CN115423795A - Still frame detection method, electronic device and storage medium - Google Patents

Abstract

The present application discloses a still frame detection method, an electronic device and a storage medium. The still frame detection method includes: acquiring an image to be detected and a first reference image located one frame before the image to be detected; The algorithm, the image to be detected and the first reference image detect a first moving object, then judge whether the first moving object is in a typing state according to the first moving direction of the first moving object; if determined The first moving object is not in the typing state, then according to the optical flow algorithm, the image to be detected and the first reference image, detect the second moving object; if the second moving object is not detected in the image to be detected target, it is determined that the currently displayed image is a still frame image. The present application solves the technical problem of low accuracy of static frame detection in the prior art.

Description

Still frame detection method, electronic device and storage medium

technical field

The present application relates to the technical field of image display, and in particular to a still frame detection method, electronic equipment and a storage medium.

Background technique

With the rapid development of industries such as radio and television and video surveillance, the degree of digitalization of video is accelerating, and the requirements for accurate, safe and continuous video playback are getting higher and higher. Detect whether there is movement of objects in the video image during video playback, that is, judge whether the video signal is static, so as to make timely remedial measures, ensure the integrity and stability of TV program broadcasting, and ensure the real-time performance of video surveillance and security. For example, in the field of video surveillance, security personnel need to observe multiple display screens one by one to ensure that no dangerous situations appear in the video. If there are too many display screens, security personnel will inevitably be unable to handle them. If the video is still and the dangerous situation is not detected in time, this will bring great safety hazards. Manual detection of whether the video is in a static state has a high risk factor, and it is prone to fatigue of the staff and cannot detect abnormal situations in time. At present, the method for detecting still frames in video signals is mainly to detect the whole picture. By partitioning the image, the number of still frame partitions is determined, and the number of still frame partitions greater than the threshold is determined as a still image. However, this method It is easy to cause misjudgment when the motion range is small or the moving target is small, resulting in a decrease in the accuracy of still frame detection.

Contents of the invention

The main purpose of the present application is to provide a still frame detection method, an electronic device and a storage medium, aiming at solving the technical problem of low accuracy of still frame detection in the prior art.

In order to achieve the above purpose, the present application provides a still frame detection method, the still frame detection method comprising:

Acquiring an image to be detected and a first reference image located one frame before the image to be detected;

If the first moving object is detected according to the inter-frame difference algorithm, the image to be detected and the first reference image, then according to the first moving direction of the first moving object, it is judged whether the first moving object is in Typing status;

If it is determined that the first moving object is not in the typing state, then detecting the second moving object according to the optical flow algorithm, the image to be detected and the first reference image;

If no second moving object is detected in the image to be detected, it is determined that the currently displayed image is a still frame image.

Optionally, if the second moving target is not detected in the image to be detected, the step of determining that the image to be detected is a still picture includes:

If no second moving object is detected in the image to be detected, then determining that the currently displayed image is a suspected still frame image;

Acquiring historical detection results of suspected still frame pictures corresponding to the historical images to be detected by a preset first number of frames before the image to be detected;

If it is determined that the number of suspected freeze-frame pictures is greater than or equal to a preset number of times threshold in the history detection results of each suspected freeze-frame picture, then it is determined that the currently displayed picture is a freeze-frame picture.

Optionally, the optical flow algorithm includes a sparse optical flow algorithm, and the step of detecting the second moving target according to the optical flow algorithm, the image to be detected and the first reference image includes:

detecting a first corner point in the first reference image;

According to the sparse optical flow algorithm, a second corner point matching the first corner point is determined from the image to be detected, and the matched first corner point and the second corner point form a corner point pair;

According to the relative positional relationship of the pair of corner points, it is judged whether there is a second moving object in the image to be detected.

Optionally, the step of judging whether there is a second moving object in the image to be detected according to the relative positional relationship between the corner point pairs includes:

On the preset canvas, determining the positions of the first corner point and the second corner point in each pair of corner points;

connecting the first corner point and the second corner point in each of the corner point pairs to obtain a connecting line between the corner point pairs;

If the pair of corner points is detected in the preset non-edge area of the canvas, it is determined that there is a second moving object in the image to be detected.

Optionally, the optical flow algorithm includes a dense optical flow algorithm, and the step of detecting the second moving target according to the optical flow algorithm, the image to be detected and the first reference image includes:

determining an optical flow vector corresponding to each pixel in the image to be detected according to an optical flow algorithm, the image to be detected, and the first reference image;

generating an optical flow image according to the optical flow mode length and the optical flow direction of the optical flow vector;

If a moving image is detected in the optical flow image, it is determined that a second moving object exists in the image to be detected.

Optionally, if a moving image is detected in the optical flow image, the step of determining the second moving object existing in the image to be detected includes:

Grayscale the optical flow image to obtain an optical flow grayscale image;

performing adaptive binarization on the optical flow grayscale image to obtain an image to be detected;

determining a second contour of the moving image in the image to be detected according to a contour finding algorithm;

If the second contour is detected in the preset non-edge area of the image to be detected, then determine the second moving object existing in the image to be detected.

Optionally, the step of judging whether the first moving object is in a typing state according to the first moving direction of the first moving object includes:

According to the contour finding algorithm, determine the respective target contours corresponding to each of the first moving targets;

determining, from each of said object contours, the largest object contour having the largest area;

determining the minimum circumscribed rectangle of the maximum target contour;

If it is determined according to the inclination angle of the minimum circumscribed rectangle that the minimum circumscribed rectangle is in a horizontal state or a vertical state, then it is determined that the first moving object is in a typing state.

Optionally, after the step of determining that the currently displayed picture is a still frame picture, it also includes:

Acquiring a second reference image that is preset a second number of frames before the image to be detected;

determining the first peak signal-to-noise ratio corresponding to the image to be detected, and the average signal-to-noise ratio of the second peak signal-to-noise ratio corresponding to each of the second reference images;

If the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is greater than or equal to a preset difference threshold, it is determined that the still-frame picture is disturbed by noise;

If the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is smaller than a preset difference threshold, it is determined that the still-frame picture is a completely still-frame picture.

The present application also provides an electronic device, the electronic device is a physical device, and the electronic device includes: a memory, a processor, and the still frame detection method stored in the memory and operable on the processor A program of the still frame detection method, when the program of the still frame detection method is executed by a processor, the steps of the above still frame detection method can be realized.

