WO2017166585A1 - Method, device, and electronic apparatus for determining video transition - Google Patents

Abstract

The present invention relates to a method, device, and electronic apparatus for determining a video transition. The method for determining a video transition comprises: calculating histograms respectively corresponding to regions obtained by dividing an image of a video frame; computing differences between adjacent video frames respectively for the histograms of the regions, removing extreme values from the difference computation results, and then obtaining an average value; and determining, according to the obtained average value, a frame number of a video transition. In the present invention, an image of a video frame is divided into multiple regions, a histogram of each region is respectively calculated, and extreme values are removed during computation of an average value of difference computation results of the histograms. In this way, the present invention eliminates interference to video transition determination from a sudden appearance or disappearance of an object on a screen, thus reducing likelihood of false determination of a transition.

Description

Video transition judging method, device and electronic device

cross reference

The present application claims priority to Chinese Patent Application No. 201610202103.5, entitled "Video Transition Judgment Method and Apparatus", filed on March 31, 2016, the entire contents of which is incorporated herein by reference. .

Technical field

The present invention relates to the field of video processing technologies, and in particular, to a video transition method, apparatus, and electronic device.

Background technique

There are multiple paragraphs and scenes in the video file. As the timeline progresses, there will be transitions and transitions between paragraphs or scenes. This switching and transition is called transition. The determination of the transition time is very important for video editing work, key frame judgment, etc. The common way is to manually view the video to determine the transition time, the efficiency is very low, and it also takes a lot of manpower, which in turn causes video processing work. The overall efficiency is reduced.

In order to solve this problem, an implementation scheme for automatic transition analysis of video has been presented. The calculation of the histogram of adjacent frames in the video sequence determines whether the difference between adjacent frames exceeds the threshold. This determines the moment when the transition occurs. However, there are some disadvantages to this type of transition analysis:

1. In the case that the actual scene has not changed, when an object suddenly appears or disappears on the screen, it may be determined that a transition occurs, thereby causing a false positive.

2. Because of the calculation of the histogram involving a large number of video frames, the amount of calculation is very large.

Summary of the invention

The object of the present invention is to provide a video transition determination method and apparatus, which can reduce the possibility of error in transition determination.

To achieve the above objective, the present invention provides a video transition determination method, including:

Calculating a histogram corresponding to each of the plurality of regions divided by the video frame on the image;

Performing a difference operation on the histograms of the respective regions of the adjacent video frames, and taking the average value after removing the extreme values from the difference result;

The video frame number at which the video transition occurs is determined according to the value of the mean.

Further, the operation of calculating the histogram corresponding to the plurality of regions divided by the video frame on the image specifically includes:

Dividing the video frame into a plurality of regions on the image;

Quantify the color of the image in each area;

A histogram of the quantized images in each region is calculated.

Further, the operation of dividing the video frame into multiple regions on the image is specifically:

The video frame is divided into a plurality of equally divided regions on the image.

Further, the operation of quantifying the color of the image in each area is specifically:

The color of the image in each area is quantified using a standard color palette.

Further, the operations of performing the difference calculation on the histograms of the respective regions of the adjacent video frames, and removing the extremum from the difference result are:

The histograms of the respective regions of adjacent video frames are respectively subjected to difference calculation by the following formula,

Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image. For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

Further, the determining, by the value of the average value, whether the video frame transition occurs in the adjacent video frame includes:

Calculating a derivative of the difference mean of each video frame, and determining a local maximum of the derivative of the difference mean;

Calculate the average of all local maxima and determine the local maximum mean;

Determining the occurrence of the video transition based on the difference between each local maximum and the local maximum Frequency frame number.

Further, the operation of calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is specifically:

Calculate the second derivative of the mean difference of each video frame, as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

Further, the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of determining the video frame number of the video transition field includes:

Determine the maximum and minimum values of all local maxima as the initial centroid of the K-means clustering algorithm, and select the K value as 2;

The difference between each local maximum value and the local maximum mean value is processed by a K-means clustering algorithm, and the video frame number corresponding to the local maximum value classified into the maximum value is determined as the video frame in which the video transition occurs. number.

To achieve the above object, the present invention provides a video transition determining apparatus, including:

a histogram calculation module, configured to calculate a histogram corresponding to each of the plurality of regions divided by the video frame on the image;

a difference calculation module, configured to perform a difference calculation on the histograms of the respective regions of the adjacent video frames;

a difference mean acquisition module, configured to remove an extreme value from the difference result and take an average value;

The transition frame number determining module is configured to determine, according to the value of the mean, a video frame number at which a video transition occurs.

Further, the histogram calculation module specifically includes:

a region dividing unit, configured to divide the video frame into multiple regions on an image;

a color quantization unit for quantizing the color of an image in each region;

A histogram calculation unit for calculating a histogram of the quantized images in the respective regions.

Further, the video transition determining module specifically includes:

a local maximum value determining unit, configured to calculate a derivative of the difference mean of each video frame, and determine a local maximum of the derivative of the difference mean;

a local maximum mean determining unit for calculating an average of all local maxima, determined to be local Maximum mean

The transition frame number determining unit is configured to determine, according to a difference between each local maximum value and the local maximum mean value, a video frame number at which a video transition occurs.

