KR102557904B1 - The Method of Detecting Section in which a Movement Frame Exists - Google Patents

Abstract

The present invention relates to a method for detecting a section in which a dynamic frame exists, and more particularly, to a method for detecting a section in which a motion exists in a high-quality real-time monitoring image by extracting and decoding only I-frames from an entire bitstream of data recorded through the decoded I-frame, and comparing a static background without motion through the decoded I-frame with a moving picture frame having a moving object to determine the presence or absence of motion in the image, thereby detecting a section in which motion exists in a high-quality real-time monitoring image.

Description

Method of detecting a section in which a dynamic frame exists {The Method of Detecting Section in which a Movement Frame Exists}

The present invention relates to a method for detecting a section in which a dynamic frame exists, and more particularly, to a method for detecting a section in which a motion exists in a high-quality real-time monitoring image by extracting and decoding only I-frames from an entire bitstream of data recorded through the decoded I-frame, and comparing a static background without motion through the decoded I-frame with a moving picture frame having a moving object to determine the presence or absence of motion in the image, thereby detecting a section in which motion exists in a high-quality real-time monitoring image.

Recently, it is common to build a video control system using CCTV to prevent crime or secure post-mortem evidence. With the development of filming equipment and filming technology, the quality of cameras used in CCTV has also improved, and the size of surveillance images filmed by CCTV has also increased dramatically.

CCTV surveillance video adopts compressed video for efficiency of storage space, and in case of high-definition CCTV video, it is common to use video compression technology with high compression rate such as H.265 HEVC. Here, HEVC (High Efficiency Video Coding) is a next-generation video compression standard developed to improve coding performance in high-definition application fields such as HD mobile, home cinema, and UHD TV.

However, since the technology of recognizing, analyzing, and tracking image data is an algorithm that requires a considerable amount of computation, that is, as the size of image data increases, a significant load is placed on the computing device. Accordingly, it takes longer and longer to analyze image data whose size has increased. Accordingly, there is a steady demand for a method capable of reducing image information analysis time.

On the other hand, as a conventional technique for detecting motion in image data, there is a technique of calculating a motion vector of a reference frame based on a histogram of motion vectors of blocks in a reference frame and determining a region type of a reference block based on a global motion vector, as in Korean Patent Registration No. 10-1409826.

However, since the above-described conventional technique has to calculate motion vectors for the entire region and histogram data for the entire region, it is difficult to obtain a speed that is capable of real-time processing in a current high-resolution image, and since motion vectors must be considered for all blocks, there is a problem in that calculations must be performed once for unnecessary blocks. In addition, there is a problem in that a large amount of basic calculation is required because basically all frames must be processed in relation to object detection.

Therefore, there is a need for a technology capable of reducing the amount of computation compared to the prior art and detecting only a section in which motion exists in a long-length video, such as a real-time monitoring video captured by CCTV.

Republic of Korea Patent Registration No. 10-1409826 (2014.06.13.)

An object of the present invention is to provide a method for detecting a section in which a dynamic frame exists, and more particularly, to provide a method for detecting a section in which a motion exists in a high-definition real-time monitoring image by extracting and decoding only I-frames from an entire bitstream of data recorded through the decoded I-frame, and determining whether there is motion in the image by comparing a static background without motion through the decoded I-frame with a moving picture frame having a moving object.

In order to solve the above problem, in one embodiment of the present invention, a method for detecting a section in which a dynamic frame exists is performed in a computing system including one or more processors and a main memory for storing instructions executable by the processor, comprising: an I-frame decoding step of detecting a NAL unit from image data, analyzing a header of the NAL unit, distinguishing only a plurality of I-frames from an entire bitstream of the image data, and decoding only the I-frames; MSE calculation step of calculating an MSE value between two adjacent I frames based on a pixel difference between the two adjacent I frames with respect to the decoded I frame; a dynamic frame determination step of determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and a dynamic frame presence section detection step of detecting a section in which the dynamic frame exists in the image data.

In one embodiment of the present invention, the video data may include video data encoded according to the HEVC protocol.

In one embodiment of the present invention, the dynamic frame determination step may include comparing each MSE value with a first threshold value for all calculated MSE values; detecting an MSE value greater than the first threshold value; and determining an I frame corresponding to an MSE value greater than the first threshold as a dynamic frame.

In one embodiment of the present invention, the first threshold is a first reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating an average value of the MSE values calculated from the first reference MSE calculation step; and a first threshold value determination step of determining the first threshold value based on the average value.

In one embodiment of the present invention, the dynamic frame determination step may include calculating a standard deviation of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values; comparing the standard deviation with a second threshold; and determining a plurality of I frames corresponding to the plurality of MSE values included in a window having a standard deviation greater than the second threshold as dynamic frames.

In one embodiment of the present invention, the second threshold is a second reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating a standard deviation of the MSE values calculated from the second reference MSE calculation step; and a second threshold value determining step of determining the second threshold based on the standard deviation.

In one embodiment of the present invention, the dynamic frame determination step may include calculating a standard deviation of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values; comparing the standard deviation with a third threshold; and determining an I-frame corresponding to the N-th MSE value as a dynamic frame, if a window in which the standard deviation of a plurality of MSE values included in the window is larger than the third threshold exists in one or more windows including the N-th MSE value.

In one embodiment of the present invention, the third threshold is a third reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating a standard deviation of the MSE values calculated from the third reference MSE calculation step; and a third threshold value determining step of determining the third threshold based on the standard deviation.

In one embodiment of the present invention, the dynamic frame determination step may include calculating an average value and standard deviation of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values; comparing the standard deviation with a fourth threshold; For one or more windows including the N-th MSE value, if a window in which the standard deviation of a plurality of MSE values included in the window is larger than the fourth threshold exists, an I frame corresponding to the N-th MSE value is determined as a first preliminary dynamic frame; comparing the average value with a fifth threshold value; For one or more windows including the N-th MSE value, if there exists a window in which the average value of the plurality of MSE values included in the window is greater than the fifth threshold value, an I frame corresponding to the N-th MSE value is determined as a second preliminary dynamic frame. A second determination step; and determining the I-frame as a dynamic frame with respect to the I-frame determined to be the first pre-dynamic frame in the first determination step and determined to be the second pre-dynamic frame in the second determination step.

In one embodiment of the present invention, the fourth threshold is a fourth reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating a standard deviation of the MSE values calculated from the fourth reference MSE calculation step; and a fourth threshold value determining step of determining the fourth threshold based on the standard deviation.