The present application also provides a storage medium, the storage medium is a computer-readable storage medium, and the computer-readable storage medium stores a program for implementing a still frame detection method, and the program for the still frame detection method is executed by a processor When realizing the steps of the still frame detection method as above.

The present application also provides a computer program product, including a computer program, when the computer program is executed by a processor, the steps of the above still frame detection method are realized.

The present application provides a still frame detection method, electronic equipment and a storage medium. By acquiring an image to be detected and a first reference image located one frame before the image to be detected, if according to the inter-frame difference algorithm, the image to be detected and the first reference image, if the first moving object is detected, then according to the first moving direction of the first moving object, it is judged whether the first moving object is in the typing state, and the front and rear through the frame difference algorithm are realized. The detection of whether the first moving object in the two frames of images is in the typing state, and then if it is determined that the first moving object is not in the typing state, then according to the optical flow algorithm, the image to be detected and the first reference image, Detecting the second moving object realizes the detection of the second moving object that is not in the typing state through the optical flow algorithm, and then if the second moving object is not detected in the image to be detected, it is determined that the current display screen is static Frame picture, realized still frame detection. Since the change of pixels in the overall picture is very small in the typing state, it is difficult to detect the still frame partition detection or the optical flow method, or it is easy to judge it as noise or error, resulting in misjudgment of the still frame picture. The difference algorithm can quickly detect the typing state, and for the second moving target that is not in the typing state, it can be further accurately detected by the optical flow method. The detection accuracy of motion changes is higher, so the accuracy of still frame detection can be effectively improved, and the technical problem of low accuracy of still frame detection in the prior art is overcome.

Description of drawings

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

In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, for those of ordinary skill in the art, In other words, other drawings can also be obtained from these drawings without paying creative labor.

FIG. 1 is a schematic structural diagram of an electronic device of a hardware operating environment involved in a still frame detection method in an embodiment of the present application;

FIG. 2 is a schematic diagram of a scene where the cursor moves in the embodiment of the present application;

FIG. 3 is a schematic flow diagram of an embodiment of the still frame detection method of the present application;

FIG. 4 is a schematic diagram of a scene of an inclination angle in an embodiment of the present application;

FIG. 5 is a schematic diagram of a scene of image storage format conversion in an implementation manner of the present application;

FIG. 6 is a schematic flowchart of another embodiment of a still frame detection method of the present application.

Explanation of reference numbers:

label name label name 1 previous frame position 2 current position 3 Minimum enclosing rectangle 4 horizontal line 5 Vertical line 6 slope

The realization of the purpose, functions and advantages of the present application will be further described in conjunction with the embodiments and with reference to the accompanying drawings.

detailed description

In order to make the above objects, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Apparently, the described embodiments are only some of the embodiments of the present invention, but not all of them. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

With the rapid development of industries such as radio and television and video surveillance, the degree of digitalization of video is accelerating, and the requirements for accurate, safe and continuous video playback are getting higher and higher. Detect whether there is movement of objects in the video image during video playback, that is, judge whether the video signal is static, so as to make timely remedial measures, ensure the integrity and stability of TV program broadcasting, and ensure the real-time performance of video surveillance and security. For example, in the field of video surveillance, security personnel need to observe multiple display screens one by one to ensure that no dangerous situations appear in the video. If there are too many display screens, security personnel will inevitably be unable to handle them. If the video is still and the dangerous situation is not detected in time, this will bring great safety hazards. Manual detection of whether the video is in a static state has a high risk factor, and it is prone to fatigue of the staff and cannot detect abnormal situations in time. The current detection method for still frames in video signals is mainly to use the similarity relationship between the front and back frames or the pixel distribution in the frame to determine whether there is a still frame in the picture. For example, the weighted sum of each frame is obtained based on the weighted accumulation of pixel YUV, and then The weighted sum is compared, but this method has a large amount of pixel data, and the weighted accumulation calculation will inevitably bring a lot of computing overhead. The algorithm is relatively complicated, takes up a lot of resources, and the detection result is not accurate. For example, to detect the whole picture, By partitioning the image, the number of still frame partitions is determined, and the number of still frame partitions greater than the threshold is judged as a still image. However, this method is likely to cause misjudgment when the motion range is small or the moving target is small, resulting in still frame detection. accuracy is reduced.

This application uses the inter-frame difference algorithm to quickly detect the first moving target in the typing state, and uses the optical flow method to detect the second moving target that is not in the typing state. For the small amplitude and small area changes caused by typing, through The inter-frame difference algorithm can quickly determine the direction of motion and perform accurate identification. For other small-amplitude and small-area changes, such as cursor movement, etc., especially small changes in complex backgrounds, although the results of visible changes in the screen are small, But the change process is not small. Referring to Figure 1, in Figure 1, the cursor moves from the position 1 of the previous frame to the current position 2, and there is no increase or decrease of any pixels in the overall picture, and the background itself is relatively complex, so the cursor moves from position 1 to current position 2. It is difficult to detect a movement change. It should be noted that in Figure 1, the font color in the background is also lightened to highlight the change of the cursor. In the actual image used for detection, the change of the cursor is more difficult to be found, but If the movement trajectory from position 1 of the previous frame to the current position 2 can be visualized, it can be easily detected. Therefore, this application detects the movement trajectory of the moving target through the optical flow method, and can more accurately detect small amplitude and A sporty target with small area changes.

Referring to FIG. 2 , FIG. 2 is a schematic structural diagram of an electronic device in a hardware operating environment involved in a still frame detection method in an embodiment of the present application. The electronic device in the embodiments of the present disclosure may include, but not limited to, a television, an OLED display, an LCD (Liquid Crystal Display, liquid crystal display), an LED (light-emitting diode, light-emitting diode) display, and the like.

As shown in FIG. 2 , the electronic device further includes: a processor 1001 , such as a central processing unit (Central Processing Unit, CPU), a communication bus 1002 , a network interface 1003 , and a memory 1004 . Wherein, the communication bus 1002 is used to realize connection and communication between these components. Optionally, the network interface 1003 may include a standard wired interface and a wireless interface (such as a Wireless-Fidelity (Wireless-Fidelity, WI-FI) interface). The memory 1004 may be a high-speed random access memory (Random Access Memory, RAM) memory, or a stable non-volatile memory (Non-Volatile Memory, NVM), such as a disk memory. Optionally, the memory 1004 may also be a storage device independent of the aforementioned processor 1001 .