Further, the local maximum value determining unit calculates the derivative of the difference mean of each video frame, and determines the local maximum value of the derivative of the difference mean value as follows:

Calculate the second derivative of the mean difference of each video frame, as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

Further, the transition frame number determining unit specifically includes:

An initial value setting subunit for determining a maximum value and a minimum value of all local maximum values as an initial centroid of the K-means clustering algorithm, and selecting a K value of 2;

a K-means clustering unit, configured to process a difference between each local maximum value and the local maximum mean value by a K-means clustering algorithm;

The transition frame number determining subunit is configured to determine the video frame number corresponding to the local maximum value classified into the maximum value as the video frame number at which the video transition occurs.

Further, the difference calculation module performs a difference calculation on the histograms of the respective regions of the adjacent video frames by using the following formula,

The difference mean value obtaining module is configured to remove the maximum value from the difference result and calculate the mean value, and the formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

An embodiment of the present invention further discloses an electronic device including at least one processor; and, The at least one processor communicatively coupled memory; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to cause the at least one processor A histogram corresponding to each of the plurality of regions divided by the video frame can be calculated; a histogram of each region of the adjacent video frame is separately subjected to a difference operation, and an average value is removed from the difference result; The value of the mean determines the video frame number at which the video transition occurs.

In the above electronic device, the operation of calculating a histogram corresponding to each of the plurality of regions divided by the video frame by the video frame comprises: dividing the video frame into a plurality of regions on the image; and The color is quantized; a histogram of the quantized image in each region is calculated.

In the above electronic device, the operation of dividing the video frame into a plurality of regions on the image is specifically: dividing the video frame into a plurality of equally divided regions on the image.

In the above electronic device, the operation of quantizing the color of the image in each area is specifically: quantizing the color of the image in each area using a standard color palette.

In the above electronic device, the operation of performing the difference calculation on the histograms of the respective regions of the adjacent video frames, and removing the extremum from the difference result is: using the following formula for the adjacent video frames The histograms of the respective regions are subjected to difference calculations,

Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

The foregoing electronic device, wherein the determining, according to the value of the mean, whether the video frame transition occurs in the adjacent video frame comprises: calculating a derivative of a mean difference of each video frame, and determining the difference The local maximum of the derivative of the mean; the average of all local maxima is calculated and determined as the local maximum mean; the video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum mean.

In the above electronic device, the operation of calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is specifically: calculating the difference mean of each video frame The derivative, the formula is as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

The above electronic device, wherein the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises: determining the maximum value and the minimum value of all the local maximum values as The initial centroid of the K-means clustering algorithm, and the K value is chosen to be 2; the difference between each local maximum and the local maximum mean is processed by the K-means clustering algorithm, and classified into the maximum class The video frame number corresponding to the local maximum is determined as the video frame number at which the video transition occurs.

The present invention also discloses a non-volatile computer storage medium, wherein the storage medium stores computer-executable instructions that, when executed by an electronic device, enable the electronic device to: calculate a video frame in an image a histogram corresponding to each of the divided regions; performing a difference operation on the histograms of the respective regions of the adjacent video frames, and taking the average value after removing the extreme values from the difference result; determining the occurrence according to the value of the average value The video frame number of the video transition.

The foregoing storage medium, wherein the operation of calculating a histogram corresponding to each of the plurality of regions divided by the video frame by the video frame comprises: dividing the video frame into a plurality of regions on the image; and The color is quantized; a histogram of the quantized image in each region is calculated.

In the above storage medium, the operation of dividing the video frame into a plurality of regions on the image is specifically: dividing the video frame into a plurality of equally divided regions on the image.

In the above storage medium, the operation of quantizing the color of the image in each area is specifically: quantifying the color of the image in each area using a standard color palette.

In the above storage medium, the operation of performing the difference operation on the histograms of the respective regions of the adjacent video frames, and removing the extremum from the difference result is: the adjacent video frame by the following formula The histograms of the respective regions are subjected to difference calculations,

Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

The foregoing storage medium, wherein the determining, according to the value of the mean, whether the video frame transition occurs in the adjacent video frame comprises: calculating a derivative of a difference mean of each video frame, and determining the difference The local maximum of the derivative of the mean; the average of all local maxima is calculated and determined as the local maximum mean; the video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum mean.

In the above storage medium, the operation of calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is specifically: calculating the second derivative of the mean difference of each video frame , the formula is as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

The above storage medium, wherein the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises: determining the maximum value and the minimum value of all the local maximum values as The initial centroid of the K-means clustering algorithm, and the K value is chosen to be 2; the difference between each local maximum and the local maximum mean is processed by the K-means clustering algorithm, and classified into the maximum class The video frame number corresponding to the local maximum is determined as the video frame number at which the video transition occurs.

Embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, when the program instructions are executed by a computer The computer is caused to perform the method of any of the above.