In one embodiment of the present invention, the fifth threshold is a fifth reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; a fifth reference mean value calculation step of calculating an average value of the MSE values calculated from the fifth reference MSE calculation step; and a fifth threshold value determination step of determining the fifth threshold value based on the average value calculated in the fifth reference mean value calculation step.

In order to solve the above problem, in one embodiment of the present invention, an apparatus for detecting a frame in which motion exists, including one or more processors and a main memory storing instructions executable by the processor, detects a NAL unit from video data, analyzes a header of the NAL unit, separates only I frames from the entire bitstream of the video data, and decodes only the I frames; an MSE calculation unit for calculating an MSE value between two adjacent I frames based on a pixel difference between two adjacent I frames with respect to the decoded I frame; a dynamic frame determination unit determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and a dynamic frame presence section detecting section for detecting a section in which the dynamic frame exists in the image data.

In one embodiment of the present invention to solve the above problem, in one embodiment of the present invention, a computer program stored in a computer-readable medium, including a plurality of instructions executed by one or more processors, wherein the computer program detects a NAL unit from image data, analyzes a header of the NAL unit, divides only a plurality of I frames from the entire bitstream of the image data, and decodes only the I frames; MSE calculation step of calculating an MSE value between two adjacent I frames based on a pixel difference between the two adjacent I frames with respect to the decoded I frame; a dynamic frame determination step of determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and a dynamic frame existence section detecting step of detecting a section in which the dynamic frame exists in the image data.

According to an embodiment of the present invention, by distinguishing and decoding only I-frames from a bitstream of the entire video data, an effect of detecting a section in which motion exists in the video data can be achieved in less time than conventional techniques.

According to an embodiment of the present invention, by performing image processing only for a minimum number of frames, it is possible to achieve an effect requiring less computational processing capability than the conventional technology.

According to an embodiment of the present invention, it is possible to achieve an effect of detecting a section in which motion exists at high speed with high accuracy despite a small amount of computation.

According to an embodiment of the present invention, it is possible to perform object detection and object analysis only for frames in which motion is determined to exist, so that the amount of calculations processed during object analysis can be reduced to a significant level.

According to an embodiment of the present invention, although a significant computational load is required for recent high-definition CCTV images, an effect of reducing the analysis target area to a considerable level can be achieved only with frames in which motion exists.

According to an embodiment of the present invention, by patterning the calculated MSE value and then learning it, the type of object can be identified, and object movement information in a specific area can be obtained based on the learned content.

According to an embodiment of the present invention, the dynamic frame determination unit can detect dynamic frames using various methods, and the various methods can be applied according to the purpose of detecting dynamic frames.

1 schematically illustrates overall steps of a method for detecting a section in which a dynamic frame exists according to an embodiment of the present invention.
2 exemplarily illustrates image data including I frames, B frames, and P frames according to an embodiment of the present invention.
3 schematically illustrates SYNTAX of a coding unit in the HEVC standard according to an embodiment of the present invention.
4 schematically shows an I frame decoding step and an MSE calculation step according to an embodiment of the present invention.
FIG. 5 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period, including a dynamic frame determination step, according to an embodiment of the present invention.
6 schematically shows a histogram and a first threshold value for calculated MSE values according to an embodiment of the present invention.
FIG. 7 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period, including a dynamic frame determination step, according to an embodiment of the present invention.
8 schematically shows a histogram and a sliding window for calculated MSE values according to an embodiment of the present invention.
9 schematically illustrates several MSE values and sliding windows according to an embodiment of the present invention.
10 schematically shows a table comparing standard deviations and second threshold values for MSE values included in several windows according to an embodiment of the present invention.
FIG. 11 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period, including a dynamic frame determination step, according to an embodiment of the present invention.
12 schematically shows a table comparing standard deviations and third thresholds for MSE values included in several windows according to an embodiment of the present invention.
FIG. 13 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period including a dynamic frame determination step according to an embodiment of the present invention.
FIG. 14 schematically shows a table comparing average values and standard deviations of MSE values included in several windows with fourth and fifth threshold values, respectively, according to an embodiment of the present invention.
15 illustratively illustrates the internal configuration of a computing device according to an embodiment of the present invention.

In the following, various embodiments and/or aspects are disclosed with reference now to the drawings. In the following description, for purposes of explanation, numerous specific details are set forth in order to facilitate a general understanding of one or more aspects. However, it will also be appreciated by those skilled in the art that such aspect(s) may be practiced without these specific details. The following description and accompanying drawings describe in detail certain illustrative aspects of one or more aspects. However, these aspects are exemplary and some of the various methods in principle of the various aspects may be used, and the described descriptions are intended to include all such aspects and their equivalents.

Moreover, various aspects and features will be presented by a system that may include a number of devices, components and/or modules, and the like. It should also be understood and appreciated that various systems may include additional devices, components and/or modules, and/or may not include all of the devices, components, modules, etc. discussed in connection with the figures.

“Example”, “example”, “aspect”, “exemplary”, etc., used herein should not be construed as preferring or advantageous to any aspect or design being described over other aspects or designs. The terms '~unit', 'component', 'module', 'system', 'interface', etc. used below generally mean a computer-related entity, and may mean, for example, hardware, a combination of hardware and software, or software.

Also, the terms "comprises" and/or "comprising" should be understood to mean that the feature and/or element is present, but does not preclude the presence or addition of one or more other features, elements and/or groups thereof.

In addition, terms including ordinal numbers, such as first and second, may be used to describe various components, but the components are not limited by the terms. These terms are only used for the purpose of distinguishing one component from another. For example, a first element may be termed a second element, and similarly, a second element may be termed a first element, without departing from the scope of the present invention. The terms and/or include any combination of a plurality of related recited items or any of a plurality of related recited items.

In addition, in the embodiments of the present invention, unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with the meaning in the context of the related art, and unless explicitly defined in the embodiments of the present invention, an ideal or excessively formal meaning. It is not to be interpreted.

As mentioned below, "dynamic frame" corresponds to a frame in which motion exists. For a plurality of frames constituting the bitstream of the entire image data, the pixel difference between two adjacent frames can be digitized, and if the pixel difference between the two adjacent frames is greater than or equal to a reference value, it can be determined that there is motion between the two frames, and the two frames correspond to the dynamic frame.

Meanwhile, the two adjacent frames do not necessarily have to be adjacent to each other on the bitstream of the entire image data, and may correspond to two adjacent frames among a plurality of frames extracted from the bitstream according to a preset rule.

1 schematically illustrates overall steps of a method for detecting a section in which a dynamic frame exists according to an embodiment of the present invention.

As shown in FIG. 1, a method for detecting a section in which a dynamic frame exists according to an embodiment of the present invention is performed in a computing system unit 1000 including one or more processors and a main memory storing instructions executable by the processor.

