Fancy screen detection method and device and storage medium

Abstract

The invention discloses a blurred screen detection method and device and a storage medium. The method comprises the steps of obtaining video image data corresponding to a to-be-detected video frame; at least one type of image domain transformation is carried out on the video image data to correspondingly obtain at least one type of image domain data, and the image domain data correspond to different blurred screen types; based on the domain value condition of each pixel in each piece of image domain data, determining a blurred screen detection result corresponding to each piece of image domain data, the blurred screen detection result representing whether the to-be-detected video frame has a blurred screen type corresponding to the image domain data. Through the method, the detection efficiency of blurred screen detection can be improved.

Description

Technical Field

[0001] The present application relates to the field of image processing technology, and in particular to a method, device and storage medium for detecting a flower screen. Background Art

[0002] With the continuous advancement of digital video technology, video quality has become one of the key factors that directly affect user experience. Therefore, video distorted screen detection is particularly important. However, traditional video distorted screen detection methods mainly rely on manual visual inspection, which requires a large amount of manpower and has low work efficiency, and cannot meet the needs of today's large-scale video surveillance systems. Summary of the invention

[0003] The main technical problem solved by the present application is to provide a method, device and storage medium for detecting a distorted screen, which can realize intelligent distorted screen detection of video files or video streams, thereby improving the detection efficiency of distorted screen detection.

[0004] In order to solve the above technical problems, a technical solution adopted in the present application is to provide a flower screen detection method, which includes: obtaining video image data corresponding to a video frame to be detected; performing at least one image domain transformation on the video image data to obtain at least one image domain data, wherein each image domain data corresponds to a different type of flower screen; based on the domain value of each pixel in each image domain data, determining the flower screen detection result corresponding to each image domain data, wherein the flower screen detection result indicates whether there is a flower screen type corresponding to the image domain data in the video frame to be detected.

[0005] In order to solve the above technical problems, another technical solution adopted by the present application is: to provide a flower screen detection device, which includes a memory and a processor, the memory stores program instructions, and the processor is used to execute the program instructions to implement the above flower screen detection method.

[0006] In order to solve the above technical problems, another technical solution adopted by the present application is: providing a computer-readable storage medium, which is used to store program instructions, and the program instructions can be executed to implement the above-mentioned screen distortion detection method.

[0007] The above scheme obtains the video image data corresponding to the video frame to be detected, performs at least one image domain transformation on the video image data, so as to obtain at least one image domain data; based on the domain value of each pixel in each image domain data, the flower screen detection result corresponding to each image domain data is determined, thereby realizing intelligent flower screen detection. Compared with manual inspection of each video separately, the present application reduces manpower costs and can improve the detection efficiency of flower screen detection. In addition, each image domain data corresponds to a different type of flower screen, and the flower screen detection result indicates whether there is a flower screen type corresponding to the image domain data in the video frame to be detected, that is, at least one type of flower screen can be detected for the video file or video stream, so detailed detection of the video file or video stream can be realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 It is a flow chart of an embodiment of a method for detecting a flower screen provided by the present application;

[0009] Figure 2 It is a flowchart of a specific embodiment of the flower screen detection method provided by the present application;

[0010] Figure 3 This is a flow chart of an embodiment of a flower screen detection device of the present application;

[0011] Figure 4 It is a schematic diagram of the framework of an embodiment of the flower screen detection device of the present application;

[0012] Figure 5 It is a schematic diagram of a framework of an embodiment of a computer-readable storage medium of the present application. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and effect of the present application clearer and more specific, the present application is further described in detail below with reference to the accompanying drawings and examples.

[0014] It should be noted that the terms "first", "second", etc. are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. The term "and / or" is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the previously associated objects are in an "or" relationship. In addition, the term "at least one" in this article represents any combination of at least two of any one or more of a plurality of types. For example, at least one of A, B, and C can represent any one or more elements selected from the set consisting of A, B, and C.

[0015] See also Figure 1 , Figure 1is a flow chart of an embodiment of the screen noise detection method provided by the present application. It should be noted that if there are substantially the same results, this embodiment is not used. Figure 1 The process sequence shown is limited. Figure 1 As shown, this embodiment includes:

[0016] Step S11: Obtain video image data corresponding to the video frame to be detected.