As can be seen from the above, the video transition determining method and apparatus provided by the present invention will have video frames in The image is divided into a plurality of regions, and the histograms of the respective regions are respectively calculated, and the extremum is removed when the mean value of the histogram difference results is obtained, so that the sudden appearance or disappearance of the object on the screen can be eliminated. The interference caused, thereby reducing the possibility of error in the judgment of the transition.

DRAWINGS

In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings to be used in the specific embodiments or the description of the prior art will be briefly described below, and obviously, the attached in the following description The drawings are some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative work.

FIG. 1 is a schematic flowchart diagram of an embodiment of a video transition determining method according to the present invention.

FIG. 2 is a schematic flowchart diagram of another embodiment of a video transition determining method according to the present invention.

FIG. 3 is a schematic flowchart diagram of still another embodiment of a video transition determining method according to the present invention.

FIG. 4 is a schematic flowchart diagram of still another embodiment of a video transition determining method according to the present invention.

FIG. 5 is a schematic structural diagram of an embodiment of a video transition determining apparatus according to the present invention.

FIG. 6 is a schematic structural diagram of another embodiment of a video transition determining apparatus according to the present invention.

FIG. 7 is a schematic structural diagram of still another embodiment of a video transition determining apparatus according to the present invention; FIG.

FIG. 8 is a schematic structural diagram of hardware of an electronic device according to an embodiment of the present invention.

detailed description

The existing video automatic transition analysis uses a histogram of adjacent frames to determine the difference between adjacent frames, but this method is difficult to distinguish global and local changes on the image of the video frame, so it is easy to cause Misjudgment. The present invention divides an image into a plurality of regions when calculating a histogram, and removes extreme values in the calculation to minimize the influence of local variations of the image on global variations.

The technical solutions of the present invention will be clearly and completely described in the following with reference to the accompanying drawings. It is obvious that the described embodiments are a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts are within the scope of the present invention.

In the description of the present invention, it should be noted that the terms "center", "upper", "lower", "left", The orientation or positional relationship of the indications of "right", "vertical", "horizontal", "inside", "outside", etc. is based on the orientation or positional relationship shown in the drawings, for convenience of description of the present invention and simplified description. Instead of indicating or implying that the device or component referred to must have a particular orientation, constructed and operated in a particular orientation, it is not to be construed as limiting the invention. Moreover, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.

In the description of the present invention, it should be noted that the terms "installation", "connected", and "connected" are to be understood broadly, and may be fixed or detachable, for example, unless otherwise explicitly defined and defined. Connection, or integral connection; may be mechanical connection or electrical connection; may be directly connected, may also be indirectly connected through an intermediate medium, or may be internal communication of two components, may be wireless connection, or may be wired connection. The specific meaning of the above terms in the present invention can be understood in a specific case by those skilled in the art.

FIG. 1 is a schematic flowchart diagram of an embodiment of a video transition determining method according to the present invention. In this embodiment, the video transition determining method includes:

Step 100: Calculate a histogram corresponding to each of the multiple regions divided by the video frame on the image;

Step 200: Perform a difference calculation on the histograms of the respective regions of the adjacent video frames, and remove the extreme values from the difference result to obtain an average value;

Step 300: Determine, according to the value of the average value, a video frame number at which a video transition occurs.

In this embodiment, the video frame is divided into a plurality of regions on the image, and the histograms of the respective regions are respectively calculated, and the extreme values are removed when the mean value of the histogram difference results is obtained, when the object in the screen suddenly appears or When it disappears, this local change can be removed when the mean value of the histogram difference result is obtained, thereby eliminating the interference caused by the sudden appearance or disappearance of the object on the screen, and thus reducing the error of the transition judgment. possibility.

FIG. 2 is a schematic flowchart diagram of another embodiment of a video transition determining method according to the present invention. Compared with the previous embodiment, the step 100 of this embodiment specifically includes:

Step 110: Divide the video frame into multiple regions on an image;

Step 120: Quantify the color of the image in each area;

Step 130: Calculate a histogram of the quantized images in the respective regions.

In this embodiment, the image of the video frame may be divided into multiple regions, and the division manner and the number of divisions of the region may be preset, for example, according to a preset number of rows and columns, or according to the figure. Divide areas like the importance of different areas. Preferably, the video frame is divided into a plurality of equally divided regions on the image, for example, divided into four equally divided regions, that is, the width and height of each of the equally divided regions are each half of the entire image. This is more convenient in the division operation. During the processing of the entire video sequence, the division mode and number of regions can be adaptively adjusted according to the historical data, so as to minimize the possibility of error in the transition determination.

The calculation of the histogram is calculated for the color of the image in the video frame, but in general, each divided area will produce a vector of RGB (256 * 256 * 256) dimensions, involving 16,777,216 colors, which obviously consumes a lot of Computing resources, and does not have a significant impact on the judgment results. Therefore, in the present embodiment, before calculating the histogram, the color of the image in each region can be quantized, and the quantized image color can be significantly reduced, thereby improving the calculation efficiency. Preferably, the color of the image in each area is quantized using a standard color palette, that is, each component is equally quantized to 6 copies, so that there are only 216 colors in total, which can significantly improve the calculation efficiency. Of course, according to the influence of the histogram calculation result on the final judgment result, other quantization methods can also be selected.