Optionally, the terminal may further include a camera, an RF (Radio Frequency, radio frequency) circuit, a sensor, an audio circuit, a WiFi module, and the like. Among them, sensors such as light sensors, motion sensors and other sensors. Specifically, the light sensor may include an ambient light sensor and a proximity sensor, wherein the ambient light sensor may adjust the brightness of the display screen according to the brightness of the ambient light, and the proximity sensor may turn off the display screen and/or backlight. As a kind of motion sensor, the gravitational acceleration sensor can detect the magnitude of acceleration in various directions (generally three axes), and can detect the magnitude and direction of gravity when it is stationary, and can be used for applications that recognize the posture of mobile terminals (such as horizontal and vertical screen switching, Related games, magnetometer posture calibration), vibration recognition related functions (such as pedometer, tap), etc.; of course, the mobile terminal can also be equipped with other sensors such as gyroscope, barometer, hygrometer, thermometer, infrared sensor, etc. No longer.

Those skilled in the art can understand that the structure shown in FIG. 2 does not constitute a limitation on the operating equipment, and may include more or less components than shown in the figure, or combine some components, or arrange different components.

As shown in FIG. 2 , the memory 1004 as a storage medium may include an operating system, a data storage module, a network communication module, and a still frame detection program.

In the running device shown in FIG. 2 , the network interface 1003 is mainly used for data communication with other devices. The processor 1001 and the memory 1004 in the running device of the present invention can be set in the running device, and the running device calls the still frame detection program stored in the memory 1004 through the processor 1001, and performs the following operations:

Acquiring an image to be detected and a first reference image located one frame before the image to be detected;

If the first moving object is detected according to the inter-frame difference algorithm, the image to be detected and the first reference image, then according to the first moving direction of the first moving object, it is judged whether the first moving object is in Typing status;

If it is determined that the first moving object is not in the typing state, then detecting the second moving object according to the optical flow algorithm, the image to be detected and the first reference image;

If no second moving object is detected in the image to be detected, it is determined that the currently displayed image is a still frame image.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

If no second moving object is detected in the image to be detected, then determining that the currently displayed image is a suspected still frame image;

Acquiring historical detection results of suspected still frame pictures corresponding to the historical images to be detected by a preset first number of frames before the image to be detected;

If it is determined that the number of suspected freeze-frame pictures is greater than or equal to a preset number of times threshold in the history detection results of each suspected freeze-frame picture, then it is determined that the currently displayed picture is a freeze-frame picture.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

detecting a first corner point in the first reference image;

According to the sparse optical flow algorithm, a second corner point matching the first corner point is determined from the image to be detected, and the matched first corner point and the second corner point form a corner point pair;

According to the relative positional relationship of the pair of corner points, it is judged whether there is a second moving object in the image to be detected.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

On the preset canvas, determining the positions of the first corner point and the second corner point in each pair of corner points;

connecting the first corner point and the second corner point in each of the corner point pairs to obtain a connecting line between the corner point pairs;

If the pair of corner points is detected in the preset non-edge area of the canvas, it is determined that there is a second moving object in the image to be detected.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

determining an optical flow vector corresponding to each pixel in the image to be detected according to an optical flow algorithm, the image to be detected, and the first reference image;

generating an optical flow image according to the optical flow mode length and the optical flow direction of the optical flow vector;

If a moving image is detected in the optical flow image, it is determined that a second moving object exists in the image to be detected.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

Grayscale the optical flow image to obtain an optical flow grayscale image;

performing adaptive binarization on the optical flow grayscale image to obtain an image to be detected;

determining a second contour of the moving image in the image to be detected according to a contour finding algorithm;

If the second contour is detected in the preset non-edge area of the image to be detected, then determine the second moving object existing in the image to be detected.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

According to the contour finding algorithm, determine the respective target contours corresponding to each of the first moving targets;

determining, from each of said object contours, the largest object contour having the largest area;

determining the minimum circumscribed rectangle of the maximum target contour;

If it is determined according to the inclination angle of the minimum circumscribed rectangle that the minimum circumscribed rectangle is in a horizontal state or a vertical state, then it is determined that the first moving object is in a typing state.

Further, the processor 1001 may call the still frame detection program stored in the memory 1004, and also perform the following operations:

Acquiring a second reference image that is preset a second number of frames before the image to be detected;

determining the first peak signal-to-noise ratio corresponding to the image to be detected, and the average signal-to-noise ratio of the second peak signal-to-noise ratio corresponding to each of the second reference images;

If the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is greater than or equal to a preset difference threshold, it is determined that the still-frame picture is disturbed by noise;

If the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is smaller than a preset difference threshold, it is determined that the still-frame picture is a completely still-frame picture.

The embodiment of the present application provides a still frame detection method. In the first embodiment of the still frame detection method of the present application, referring to FIG. 3 , the still frame detection method includes the following steps:

Step S10, acquiring an image to be detected and a first reference image located one frame before the image to be detected;

In this embodiment, it should be noted that the still frame detection method is applied to a display device, and the display device may include but not limited to a television or a display linearly connected to a user device, such as an OLED display, an LCD, an LED display etc., the display can be linearly connected to the user equipment through HDMI (High Definition Multimedia Interface, high-definition multimedia interface), VGA (Video Graphics Array, video graphics array) interface or DP (DisplayPort, display interface), and then receive and output A video signal transmitted by user equipment is displayed, wherein the user equipment includes portable terminal equipment such as a notebook computer, a computer, a tablet computer, and a mobile phone.

Freeze frame refers to the situation that the image of the current frame does not change compared with the image of the previous frame. When the video data played by the display device is still frame video data, it means that the video data played by the display device is abnormally played, causing an abnormality. The reason may be hardware abnormality, software abnormality, or abnormality in video data stream transmission, etc. If there are continuous still frames, it may cause damage to the display screen.

Specifically, a frame of image is acquired from the video as the image to be detected, and an image of the previous frame of the image to be detected is acquired as the first reference image, wherein the method of acquiring the image to be detected includes acquiring from the video stream, Take a screenshot or take a photo, etc.