[0017] The video frame to be detected can be obtained from a video file or a video stream. For example, an open source computer vision library can be used to read the video to be detected, and a loop structure can be used to read the video content of the video file frame by frame, and save it as each video frame to be detected. Of course, the video content of the video file can also be read frame by frame using a built-in function to obtain the corresponding video frames to be detected, for example, the VideoReader function. There is no restriction on the method of obtaining the video frame to be detected.

[0018] The video file or video stream may be provided by the target user or determined by the flower screen detection device itself. Specifically, the flower screen detection device may be provided with a human-computer interaction interface, and then the video file or video stream input by the user may be obtained through the human-computer interaction interface. Of course, in other embodiments, the flower screen detection device has a communication module, and the flower screen detection device receives the video file or video stream sent by the terminal device through the communication module. The terminal device has a human-computer interaction function, and the user uses the human-computer interaction function to input the video file or video stream to be sent to the flower screen detection device on the terminal device.

[0019] The video frame to be detected can be converted into data on different channels as video image data to meet different image processing requirements. In one embodiment, the video image data is BGR data.

[0020] BGR data can be obtained by converting the pixel data corresponding to the video frame to be detected into a color space, specifically, reordering the color channels of each pixel. In a specific embodiment, the video frame to be detected is extracted from the received video stream or video file. The RGB data corresponding to the video frame to be detected is obtained. The RGB data is converted into BGR data.

[0021] The method for obtaining RGB data can be selected according to the actual scene requirements. In a specific implementation, the data corresponding to the video frame to be detected is extracted from the received video stream or video file. The data corresponding to the video frame to be detected is decoded and rendered to obtain RGB data.

[0022] In yet another specific embodiment, the step of converting RGB data into BGR data is implemented by an open source computer vision library.

[0023] Step S12: performing at least one image domain transformation on the video image data to obtain at least one image domain data.

[0024] Among them, each image domain data corresponds to a different type of flower screen.

[0025] Image domain transformation refers to the process of converting video image data from one representation domain to another. This transformation can better analyze and process video image data, extract effective information from the image, and implement certain specific image processing operations.

[0026] In one embodiment, the image domain transformation includes at least one of HSV color domain transformation, gray domain transformation and binary domain transformation. Correspondingly, the image domain data includes at least one of HSV color domain data, gray domain data and binary domain data.

[0027] Abnormalities in different image domain data will lead to different types of screen distortion phenomena, that is, HSV color domain data, gray domain data, and binary domain data are used to detect different types of screen distortion, respectively. In one embodiment, the type of screen distortion corresponding to the HSV color domain data is a green screen, the type of screen distortion corresponding to the gray domain is a black-gray screen, and the type of screen distortion corresponding to the binary domain is a striped screen distortion. For example, the HSV color domain, that is, the color space composed of hue, saturation, and brightness, any data fluctuations in this space may cause deviations in color expression, causing the screen to appear to be too green, that is, a green screen, so the HSV color domain data can be used for green screen detection.

[0028] Step S13: based on the domain value of each pixel in each image domain data, determine the flower screen detection result corresponding to each image domain data.

[0029] The screen noise detection result indicates whether the video frame to be detected has a screen noise type corresponding to the image domain data.

[0030] As mentioned above, the image domain data includes at least one of HSV color domain data, grayscale domain data, and binary domain data, that is, the corresponding screen noise detection result is determined using at least one of HSV color domain data, grayscale domain data, and binary domain data.

[0031] In one embodiment, HSV color gamut data may be used to perform green screen detection. For example, in response to the image domain data including HSV color gamut data, based on the domain value of each pixel in the HSV color gamut data, the first number of green pixels in the HSV color gamut data is obtained by counting, and it is detected that the first number of green pixels meets the green screen condition, and it is determined that the video frame to be detected has a green screen.

[0032] The first number of green pixels can be obtained by extracting colors within a specific hue range. In a specific implementation, in the HSV color gamut data, pixels whose hue dimension is within the green hue range are extracted, and the number of extracted pixels is counted as the first number of green pixels.