By performing histogram calculation on the quantized color, the result of obtaining the histogram is

t is the current time and also represents the video frame number. i is the i-th region in the image of the video frame. In step 200, the histograms of the respective regions of the adjacent video frames are respectively subjected to difference calculation by the following formula.

Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

In this embodiment, the method of removing the maximum value is adopted. In other embodiments, more than one extreme value may be selected according to the interference condition, for example, the largest first and second results in the difference result are removed. . In calculating the mean of the differences, it is preferred to use the squared mean value shown in the previous formula, and geometrical averages or arithmetic mean values may also be employed in other embodiments.

For all video frames, a D _mean can be calculated for each frame. Simply put, the greater the value of D _mean , the greater the difference between the two frames. According to this difference, the video frame number at which the video transition occurs can be determined based on the value of the mean.

FIG. 3 is a schematic flowchart diagram of still another embodiment of a video transition determining method according to the present invention. Compared with the previous embodiment, the step 300 of this embodiment specifically includes:

Step 310: Calculate a derivative of the difference mean of each video frame, and determine a local maximum of the derivative of the difference mean;

Step 320: Calculate an average value of all local maximum values, and determine a local maximum mean value;

Step 330: Determine, according to a difference between each local maximum value and the local maximum mean value, a video frame number at which a video transition occurs.

In this embodiment, the local maximum is determined by calculating the derivative, and the average of all the local maximums is determined to determine the local maximum mean, and the video transition can be determined according to the difference between each local maximum and the local maximum mean. The video frame number of the field, thereby implementing an adaptive threshold to determine the transition.

The step 310 preferably calculates the second derivative, which can avoid calculating too many transition frames that may be detected by too many first derivatives, and can also prevent the third derivative from detecting too few results. That is, the second derivative of the mean difference of each video frame is calculated, and the formula is as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

After determining the local maximum of the derivative of the difference mean of all video frames, the average value can be varied by calculating the average value, thereby achieving an adaptive determination of the most appropriate threshold.

FIG. 4 is a schematic flowchart diagram of still another embodiment of a video transition determining method according to the present invention. Compared with the previous embodiment, in the embodiment, step 330 specifically includes:

Step 331, determining a maximum value and a minimum value of all local maximum values as the initial centroid of the K-means clustering algorithm, and selecting a K value of 2;

Step 332: Process a difference between each local maximum value and the local maximum mean value by using a K-means clustering algorithm;

Step 333: Determine a video frame number corresponding to a local maximum value classified into a maximum value as a video frame number at which a video transition occurs.

In this embodiment, the K-means clustering algorithm is adopted. The advantage of this algorithm is that the algorithm is simple and fast. Speed, and can avoid its biggest shortcoming, the uncertainty of the initial K. Since the K value is predetermined to be 2, and the initial centroid is also determined, the video frame number corresponding to the local maximum value classified into the maximum value is the video frame number at which the video transition occurs.

One of ordinary skill in the art will appreciate that all or part of the steps to implement the various method embodiments described above may be accomplished by hardware associated with the program instructions. The aforementioned program can be stored in a computer readable storage medium. The program, when executed, performs the steps including the foregoing method embodiments; and the foregoing storage medium includes various media that can store program codes, such as a ROM, a RAM, a magnetic disk, or an optical disk.

FIG. 5 is a schematic structural diagram of an embodiment of a video transition determining apparatus according to the present invention. In this embodiment, the video transition determining apparatus includes: a histogram calculation module 1, a difference calculation module 2, a difference mean acquisition module 3, and a transition frame number determination module 4. The histogram calculation module 1 is configured to calculate a histogram corresponding to each of the multiple regions of the video frame divided by the image; the difference calculation module 2 is configured to perform a difference calculation on the histograms of the respective regions of the adjacent video frames; The difference mean value obtaining module 3 is configured to take an average value after removing the extreme value from the difference result; the transition frame number determining module 4 is configured to determine, according to the value of the average value, a video frame number at which a video transition occurs.

In this embodiment, the video frame is divided into a plurality of regions on the image, and the histograms of the respective regions are respectively calculated, and the extreme values are removed when the mean value of the histogram difference results is obtained, when the object in the screen suddenly appears or When it disappears, this local change can be removed when the mean value of the histogram difference result is obtained, thereby eliminating the interference caused by the sudden appearance or disappearance of the object on the screen, and thus reducing the error of the transition judgment. possibility.

FIG. 6 is a schematic structural diagram of another embodiment of a video transition determining apparatus according to the present invention. Compared with the previous embodiment, the histogram calculation module 1 of the present embodiment specifically includes an area dividing unit 11, a color quantization unit 12, and a histogram calculation unit 13. The area dividing unit 11 is configured to divide the video frame into a plurality of areas on the image; the color quantization unit 12 is configured to quantize the color of the image in each area; the histogram calculation unit 13 is configured to calculate the quantized areas. The histogram of the image.