In an implementable manner, the specification adjustment method is to intercept the part corresponding to the region to be detected in the screenshot image, wherein the region to be detected can be determined according to actual needs, for example, according to the information The position of the prompt pop-up window when it appears on the screen is determined, and this type of information prompt pop-up window usually appears in the peripheral area of the screen, so the area to be detected can be a part of the center of the screenshot image, or it can be The area above the nth row of pixels from bottom to top in the screenshot image, etc. When the user equipment is unmanned, a pop-up window may also appear with a message prompt. This type of message prompt is usually irrelevant to whether the user equipment is operated. The screen is a dynamic screen, and the screen saver screen is exited. At this time, because the release of the screen saver screen is not due to user operation, it may re-enter the screen saver screen after a period of time, that is, the screen saver screen is raised due to a message prompt The process is essentially an invalid operation, and it will also consume the computing power and power of the device. By intercepting the area to be detected, the area that may cause misoperation is removed in advance, which can not only reduce the amount of calculation for subsequent moving target detection, It can also effectively avoid the above-mentioned situation of invalidly exiting the screen saver.

In this embodiment, through specification adjustment and grayscale adjustment, the calculation amount of subsequent moving object detection can be effectively reduced, and the detection efficiency of still frame detection can be improved.

Step S20, if a first moving object is detected according to the inter-frame difference algorithm, the image to be detected and the first reference image, then judge the first moving object according to the first moving direction of the first moving object Whether the target is in typing state;

In this embodiment, a differential operation is performed on the gray values of the image to be detected and the first reference image according to the inter-frame difference algorithm to obtain the absolute value of the gray difference between the two frames of images. If the gray value of a certain point If the absolute value of the difference is 0 or less than or equal to the preset threshold, it can be considered that there is no moving object at this point. If the absolute value of the gray level difference of a certain point is not 0 or greater than the preset threshold, it can be considered that there is a moving object at this point After that, in an implementable manner, the change between the kth frame and the k+1 frame image f _k (x, y), f _k+1 (x, y) uses a binary difference image D(x ,y) indicates that D(x,y)=1, if|f _k+1 (x,y)-f _k (x,y)|>T,else,0.

If the first moving object is not detected from the image to be detected according to the inter-frame difference algorithm, the image to be detected and the first reference image, it means that the two frames of images are almost completely overlapped, that is, the picture is in a static state , then it is determined that the currently displayed image is a still frame image.

If a first moving object is detected from the image to be detected according to the inter-frame difference algorithm, the image to be detected and the first reference image, then determine the first contour of the first moving object, according to the For the first contour, judge the first moving direction of the first moving object by means of image recognition technology or the inclination angle of the circumscribed rectangle, and then match the first moving direction with the preset typing moving direction, if the If the first movement direction matches the preset typing movement direction, then it is determined that the first movement object is in the typing state; if the first movement direction does not match the preset typing movement direction, then it is determined that the first movement The target is not in the typing state, wherein, the preset typing movement direction can be the horizontal direction or the vertical direction, and there is generally no regular horizontal movement or vertical movement in other movement states, so it can be based on the first Whether the moving direction is horizontal or vertical, quickly and easily judge whether the first moving object is in a typing state.

In an implementable manner, the step of determining the first contour of the first moving object includes: obtaining an initial contour obtained after an inter-frame difference algorithm operation, and using a morphological operation to enlarge the initial contour to obtain the The first contour of the first moving object, wherein the morphological operation includes a dilation operation, a dilation operation after erosion, or an erosion operation after dilation, and the like. When detecting by the inter-frame difference algorithm, if the object is almost completely overlapped in the two frames of images, the object cannot be detected, that is, the picture is in a static state. If a moving target can be detected, the outline of the moving target will be obtained, but the moving There will be holes inside the target, so more accurate boundary contours can be obtained by means of expansion, erosion, etc., especially for the first moving target with a small motion range, the initial contour obtained by the inter-frame difference algorithm may be small, and the size of the defined structural element Then use the expansion operation to enlarge the initial contour, which can make the display of the first moving target in the image more obvious and accurate.

In an implementable manner, the step of determining the first contour of the first moving object includes: determining a binary image of the first moving object through adaptive binarization, and performing a contour finding algorithm on the first contour of the first moving object. Extracting the contour in the binary image to obtain the first contour of the first moving object. Through adaptive binarization, according to the distribution characteristics of image pixel neighborhood blocks, the binarization threshold of the determined area is adaptively determined, and the binarization threshold is automatically used to segment the foreground and the background, the foreground is the outline of the moving object, and the Adaptive binarization includes a Gaussian weighted sum algorithm. The Gaussian weighted sum algorithm weights the pixels around the center point of the region according to the Gaussian function to calculate their distance from the center point, which can effectively cover the scene of pixel variation difference.

Optionally, the step of judging whether the first moving object is in a typing state according to the first moving direction of the first moving object includes:

Step A10, according to the contour finding algorithm, determine the target contour corresponding to each of the first moving objects;

In this embodiment, the binary difference image generated after the operation of the inter-frame difference algorithm is acquired, and the target contour corresponding to each of the first moving objects is extracted from the binary difference image according to the contour finding algorithm.

Step A20, determining the largest target contour with the largest area from each of the target contours;

In this embodiment, the area sizes of the target contours are compared, and the target contour with the largest area is determined as the largest target contour.

In an implementable manner, the step of determining the largest target contour with the largest area from each of the target contours includes: determining any target contour as the largest target contour, traversing the areas of all target contours, if there is A second target contour with a larger area than the maximum target contour is determined as the maximum target contour until all target contours are traversed, thereby determining the maximum target contour.

Step A30, determining the minimum circumscribed rectangle of the maximum target contour;

Step A40, if it is determined according to the inclination angle of the minimum circumscribed rectangle that the minimum circumscribed rectangle is in a horizontal state or a vertical state, then it is determined that the first moving object is in a typing state.

