currently being processed) stored in the row buffer.

[0092] For example, in this embodiment of the application, when the controller is equipped with a row buffer, the input pixel stream of the current frame image can be processed using the following streaming hash update method, and the corresponding hash update formula is as follows:

[0093] .

[0094] in, After the current pixel is processed, the hash register is updated with a new hash value. The hash value of the current hash register. Left circular shift Bit, that is, the amount of shift in the hash value of the current hash register. The pixel value of the current pixel to be processed. ; The current pixel to be processed The pixel at the position corresponding to the kth previous row belongs to the same column as the current pixel to be processed and is offset upwards by k rows; This refers to left circular shift operations; For the k-th preceding row pixel, i.e. The set number of bits for left circular shift. The perturbation term generated for the pixel position of the current pixel to be processed.

[0095] Understandably, the controller extracts feature values ​​by partition hashing and combines them with real-time shift XOR scrambling to calculate pixel feature values ​​in the row unit region of the current frame image, thereby achieving pixel-by-pixel level complete consistency detection.

[0096] In this embodiment, if the controller enables row caching, it introduces data perturbation of the preceding row unit region. Specifically, after receiving the input pixel stream of the current frame image, the controller generates pixel feature values ​​at the row unit level with pixel feature value generation granularity. During the pixel-by-pixel streaming generation of pixel feature values ​​of the row unit region in the current frame image, the controller uses the hash value of the current hash register as the base operation data, combined with the preset hash value shift amount, the pixel value of the current pixel to be processed, the perturbation term generated by the pixel position of the current pixel to be processed, and the data perturbation of the preceding row unit region introduced through row caching. The hash value is iteratively updated pixel by pixel according to the preset operation rules corresponding to the above parameters to generate the pixel feature value corresponding to the row unit region, thereby improving the spatial sensitivity and anti-collision capability of the pixel feature value. This processing method effectively enhances the sensitivity to the spatial structure of the image, breaks the periodicity, and reduces structural collisions by incorporating the state of the previous row or several preceding rows into the streaming hash operation for perturbation and mixing, while maintaining the real-time performance and low memory consumption characteristics of streaming processing.

[0097] In this embodiment, the controller calculates the pixel feature values ​​of the row unit region in the current frame image through pure logic methods (such as shift, XOR, and addition operation rules), without the need for complex calculation units, and has the advantages of low power consumption and small area.

[0098] In some embodiments of this application, the controller is specifically configured to: receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the target row cell region in the current frame image according to the pixel values ​​and weights of the pixels in the target row cell region.

[0099] The weight of a pixel can be determined based on its position or the grayscale range in which its pixel value falls.

[0100] In this embodiment of the application, taking the target row cell region as a single row as an example, the pixel feature value of the target row cell region can be obtained by multiplying the pixel value of each pixel in the target row cell region with its corresponding weight one by one, and then performing successive XOR operations on the product result.

[0101] In some embodiments of this application, the controller is specifically configured to: receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions using a multi-path hash combination method.

[0102] One approach to multi-path hashing is to construct at least two independent, parallel hashing paths to simultaneously extract features from the input pixel stream. Each path independently generates a hash result of a fixed number of bits. By concatenating and integrating the independent hash results of all paths, a complete pixel feature value is formed. For example, a single path generates a 32-bit hash value, two paths generate a 64-bit hash value, and three paths generate a 96-bit hash value.

[0103] In this embodiment, the controller uses a multi-path disturbance anti-collision mechanism to achieve detection through multi-path hash combination, which not only has a low collision probability but also accurately detects 1-bit changes in any pixel.

[0104] In some embodiments of this application, multiple independent hashes are generated in parallel by configuring multiple different perturbation seeds (such as the perturbation terms mentioned above) and shift amounts.

[0105] In some embodiments of this application, the controller is specifically configured to: receive the input pixel stream of the current frame image, determine a multi-path hash combination based on the independent hash channels corresponding to the color channels, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions using the multi-path hash combination method.

[0106] Among them, the independent hash channel corresponding to the color channel can be configured by setting an independent hash channel for each of the RGB color channels of the current pixel to be processed.

[0107] In this embodiment, a multi-input source parallel computing mode is adopted, and independent hash channels are configured for different input sources. Each channel operates synchronously and does not interfere with each other. As a result, when any pixel changes, at least one hash result will change accordingly, thereby effectively reducing collisions, improving sensitivity, and enhancing robustness and diagnosability.

[0108] In some embodiments of this application, the multi-path hash combination can further enrich the dimensions of hash combination by configuring corresponding independent hash channels for each color channel of RGB, based on the above-mentioned independent hash channels generated in parallel by configuring multiple different perturbation seeds (such as the perturbation terms mentioned above) and shift amounts.