In this embodiment, the image of the video frame may be divided into multiple regions, and the division manner and the number of divisions of the region may be preset, for example, according to a preset number of rows and columns, or according to different regions in the image. Importance to divide areas and so on. Preferably, the video frame is divided into a plurality of equally divided regions on the image, for example, divided into four equally divided regions, that is, the width and height of each of the equally divided regions are each half of the entire image. This is more convenient in the division operation. During the processing of the entire video sequence, the division mode and number of regions can be adaptively adjusted according to the historical data, so as to minimize the possibility of error in the transition determination.

The calculation of the histogram is calculated for the color of the image in the video frame, but in general, each divided area will produce a vector of RGB (256 * 256 * 256) dimensions, involving 16,777,216 colors, which obviously consumes a lot of Computing resources, and does not have a significant impact on the judgment results. Therefore, in the present embodiment, before calculating the histogram, the color of the image in each region can be quantized, and the quantized image color can be significantly reduced, thereby improving the calculation efficiency. Preferably, the color of the image in each area is quantized using a standard color palette, that is, each component is equally quantized to 6 copies, so that there are only 216 colors in total, which can significantly improve the calculation efficiency. Of course, according to the influence of the histogram calculation result on the final judgment result, other quantization methods can also be selected.

By performing histogram calculation on the quantized color, the result of obtaining the histogram is t is the current time and also represents the video frame number. i is the i-th region in the image of the video frame. The difference calculation module 2 performs a difference operation on the histograms of the respective regions of the adjacent video frames by the following formula,

The difference mean acquisition module 3 removes the maximum value from the difference result and calculates the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

In this embodiment, the method of removing the maximum value is adopted. In other embodiments, more than one extreme value may be selected according to the interference condition, for example, the largest first and second results in the difference result are removed. . In calculating the mean of the differences, it is preferred to use the squared mean value shown in the previous formula, and geometrical averages or arithmetic mean values may also be employed in other embodiments.

For all video frames, a D _mean can be calculated for each frame. Simply put, the greater the value of D _mean , the greater the difference between the two frames. According to this difference, the video frame number at which the video transition occurs can be determined according to the value of the mean.

FIG. 7 is a schematic structural diagram of still another embodiment of a video transition determining apparatus according to the present invention. Compared with the previous embodiment, the video transition determining module 4 of the embodiment specifically includes a local maximum value determining unit 41, a local maximum mean value determining unit 42, and a transition field number determining unit 43. The local maximum value determining unit 41 is configured to calculate a derivative of the difference mean of each video frame, and determine the mean value of the difference a local maximum of the derivative; the local maximum mean determining unit 42 is configured to calculate an average of all local maxima, determined as a local maximum mean; the transition frame number determining unit 43 is configured to use the local maxima and the local maximal mean The difference determines the video frame number at which the video transition occurs.

In this embodiment, the local maximum is determined by calculating the derivative, and the average of all the local maximums is determined to determine the local maximum mean, and the video transition can be determined according to the difference between each local maximum and the local maximum mean. The video frame number of the field, thereby implementing an adaptive threshold to determine the transition.

The local maximum value determining unit 41 preferably calculates the second-order derivative, which can avoid calculating too many transition frames that may be detected by excessive first-order derivatives, and can also prevent the third-order derivative from detecting too few results. That is, the second derivative of the mean difference of each video frame is calculated, and the formula is as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

After the local maximum value determining unit 41 determines the local maximum value of the derivative of the difference mean of all the video frames, the local maximum mean value determining unit 42 can be changed by the calculation of the average value as the overall local maximum value, thereby realizing Adaptation determines the most appropriate threshold.

In a further embodiment, the transition frame number determining unit 43 may specifically include:

An initial value setting subunit for determining a maximum value and a minimum value of all local maximum values as an initial centroid of the K-means clustering algorithm, and selecting a K value of 2;

a K-means clustering unit, configured to process a difference between each local maximum value and the local maximum mean value by a K-means clustering algorithm;

The transition frame number determining subunit is configured to determine the video frame number corresponding to the local maximum value classified into the maximum value as the video frame number at which the video transition occurs.

In this embodiment, the K-means clustering algorithm is adopted. The advantage of this algorithm is that the algorithm is simple and fast, and can avoid its biggest shortcoming, namely the uncertainty of the initial K. Since the K value is predetermined to be 2, and the initial centroid is also determined, the video frame number corresponding to the local maximum value classified into the maximum value is the video frame number at which the video transition occurs.

As shown in FIG. 8, an embodiment of the present invention further discloses an electronic device including at least one processor 810; and a memory 800 communicably connected to the at least one processor 810; wherein the memory 800 stores An instruction executed by the at least one processor 810, the instruction being At least one processor 810 is configured to enable the at least one processor 810 to calculate a histogram corresponding to each of the plurality of regions of the video frame divided by the image; and perform a difference operation on the histograms of the respective regions of the adjacent video frames And taking the average value after removing the extreme value from the difference result; determining the video frame number at which the video transition occurs according to the value of the average value. The electronic device also includes an input device 830 and an output device 840 that are electrically coupled to the memory 800 and the processor, the electrical connections preferably being connected by a bus.

In the electronic device of the embodiment, preferably, the operation of calculating a histogram corresponding to each of the plurality of regions divided by the video frame by the video frame comprises: dividing the video frame into multiple regions on the image; The color of the inner image is quantized; the histogram of the quantized image in each region is calculated.