In this embodiment, the minimum circumscribed rectangle of the maximum target contour is determined, and the side of the minimum circumscribed rectangle is compared with a horizontal line or a vertical line to determine the inclination angle of the minimum circumscribed rectangle. For example, referring to FIG. 4 , it can be The angle between the broadside of the minimum circumscribed rectangle 3 and the horizontal line 4 is determined as the angle of inclination 6, and the angle between the broadside of the minimum circumscribed rectangle 3 and the vertical line 5 can also be determined is the angle of inclination (not shown in the figure). According to whether the inclination angle is in a preset angle range, it is judged whether the minimum circumscribed rectangle is in a horizontal state or a vertical state, and if the inclination angle is in a preset angle range, then it is determined that the minimum circumscribed rectangle is in a horizontal state or a vertical state , it is determined that the first moving object is in the typing state, if the tilt angle is not in the preset angle range, it is determined that the minimum circumscribed rectangle is not in the horizontal state or vertical state, then it is determined that the first moving object is not in the Typing state, wherein, the inclination angle is the size of the angle formed between the longer side or the shorter side of the minimum circumscribed rectangle and the horizontal line or vertical line, because the minimum circumscribed rectangle detected under the typing state The long side may be horizontal or vertical, so the preset angle range can be 0 degrees or 90 degrees, and a certain error range can also be set, that is, the preset angle range can be 0 to 5 degrees or 85 degrees ~90 degrees, such as 0 degrees, 3 degrees, 5 degrees, 85 degrees, 88 degrees, 90 degrees, etc.

Step S30, if it is determined that the first moving object is not in the typing state, then detect the second moving object according to the optical flow algorithm, the image to be detected and the first reference image;

In this embodiment, it should be noted that although the calculation speed of the inter-frame difference algorithm is fast, the detection accuracy is low for the case of a small motion range or a small moving target. If the first The moving object is not in the typing state, and the first moving object detected at this time may be generated by a dynamic picture, or may be generated by noise in a still frame picture, so further judgment is required.

Specifically, if it is determined that the first moving object is not in the typing state, the second moving object in the image to be detected and the first reference image is detected according to an optical flow algorithm, wherein the optical flow algorithm Including dense optical flow algorithm and sparse optical flow algorithm.

If it is determined that the first moving object is in a typing state, then it is determined that the currently displayed image is a dynamic image.

It should be noted that the number of first moving objects detected from the image to be detected may be zero, one or more, and the more the number of the first moving objects is, the more time-consuming it is to judge the typing state. long, and may be a misjudgment caused by noise. Therefore, if more than one first moving object is detected, each first moving object can be judged according to the first moving direction corresponding to each of the first moving objects. Whether the first moving object is in the typing state, if it is determined that any one of the first moving objects is in the typing state, then determine that the current display screen is a dynamic picture, if it is determined that each of the first moving objects is not in the typing state, then according to The optical flow algorithm, the image to be detected and the first reference image are used to detect a second moving object; if more than one first moving object is detected, it is also possible to select from each of the first moving objects Select a preset number (such as 1, 2, etc.) of the first moving objects to be detected, and determine each of the first moving objects to be detected according to the first moving directions corresponding to each of the first moving objects to be detected Whether it is in the typing state, if it is determined that any one of the first moving objects to be detected is in the typing state, then it is determined that the current display screen is a dynamic picture, if it is determined that each of the first moving objects to be detected is not in the typing state, then according to the light flow algorithm, the image to be detected and the first reference image to detect a second moving object, wherein the method of selecting a preset number of first moving objects to be detected from each of the first moving objects may be Selecting a preset number of first moving objects to be detected according to the contour area and position corresponding to each of the first moving objects, for example, selecting a first moving object whose contour area is greater than a preset area threshold as the first moving object to be detected Target, select the first moving object whose position is located in the preset non-edge area as the first moving object to be detected, etc., the first moving object whose contour area is smaller than the preset area threshold, or whose position is located in the edge area of the image may be noise or detection When errors occur, pre-filtering can not only reduce the calculation amount, but also improve the detection accuracy.

Optionally, the optical flow algorithm includes a dense optical flow algorithm, and the step of detecting the second moving target according to the optical flow algorithm, the image to be detected and the first reference image includes:

Step B10, according to the optical flow algorithm, the image to be detected and the first reference image, determine the optical flow vector corresponding to each pixel in the image to be detected;

In this embodiment, it should be noted that the optical flow refers to the change of the gray value between pixels on the video image, that is, the instantaneous velocity of the moving object. The optical flow method studies the optical flow field of the video image sequence, using The difference between the optical flow information of the moving object in the image and the optical flow information of the background is used to determine the position of the moving object, and then detect the moving target. The dense optical flow algorithm uses two adjacent frames of images to estimate the optical flow vector of the object. Its implementation method It is: first use the method of polynomial expansion, use a quadratic polynomial to approximate the expression of the neighborhood of each pixel, and then estimate the displacement vector of the optical flow field by analyzing the polynomial expansion coefficients of the two frames of pixels before and after. The instantaneous speed of all pixels in the image has high accuracy, high robustness and reliability, and meets the actual needs of video still frame detection.

Specifically, according to the dense optical flow algorithm, compare the pixels in the image to be detected and the first reference image one by one, determine the optical flow field corresponding to the image to be detected, and determine the The optical flow vector corresponding to each pixel, it should be noted that the optical flow vector corresponding to the changed pixel has the optical flow mode length and the optical flow direction, and the optical flow mode of the optical flow vector corresponding to the unchanged pixel The length is 0 or a minimum value, and the optical flow vector at this time is a point.

In an implementable manner, if the image to be detected and/or the first reference image is an image in RGBA storage format, since the transparency of the A channel has little influence in the still frame detection process, and will bring Therefore, the image to be detected and/or the first reference image in RGBA storage format can be converted into RGB storage format. Referring to Fig. 5 , the method of converting the image to be detected and/or the first reference image in the RGBA storage format into the RGB storage format includes: from the image data in the RGBA storage format, take out four positions at intervals The single-channel data of R channel, G channel and B channel are merged with the single-channel data of R channel, G channel and B channel corresponding to each pixel to obtain an image in RGB storage format.

In an implementable manner, before the step of determining the optical flow vector corresponding to each pixel in the image to be detected according to the optical flow algorithm, the image to be detected and the first reference image, further It may include: performing grayscale conversion on the image to be detected and the first reference image, so as to reduce the amount of computation.

In an implementable manner, before the step of determining the optical flow vector corresponding to each pixel in the image to be detected according to the optical flow algorithm, the image to be detected and the first reference image, further It may include: performing compression processing on the image to be detected and the first reference image, reducing the amount of computation by reducing the size of the image or reducing the quality of the image.