A method for detecting a section in which such a dynamic frame exists is an I frame decoding step (S100) of detecting a NAL unit from video data, analyzing a header of the NAL unit, distinguishing only a plurality of I frames from the entire bitstream of the video data, and decoding only the I frames; MSE calculation step (S200) of calculating an MSE value between two adjacent I frames based on a pixel difference between the two adjacent I frames with respect to the decoded I frame; a dynamic frame determination step (S300) of determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and a dynamic frame existence section detection step (S400) of detecting a section in which the dynamic frame exists in the image data.

In one embodiment of the present invention, the steps S100, S200, S300, and S400 are performed by the I frame decoding unit 100, the MSE calculation unit 200, the dynamic frame determination unit 300, and the dynamic frame existence section detection unit 400, respectively.

Preferably, the video data may include video data encoded according to the HEVC protocol.

2 exemplarily illustrates image data including I frames, B frames, and P frames.

The video portion of a general motion picture is composed of I frames (frames indicated by “I”), P frames (frames indicated by “P”), and B frames (frames indicated by “B”).

The I-frame includes the entire image as a key frame, can function as an access point in a video file, corresponds to an independently encoded frame, and has a low compression rate.

Meanwhile, in the case of a P frame, a frame created by forward prediction by referring to a previous I frame or P frame does not correspond to an independently encoded frame. The P frame has a higher compression ratio than the I frame. Here, the “previous” frame means not only the immediately preceding frame but also one of a plurality of frames existing before the corresponding frame, and the “post” frame means not only the immediately following frame but also one of a plurality of frames existing after the corresponding frame.

Meanwhile, in the case of a B frame, a frame made by forward and backward prediction with reference to previous and subsequent frames does not correspond to an independently encoded frame. The B frame has a higher compression ratio than the I frame and the P frame. Accordingly, the independently encoded frames may correspond to the I-frames, and the non-independently encoded frames may correspond to the remaining B-frames or P-frames.

3 shows SYNTAX of a coding unit in the HEVC standard.

Schematically, (a) of FIG. 3 shows a SYNTAX of a NAL unit, and (b) of FIG. 3 shows a SYNTAX of a header of a NAL unit.

Specifically, as shown in (a) of FIG. 3, according to the HEVC standard document “ITU-T H.265 TELECOMMUNICATION STANDARDIZATION SECTOR OF ITU (11/2019) SERIES H: AUDIOVISUAL AND MULTIMEDIA SYSTEMS Infrastructure of audiovisual services - Coding of moving video”, encoded H.265 data is a series of packets known as Network Abstraction Layer Units (NAL Unit). stored or transmitted.

The HEVC/H.265 codec applied to the present invention performs compression by calculating subsequent B-frames and P-frames based on I-frames generated at regular intervals.

In (b) of FIG. 3, SYNTAX of the header of the NAL unit included in nal_unit in (a) of FIG. 3 is shown. In the present invention, by using nal_unit_type included in the header of the NAL unit, only each I frame can be distinguished from the bitstream of the entire video data, and only the I frame is decoded. By analyzing only the decoded I-frame, the event of interest can be summarized without the need to decode the entire bitstream, and through this, a dynamic frame can be detected from the bitstream in less time than in the prior art, and a dynamic frame can be detected with a smaller amount of computation than in the prior art.

4 schematically shows an I frame decoding step (S100) and an MSE calculation step (S200) according to an embodiment of the present invention.

Schematically, video data described below as an embodiment of the present invention corresponds to video data including I-frames, B-frames, and P-frames with reference to FIG.

Specifically, with reference to the description of FIG. 3, the I frame decoding unit 100 extracts and decodes only each I frame from the entire bitstream of the video data using nal_unit_type included in the header of the NAL unit from the entire bitstream of the video data.

The plurality of I frames extracted by the I frame decoding unit 100 are transmitted to the MSE calculating unit 200, and the MSE calculating unit 200 performs an MSE calculating step (S200) of calculating an MSE value for the plurality of I frames received from the I frame decoding unit 100 based on a pixel difference between two adjacent frames.

The larger the MSE value, the larger the pixel difference between the two frames. When the MSE value is greater than or equal to a predetermined threshold value, it can be determined that motion has occurred between the two frames, and the two frames can be determined as dynamic frames. In the opposite case, when there is almost no pixel difference between the two frames, the MSE value between the two frames is calculated to be less than the threshold value, and it can be determined that there is no motion between the two frames.

Meanwhile, the MSE value can be calculated in the following manner. The MSE value can be obtained by obtaining a difference value between a plurality of pixels corresponding to each other in two adjacent frames, squaring the difference value, summing all the squared values of the difference value, and then dividing by the total number of pixels in a single frame.

FIG. 5 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period including a dynamic frame determining step (S300) according to an embodiment of the present invention.

As shown in FIG. 5, the dynamic frame determination step (S300) includes a step of comparing each MSE value with a first threshold value for all calculated MSE values (S310); Detecting an MSE value greater than the first threshold (S311); And determining an I frame corresponding to an MSE value greater than the first threshold as a dynamic frame (S312);

The first threshold value includes a first reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating an average value of the MSE values calculated from the first reference MSE calculation step; and a first threshold value determination step of determining the first threshold value based on the average value.

Specifically, the dynamic frame determination step (S300) shown in FIG. 5 may be performed as an embodiment of the present invention. Before the dynamic frame determination step (S300) is performed, referring to the description of FIG. 4, the I frame decoding step (S100) and the MSE calculating step (S200) are performed by the I frame decoding unit 100 and the MSE calculating unit 200, respectively. After a plurality of MSE values are calculated through the MSE calculation step (S200), the steps S310 to S312 are performed by the dynamic frame determination unit 300.

The first threshold may be set in advance through learning image data. The learning image data, as an embodiment of the present invention, may be generated from image data including an image section in which motion does not exist, and preferably, the image data may include both a video section in which motion does not exist and an image section in which motion exists. The learning image data may be generated by extracting an I frame of an image section in which motion does not exist from the image data.

After performing the first reference MSE calculation step of calculating the MSE value for each I-frame of the learning image data, the step of calculating the average value of all MSE values of the learning image data is performed, and based on the average value, the first threshold value determining step of determining the first threshold value can be performed.

On the other hand, as an embodiment of the present invention, the first threshold may be set by referring not only to the average value of MSE values for I-frames of the section in which motion does not exist, but also to the average value of MSE values of I-frames in the section in which motion exists from the image data.