In the electronic device of this embodiment, preferably, the operation of dividing the video frame into a plurality of regions on the image is specifically: dividing the video frame into a plurality of equally divided regions on the image.

In the electronic device of the embodiment, preferably, the operation of quantizing the color of the image in each region is specifically: quantizing the color of the image in each region using a standard color palette.

In the electronic device of the embodiment, preferably, the histograms of the respective regions of the adjacent video frames are respectively subjected to a difference operation, and the operation of removing the extremum from the difference result is performed by using the following formula: The histograms of the respective regions of the video frame are respectively subjected to difference calculation,

Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

In the electronic device of this embodiment, preferably, the determining, according to the value of the mean, whether the video frame transition occurs in the adjacent video frame comprises: calculating a derivative of the difference mean of each video frame, and determining The local maximum of the derivative of the difference mean is calculated; the average of all local maximums is calculated and determined as the local maximum mean; and the video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum mean.

In the electronic device of this embodiment, preferably, the calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is: calculating the difference of each video frame The second derivative of the value, the formula is as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

In the electronic device of this embodiment, preferably, the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises: determining a maximum value of all local maximum values and The minimum value is taken as the initial centroid of the K-means clustering algorithm, and the K value is selected as 2; the difference between each local maximum and the local maximum mean is processed by the K-means clustering algorithm, and the maximum value is classified. The video frame number corresponding to a local maximum of one type is determined as the video frame number at which the video transition occurs.

The present invention also discloses a non-volatile computer storage medium, wherein the storage medium stores computer-executable instructions that, when executed by an electronic device, enable the electronic device to: calculate a video frame in an image a histogram corresponding to each of the divided regions; performing a difference operation on the histograms of the respective regions of the adjacent video frames, and taking the average value after removing the extreme values from the difference result; determining the occurrence according to the value of the average value The video frame number of the video transition.

In the storage medium of the embodiment, preferably, the operation of calculating a histogram corresponding to each of the plurality of regions divided by the video frame by the video frame comprises: dividing the video frame into multiple regions on the image; The color of the inner image is quantized; the histogram of the quantized image in each region is calculated.

In the storage medium of the embodiment, the operation of dividing the video frame into a plurality of regions on the image is specifically: dividing the video frame into a plurality of equally divided regions on the image.

In the storage medium of the embodiment, preferably, the operation of quantizing the color of the image in each area is specifically: quantizing the color of the image in each area using a standard color palette.

In the storage medium of this embodiment, preferably, the histograms of the respective regions of the adjacent video frames are respectively subjected to a difference operation, and the operation of removing the extremum from the difference result is performed by using the following formula: The histograms of the respective regions of the video frame are respectively subjected to difference calculation,

Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

among them,

The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.

The storage medium of the embodiment, preferably, the determining, according to the value of the mean, whether the video transition occurs in the adjacent video frame comprises: calculating a derivative of the mean difference of each video frame, and determining The local maximum of the derivative of the difference mean is calculated; the average of all local maximums is calculated and determined as the local maximum mean; and the video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum mean.

In the storage medium of this embodiment, preferably, the calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is: calculating the mean of the difference of each video frame. The second derivative, the formula is as follows:

D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.

The storage medium of this embodiment, preferably, the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises: determining a maximum value of all local maximum values and The minimum value is taken as the initial centroid of the K-means clustering algorithm, and the K value is selected as 2; the difference between each local maximum and the local maximum mean is processed by the K-means clustering algorithm, and the maximum value is classified. The video frame number corresponding to a local maximum of one type is determined as the video frame number at which the video transition occurs.

Embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-transitory computer readable storage medium, the computer program comprising program instructions, when the program instructions are executed by a computer The computer is caused to perform the method described in the above embodiments.

Those skilled in the art will appreciate that embodiments of the present invention may be provided as a method, system, or Computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or a combination of software and hardware. Moreover, the invention can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) including computer usable program code.

The present invention has been described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (system), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or FIG. These computer program instructions can be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing device to produce a machine for the execution of instructions for execution by a processor of a computer or other programmable data processing device. Means for implementing the functions specified in one or more of the flow or in a block or blocks of the flow chart.

The computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing device to operate in a particular manner, such that the instructions stored in the computer readable memory produce an article of manufacture comprising the instruction device. The apparatus implements the functions specified in one or more blocks of a flow or a flow and/or block diagram of the flowchart.

These computer program instructions can also be loaded onto a computer or other programmable data processing device such that a series of operational steps are performed on a computer or other programmable device to produce computer-implemented processing for execution on a computer or other programmable device. The instructions provide steps for implementing the functions specified in one or more of the flow or in a block or blocks of a flow diagram.

It is apparent that the above-described embodiments are merely illustrative of the examples, and are not intended to limit the embodiments. Other variations or modifications of the various forms may be made by those skilled in the art in light of the above description. There is no need and no way to exhaust all of the implementations. By Obvious changes or variations resulting from this are still within the scope of the invention.