Step B20, generating an optical flow image according to the optical flow mode length and the optical flow direction of the optical flow vector;

In this embodiment, specifically, the optical flow component in the x direction and the optical flow component in the y direction of each pixel are determined, based on the size and direction of the optical flow component in the x direction and the optical flow component in the y direction, that is converting the coordinates of the optical flow component in the x direction and the optical flow component in the y direction from Cartesian coordinates to polar coordinates, and then determining the optical flow mode length and the optical flow direction of the optical flow vector corresponding to each pixel in the image to be detected, The optical flow direction is expressed in radians, the optical flow direction is used as hue information, the optical flow modulus length is used as saturation information, and the brightness information is determined as a preset value, or after normalizing the optical flow modulus length , as the luminance information, the luminance of the optical flow image is made to float within a certain range, the color of the HSV color space corresponding to each of the optical flow vectors can be determined, and the optical flow image is obtained. In the optical flow image, there is no The moving parts are almost the same, and the colors presented are almost the same, which can be used as the background color of the optical flow image, while the moving parts, due to their different speeds of movement, show different colors, which can be displayed on the background color. A corresponding moving image is formed.

Step B30, if a moving image is detected in the optical flow image, then determine a second moving object existing in the image to be detected.

In this embodiment, specifically, the moving image in the optical flow image is identified through image recognition technology, contour recognition algorithm, etc., and if a moving image is detected in the optical flow image, it is determined that the pending Detecting the second moving object existing in the image, and determining the second moving object not existing in the image to be detected if no moving image is detected in the optical flow image.

Optionally, if a moving image is detected in the optical flow image, the step of determining the second moving object existing in the image to be detected includes:

Step B41, grayscale the optical flow image to obtain an optical flow grayscale image;

Step B42, performing adaptive binarization on the optical flow grayscale image to obtain an image to be detected;

Step B43, determining a second contour of the moving image in the image to be detected according to a contour finding algorithm;

Step B44, if the second contour is detected in the preset non-edge area of the image to be detected, then determine the second moving object existing in the image to be detected.

In this embodiment, specifically, the optical flow image is grayscaled to obtain an optical flow grayscale image, and the optical flow grayscale image is adaptively binarized to obtain an image to be detected. The algorithm performs contour extraction on the image to be detected. If no second contour can be extracted, it is determined that there is no second moving object in the image to be detected. If at least one second contour of the moving image can be extracted, then Traversing each of the second contours in turn, detecting whether the position of each of the second contours is in the preset non-edge area of the image to be detected, if the second contour is detected in the preset non-edge area of the image to be detected If the second contour is not detected in the preset non-edge area of the image to be detected, it is determined that there is no second moving object in the image to be detected. The second moving object, wherein, the binary grayscale image only contains image blocks of moving pixels, and other regions except the moving pixels are black backgrounds, and the preset non-edge region may be the An area in the image that is greater than or equal to preset pixels (for example, 20 pixels, 30 pixels, etc.) from the edge of the image.

In an implementable manner, the step of detecting whether each of the second contours is in a preset non-edge area of the binarized image includes: traversing all the second contours, drawing each of the second contours respectively The minimum circumscribed rectangle of the contour, determine the center point coordinates of each of the minimum circumscribed moments, and judge whether each of the center point coordinates is in the preset non-edge area of the binarized image.

In the actual development, it is found that there will be noise interference at the edge of the image, which is not conducive to the judgment of the still frame picture. Filtering out the moving target at the edge of the image can effectively reduce the noise interference.

Step S40, if the second moving object is not detected in the image to be detected, it is determined that the currently displayed image is a still frame image.

In this embodiment, if the second moving object is not detected in the image to be detected, it is determined that the currently displayed image is a still frame image; if the second moving object is detected in the image to be detected, it is determined that the current The display screen is a dynamic screen.

Optionally, if the second moving target is not detected in the image to be detected, the step of determining that the image to be detected is a still picture includes:

Step S41, if no second moving object is detected in the image to be detected, then determine that the currently displayed image is a suspected still-frame image;

Step S42, acquiring historical detection results of suspected still frames corresponding to historical images to be detected that are preset for a first number of frames before the image to be detected;

Step S43 , if the number of suspected freeze-frame images in the history detection results of each of the suspected freeze-frame images is greater than or equal to a preset number threshold, then it is determined that the currently displayed image is a freeze-frame image.

In this embodiment, specifically, if no second moving object is detected in the image to be detected, it is determined that the currently displayed image is a suspected still frame image, and a first number of objects is preset before acquiring the image to be detected. The historical detection results of suspected still frame pictures corresponding to the historical image to be detected of the frame, determine the number of times of suspected still frame pictures in each of the suspected still frame picture history detection results, if the number of times of the suspected still frame pictures is determined to be greater than or equal to the preset number of times threshold, then it is determined that the currently displayed picture is a still frame picture, and if the number of times it is determined to be a suspected still frame picture is less than the preset number of times threshold, then it is determined that the current displayed picture is a dynamic picture, which can be effectively reduced by increasing the number of determination times Misjudgment, improve the accuracy of the test results.

In an implementable manner, the preset number of thresholds is half of the preset first number, that is, if the number of times of the suspected freeze-frame picture is greater than or equal to the preset first number half, then it is determined that the currently displayed picture is a still frame picture. For example, if the preset first number is 4, a total of 4 historical detection results of suspected still frame pictures can be obtained. The result is a suspected still frame picture, so among the 5 judgment results, at least 3 suspected still frame pictures were detected, more than half of them, that is, more than half of the judgment results were suspected still frame in the historical detection results and this detection result Therefore, it can be judged that the currently displayed picture is a still frame picture.

Optionally, after the step of determining that the currently displayed picture is a still frame picture, it also includes:

Step S50, acquiring a second reference image that is preset a second number of frames before the image to be detected;

Step S60, determining the first peak signal-to-noise ratio corresponding to the image to be detected, and the average signal-to-noise ratio of the second peak signal-to-noise ratio corresponding to each of the second reference images;

Step S70, if the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is greater than or equal to a preset difference threshold, it is determined that the still-frame picture is disturbed by noise;

Step S80, if the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is smaller than a preset difference threshold, determine that the still-frame picture is a completely still-frame picture.