[0033] The corresponding green screen detection result can be determined by using the HSV color gamut data corresponding to the current video frame to be detected, or by using the HSV color gamut data corresponding to the current video frame to be detected and the HSV color gamut data corresponding to the previous video frame.

[0034] In a specific implementation, the green screen condition is that the number of first green pixels is greater than a first threshold.

[0035] In another specific implementation, the green screen condition is that the first green pixel number is greater than a first threshold and the adjacent number difference is greater than a second threshold, and the adjacent number difference is the difference between the first green pixel number and the second green pixel number corresponding to the previous video frame.

[0036] For example, the green screen condition is that the number of first green pixels is greater than a first threshold value, and the difference between adjacent numbers is greater than a second threshold value. The number of first green pixels corresponding to the current video frame to be detected is a j , the number of the second green pixels corresponding to the previous video frame is a i , wherein the first number of green pixels and the second number of green pixels can be recorded in the element set a(a i , a j ). The difference in the number of adjacent numbers Δ1 = |a j -a i |. In a j When Δ1 is greater than the first threshold value T1 and Δ1 is greater than the second threshold value T2, it is determined that a green screen exists in the video frame to be detected.

[0037] In another embodiment, grayscale domain data may be used to perform black-gray screen detection. For example, in response to the image domain data including grayscale domain data, the domain value of each pixel in the grayscale domain data is counted to obtain at least one first statistical value, and it is detected that at least one first statistical value meets the black-gray screen condition, and it is determined that the video frame to be detected has a black-gray screen, wherein the at least one first statistical value includes at least one of a central tendency statistical value and a discrete degree statistical value.

[0038] The corresponding black-gray screen detection result can be determined by using the grayscale domain data corresponding to the current video frame to be detected, or by using the grayscale domain data corresponding to the current video frame to be detected and the grayscale domain data corresponding to the previous video frame.

[0039] In a specific implementation, the black-gray screen condition is that each first statistical value exceeds a corresponding statistical value range.

[0040] In another specific implementation, the black-gray screen condition is that each first statistical value exceeds the corresponding statistical value range, and each adjacent statistical difference exceeds the corresponding statistical difference range, and the adjacent statistical difference is the difference between a first statistical value and a second statistical value corresponding to the previous video frame.

[0041] The central tendency value can represent the statistic of the distribution center position of each pixel domain value in the gray domain data. For example, the mean, median, mode, etc. The dispersion degree statistic can represent the measurement value of the difference between each pixel domain value in the gray domain data and the mean or median. For example, the standard deviation, variance, range, etc. In a specific embodiment, the central tendency statistic is the mean, and the dispersion degree statistic is the standard deviation.

[0042] For example, the pixel value range in the grayscale domain of the image is 0 to 255. When the first statistical value includes the mean and the standard deviation, the mean of the grayscale domain data corresponding to the current video frame to be detected is μ j , the mean value of the grayscale data corresponding to the previous video frame is μ i , where the mean can be recorded in the element set μ(μ i , μ j ). The adjacent statistical difference is Δ2, where Δ3 = |μ j -μ i |. In μ j If the statistical value range is exceeded and Δ3 exceeds the statistical difference range, the video frame to be detected has a black screen, that is, a black-gray screen. The standard deviation of the grayscale domain data corresponding to the current video frame to be detected is σ j , the standard deviation of the grayscale data corresponding to the previous video frame is σ i , where the standard deviation can be recorded in the element set σ(σ i , σ j ). The adjacent statistical difference is Δ4, where Δ4 = |σ j -σ i |. In σ j If the statistical value range is exceeded and Δ4 exceeds the statistical difference range, the video frame to be detected has a gray screen, that is, a black-gray screen. i =(1 / (M*N)*∑∑l(x,y)), standard deviation σ i =sqrt((1 / (M*N))*∑∑[l(x,y)-μ i ] 2), M and N are the width and height of the grayscale image corresponding to the video frame to be detected, and the grayscale value corresponding to each pixel is l(x, y). ∑∑ represents the sum of all pixels, that is, ∑∑l(x, y) represents the sum of the grayscale values ​​corresponding to all pixels. It can be understood that the mean corresponding to the grayscale domain data can be used for black screen detection, and the standard deviation corresponding to the grayscale domain data can be used for gray screen detection. When it is determined that there is a black screen or a gray screen, it can be confirmed that the video frame to be detected has a black and gray screen.