Claims (31)

1. A video transition judging method is applied to a terminal, and is characterized in that:

   Calculating a histogram corresponding to each of the plurality of regions divided by the video frame on the image;

   Performing a difference operation on the histograms of the respective regions of the adjacent video frames, and taking the average value after removing the extreme values from the difference result;

   The video frame number at which the video transition occurs is determined according to the value of the mean.
2. The video transition determining method according to claim 1, wherein the calculating the histogram corresponding to the plurality of regions divided by the video frame on the image comprises:

   Dividing the video frame into a plurality of regions on the image;

   Quantify the color of the image in each area;

   A histogram of the quantized images in each region is calculated.
3. The video transition determining method according to claim 2, wherein the operation of dividing the video frame into a plurality of regions on the image is specifically:

   The video frame is divided into a plurality of equally divided regions on the image.
4. The video transition determining method according to claim 2, wherein the operation of quantizing the color of the image in each area is specifically:

   The color of the image in each area is quantified using a standard color palette.
5. The video transition determining method according to claim 1, wherein the operation of performing a difference operation on the histograms of the respective regions of the adjacent video frames, and removing the extremum from the difference result is :

   The histograms of the respective regions of adjacent video frames are respectively subjected to difference calculation by the following formula,

   

   Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

   

   among them,

   

   The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

   

   For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.
6. The video transition determining method according to claim 1, wherein the determining, by the value of the mean, whether the video frame transition occurs in the adjacent video frame comprises:

   Calculating a derivative of the difference mean of each video frame, and determining a local maximum of the derivative of the difference mean;

   Calculate the average of all local maxima and determine the local maximum mean;

   The video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum.
7. The video transition determining method according to claim 6, wherein the calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is:

   Calculate the second derivative of the mean difference of each video frame, as follows:

   D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

   Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

   D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

   Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.
8. The video transition determining method according to claim 6, wherein the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises:

   Determine the maximum and minimum values of all local maxima as the initial centroid of the K-means clustering algorithm, and select the K value as 2;

   The difference between each local maximum value and the local maximum mean value is processed by a K-means clustering algorithm, and the video frame number corresponding to the local maximum value classified into the maximum value is determined as the video frame in which the video transition occurs. number.
9. A video transition judging device, comprising:

   a histogram calculation module, configured to calculate a histogram corresponding to each of the plurality of regions divided by the video frame on the image;

   a difference calculation module, configured to perform a difference calculation on the histograms of the respective regions of the adjacent video frames;

   a difference mean acquisition module, configured to remove an extreme value from the difference result and take an average value;

   The transition frame number determining module is configured to determine, according to the value of the mean, a video frame number at which a video transition occurs.
10. The video transition judging device according to claim 9, wherein the histogram calculation module specifically comprises:

    a region dividing unit, configured to divide the video frame into multiple regions on an image;

    a color quantization unit for quantizing the color of an image in each region;

    A histogram calculation unit for calculating a histogram of the quantized images in the respective regions.
11. The video transition determining apparatus according to claim 9, wherein the video transition determining module specifically comprises:

    a local maximum value determining unit, configured to calculate a derivative of the difference mean of each video frame, and determine a local maximum of the derivative of the difference mean;

    a local maximum mean determining unit for calculating an average value of all local maximum values, and determining the local maximum mean value;

    The transition frame number determining unit is configured to determine, according to a difference between each local maximum value and the local maximum mean value, a video frame number at which a video transition occurs.
12. The video transition determining apparatus according to claim 11, wherein said local maximum value determining unit calculates a derivative of a difference mean of each video frame, and determines a local maximum value of a derivative of said difference mean The operation is specifically as follows:

    Calculate the second derivative of the mean difference of each video frame, as follows:

    D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

    Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

    D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

    Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.
13. The video transition determining device according to claim 11, wherein the transition frame number determining unit specifically comprises:

    An initial value setting subunit for determining a maximum value and a minimum value of all local maximum values as an initial centroid of the K-means clustering algorithm, and selecting a K value of 2;

    a K-means clustering unit, configured to process a difference between each local maximum value and the local maximum mean value by a K-means clustering algorithm;

    The transition frame number determining subunit is configured to determine the video frame number corresponding to the local maximum value classified into the maximum value as the video frame number at which the video transition occurs.
14. The video transition judging device according to claim 9, wherein:

    The difference calculation module performs a difference calculation on the histograms of the respective regions of the adjacent video frames by using the following formula,

    

    The difference mean value obtaining module is configured to remove the maximum value from the difference result and calculate the mean value, and the formula is as follows:

    

    among them,

    

    The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

    

    For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.
15. An electronic device, comprising: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions Executed by the at least one processor to enable the at least one processor to

    Calculating a histogram corresponding to each of the plurality of regions divided by the video frame on the image;

    Performing a difference operation on the histograms of the respective regions of the adjacent video frames, and taking the average value after removing the extreme values from the difference result;

    The video frame number at which the video transition occurs is determined according to the value of the mean.
16. The electronic device according to claim 15, wherein the operation of calculating the histogram corresponding to the plurality of regions divided by the video frame on the image comprises:

    Dividing the video frame into a plurality of regions on the image;

    Quantify the color of the image in each area;