In this embodiment, specifically, the second reference image that is preset a second number of frames before the image to be detected is obtained, the first peak signal-to-noise ratio corresponding to the image to be detected is determined, and each of the second calculating the difference between the first peak signal-to-noise ratio and the mean signal-to-noise ratio with reference to the second peak signal-to-noise ratio corresponding to the image, if the difference is greater than or equal to a preset difference threshold, it is determined that the still frame picture is disturbed by noise. In an implementable manner, noise prompt information can be further generated to remind the user that the current detection result may be disturbed by noise; if the difference is less than the preset difference value threshold, then it is determined that the still picture is a completely still picture, wherein the preset second number may be the same as or different from the preset first number, and the peak signal-to-noise ratio represents the maximum possible signal power and The ratio of the destructive noise power that affects its representation accuracy can be used to judge the similarity of two images. The larger the peak signal-to-noise ratio between the images, the more similar they are. The corresponding peak value of the image to be detected in this embodiment The signal-to-noise ratio refers to the peak signal-to-noise ratio between the image to be detected and its previous frame image.

In an implementable manner, after it is determined that the still-frame picture is disturbed by noise, it is also possible not to judge whether the current display picture is a still-frame picture temporarily, and return to the execution step: acquire the image to be detected and locate the image to be detected The first reference image of the previous frame of the image is used to obtain the next frame of image as the image to be detected, and the number of samples is increased to reduce the impact of noise or misjudgment on the final detection result.

In an implementable manner, after the step of determining that the currently displayed image is a dynamic image, it further includes:

Acquiring a third reference image that is preset a third number of frames before the image to be detected;

determining the first peak signal-to-noise ratio corresponding to the image to be detected, and the average signal-to-noise ratio of the second peak signal-to-noise ratio corresponding to each of the second reference images;

If the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is greater than or equal to a preset difference threshold, it is determined that the dynamic picture is a dynamic picture with a large change;

If the difference between the first peak signal-to-noise ratio and the average value of the signal-to-noise ratio is smaller than a preset difference threshold, it is determined that the dynamic picture is a dynamic picture with small amplitude changes.

In this embodiment, specifically, a third reference image that is preset a third number of frames before the image to be detected is obtained, the first peak signal-to-noise ratio corresponding to the image to be detected is determined, and each of the third Referring to the SNR average of the third PSNR corresponding to the image, calculating the difference between the first PSNR and the SNR average of the third PSNR, if the first The difference between the peak signal-to-noise ratio and the signal-to-noise ratio average value of the third peak signal-to-noise ratio is greater than or equal to a preset difference threshold, then it is determined that the dynamic picture is a dynamic picture with a large change; if the first If the difference between the peak signal-to-noise ratio and the average value of the signal-to-noise ratio of the third peak signal-to-noise ratio is less than a preset difference threshold, it is determined that the dynamic picture is a dynamic picture with a small change, wherein the preset first The third quantity may be the same as or different from the preset first quantity or the preset second quantity.

In this embodiment, by acquiring the image to be detected and the first reference image located one frame before the image to be detected, if the first reference image is detected according to the inter-frame difference algorithm, the image to be detected and the first reference image A moving object, then according to the first moving direction of the first moving object, it is judged whether the first moving object is in the typing state, and it is realized whether the first moving object in the two frames of images before and after is in the typing state through the frame difference algorithm. The detection of the typing state, and then if it is determined that the first moving object is not in the typing state, then according to the optical flow algorithm, the image to be detected and the first reference image, detect the second moving object, and realize the detection of the second moving object through optical flow. The algorithm detects the second moving object that is not in the typing state, and then if the second moving object is not detected in the image to be detected, it is determined that the current display image is a still frame image, and the still frame detection is realized. Since the change of pixels in the overall picture is very small in the typing state, it is difficult to detect the still frame partition detection or the optical flow method, or it is easy to judge it as noise or error, resulting in misjudgment of the still frame picture. The difference algorithm can quickly detect the typing state, and for the second moving target that is not in the typing state, it can be further accurately detected by the optical flow method. The detection accuracy of motion changes is higher, so the accuracy of still frame detection can be effectively improved, and the technical problem of low accuracy of still frame detection in the prior art is overcome.

Further, in another embodiment of the still frame detection method of the present application, referring to FIG. 5 , the optical flow algorithm includes a sparse optical flow algorithm, and the optical flow algorithm, the image to be detected and the first reference image, the step of detecting the second moving target includes:

Step C10, detecting a first corner point in the first reference image;

In this embodiment, the first corner in the first reference image is detected by a corner detector, wherein the corner detector includes a Shi-Tomasi corner detector, etc., specifically, a fixed The window slides in any direction on the first reference image, and compares the degree of grayscale change of pixels in the fixed window before and after sliding. If the fixed window slides in any direction, the If the variation of the grayscale value exceeds the preset grayscale threshold, it can be considered that there are corner points in the window, and thus the positions of the corner points in the first reference image can be obtained, and a list of initial corner point positions can be generated.

In an implementable manner, before the step of detecting the first corner point in the first reference image, it may further include: performing grayscale conversion on the image to be detected and the first reference image.

Step C20, according to the sparse optical flow algorithm, determine a second corner point that matches the first corner point from the image to be detected, and the matched first corner point and second corner point form a corner point pair;

In this embodiment, specifically, each of the first corner points is tracked according to the sparse optical flow algorithm to determine the new corner position of each of the first corner points in the image to be detected, and according to the new corner The point position determines the second corner point matching the first corner point from the image to be detected, and the matched first corner point and second corner point form a corner point pair. It should be noted that, according to the sparse light The new corner position corresponding to each of the first corner points determined by the flow algorithm may exceed the image range of the image to be detected, and the new corner position beyond the image range will disappear from the image to be detected, so from No matching second corner point can be determined in the image to be detected.

Step C30, judging whether there is a second moving object in the image to be detected according to the relative positional relationship between the corner point pairs.

In this embodiment, specifically, according to whether the position of the second corner point in the corner point pair changes relative to the position of the first corner point in the same corner point pair, it can be judged whether there is For the second moving object, if the position of the second corner point in the corner point pair changes relative to the position of the first corner point in the same corner point pair, it is determined that there is a second moving object in the image to be detected, If the position of the second corner point in the corner point pair does not change relative to the position of the first corner point in the same corner point pair, it is determined that there is no second moving object in the image to be detected.

Optionally, the step of judging whether there is a second moving object in the image to be detected according to the relative positional relationship between the corner point pairs includes:

Step C31, on the preset canvas, determine the positions of the first corner point and the second corner point in each pair of corner points;

Step C32, connecting the first corner point and the second corner point in each of the corner point pairs to obtain a connecting line of the corner point pairs;

Step C33, if the pair of corners is detected in the preset non-edge area of the canvas, then it is determined that there is a second moving object in the image to be detected.