[0109] In some embodiments of this application, the controller is specifically configured to: receive the input pixel stream of the current frame image, segment the target row cell region of the current frame image through a moving hash window of a preset size, generate a pixel feature value sequence of the target row cell region, compare the pixel feature value sequence with the pixel feature value sequence of the target row cell region corresponding to the adjacent previous frame image of the current frame image, and obtain a comparison result.

[0110] The preset-size moving hash window can be a fixed-width moving hash window. Specifically, the moving hash window uses a fixed-length observation window that slides along the row direction (horizontal direction) with a specified step size. This expands the pixel feature values ​​of an entire row or frame into a sequence of locally sensitive pixel feature values ​​for change detection, similarity comparison, and location. The moving hash window has a fixed window size but an adjustable step size. The fixed window size determines the coverage of each hash; a larger window results in more robust features and stronger noise resistance, but coarser localization granularity; a smaller window is more sensitive and provides finer localization, but is more sensitive to noise. The size can be determined based on the actual memory space and application scenario. The moving hash window refines a single pixel feature value in an entire row into a sequence of pixel feature values ​​for consecutive windows. By comparing the corresponding pixel value sequence of the previous frame, the earliest inconsistent window index provides the approximate column coordinates of the non-static starting point, enabling rapid determination of which column interval has changed without decoding or complex matching.

[0111] The step size of the moving hash window is adjustable to control the degree of overlap and output frequency between adjacent moving hash windows. A smaller step size results in greater overlap, better continuity and positioning accuracy, but higher computational or bandwidth overhead; a larger step size results in less overlap and better throughput. When the step size equals the window width, there is no overlap between adjacent moving hash windows, and the algorithm transforms into non-overlapping block hashing, where entire rows or multiple rows are cut into equal-length blocks, each outputting a pixel feature value.

[0112] In this embodiment, the target row cell region can be a specific horizontal row cell region in the current frame image selected for performing moving hash window segmentation and feature extraction operations. The pixel feature value sequence can be a sequence formed by multiple consecutive pixel blocks obtained after the target row cell region is segmented by the moving hash window, with each pixel block generating a corresponding pixel feature value, and all pixel feature values ​​arranged in the window sliding order.

[0113] In some embodiments of this application, the controller is further configured to: determine the average intra-row gradient value of the target region based on the pixel values ​​of adjacent pixels in the target region in the current frame image, and determine the pixel value variance of the target region based on the pixel values ​​of the pixels in the target region, and determine the classification result of the target region based on the average intra-row gradient value and the pixel value variance; if the classification result is that the target region is a natural region, then the pixel stream of the target region is used as the input pixel stream of the current frame image.

[0114] The target region can be any region divided by rows in the current frame image. The average intra-row gradient value can be obtained by averaging the absolute values ​​of the differences in pixel values ​​between adjacent pixels in all rows within the target region. The pixel value variance can be a statistical measure calculated based on the pixel values ​​of all pixels within the target region. The classification result can be the type determination result of the target region. Specifically, the target result can be divided into two categories: natural regions and noisy regions. Natural regions refer to areas in the image that conform to the laws of natural scenes, where pixel changes are reasonable and continuous. Their average intra-row gradient value and pixel value variance are within a preset reasonable range, and they often correspond to real physical scenes. Noisy regions refer to images with no effective content or extremely weak textures, such as black points, gray points, near-solid colors, and weak textures below the noise threshold. In this case, the texture intensity and pixel value variance are low.

[0115] In this embodiment of the application, the local gradient value of the corresponding pixel is determined based on the pixel values ​​of adjacent pixels in the target region in the current frame image, and the average local gradient value of the target region is determined based on the local gradient values ​​of each pixel in the target region.

[0116] Specifically, for feature extraction of local feature values ​​within a row, the following calculation formula can be used:

[0117] .

[0118] in, This represents the local gradient value within a single pixel. Indicates coordinates as The pixel value of the pixel, Indicates coordinates as The pixel value of the pixel.

[0119] The following formula can be used to calculate the variance of pixel values ​​in the target region:

[0120] .

[0121] in, E represents the variance of pixel values ​​in the target region, E represents the expected value, and P represents the pixel value of a pixel within the target region.

[0122] In this embodiment, the average intra-row gradient value of the target region is used as the texture intensity of the target region. If the texture intensity of the target region is less than the texture intensity threshold and the pixel value variance of the target region is less than the pixel value variance threshold, then the target region is determined to be a noise region; otherwise, it is a natural region. In this case, the noise region is masked and does not proceed to the subsequent hash stillness detection.