[0043] In another embodiment, the binary domain data may be used to perform stripe screen detection. For example, in response to the image domain data including the binary domain data, contour detection is performed on the binary domain data to obtain a number of contours in the video frame to be detected, the number of first contours of a preset shape among the plurality of contours is counted, and when the number of first contours is detected to meet the stripe screen condition, it is determined that the video frame to be detected has a stripe screen.

[0044] The corresponding striped screen detection result can be determined by using the binary domain data corresponding to the current video frame to be detected, or by using the binary domain data corresponding to the current video frame to be detected and the binary domain data corresponding to the previous video frame.

[0045] In a specific implementation, the stripe screen condition is that the number of first contours is greater than a third threshold.

[0046] In another specific implementation, the stripe screen condition is that the difference in the number of adjacent contours is greater than a fourth threshold, wherein the difference in the number of contours is the difference between the number of first contours and the number of second contours of a preset shape in the previous video frame.

[0047] In another specific embodiment, the striped screen condition is that the number of first contours is greater than a third threshold, and the difference in the number of adjacent contours is greater than a fourth threshold, wherein the difference in the number of contours is the difference between the number of first contours and the number of second contours of a preset shape in the previous video frame.

[0048] In yet another specific implementation, the preset shape is a rectangle.

[0049] In another specific implementation, contour detection is implemented by an open source computer vision library, such as OpenCV library, Pillow library, etc., which are not limited here.

[0050] For example, the stripe screen condition is that the number of first contours is greater than the third threshold or the difference in the number of adjacent contours is greater than the fourth threshold. The number of first contours corresponding to the current video frame to be detected is The number of second contours corresponding to the previous video frame is The first contour quantity and the second contour quantity can be recorded in the element set Middle. Adjacent quantity difference exist When Δ5 is greater than the third threshold value T3, it is determined that the video frame to be detected has striped screen. When Δ5 is greater than the fourth threshold value T4, it is determined that the video frame to be detected has striped screen. When Δ5 is greater than the third threshold value T3 and Δ5 is greater than the fourth threshold value T4, it is determined that the video frame to be detected has striped screen.

[0051] In another embodiment, HSV color gamut data and grayscale domain data may be used to perform corresponding green screen and black-gray screen detection. For example, in response to the image domain data including HSV color gamut data and grayscale domain data, corresponding green screen detection and gray-black screen detection are performed based on the domain value of each pixel in the HSV color gamut data and grayscale domain data. For another example, HSV color gamut data and binary domain data may be used to perform corresponding green screen and striped screen detection. Of course, corresponding flower screen detection results may also be determined using at least one of the HSV color gamut data, grayscale domain data, and binary domain data. Among them, for the specific instructions on performing flower screen detection using each image domain data, please refer to the relevant description in this step, which will not be repeated here.

[0052] The video image data may be sequentially subjected to image domain transformation to obtain corresponding image domain data, that is, the flower screen detection corresponding to each image domain data is sequentially performed. In a specific embodiment, the image domain data includes HSV color domain data, gray domain data, and binary domain data. The video image data is transformed into the HSV color domain data. Based on the domain value of each pixel in the HSV color domain data, the flower screen detection result corresponding to the HSV color domain data is determined, and the flower screen detection result corresponding to the HSV color domain data indicates whether there is a green screen in the video frame to be detected. In response to the flower screen detection result corresponding to the HSV color domain data being that there is no green screen, the video image data is transformed into gray domain data. Based on the domain value of each pixel in the gray domain data, the flower screen detection result corresponding to the gray domain data is determined, and the flower screen detection result corresponding to the gray color domain data indicates whether there is a black-gray screen in the video frame to be detected. In response to the flower screen detection result corresponding to the gray domain data being that there is no black-gray screen, the gray domain data is binarized to obtain binary domain data. Based on the domain value of each pixel in the binary domain data, a striped screen detection result corresponding to the binary domain data is determined, and the striped screen detection result corresponding to the binary domain data indicates whether there is a striped screen in the video frame to be detected.