In this embodiment, specifically, a new canvas is created, and the first corner point and the second corner point in each pair of corner points are drawn on the canvas according to their respective corresponding positions, and each pair of corner points is connected The first corner point and the second corner point in the corner point are obtained to connect the corner points, and it is judged whether each of the corner point pairs is in the preset non-edge area of the canvas, if it is in the preset non-edge area of the canvas If the pair of corner points is detected in the area, it is determined that there is a second moving object in the image to be detected; if the pair of corner points is not detected in the preset non-edge area of the canvas, it is determined that the pair of corner points is connected There is no second moving object in the image to be detected.

In an implementable manner, the canvas can be a black canvas, and the pair of corner points can be drawn as a white line. Therefore, if a white motion track curve is detected on the canvas according to the color tracking algorithm, then It may be determined that there is a second moving object in the image to be detected.

In this embodiment, the sparse optical flow does not need to calculate each pixel of the image point by point, but only needs to specify a set of feature points for tracking. In this embodiment, the corner points in the image are used to represent the image features , the way of local representation of features greatly saves computational overhead and effectively improves the running speed of still frame detection.

Further, the present application also provides a computer program product, including a computer program, when the computer program is executed by a processor, the steps of the above still frame detection method are realized.

The computer program product provided by the present application solves the technical problem of low accuracy of static frame detection in the prior art. Compared with the prior art, the beneficial effect of the computer program product provided by the embodiment of the present invention is the same as the beneficial effect of the still frame detection method provided by the above embodiment, and will not be repeated here.

The above are only preferred embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structures or equivalent process transformations made by using the description of the application and the accompanying drawings are directly or indirectly used in other related technical fields. , are all included in the patent processing scope of the present application in the same way.

Claims (10)

1. A static frame detection method is characterized by comprising the following steps:

acquiring an image to be detected and a first reference image positioned in a frame before the image to be detected;

if a first moving target is detected according to an interframe difference algorithm, the image to be detected and the first reference image, judging whether the first moving target is in a typing state or not according to a first moving direction of the first moving target;

if the first moving object is determined not to be in a typing state, detecting a second moving object according to an optical flow algorithm, the image to be detected and the first reference image;

and if the second moving target is not detected in the image to be detected, judging that the current display picture is a static frame picture.

2. The method as claimed in claim 1, wherein said step of determining that said image to be detected is a still frame picture if said second moving object is not detected in said image to be detected comprises:

if a second moving target is not detected in the image to be detected, judging that the current display picture is a suspected static frame picture;

acquiring a suspected static frame picture historical detection result corresponding to a historical to-be-detected image of a first number of frames preset in front of the to-be-detected image;

and if the frequency of the suspected static frame pictures is judged to be greater than or equal to a preset frequency threshold value in each historical detection result of the suspected static frame pictures, judging that the current display picture is a static frame picture.

3. The method of claim 1, wherein the optical flow algorithm comprises a sparse optical flow algorithm, and the step of detecting a second moving object based on the optical flow algorithm, the image to be detected and the first reference image comprises:

detecting a first corner in the first reference image;

according to a sparse optical flow algorithm, determining a second angular point matched with the first angular point from the image to be detected, and mutually matched first angular point and second angular point group angular point pairs;

and judging whether a second moving target exists in the image to be detected or not according to the relative position relation of the angle point pair.

4. The method as claimed in claim 3, wherein the step of determining whether the second moving object exists in the image to be detected according to the relative position relationship of the pair of angle points comprises:

determining the positions of a first corner point and a second corner point in each corner point pair on a preset canvas;

connecting a first corner point and a second corner point in each corner point pair to obtain a corner point pair connecting line;

and if the corner point pair connection line is detected in the preset non-edge area of the canvas, judging that a second moving target exists in the image to be detected.

5. The method of claim 1, wherein the optical flow algorithm comprises a dense optical flow algorithm, and the step of detecting a second moving object based on the optical flow algorithm, the image to be detected, and the first reference image comprises:

determining an optical flow vector corresponding to each pixel point in the image to be detected according to an optical flow algorithm, the image to be detected and the first reference image;

generating an optical flow image according to the optical flow modular length and the optical flow direction of the optical flow vector;

and if the moving image is detected in the optical flow image, judging a second moving object existing in the image to be detected.

6. The still frame detection method according to claim 5, wherein the step of determining, if a moving image is detected in the optical flow image, a second moving object present in the image to be detected comprises:

graying the optical flow image to obtain an optical flow grayscale image;

carrying out self-adaptive binarization on the light stream gray-scale image to obtain an image to be detected;

determining a second contour of the moving image in the image to be detected according to a contour finding algorithm;

and if a second contour is detected in the preset non-edge area of the image to be detected, judging a second moving target existing in the image to be detected.

7. The method of claim 1, wherein the step of determining whether the first moving object is typing according to the first moving direction of the first moving object comprises:

determining a target contour corresponding to each first moving target according to a contour finding algorithm;

determining a maximum target contour with a maximum area from each target contour;

determining a minimum bounding rectangle of the maximum target contour;

and if the minimum circumscribed rectangle is determined to be in a horizontal state or a vertical state according to the inclination angle of the minimum circumscribed rectangle, determining that the first moving target is in a typing state.

8. The method of claim 1, wherein the step of determining that the currently displayed picture is a still frame picture is followed by the step of:

acquiring a second reference image which is positioned in front of the image to be detected and is preset with a second number of frames;

determining a first peak signal-to-noise ratio corresponding to the image to be detected and a signal-to-noise ratio mean value of second peak signal-to-noise ratios corresponding to the second reference images;

if the difference value between the first peak signal-to-noise ratio and the signal-to-noise ratio mean value is larger than or equal to a preset difference value threshold value, determining that the static frame picture is interfered by noise;

and if the difference value between the first peak signal-to-noise ratio and the mean value of the signal-to-noise ratios is smaller than a preset difference value threshold value, determining that the static frame picture is a complete static frame picture.

9. An electronic device, characterized in that the electronic device comprises:

at least one processor; and the number of the first and second groups,

a memory communicatively coupled to the at least one processor; wherein,

the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the method of detecting static frames according to any of claims 1 to 8.

10. A storage medium, characterized in that the storage medium is a computer-readable storage medium having stored thereon a program for implementing a static frame detection method, the program being executed by a processor to implement the steps of the static frame detection method according to any one of claims 1 to 8.

Images