[0123] In some embodiments of this application, the controller is deployed with a row cache. Specifically, the controller is configured to: determine the average inter-row gradient value of the target region based on the pixel values ​​of the corresponding columns of adjacent rows in the target region in the current frame image; and determine the classification result of the target region based on the average intra-row gradient value, the pixel value variance, and the average inter-row gradient value.

[0124] In this embodiment of the application, the inter-row local gradient value of the corresponding pixel is determined based on the pixel of the adjacent row corresponding column of the target region in the current frame image, and the average inter-row gradient value of the target region is determined based on the inter-row local gradient value of each pixel in the target region.

[0125] The following calculation formula can be used for extracting local gradient values ​​between rows.

[0126] .

[0127] in, This represents the inter-row local gradient value of a single pixel. Indicates coordinates as The pixel value of the pixel, Indicates coordinates as The pixel value of the pixel.

[0128] In this embodiment of the application, after determining the average inter-row gradient value of the target region, the classification result of the target region is determined based on the average intra-row gradient value, pixel value variance and average inter-row gradient value. If the classification result is that the target region is a natural region, the pixel stream of the target region is used as the input pixel stream of the current frame image.

[0129] In this embodiment, the sum of the average intra-row gradient value and the average inter-row gradient value of the target region is used as the texture intensity of the target region. If the texture intensity of the target region is less than the texture intensity threshold and the pixel value variance of the target region is less than the pixel value variance threshold, then the target region is determined to be a noise region (i.e., a non-natural region); otherwise, it is a natural region. In this case, the noise region is masked and does not enter the subsequent hash stillness detection, thereby avoiding misjudgment of the noise region.

[0130] In the embodiments of this application, wherein and The first-order difference amplitude in the horizontal and vertical directions directly measures the strength of local edges or textures; pixel value variance is the overall variance of the entire region, a second-order statistic of intensity fluctuations, reflecting the overall contrast and distribution width of the image. Solid or near-solid color images have very low variance, while structured scenes have higher variance. Gradients emphasize high-frequency structures (edges or textures), while variance is sensitive to both low-frequency contrast and high-frequency noise.

[0131] In some embodiments of this application, the controller is further configured to: receive a comparison result, and if the comparison result indicates that the difference state of the corresponding row cell region in the current frame image and its adjacent previous frame image is a static state, then output a static signal.

[0132] The difference state can be the result of comparing the pixel feature values ​​of the corresponding row unit regions in the current frame image and its adjacent previous frame image, and is used to reflect the difference state of the corresponding row unit regions in the two frames. Specifically, the difference state includes static state and dynamic state.

[0133] In some embodiments of this application, if the pixel feature value of the row cell region in the current frame image is the same as the pixel feature value of the corresponding row cell region in the adjacent previous frame image, then the corresponding row cell region is determined to be in a static state, and a static signal is output.

[0134] In some embodiments of this application, if the pixel feature value sequence of the target row cell region in the current frame image is the same as the pixel feature value sequence of the target row cell region in the adjacent previous frame image, then the target row cell region of the current frame image is determined to be in a stationary state, and a stationary signal is output.

[0135] In this embodiment, a pre-noise filtering and streaming partitioning high-precision hashing method are adopted to first identify natural and noisy regions at the input end. Then, bit-level real-time feature generation and multi-path anti-collision mechanism are used to determine pixel-by-pixel consistency, significantly reducing storage and computational overhead. It also maintains high accuracy, low latency, and strong robustness under complex noise and backlight drift conditions, making it feasible for engineering implementation. Specifically, it does not require caching the pixel values ​​of the entire frame, but only the pixel feature values ​​at the row unit level. It can calculate in real time whether the corresponding row or m rows of two frames are the same, with low resource consumption. In addition, the multi-path hashing combination method greatly reduces the probability of hash collisions and supports flexible expansion according to the scenario. It has stronger detection stability, is sensitive to 1-bit pixel changes, and is not sensitive to compression artifacts and brightness perturbations, achieving a low false alarm rate and fast response. The pure logic implementation combined with the optional row caching method greatly reduces the system power consumption and area occupation, and is easy to deploy on multiple platforms. In complex noise and flicker environments, it can maintain excellent performance with near-zero false detection and zero false locking. Compared with traditional methods, the embodiments of this application have higher accuracy, stronger robustness, and significantly improved adaptability and real-time performance.