[0053] In the case where the distorted screen detection result is abnormal, a corresponding distorted screen detection report may be output. In one embodiment, each distorted screen type may correspond to a distorted screen detection report. For example, in the case where it is determined that a green screen exists in the video frame to be detected, a green screen detection report is output.

[0054] The screen distortion detection report may be an image corresponding to the video frame to be detected in which the abnormality is detected, or may be image domain data corresponding to the video frame to be detected. Of course, it may also be other images or data that can represent the abnormality, which is not limited here.

[0055] See also Figure 2 , Figure 2 It is a flow chart of a specific embodiment of the flower screen detection method provided in the present application, wherein the image domain data includes HSV color domain data, gray domain data, and binary domain data.

[0056] Step S21: Obtain the BGR data corresponding to the video frame to be detected. Step S22: Perform HSV color gamut transformation on the BGR data to obtain HSV color gamut data. Step S23: Use the HSV color gamut data to determine whether there is a green screen in the video frame to be detected. If the judgment result of step S23 is yes, execute step S24: Output a flower screen detection report. If the judgment result of step S23 is no, execute step S25: Perform grayscale domain conversion on the BGR data to obtain grayscale domain data. After step S25, execute step S26: Use the grayscale domain data to determine whether there is a black and gray screen in the video frame to be detected. If the judgment result of step S26 is yes, execute step S24: Output a flower screen detection report. If the judgment result of step S26 is no, execute step S27: Binarize the grayscale domain to obtain binary domain data. After step S27, execute step S28: Use the binary domain data to determine whether there is a striped flower screen in the video frame to be detected. If the judgment result of step S28 is yes, step S24 is executed: outputting the screen distortion detection result. If the judgment result of step S28 is no, step S21 is re-executed. After step S24 is executed, step S21 is re-executed.

[0057] The above scheme obtains the video image data corresponding to the video frame to be detected, performs at least one image domain transformation on the video image data, so as to obtain at least one image domain data; based on the domain value of each pixel in each image domain data, the flower screen detection result corresponding to each image domain data is determined, thereby realizing intelligent flower screen detection. Compared with manual inspection of each video separately, the present application reduces manpower costs and can improve the detection efficiency of flower screen detection. In addition, each image domain data corresponds to a different type of flower screen, and the flower screen detection result indicates whether there is a flower screen type corresponding to the image domain data in the video frame to be detected, that is, at least one type of flower screen can be detected for the video file or video stream, so a detailed analysis of the video file or video stream can be realized.

[0058] See also Figure 3 , Figure 3It is a flow chart of an embodiment of the flower screen detection device of the present application. The flower screen detection device 300 includes a video image data acquisition module 310, an image domain data acquisition module 320, and a flower screen detection result acquisition module 330. The video image data acquisition module 310 is used to acquire video image data corresponding to the video frame to be detected. The image domain data acquisition module 320 is used to perform at least one image domain transformation on the video image data to obtain at least one image domain data, wherein each image domain data corresponds to a different type of flower screen. The flower screen detection result acquisition module 330 is used to determine the flower screen detection result corresponding to each image domain data based on the domain value of each pixel in each image domain data, wherein the flower screen detection result characterizes whether the video frame to be detected has a flower screen type corresponding to the image domain data.

[0059] In some embodiments, the image domain data includes at least one of HSV color domain data, grayscale domain data, and binary domain data. The type of flower screen corresponding to the HSV color domain data is a green screen, the type of flower screen corresponding to the grayscale domain data is a black-gray screen, and the type of flower screen corresponding to the binary domain data is a striped flower screen.