In step S310, the plurality of MSE values calculated in the MSE calculation step (S200) are compared with the first threshold, respectively, and in step S311, an MSE greater than the first threshold is detected. In step S312, I frames corresponding to an MSE greater than the first threshold are determined as dynamic frames, and steps S310 to S312 may be performed by the dynamic frame determination unit 300. After the dynamic frame is determined by the dynamic frame determination unit 300, in the step S400, it is possible to detect a section in which the dynamic frame exists, that is, a section in which motion exists in the image data.

In addition, referring to the description of FIG. 4 described above, the section in which the motion exists includes two I frames corresponding to an MSE value greater than the first threshold and one or more B frames dependent on the two I frames. It includes a section corresponding to one or more P frames.

6 schematically shows a histogram and a first threshold for calculated MSE values according to an embodiment of the present invention.

Schematically, the histogram is a graph showing a plurality of MSE values calculated in the MSE calculation step (S200). The x-axis of the histogram means the number of the MSE values, and the y-axis means the MSE value corresponding to the number of each MSE value.

The number of the MSE value corresponds to the serial number given by the MSE calculation unit 200 to the MSE values, and preferably, the serial number can be assigned to the MSE values in order of time.

Meanwhile, the histogram is shown for explanation of the embodiment dealt with in the description of FIG. 5 in the present invention, and may differ from the result when the present invention is actually applied.

As shown in FIG. 6, the section corresponding to the MSE value below the first threshold is determined to be a section without motion, and the section corresponding to the MSE value above the first threshold is determined to be a section in which motion exists. Comparing the MSE values with the first threshold may be performed in steps S310 and S311 with reference to the description of FIG.

Meanwhile, one embodiment of the dynamic frame determination step (S300) dealt with in the description of FIGS. 5 and 6 corresponds to one of several embodiments of the present invention, and descriptions of other embodiments will be described later.

The embodiment of the dynamic frame determination step (S300) described in FIGS. 5 and 6 has a relatively simple logic and is easy to implement among various embodiments of the present invention. However, since each MSE value must be compared with the first threshold value, it may take more time to detect a dynamic frame compared to the method described later. However, when the learning image data for setting the first threshold is appropriately selected, an effect of detecting motion from the image data with high accuracy can be exerted. Therefore, the embodiments in FIGS. 5 and 6 are preferably used to analyze CCTV images recording objects with relatively small movements.

FIG. 7 schematically illustrates the overall execution steps of a dynamic frame existence period detection method including a dynamic frame determination step (S300) according to an embodiment of the present invention.

As shown in FIG. 7, the dynamic frame determination step (S300) includes a step of calculating a standard deviation of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values (S320); Comparing the standard deviation with a second threshold (S321); And determining a plurality of I frames corresponding to the plurality of MSE values included in a window having a standard deviation greater than the second threshold as dynamic frames (S322);

The second threshold value includes a second reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating a standard deviation of the MSE values calculated from the second reference MSE calculation step; and a second threshold determining step of determining the second threshold based on the standard deviation.

Meanwhile, the dynamic frame determination step (S300) shown in FIG. 7 may be performed as an embodiment of the present invention, and may be performed as an embodiment different from the embodiment described above in the description of FIG. 5 . Before the dynamic frame determination step (S300) is performed, referring to the description of FIG. 4, the I frame decoding step (S100) and the MSE calculation step (S200) are performed.

Specifically, in step S320, the sliding window method is applied to the plurality of MSE values calculated in the MSE calculation step (S200). Here, the window means giving a serial number to each of the plurality of MSE values calculated in the MSE calculation step (S200), and grouping them into a predetermined size according to the serial number.

For example, if the number of MSE values calculated in the MSE calculation step (S200) is 100, serial numbers from 1 to 100 can be assigned to each MSE value, and the criteria for assigning the serial numbers are set in chronological order. It is preferable to set. If a window having a size of 10 is applied to the MSE values from 1 to 100, the first window includes MSE values 1 to 10, the second window includes MSE values 2 to 11, and the third window includes MSE values 3 to 12. Additional description of the sliding window method will be described later.

As such, if the sliding window method is applied after setting the window size to 10, in step S320, the standard deviation of the MSE corresponding to Nos. 1 to 10 corresponding to the first window is calculated, and the standard deviation corresponding to the MSE corresponding to Nos. 2 to 11 corresponding to the second window is calculated, thereby calculating the standard deviation corresponding to each window.

In step S321, a window having a standard deviation higher than the second threshold may be detected by comparing the standard deviation corresponding to each window calculated in step S320 with the second threshold. In step S322, all I-frames corresponding to MSE values included in a window having a standard deviation higher than the second threshold are determined as dynamic frames, and steps S320 to 322 may be performed by the dynamic frame determination unit 300. After the dynamic frame is determined by the dynamic frame determination unit 300, in the step S400, it is possible to detect a section in which the dynamic frame exists, that is, a section in which motion exists in the image data.

The dynamic frame detected by the above method may include one or more I-frames, one or more B-frames, and one or more P-frames existing between the two I-frames for calculating the last MSE value of the previous I-frame among the two I-frames for calculating the first MSE value of the window having a standard deviation higher than the second threshold and the last MSE value of the corresponding window.

Meanwhile, the second threshold, like the first threshold, may be set through the learning image data. The learning image data corresponds to data generated by extracting an I frame of a section in which motion does not exist from image data including an image in which motion does not exist, with reference to the description of FIG. 5 .

The second reference MSE calculation step of calculating the MSE value for each I-frame of the training image data is performed, and then the standard deviation of all MSE values of the training image data is calculated. After calculating the standard deviation, the second threshold value determining step of determining the second threshold value based on the standard deviation may be performed.

On the other hand, the embodiment of the dynamic frame determination step (S300) described above in the description of FIG. 7 can exhibit an effect of greatly improving the operation speed compared to the embodiment described above in the description of FIG. 5 . Since each MSE is grouped and compared in a window rather than individually compared, a section in which a dynamic frame exists can be quickly detected, but accuracy may be somewhat lower than that of the embodiment in FIG. 5 . Accordingly, the embodiment in FIG. 7 is preferably used to analyze CCTV images of vehicles photographed on, for example, suburban highways or general roads with low traffic volume.

8 schematically shows a histogram and a sliding window for calculated MSE values according to an embodiment of the present invention, and FIG. 9 schematically shows several MSE values and a sliding window according to an embodiment of the present invention.

Schematically, FIGS. 8 and 9 are shown to explain the sliding window method.

Specifically, FIG. 8, as an embodiment of the present invention, shows a histogram of a plurality of MSE values calculated by the MSE calculation unit 200, and shows an embodiment in which a window having a size of 10 is applied to the histogram. As shown in FIG. 8, windows including MSE values corresponding to numbers 51 to 60 are shown as solid line boxes, and the window can be moved to the right by 5 spaces by setting the movement interval to 5.