[0136] To illustrate this solution in more detail, the following will use examples to illustrate it. Figures 4 to 10 To explain, it is understandable that Figures 4 to 10The steps involved may include more or fewer steps in actual implementation, and the order of these steps may also differ, as long as the feature value extraction method for still frame detection provided in this application embodiment can be implemented. The executing entity of the feature value extraction method for still frame detection can be a display device, or a functional module or functional entity within the display device that can implement the feature value extraction method for still frame detection; no limitation is made here. Moreover, the specific description of the feature value extraction method for still frame detection provided in this application embodiment can refer to the relevant description of the above-mentioned display device, and the same or similar technical effects can be achieved; further details are omitted here.

[0137] Figure 4 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S401 to S404.

[0138] S401. Based on the pixel values ​​of adjacent pixels within the target region in the current frame image, determine the average intra-row gradient value of the target region, and based on the pixel values ​​of pixels within the target region, determine the pixel value variance of the target region.

[0139] S402. Based on the average intra-row gradient value and pixel value variance, determine the classification result of the target region. If the classification result is that the target region is a natural region, then use the pixel stream of the target region as the input pixel stream of the current frame image.

[0140] S403. Receive the input pixel stream of the current frame image and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions.

[0141] S404. The pixel feature value of the row cell region in the current frame image is compared with the pixel feature value of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

[0142] In this embodiment, the feature value extraction method for still frame detection can be applied to a chip.

[0143] Figure 5 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S501 to S504.

[0144] S501. Based on the pixel values ​​of adjacent pixels within the same row of the target region in the current frame image, determine the average intra-row gradient value of the target region; based on the pixel values ​​of pixels in the target region, determine the pixel value variance of the target region; and based on the pixel values ​​of pixels in the corresponding columns of adjacent rows of the target region in the current frame image, determine the average inter-row gradient value of the target region.

[0145] S502. Based on the average intra-row gradient value, pixel value variance, and average inter-row gradient value, determine the classification result of the target region. If the classification result is that the target region is a natural region, then use the pixel stream of the target region as the input pixel stream of the current frame image.

[0146] S503: Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions.

[0147] S504. The pixel feature value of the row cell region in the current frame image is compared with the pixel feature value of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

[0148] Figure 6 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S601 to S602.

[0149] S601. Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell region in the current frame image according to the operation rules of the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, and the perturbation term generated by the pixel position of the current pixel to be processed.

[0150] S602. The pixel feature value of the row cell region in the current frame image is compared with the pixel feature value of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

[0151] Figure 7 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S701 to S702.

[0152] S701: Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row unit region in the current frame image according to the operation rules of the operation rules of the perturbation term generated by the pixel position of the current pixel and the pixel value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, and the perturbation of the data in the preceding row unit region.

[0153] S702. The pixel feature value of the row cell region in the current frame image is compared with the pixel feature value of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

[0154] Figure 8 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S801 to S802.

[0155] S801: Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row cell regions in the current frame image according to the row cell regions using a multi-path hash combination method.

[0156] S802. The pixel feature value of the row cell region in the current frame image is compared with the pixel feature value of the corresponding row cell region in the adjacent previous frame image to obtain the comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

[0157] Figure 9 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S901 to S902.

[0158] S901: Receive the input pixel stream of the current frame image, segment the target row cell region of the current frame image through a moving hash window of a preset size, and generate a pixel feature value sequence of the target row cell region.

[0159] S902. The pixel feature value sequence is compared with the pixel feature value sequence of the target row unit region corresponding to the previous frame image of the current frame image to obtain the comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

[0160] Figure 10 The flowchart illustrates the steps of a feature value extraction method for still frame detection according to one or more embodiments of this application, applied to a display device. The feature value extraction method for still frame detection may include the following steps S1001 to S1006.

[0161] S1001. Obtain the classification result of the target region in the current frame image by feature value extraction.

[0162] S1002, Scene Classification: Based on the classification results, determine whether the target region is a natural region or a noisy image, and output the scene label.

[0163] S1003. If the scene is identified as noise, output the non-static state and exit the process.

[0164] S1004. Otherwise, perform hash feature extraction (optionally move the hash window) and store the hash feature values ​​(such as the pixel feature values ​​or pixel feature value sequences mentioned above).

[0165] S1005. Perform hash comparison on the corresponding row unit regions of the two frames of images to obtain the comparison result.

[0166] S1006. If the comparison result is 0, it is determined to be a static state; otherwise, it is determined to be a dynamic region.

[0167] The present invention also provides a computer-readable storage medium storing a computer program. When the computer program is executed by a processor, it implements the various processes of the feature value extraction method for still frame detection described above and achieves the same technical effect. To avoid repetition, it will not be described again here.

[0168] The computer-readable storage medium can be a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk, etc.