[0060] In some embodiments, the flower screen detection result acquisition module 330 performs the domain value of each pixel in each image domain data to determine the flower screen detection result corresponding to each image domain data, including at least one of the following steps: in response to the image domain data including HSV color domain data, based on the domain value of each pixel in the HSV color domain data, the number of first green pixels in the HSV color domain data is obtained by counting, and it is detected that the first number of green pixels meets the green screen condition, and it is determined that the video frame to be detected has a green screen; in response to the image domain data including grayscale domain data, the domain value of each pixel in the grayscale domain data is counted to obtain at least one first statistical value, and it is detected that the at least one first statistical value meets the black-gray screen condition, and it is determined that the video frame to be detected has a black-gray screen, wherein the at least one first statistical value includes at least one of a central tendency statistical value and a discrete degree statistical value; in response to the image domain data including binary domain data, contour detection is performed on the binary domain data to obtain a plurality of contours in the video frame to be detected, the number of first contours of a preset shape among the plurality of contours is counted, and it is detected that the first number of contours meets the striped flower screen condition, and it is determined that the video frame to be detected has a striped flower screen.

[0061] In some embodiments, the green screen condition is that the first number of green pixels is greater than a first threshold.

[0062] In some embodiments, the green screen condition is that the first green pixel number is greater than a first threshold and the adjacent number difference is greater than a second threshold, and the adjacent number difference is the difference between the first green pixel number and the second green pixel number corresponding to the previous video frame.

[0063] In some embodiments, the flower screen detection result acquisition module 330 specifically executes the domain value of each pixel in the HSV color gamut data to count the first green pixel number in the HSV color gamut data, including: in the HSV color gamut data, extracting pixels whose hue dimension is in the green hue range, and counting the number of extracted pixels as the first green pixel number.

[0064] In some embodiments, the black-gray screen condition is that each first statistical value exceeds the corresponding statistical value range.

[0065] In some embodiments, the black-gray screen condition is that each first statistical value exceeds the corresponding statistical value range, and each adjacent statistical difference exceeds the corresponding statistical difference range, and the adjacent statistical difference is the difference between the first statistical value and the second statistical value corresponding to the previous video frame.

[0066] In some embodiments, the central tendency statistic is the mean, and the dispersion statistic is the standard deviation.

[0067] In some embodiments, the stripe screen condition is that the number of first contours is greater than a third threshold.

[0068] In some embodiments, the striped screen condition is that the difference in the number of adjacent contours is greater than a fourth threshold, wherein the difference in the number of contours is the difference between the number of the first contours and the number of the second contours of a preset shape in the previous video frame.

[0069] In some embodiments, the striped screen condition is that the number of first contours is greater than a third threshold, and the difference in the number of adjacent contours is greater than a fourth threshold, wherein the difference in the number of contours is the difference between the number of the first contours and the number of the second contours of a preset shape in the previous video frame.

[0070] In some embodiments, the preset shape is a rectangle.

[0071] In some embodiments, contour detection is implemented by an open source computer vision library.

[0072] In some embodiments, the image domain data includes HSV color domain data, gray domain data, and binary domain data. The image domain data acquisition module 320 performs at least one image domain transformation on the video image data to obtain at least one image domain data, and the flower screen detection result acquisition module 330 performs the domain value of each pixel in each image domain data to determine the flower screen detection result corresponding to each image domain data, including: transforming the video image data into HSV color domain data. Based on the domain value of each pixel in the HSV color domain data, the flower screen detection result corresponding to the HSV color domain data is determined, and the flower screen detection result corresponding to the HSV color domain data indicates whether the video frame to be detected has a green screen. In response to the flower screen detection result corresponding to the HSV color domain data being that there is no green screen, the video image data is transformed into gray domain data; based on the domain value of each pixel in the gray domain data, the flower screen detection result corresponding to the gray domain data is determined, and the flower screen detection result corresponding to the gray color domain data indicates whether the video frame to be detected has a black and gray screen. In response to the grayscale domain data corresponding to the flower screen detection result indicating that there is no black-gray screen, the grayscale domain data is binarized to obtain binary domain data. Based on the domain value of each pixel in the binary domain data, the flower screen detection result corresponding to the binary domain data is determined, and the flower screen detection result corresponding to the binary domain data indicates whether there is a striped flower screen in the video frame to be detected.