9 is a histogram and a window shown as an embodiment of the present invention, which are shown for explanation, and may differ from the actual applied result of the present invention. The size of the window shown in FIG. 9 is 10, and the histogram shows MSE values corresponding to times 1 to (N+12).

As shown in FIG. 9, a serial number may be assigned to each window, and the serial number of the window follows the smallest MSE number included in the window. That is, window #1 includes MSE values corresponding to times 1 to 10, and window #N includes MSE values corresponding to times N to (N+9). In this way, as the order of windows increases, it can be named as a sliding window in the sense that the window moves one by one in a sliding manner. On the other hand, the above-mentioned N corresponds to an arbitrary natural number of 1 or more, and N described below is also the same.

In the step S320, as shown in FIG. 9, the standard deviation of the MSE values included in each window may be calculated while moving to the right by a predetermined movement interval according to the serial number of the MSE. For example, when the MSE value calculated by the MSE calculation unit 200 is 1726, the size of the window is set to 10, and the movement interval of the window is set to 1, the dynamic frame determination unit 300 can generate windows #1 to #1717, and can calculate 1717 standard deviations corresponding to each window.

Meanwhile, as an embodiment of the present invention, the size of the window and the interval at which the window moves may be set by an operator of the computing system unit 1000. As the size of the window increases, the detection accuracy of the section where the dynamic frame exists decreases, but the operation speed increases. As the size of the window decreases, the detection accuracy increases, but the operation speed decreases and the operation load may increase. In addition, the movement interval of the window is preferably set to be less than the size of the window, and the larger the movement interval of the window, the higher the calculation speed and the lower the calculation load, but the lower the detection accuracy.

In the following description, the size of the window is set to 10 and the movement interval of the window is preset to 1 for better understanding. However, such preset values are only one embodiment of the present invention, and according to various embodiments of the present invention, various preset values may be set by an operator of the computing system unit 1000 .

10 schematically shows a table comparing standard deviations and second threshold values for MSE values included in several windows according to an embodiment of the present invention.

As shown in FIG. 10, according to an embodiment of the present invention, the dynamic frame determining unit 300 compares the standard deviation corresponding to each of windows #1 to #(N+1) with the second threshold, and determines whether a dynamic frame exists in each window based on the comparison result.

Meanwhile, the table shown in FIG. 10 is shown for description of the embodiment dealt with in the description of FIG. 7 in the present invention.

As shown in FIG. 10, as an embodiment of the present invention, window #1 includes MSE values corresponding to Nos. 1 to 10, and the standard deviation of the MSE values corresponding to #1 to #10 is calculated. The result is 3.7. Meanwhile, the value of the second threshold calculated with reference to the description of FIG. 7 is 19.48. Since the standard deviation corresponding to the window #1 is determined to be less than the second threshold, no dynamic frame exists in the window #1, that is, I frames corresponding to MSE values 1 to 10 are not determined as dynamic frames.

In the same way, since the standard deviation corresponding to window # 2 is calculated to be less than the second threshold, as shown in FIG. 10, I frames corresponding to MSE values 2 to 11 are not determined as dynamic frames.

On the other hand, when the standard deviation of the MSE values corresponding to #N to #(N+9) included in window #N corresponding to the Nth window is calculated to be 25.1, the standard deviation higher than the second threshold is calculated. I frames corresponding to MSE values corresponding to #N to #(N + 9) are determined to be dynamic frames.

In addition, when one MSE value is not determined as a dynamic frame or determined as a dynamic frame, that is, when one MSE value is included in each of a plurality of windows, and the standard deviation of one of the plurality of windows is greater than the second threshold or the standard deviation of another window is less than the second threshold, the I frame corresponding to the corresponding MSE value is determined to be a dynamic frame. For example, in the case of the MSE value corresponding to number 123, the standard deviation of window #122 is less than the second threshold, but the standard deviation of window #123 may be greater than or equal to the second threshold. At this time, the I frame corresponding to the MSE value corresponding to number 123 included in both window #122 and window #123 is determined as a dynamic frame.

FIG. 11 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period including a dynamic frame determining step (S300) according to an embodiment of the present invention.

As shown in FIG. 11, the dynamic frame determination step (S300) includes a step of calculating a standard deviation of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values (S330); Comparing the standard deviation with a third threshold (S331); And for one or more windows including the N-th MSE value, if a window in which the standard deviation of a plurality of MSE values included in the window is greater than the third threshold exists, determining an I frame corresponding to the N-th MSE value as a dynamic frame (S332);

The third threshold is a third reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating a standard deviation of the MSE values calculated from the third reference MSE calculation step; and a third threshold value determining step of determining the third threshold based on the standard deviation.

Meanwhile, the dynamic frame determination step (S300) shown in FIG. 11 may be performed as an embodiment of the present invention, and may be performed as an embodiment different from the embodiments described above in FIGS. 5 and 7. Before the dynamic frame determination step (S300) is performed, referring to the description of FIG. 4, the I frame decoding step (S100) and the MSE calculation step (S200) are performed.

Specifically, in step S330, the sliding window method is applied to the plurality of MSE values calculated in the MSE calculation step (S200). The sliding window method applied here proceeds in the same manner as described above in the description of FIGS. 8 to 9 . After applying the sliding window method, the standard deviation of MSE values included in the window including the Nth MSE may be calculated by the dynamic frame determining unit 300 . For example, the window including the 324th MSE corresponds to window #313 to window #324. That is, in the step S330, the standard deviation of the MSE values included in each of the windows #313 to #324 can be obtained by the dynamic frame determining unit 300.

In step S331, a window having a standard deviation exceeding the third threshold is detected among the standard deviations corresponding to window #(N-9) to window #N calculated in step S330. In step S332, if a window having a standard deviation exceeding the third threshold is detected among the standard deviations corresponding to the window #(N-9) to window #N, the N-th MSE is determined as a dynamic frame.

As described above, the biggest difference between the embodiment in FIG. 11 and the embodiment in FIG. 7 is that the I frame corresponding to the MSE value is determined as a dynamic frame based on the results of a plurality of windows including the MSE value for one MSE value.

That is, in the embodiment of FIG. 7, it is determined whether all I frames corresponding to a plurality of MSE values included in one window are dynamic frames at once, whereas in the embodiment of FIG.

Therefore, the embodiment in FIG. 11 can derive results with higher accuracy than the embodiment in FIG. 7, and by using the sliding window method, a faster calculation speed can be expected compared to the embodiment in FIG. 5. Can be. Based on these effects, the embodiment in FIG. 11 has characteristics that can be used for various purposes compared to the embodiments in FIGS. 5 and 7.