[0169] The present invention provides a computer program product, comprising: when the computer program product is run on a computer, causing the computer to implement the above-mentioned feature value extraction method for still frame detection.

[0170] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features therein. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0171] For ease of explanation, the above description has been provided in conjunction with specific embodiments. However, the above exemplary discussion is not intended to be exhaustive or to limit the embodiments to the specific forms disclosed above. Various modifications and variations can be obtained based on the above teachings. The selection and description of the above embodiments are for the purpose of better explaining the principles and practical applications, thereby enabling those skilled in the art to better utilize the embodiments and various different variations of embodiments suitable for specific application considerations.

Claims

1. A display device, characterized in that, include: The controller is configured as follows: Based on the pixel values ​​of adjacent pixels within a row in the target region of the current frame image, the average intra-row gradient value of the target region is determined, and based on the pixel values ​​of the pixels in the target region, the pixel value variance of the target region is determined. The average intra-row gradient value of the target region is used as the texture intensity of the target region. When the texture intensity of the target region is greater than or equal to the texture intensity threshold, and / or the pixel value variance of the target region is greater than or equal to the pixel value variance threshold, the pixel stream of the target region is used as the input pixel stream of the current frame image. Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row unit regions in the current frame image according to the row unit regions; The pixel feature value of the row unit region in the current frame image is compared with the pixel feature value of the corresponding row unit region in the adjacent previous frame image to obtain a comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

2. The display device according to claim 1, characterized in that, The controller is deployed with a row cache, and the controller is specifically configured as follows: The average inter-row gradient value of the target region is determined based on the pixel values ​​of the corresponding columns of adjacent rows of the target region in the current frame image. The average intra-row gradient value of the target region, or the sum of the average inter-row gradient value and the average intra-row gradient value of the target region, is taken as the texture intensity of the target region.

3. The display device according to claim 1, characterized in that, The controller is specifically configured as follows: The system receives the input pixel stream of the current frame image and extracts the pixel feature values ​​of the row cell region in the current frame image according to the operation rules of the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, and the perturbation term generated by the pixel position of the current pixel to be processed.

4. The display device according to claim 1, characterized in that, The controller is deployed with a row cache, and the controller is specifically configured as follows: The system receives the input pixel stream of the current frame image and extracts the pixel feature values ​​of the row unit region in the current frame image according to the operation rules of the row unit region, based on the hash value of the current hash register, the shift amount of the hash value, the pixel value of the current pixel to be processed, the perturbation term generated by the pixel position of the current pixel to be processed, and the data perturbation of the preceding row unit region.

5. The display device according to claim 1, characterized in that, The controller is specifically configured as follows: The system receives the input pixel stream of the current frame image and extracts the pixel feature values ​​of the row unit regions in the current frame image according to the row unit regions using a multi-path hash combination method.

6. The display device according to claim 5, characterized in that, The controller is specifically configured as follows: The system receives the input pixel stream of the current frame image, determines a multi-path hash combination based on the independent hash channels corresponding to the color channels, and uses the multi-path hash combination to extract the pixel feature values ​​of the row unit regions in the current frame image according to the row unit regions.

7. The display device according to claim 1, characterized in that, The controller is specifically configured as follows: Receive the input pixel stream of the current frame image, segment the target row unit region of the current frame image through a moving hash window of a preset size, and generate a pixel feature value sequence of the target row unit region; The pixel feature value sequence is compared with the pixel feature value sequence of the target row unit region corresponding to the previous frame image of the current frame image to obtain the comparison result.

8. The display device according to claim 1 or 7, characterized in that, The controller is also configured to: Receive the comparison result; If the comparison result indicates that the difference between the corresponding row unit region in the current frame image and its adjacent previous frame image is in a static state, then a static signal is output.

9. A feature value extraction method for still frame detection, characterized in that, Applied to display devices, including: Based on the pixel values ​​of adjacent pixels within a row in the target region of the current frame image, the average intra-row gradient value of the target region is determined, and based on the pixel values ​​of the pixels in the target region, the pixel value variance of the target region is determined. The average intra-row gradient value of the target region is used as the texture intensity of the target region; when the texture intensity of the target region is greater than or equal to the texture intensity threshold, and / or the pixel value variance of the target region is greater than or equal to the pixel value variance threshold, the pixel stream of the target region is used as the input pixel stream of the current frame image. Receive the input pixel stream of the current frame image, and extract the pixel feature values ​​of the row unit regions in the current frame image according to the row unit regions; The pixel feature value of the row unit region in the current frame image is compared with the pixel feature value of the corresponding row unit region in the adjacent previous frame image to obtain a comparison result; wherein, the comparison result is used to determine whether to output a stationary signal.

Images