[0073] In some embodiments, the video image data is BGR data. The video image data acquisition module 310 acquires the video image data corresponding to the video frame to be detected, including: extracting the video frame to be detected from the received video stream or video file. Acquiring RGB data corresponding to the video frame to be detected. Converting the RGB data to BGR data.

[0074] In some embodiments, the video image data acquisition module 310 specifically performs the acquisition of RGB data corresponding to the video frame to be detected, including: extracting data corresponding to the video frame to be detected from the received video stream or video file, decoding and rendering the data corresponding to the video frame to be detected, and obtaining the RGB data.

[0075] In some embodiments, the step of converting RGB data to BGR data is implemented by an open source computer vision library.

[0076] See also Figure 4 , Figure 4It is a schematic diagram of the framework of an embodiment of the flower screen detection device of the present application. The flower screen detection device 40 includes a memory 41 and a processor 42 coupled to each other, and the processor 42 is used to execute the program instructions stored in the memory 41 to implement the steps in any of the above flower screen detection method embodiments. In a specific implementation scenario, the flower screen detection device 40 may include but is not limited to: a microcomputer, a server, and in addition, the flower screen detection device 40 may also include mobile devices such as laptops and tablet computers, which are not limited here.

[0077] Specifically, the processor 42 is used to control itself and the memory 41 to implement the steps in any of the above-mentioned flower screen detection method embodiments. The processor 42 can also be called a CPU (Central Processing Unit). The processor 42 may be an integrated circuit chip with signal processing capabilities. The processor 42 can also be a general-purpose processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field-programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. In addition, the processor 42 can be implemented by an integrated circuit chip.

[0078] See also Figure 5 , Figure 5 The computer-readable storage medium 50 stores program instructions 51 that can be executed by a processor, and the program instructions 51 are used to implement the steps in any of the above-mentioned embodiments of the screen distorting detection method.

[0079] In some embodiments, the functions or modules included in the device provided by the embodiments of the present disclosure can be used to execute the method described in the above method embodiments. The specific implementation can refer to the description of the above method embodiments, and for the sake of brevity, it will not be repeated here.

[0080] The above description of various embodiments tends to emphasize the differences between the various embodiments. The same or similar aspects can be referenced to each other, and for the sake of brevity, they will not be repeated herein.

[0081] In the several embodiments provided in the present application, it should be understood that the disclosed methods and devices can be implemented in other ways. For example, the device implementation described above is only schematic. For example, the division of modules or units is only a logical function division. There may be other division methods in actual implementation, such as units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0082] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0083] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) or a processor (processor) to perform all or part of the steps of each implementation method of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program code.

Claims

1. A method for detecting a distorted screen, characterized in that: The method comprises: Obtaining video image data corresponding to the video frame to be detected; Performing at least one image domain transformation on the video image data to obtain at least one image domain data, wherein each image domain data corresponds to a different type of screen distortion; Based on the domain value of each pixel in each of the image domain data, the flower screen detection result corresponding to each of the image domain data is determined, wherein the flower screen detection result represents whether the video frame to be detected has the flower screen type corresponding to the image domain data.

2. The method according to claim 1, characterized in that The image domain data includes at least one of HSV color domain data, grayscale domain data, and binary domain data. The type of flower screen corresponding to the HSV color domain data is a green screen, the type of flower screen corresponding to the grayscale domain data is a black-gray screen, and the type of flower screen corresponding to the binary domain data is a striped flower screen.

3. The method according to claim 2, characterized in that The step of determining the flower screen detection result corresponding to each image domain data based on the domain value of each pixel in each image domain data comprises at least one of the following steps: In response to the image domain data including HSV color domain data, based on the domain value of each pixel in the HSV color domain data, a first number of green pixels in the HSV color domain data is obtained by counting, and it is detected that the first number of green pixels meets the green screen condition, and it is determined that the video frame to be detected has a green screen; In response to the image domain data including grayscale domain data, the domain value of each pixel in the grayscale domain data is counted to obtain at least one first statistical value, and when the at least one first statistical value is detected to meet a black-gray screen condition, it is determined that the to-be-detected video frame has a black-gray screen, wherein the at least one first statistical value includes at least one of a central tendency statistical value and a discrete degree statistical value; In response to the image domain data including binary domain data, contour detection is performed on the binary domain data to obtain a number of contours in the video frame to be detected, the number of first contours of preset shapes among the several contours is counted, and when it is detected that the first number of contours meets the striped screen condition, it is determined that the video frame to be detected has a striped screen.