Meanwhile, the third threshold, like the second threshold, may be set through the learning image data. The learning image data corresponds to data generated by extracting an I frame of a section in which motion does not exist from image data including an image in which motion does not exist, with reference to the description of FIG. 5 .

The third standard MSE calculation step of calculating the MSE value for each I-frame of the training image data is performed, and then the standard deviation of all MSE values of the training image data is calculated. Preferably, the step of calculating the average of all the MSE values is performed before calculating the standard deviation. After calculating the standard deviation, the third threshold value determining step of determining the third threshold value based on the standard deviation may be performed.

12 schematically shows a table comparing standard deviations and third thresholds for MSE values included in several windows according to an embodiment of the present invention.

As shown in FIG. 12, according to an embodiment of the present invention, the dynamic frame determination unit 300 compares the standard deviation corresponding to each of windows #(N-9) to window #N with the third threshold, and based on the comparison result, it can be determined whether the Nth MSE corresponds to a dynamic frame. In addition, the table shown in FIG. 12 is based on the embodiment of the dynamic frame determination step (S300) dealt with in the description of FIG. 11.

Meanwhile, the table shown in FIG. 12 is shown for description of the embodiment dealt with in the description of FIG. 11 in the present invention.

As shown in FIG. 12, as an embodiment of the present invention, window # (N-9) includes MSE values corresponding to (N-9) to N, and the standard deviation of the MSE values corresponding to (N-9) to N is calculated and the result value is 5.12. Meanwhile, the value of the third threshold calculated with reference to the description of FIG. 11 is 18.36. That is, the standard deviation corresponding to window # (N-9) has a value less than the third threshold. In the same way, the standard deviation corresponding to window # (N-8) was calculated to be 6.38, which is less than the third threshold value, as shown in FIG. 12 .

On the other hand, the standard deviation corresponding to window #N is calculated as 23.31 and corresponds to the third threshold or higher. That is, since there exists a window having a standard deviation greater than the third threshold among windows including the N-th MSE value, the I-frame corresponding to the N-th MSE value is determined to be a dynamic frame.

That is, for example, in order to determine whether an I frame corresponding to MSE corresponding to No. 24 corresponds to a dynamic frame, the dynamic frame determination unit 300 calculates the standard deviation of all windows including the MSE value corresponding to No. 24. In this case, window #15 to window #24 correspond to the window including the MSE value corresponding to number 24. If any one of the standard deviations corresponding to the windows #15 to #24 is greater than or equal to the third threshold, the I frame corresponding to the MSE value corresponding to the number 24 corresponds to a dynamic frame. If all of the standard deviations corresponding to windows #15 to #24 are calculated to be less than the third threshold, the I frame corresponding to the MSE value corresponding to the number 24 is not determined as a dynamic frame.

FIG. 13 schematically illustrates the overall execution steps of a method for detecting a dynamic frame existence period including a dynamic frame determining step (S300) according to an embodiment of the present invention.

As shown in FIG. 13, the dynamic frame determination step (S300) includes calculating an average value and standard deviation of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values (S340); Comparing the standard deviation with a fourth threshold (S341); For one or more windows including the N-th MSE value, if there exists a window in which the standard deviation of the plurality of MSE values included in the window is greater than the fourth threshold, an I frame corresponding to the N-th MSE value is determined as a first preliminary dynamic frame. A first determination step (S342); Comparing the average value with a fifth threshold (S343); For one or more windows including the N-th MSE value, if there exists a window in which the average value of the plurality of MSE values included in the window is greater than the fifth threshold, an I frame corresponding to the N-th MSE value is determined as a second preliminary dynamic frame. A second determination step (S344); And determining the I-frame as a dynamic frame with respect to the I-frame determined as the first pre-dynamic frame in the first determination step and the second preliminary-dynamic frame in the second determination step (S345);

The fourth threshold value includes a fourth reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; Calculating a standard deviation of the MSE values calculated from the fourth reference MSE calculation step; and a fourth threshold value determining step of determining the fourth threshold based on the standard deviation.

The fifth threshold value includes a fifth reference MSE calculation step of calculating an MSE value for each frame of training image data including an I frame labeled as having no motion; a fifth reference mean value calculation step of calculating an average value of the MSE values calculated from the fifth reference MSE calculation step; and a fifth threshold value determination step of determining the fifth threshold value based on the average value calculated in the fifth reference mean value calculation step.

Meanwhile, the dynamic frame determination step (S300) shown in FIG. 13 may be performed as an embodiment of the present invention, and may be performed as an embodiment different from the embodiment described above in the description of FIGS. 5, 7, and 11. Before the dynamic frame determination step (S300) is performed, referring to the description of FIG. 4, the I frame decoding step (S100) and the MSE calculation step (S200) are performed.

Specifically, in step S340, the sliding window method is applied to the plurality of MSE values calculated in the MSE calculation step (S200). The sliding window method applied here proceeds in the same manner as described above in the description of FIGS. 8 to 9 . After applying the sliding window method, the average value and standard deviation of the MSE values included in the window including the Nth MSE value may be calculated by the dynamic frame determining unit 300 . Preferably, calculating the average of the MSE values is performed before calculating the standard deviation.

In step S341, a window having a standard deviation exceeding the fourth threshold is detected among the standard deviations corresponding to window #(N-9) to window #N calculated in step S340.

In step S342, if a window having a standard deviation exceeding the fourth threshold is detected among standard deviations corresponding to windows # (N-9) to window # N, the I frame corresponding to the N-th MSE value is determined as the first preliminary dynamic frame. The first determination step is performed. As an embodiment of the present invention, the first determination step performed in steps S340 to S342 may be performed through the same steps as the dynamic frame determination step (S300) in the embodiment described in the description of FIG. 11 .

In step S343, a window having an average value exceeding the fifth threshold among average values corresponding to window #(N-9) to window #N calculated in step S340 is detected.

In step S344, if a window having an average value exceeding the fifth threshold is detected among the average values corresponding to windows # (N-9) to window # N, the I frame corresponding to the N-th MSE value is the second pre-dynamic. The second determination step of determining the frame is performed.

In step S345, the I frame determined to be the first preliminary dynamic frame in the first determination step and the second preliminary dynamic frame in the second determination step is determined to be a dynamic frame.

As described above, in the embodiment of the dynamic frame determination step (S300) in the description of FIG. 13, by further performing the second determination step in the embodiment in FIG. That is, it is preferable to use the embodiment in FIG. 13 when it is desired to detect only a section where a vehicle has passed for an image recorded by a CCTV installed in a place such as an alley where people and cars go together, for example.