    A histogram of the quantized images in each region is calculated.
17. The electronic device according to claim 16, wherein the operation of dividing the video frame into a plurality of regions on the image is specifically:

    The video frame is divided into a plurality of equally divided regions on the image.
18. The electronic device according to claim 16, wherein the operation of quantizing the color of the image in each area is specifically:

    The color of the image in each area is quantified using a standard color palette.
19. The electronic device according to claim 15, wherein the operation of performing the difference operation on the histograms of the respective regions of the adjacent video frames and extracting the extremum from the difference result is:

    The histograms of the respective regions of adjacent video frames are respectively subjected to difference calculation by the following formula,

    

    Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

    

    among them,

    

    The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

    

    For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.
20. The electronic device according to claim 15, wherein the determining, according to the value of the mean, whether the video frame transition occurs in the adjacent video frame comprises:

    Calculating a derivative of the difference mean of each video frame, and determining a local maximum of the derivative of the difference mean;

    Calculate the average of all local maxima and determine the local maximum mean;

    The video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum.
21. The electronic device according to claim 20, wherein the calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is:

    Calculate the second derivative of the mean difference of each video frame, as follows:

    D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

    Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

    D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

    Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.
22. The electronic device according to claim 20, wherein the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises:

    Determine the maximum and minimum values of all local maxima as the initial centroid of the K-means clustering algorithm, and select the K value as 2;

    The difference between each local maximum value and the local maximum mean value is processed by a K-means clustering algorithm, and the video frame number corresponding to the local maximum value classified into the maximum value is determined as the video frame in which the video transition occurs. number.
23. A non-volatile computer storage medium characterized by the storage medium storing computer-executable instructions that, when executed by an electronic device, enable the electronic device to:

    Calculating a histogram corresponding to each of the plurality of regions divided by the video frame on the image;

    Performing a difference operation on the histograms of the respective regions of the adjacent video frames, and taking the average value after removing the extreme values from the difference result;

    The video frame number at which the video transition occurs is determined according to the value of the mean.
24. The storage medium according to claim 23, wherein the operation of calculating the histogram corresponding to the plurality of regions divided by the video frame on the image comprises:

    Dividing the video frame into a plurality of regions on the image;

    Quantify the color of the image in each area;

    A histogram of the quantized images in each region is calculated.
25. The storage medium according to claim 24, wherein the operation of dividing the video frame into a plurality of regions on the image is specifically:

    The video frame is divided into a plurality of equally divided regions on the image.
26. The storage medium according to claim 24, wherein the operation of quantizing the color of the image in each area is specifically:

    The color of the image in each area is quantified using a standard color palette.
27. The storage medium according to claim 23, wherein the operation of performing a difference operation on the histograms of the respective regions of the adjacent video frames and extracting the extremum from the difference result is:

    The histograms of the respective regions of adjacent video frames are respectively subjected to difference calculation by the following formula,

    

    Remove the maximum value from the difference result and calculate the mean value. The formula is as follows:

    

    among them,

    

    The histograms of the t-th frame and the t-1th frame of the jth color in the i-th region, respectively, Nc is the number of colors in the divided region, and N is the number of regions divided in the image.

    

    For the difference between the t-th frame and the t-1th frame in the i-th region, D _mean (t) is the mean value of the difference between the t-th frame and the t-1th frame, that is, the difference mean.
28. The storage medium according to claim 23, wherein the determining, according to the value of the mean, whether the video frame transition occurs in the adjacent video frame comprises:

    Calculating a derivative of the difference mean of each video frame, and determining a local maximum of the derivative of the difference mean;

    Calculate the average of all local maxima and determine the local maximum mean;

    The video frame number at which the video transition occurs is determined according to the difference between each local maximum and the local maximum.
29. The storage medium according to claim 28, wherein the calculating the derivative of the difference mean of each video frame and determining the local maximum of the derivative of the difference mean is:

    Calculate the second derivative of the mean difference of each video frame, as follows:

    D′′ _mean (t)=D _mean (t)-2*D _mean (t+1)+D _mean (t+2);

    Determining the local maximum of the second derivative of the mean of the differences of all video frames that satisfy the following formula,

    D′′ _mean (t)>D′′ _mean (t-1), and D′′ _mean (t)>D′′ _mean (t+1);

    Among them, D _mean (t-1), D _mean (t), D _mean (t+1), and D _mean (t+2) are the requirements of the t-1, t, t+1, and t+2 frames, respectively. The difference mean.
30. The storage medium according to claim 28, wherein the determining, according to the difference between each local maximum value and the local maximum mean value, the operation of the video frame number of the video transition field comprises:

    Determine the maximum and minimum values of all local maxima as the initial centroid of the K-means clustering algorithm, and select the K value as 2;

    The difference between each local maximum value and the local maximum mean value is processed by a K-means clustering algorithm, and the video frame number corresponding to the local maximum value classified into the maximum value is determined as the video frame in which the video transition occurs. number.
31. A computer program product comprising a non-transitory computer A computer program on a readable storage medium, the computer program comprising program instructions, wherein the computer program, when executed by a computer, causes the computer to perform the method of any of the preceding claims.

Images