4. The method according to claim 3, characterized in that The green screen condition is that the first number of green pixels is greater than a first threshold, or the green screen condition is that the first number of green pixels is greater than the first threshold and the adjacent number difference is greater than a second threshold, and the adjacent number difference is the difference between the first number of green pixels and the second number of green pixels corresponding to the previous video frame; And / or, the obtaining the number of first green pixels in the HSV color gamut data by counting the domain value of each pixel in the HSV color gamut data comprises: In the HSV color gamut data, pixels whose hue dimension is within the green hue range are extracted, and the number of the extracted pixels is counted as the first green pixel number.

5. The method according to claim 3, characterized in that: The black-gray screen condition is that each of the first statistical values ​​exceeds a corresponding statistical value range, or the black-gray screen condition is that each of the first statistical values ​​exceeds a corresponding statistical value range and each adjacent statistical difference exceeds a corresponding statistical difference range, and the adjacent statistical difference is a difference between the first statistical value and a second statistical value corresponding to a previous video frame; And / or, the central tendency statistic is the mean, and the dispersion statistic is the standard deviation.

6. The method according to claim 3, characterized in that The stripe screen condition includes at least one of the following: the number of the first contours is greater than a third threshold, and the difference in the number of adjacent contours is greater than a fourth threshold, wherein the difference in the number of contours is the difference between the number of the first contours and the number of the second contours of a preset shape in the previous video frame; And / or, the preset shape is a rectangle; And / or, the contour detection is implemented by an open source computer vision library.

7. The method according to claim 2, characterized in that The image domain data includes HSV color domain data, gray domain data, and binary domain data; The performing at least one image domain transformation on the video image data to obtain at least one image domain data, and determining the flower screen detection result corresponding to each image domain data based on the domain value of each pixel in each image domain data, includes: Convert the video image data into the HSV color gamut data; Based on the domain value of each pixel in the HSV color gamut data, determine the flower screen detection result corresponding to the HSV color gamut data, and the flower screen detection result corresponding to the HSV color gamut data indicates whether there is a green screen in the video frame to be detected; In response to the color distorted screen detection result corresponding to the HSV color domain data being that there is no green screen, transforming the video image data into the gray domain data; Based on the domain value of each pixel in the grayscale domain data, determining the flower screen detection result corresponding to the grayscale domain data, wherein the flower screen detection result corresponding to the grayscale domain data indicates whether there is a black-gray screen in the video frame to be detected; In response to the color-distortion screen detection result corresponding to the grayscale domain data being that a black-gray screen does not exist, binarizing the grayscale domain data to obtain the binary domain data; Based on the domain value of each pixel in the binary domain data, a striped screen detection result corresponding to the binary domain data is determined, and the striped screen detection result corresponding to the binary domain data indicates whether the video frame to be detected has a striped screen.

8. The method according to claim 1, characterized in that The video image data is BGR data; The step of obtaining video image data corresponding to the video frame to be detected includes: Extracting a video frame to be detected from a received video stream or video file; Obtaining RGB data corresponding to the video frame to be detected; Convert the RGB data to BGR data.

9. The method according to claim 8, characterized in that The obtaining of RGB data corresponding to the to-be-detected video frame includes: Extracting data corresponding to the video frame to be detected from the received video stream or video file; Decoding and rendering the data corresponding to the video frame to be detected to obtain the RGB data; And / or, the step of converting the RGB data into BGR data is implemented by an open source computer vision library.

10. A screen flower detection device, characterized in that: The flower screen detection device includes a memory and a processor, the memory stores program instructions, and the processor is used to execute the program instructions to implement the flower screen detection method according to any one of claims 1 to 9.

11. A computer-readable storage medium, characterized in that: The computer-readable storage medium is used to store program instructions, and the program instructions can be executed to implement the screen distortion detection method according to any one of claims 1 to 9.

Images