Meanwhile, the fourth threshold, like the third threshold, may be set through the learning image data. Referring to the description of FIG. 5, the learning image data corresponds to data generated by extracting an I frame of a section in which motion does not exist from image data including an image in which motion does not exist. The fourth reference MSE calculation step of calculating a plurality of MSE values corresponding to each I frame of the learning image data is performed, and then the standard deviation of the MSE value is calculated. After calculating the standard deviation, the fourth threshold value determining step of determining the fourth threshold value based on the standard deviation may be performed.

Also, the fifth threshold, like the first threshold, may be set through the learning image data. The learning image data corresponds to data generated by extracting an I frame of a section in which motion does not exist from image data including an image in which motion does not exist, with reference to the description of FIG. A fifth threshold value determination step may be performed.

FIG. 14 schematically shows a table comparing average values and standard deviations of MSE values included in several windows with fourth and fifth threshold values, respectively, according to an embodiment of the present invention.

Schematically, FIG. 14(a) shows a table comparing standard deviations corresponding to windows #(N-9) to window #N with the fourth threshold as an embodiment of the present invention in order to perform the first judgment step, and FIG. show

Meanwhile, the table shown in FIG. 14 is shown for description of the embodiment dealt with in the description of FIG. 13 in the present invention.

As shown in (a) of FIG. 14, as an embodiment of the present invention, window # (N-9) includes MSE values corresponding to (N-9) to N, and the standard deviation of the MSE values corresponding to (N-9) to N is 5.12. Meanwhile, the value of the fourth threshold calculated with reference to the description of FIG. 13 is 18.36. That is, the standard deviation corresponding to window #(N-9) has a value less than the fourth threshold. In the same manner, the standard deviation corresponding to window # (N-8) was calculated to be less than the fourth threshold, as shown in (a) of FIG. 14 .

On the other hand, the standard deviation corresponding to window #N is calculated as 23.31 and corresponds to the fourth threshold or higher. That is, since there is a window having a standard deviation greater than the fourth threshold among windows including the N-th MSE value, the I-frame corresponding to the N-th MSE value is determined to be the first preliminary dynamic frame. If, as another embodiment of the present invention, if the standard deviations corresponding to the windows # (N-9) to windows # N are all less than the fourth threshold, the I frame corresponding to the N-th MSE value is not determined as a first preliminary dynamic frame.

As shown in (b) of FIG. 14, as an embodiment of the present invention, window # (N-9) includes MSE values corresponding to (N-9) to N, and the result of calculating the average value of the MSE values corresponding to (N-9) to N is 2.8. Meanwhile, the value of the fifth threshold calculated with reference to the description of FIG. 13 is 12.41. That is, the average value corresponding to window #(N-9) has a value less than the fifth threshold. In the same way, the average value corresponding to window #(N-8) was calculated to be less than the fifth threshold, as shown in FIG. 14(b).

On the other hand, the average value of the MSE values corresponding to window #N, that is, N to (N+9), is calculated as 132.4 and corresponds to the fifth threshold or higher. That is, since there is a window having a standard deviation greater than the fifth threshold among windows including the N-th MSE value, the I-frame corresponding to the N-th MSE value is determined to be the second preliminary dynamic frame.

In step S345, as a result of the determination in the first determination step and the determination result in the second determination step, the I frame corresponding to the Nth MSE value is determined to be a dynamic frame.

15 illustratively illustrates the internal configuration of a computing device 11000 according to an embodiment of the present invention.

All or some of the components shown in FIG. 1 may include components of the computing device shown in FIG. 15 to be described later.

As shown in FIG. 15, a computing device 11000 may include at least one processor 11100, memory 11200, peripheral interface 11300, I/O subsystem 11400, power circuit 11500, and communication circuit 11600.

Specifically, the memory 11200 may include, for example, a high-speed random access memory, a magnetic disk, SRAM, DRAM, ROM, flash memory, or non-volatile memory. The memory 11200 may include a software module, a command set, or other various data necessary for the operation of the computing device 11000.

In this case, access to the memory 11200 from other components, such as the processor 11100 or the peripheral device interface 11300, may be controlled by the processor 11100. The processor 11100 may be composed of single or multiple processors, and may include GPU and TPU type processors in order to improve calculation processing speed.

The peripheral interface 11300 may couple input and/or output peripherals of the computing device 11000 to the processor 11100 and the memory 11200 . The processor 11100 may execute a software module or instruction set stored in the memory 11200 to perform various functions for the computing device 11000 and process data.

The input/output subsystem 11400 can couple various input/output peripherals to the peripheral interface 11300. For example, the input/output subsystem 11400 may include a controller for coupling a peripheral device such as a monitor, keyboard, mouse, printer, or touch screen or sensor to the peripheral device interface 11300 as needed. According to another aspect, the peripheral input/output devices may be coupled to the peripheral device interface 11300 without going through the input/output subsystem 11400.

The power circuit 11500 may supply power to all or some of the terminal's components. For example, the power circuit 11500 may include a power management system, one or more power sources such as a battery or alternating current (AC), a charging system, a power failure detection circuit, a power converter or inverter, a power status indicator, or any other component for generating, managing, or distributing power.

The communication circuit 11600 may enable communication with another computing device using at least one external port. Alternatively, as described above, the communication circuit 11600 may include an RF circuit and transmit/receive an RF signal, also known as an electromagnetic signal, to enable communication with another computing device.

The embodiment of FIG. 15 is just one example of the computing device 11000, and the computing device 11000 may omit some of the components shown in FIG. 15, may further include additional components not shown in FIG. 15, or may have a configuration or arrangement in which two or more components are combined. For example, a computing device for a communication terminal in a mobile environment may further include a touch screen or a sensor in addition to the components shown in FIG. Components that may be included in the computing device 11000 may be implemented as hardware including one or more signal processing or application-specific integrated circuits, software, or a combination of both hardware and software.

Methods according to embodiments of the present invention may be implemented in the form of program instructions that can be executed through various computing devices and recorded in computer readable media. In particular, the program according to the present embodiment may be composed of a PC-based program or a mobile terminal-specific application. An application to which the present invention is applied may be installed in a user terminal through a file provided by a file distribution system. For example, the file distribution system may include a file transmission unit (not shown) that transmits the file according to a request of a user terminal.

According to an embodiment of the present invention, by distinguishing and decoding only I-frames from a bitstream of the entire video data, an effect of detecting a section in which motion exists in the video data can be achieved in less time than conventional techniques.

According to an embodiment of the present invention, by performing image processing only for a minimum number of frames, it is possible to achieve an effect requiring less computational processing capability than the conventional technology.

According to an embodiment of the present invention, it is possible to achieve an effect of detecting a section in which motion exists at high speed with high accuracy despite a small amount of computation.

According to an embodiment of the present invention, it is possible to perform object detection and object analysis only for frames in which motion is determined to exist, so that the amount of calculations processed during object analysis can be reduced to a significant level.

According to an embodiment of the present invention, although a significant computational load is required for recent high-definition CCTV images, an effect of reducing the analysis target area to a considerable level can be achieved only with frames in which motion exists.

According to an embodiment of the present invention, by patterning the calculated MSE value and then learning it, the type of object can be identified, and object movement information in a specific area can be obtained based on the learned content.

According to an embodiment of the present invention, the dynamic frame determination unit can detect dynamic frames using various methods, and the various methods can be used according to the purpose of detecting dynamic frames.

The device described above may be implemented as a hardware component, a software component, and/or a combination of hardware components and software components. For example, the devices and components described in the embodiments may be implemented using one or more general purpose computers or special purpose computers, such as, for example, a processor, controller, arithmetic logic unit (ALU), digital signal processor, microcomputer, field programmable gate array (FPGA), programmable logic unit (PLU), microprocessor, or any other device capable of executing and responding to instructions. The processing device may run an operating system (OS) and one or more software applications running on the operating system. A processing device may also access, store, manipulate, process, and generate data in response to execution of software. For convenience of understanding, there are cases in which one processing device is used, but those skilled in the art will recognize that the processing device may include a plurality of processing elements and/or multiple types of processing elements. For example, a processing device may include a plurality of processors or a processor and a controller. Other processing configurations are also possible, such as parallel processors.

Software may include a computer program, code, instructions, or a combination of one or more of these, and may configure a processing device to operate as desired, or may independently or collectively direct a processing device. Software and/or data may be permanently or temporarily embodied in any tangible machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, to be interpreted by or to provide instructions or data to a processing device. Software may be standardized on a networked computing device and stored or executed in a standardized manner. Software and data may be stored on one or more computer readable media.

The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer readable medium. The computer readable medium may include program instructions, data files, data structures, etc. alone or in combination. Program commands recorded on the medium may be specially designed and configured for the embodiment or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program instructions such as ROM, RAM, and flash memory. Examples of program instructions include high-level language codes that can be executed by a computer using an interpreter, as well as machine language codes such as those produced by a compiler. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the embodiments, and vice versa.

As described above, although the embodiments have been described with limited examples and drawings, those skilled in the art can make various modifications and variations from the above description. For example, even if the described techniques are performed in an order different from the described method, and/or components of the described system, structure, device, circuit, etc. are combined or combined in a different form from the described method, or replaced or substituted by other components or equivalents, appropriate results can be achieved.

Therefore, other implementations, other embodiments, and equivalents of the claims are within the scope of the following claims.

Claims (13)

A method for detecting a section in which a dynamic frame exists for video data encoded according to the HEVC standard performed in a computing system including one or more processors and a main memory storing instructions executable by the processor,
an I-frame decoding step of detecting a NAL unit from video data, analyzing a header of the NAL unit, distinguishing only a plurality of I-frames from the entire bitstream of the video data, and decoding only the I-frames;
MSE calculation step of calculating an MSE value between two adjacent I frames based on a pixel difference between the two adjacent I frames with respect to the decoded I frame;
a dynamic frame determination step of determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and
A dynamic frame existence section detection step of detecting a section in which the dynamic frame exists in the image data;
The dynamic frame determination step,
Calculating standard deviations of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values;
comparing the standard deviation with a second threshold; and
Determining a plurality of I frames corresponding to the plurality of MSE values included in a window having a standard deviation greater than the second threshold as dynamic frames;
The second threshold,
A second reference MSE calculation step of calculating an MSE value for each I frame included in the training image data generated by extracting an I frame labeled as having no motion from image data including both video sections with or without motion;
Calculating a standard deviation of the MSE values calculated from the second reference MSE calculation step; and
A second threshold value determining step of determining the second threshold based on the standard deviation; a method for detecting a section in which a dynamic frame exists.
delete delete delete delete delete delete delete delete delete delete An apparatus for detecting a frame in which motion exists for image data encoded according to the HEVC standard, including one or more processors and a main memory storing instructions executable by the processor,
an I frame decoding unit that detects a NAL unit from video data, analyzes a header of the NAL unit, distinguishes only I frames from the entire bitstream of the video data, and decodes only the I frames;
an MSE calculation unit for calculating an MSE value between two adjacent I frames based on a pixel difference between two adjacent I frames with respect to the decoded I frame;
a dynamic frame determination unit determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and
A dynamic frame existence section detecting section for detecting a section in which the dynamic frame exists in the image data;
The dynamic frame determination unit,
Calculating standard deviations of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values;
comparing the standard deviation with a second threshold; and
Determining a plurality of I frames corresponding to the plurality of MSE values included in a window having a standard deviation greater than the second threshold as dynamic frames;
The second threshold,
A second reference MSE calculation step of calculating an MSE value for each I frame included in the training image data generated by extracting an I frame labeled as having no motion from image data including both video sections with or without motion;
Calculating a standard deviation of the MSE values calculated from the second reference MSE calculation step; and
A second threshold value determining step for determining the second threshold value based on the standard deviation;
A computer program for detecting a section in which a dynamic frame exists, stored in a computer-readable medium, including a plurality of instructions executed by one or more processors, and performed on image data encoded according to the HEVC standard,
The computer program,
an I-frame decoding step of detecting a NAL unit from video data, analyzing a header of the NAL unit, distinguishing only I-frames from an entire bitstream of the video data, and decoding only the I-frames;
MSE calculation step of calculating an MSE value between two adjacent I frames based on a pixel difference between the two adjacent I frames with respect to the decoded I frame;
a dynamic frame determination step of determining whether the I-frame is a dynamic frame in which motion exists, based on the MSE value; and
A dynamic frame existence section detection step of detecting a section in which the dynamic frame exists in the image data;
The dynamic frame determination step,
Calculating standard deviations of a plurality of MSE values included in each window by applying a sliding window method to the calculated MSE values;
comparing the standard deviation with a second threshold; and
Determining a plurality of I frames corresponding to the plurality of MSE values included in a window having a standard deviation greater than the second threshold as dynamic frames;
The second threshold,
A second reference MSE calculation step of calculating an MSE value for each I frame included in the training image data generated by extracting an I frame labeled as having no motion from image data including both video sections with or without motion;
Calculating a standard deviation of the MSE values calculated from the second reference MSE calculation step; and
A second threshold value determination step of determining the second threshold value based on the standard deviation; determined by, a computer program.