WO2020082866A1

WO2020082866A1 - Motion trajectory synthesising method and electronic device

Info

Publication number: WO2020082866A1
Application number: PCT/CN2019/101435
Authority: WO
Inventors: 魏祺
Original assignee: 中兴通讯股份有限公司
Priority date: 2018-10-25
Filing date: 2019-08-19
Publication date: 2020-04-30
Also published as: CN111105434A

Abstract

Disclosed in the embodiments of the present application are a motion trajectory synthesising method and an electronic device, relating to the technical field of image processing. In an embodiment of the present invention, the motion trajectory synthesising method comprises: acquiring a plurality of image frames, the plurality of image frames being ordered according to acquisition time; according to a preset method, respectively dividing the plurality of image frames into a plurality of grid areas; selecting several image frames meeting preset conditions from the plurality of image frames, the preset conditions comprising: the several image frames all include a moving body and the grid areas occupied by the moving body in any two image frames do not overlap; and, on the basis of the selected several image frames, synthesising the motion trajectory of the moving body. Also provided in the embodiments of the present application is an electronic device.

Description

Motion track synthesis method and electronic equipment

cross reference

The present invention requires the priority of a Chinese patent application filed on October 25, 2018 in the Chinese Patent Office with the application number 201811250802.2 and the invention titled "Motion Track Synthesis Method and Electronic Equipment". The entire contents of this application are incorporated herein by reference Invented.

Technical field

The embodiments of the present invention relate to the technical field of image processing, and in particular, to a motion track synthesis method and an electronic device.

Background technique

Taking pictures and videos of scenes of moving subjects is one of the common life scenes of taking pictures with cameras. Wonderful sports scenes are always fleeting, such as slam dunks during basketball games, dribbling shots during football games, high jumps, long jumps, etc. When the main body of the activity is active, a sports silhouette is generated, which combines multiple sports images into one. In the image, a series of silhouette images of the trajectory of the moving subject can be synthesized.

In some cases, the inventor found that there are at least the following problems: the moving speed of the moving subject is different, the direction of the movement is uncertain, and the size of the moving subject is different. In the composite image, the overlapping of the moving subjects may appear, and the complete motion cannot be seen. The action of the subject is shown in Figure 1, so the synthesis effect is not very good.

Summary of the invention

The purpose of the embodiments of the present invention is to provide a method for synthesizing a motion track and an electronic device, so that the motion subjects in the synthesized motion track do not overlap, and the synthesis effect is better.

To solve the above technical problems, an embodiment of the present invention provides a method for synthesizing a motion track, including: acquiring multiple frames of images; wherein the multiple frames of images are arranged according to the acquisition time; the multiple frames of images are divided according to a preset manner Into a plurality of grid regions; selecting from the multi-frame images several frames of images that meet a preset condition; wherein, the preset conditions include that the several frames of images all contain a moving subject and motion in any two frames of images The grid areas occupied by the main body do not overlap; the motion trajectory of the moving main body is synthesized according to the selected frames of images.

An embodiment of the present invention further provides an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; a camera connected to the at least one processor and the memory; wherein , The camera is used to collect images; the memory stores instructions executable by the at least one processor, the instructions are executed by the at least one processor, so that the at least one processor can perform the above movement Track synthesis method.

Embodiments of the present invention also provide a computer-readable storage medium that stores a computer program, and when the computer program is executed by a processor, the above-mentioned motion track synthesis method is implemented.

BRIEF DESCRIPTION

One or more embodiments are exemplified by the pictures in the corresponding drawings. These exemplary descriptions do not limit the embodiments. Elements with the same reference numerals in the drawings are represented as similar elements. Unless otherwise stated, the figures in the drawings do not constitute a scale limitation.

FIG. 1 is a schematic diagram of a motion trajectory synthesized in the prior art;

2 is a flowchart of a method for synthesizing a motion track according to the first embodiment of the present invention;

FIG. 3 is a schematic diagram of the synthesized motion track according to the first embodiment of the present invention;

FIG. 4 is a flowchart of a motion track synthesis method according to a second embodiment of the present invention;

FIG. 5 is a flowchart of a motion track synthesis method according to a third embodiment of the present invention;

6 is a schematic diagram of a grid area divided according to the third embodiment of the present invention;

7 is a flowchart of a screening method in sub-step 3032 according to the third embodiment of the present invention;

8 is a flowchart of a screening method in sub-step 3032 according to the fourth embodiment of the present invention;

9 is a flowchart of a motion track synthesis method according to a fifth embodiment of the present invention;

10 is a schematic diagram of an electronic device according to a sixth embodiment of the present invention.

detailed description

To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the following describes the embodiments of the present invention in detail with reference to the accompanying drawings. However, those of ordinary skill in the art may understand that, in various embodiments of the present invention, many technical details are proposed in order to enable the reader to better understand the application. However, even without these technical details and various changes and modifications based on the following embodiments, the technical solutions claimed in this application can be realized.

The first embodiment of the present invention relates to a motion track synthesis method. The specific process is shown in Figure 2.

Step 101: Acquire multi-frame images; wherein, the multi-frame images are arranged according to the acquisition time;

Step 102: Divide the multi-frame images into multiple grid areas according to a preset method;

Step 103: Select a plurality of frames of images satisfying preset conditions from the multi-frame images; wherein the preset conditions include that several frames of images include moving subjects and the grid area occupied by the moving subjects in any two frames of images does not overlap;

Step 104: Synthesize the motion trajectory of the moving subject according to the selected frames of images.

Compared with the prior art, this embodiment divides the image into a plurality of grid areas according to a preset method; selects from a plurality of frames of images that meet a preset condition for synthesis, where the preset condition includes an image The grid areas occupied by moving subjects in any two frames of images that contain moving subjects do not overlap. Therefore, the grid method is used to select the image for synthesis, which is not affected by the movement speed of the moving subject, the direction of movement, and the change in the size of the moving subject. it is good.

The implementation details of the motion trajectory synthesis method of this embodiment are described below in detail. The following content is merely implementation details provided for easy understanding, and is not necessary for implementing this solution.

The motion trajectory synthesis method of this embodiment can be applied to an electronic device, and the electronic device is, for example, a personal computer, a mobile phone, etc., and is not limited thereto. Please refer to FIG. 2 and FIG. 3 together, where FIG. 3 is a schematic diagram of the synthesized motion track in this embodiment.

In step 101, the electronic device may collect multiple frames of images through a camera; the camera may be integrated on the electronic device or may be externally connected. The camera may acquire images at a preset frequency, where the number of frames of the acquired image may be preset, for example, preset 60 frames, the camera may acquire 60 frames of images at a preset frequency; or, the time for acquiring images may be preset, for example If the preset time is 10 seconds, the camera will collect images at a preset frequency, and the acquisition will stop after 10 seconds. The number of images collected is related to the preset time and preset frequency. The collected multi-frame images are arranged according to the acquisition time.

In step 102, each frame of image may be divided into multiple grid areas in the preset manner. Each area divided by a dotted line in FIG. 3 is a grid area.

The preset method can be set as required. For example, it is preset that the image is divided into X grid areas, then the width W1 and height h1 of each grid area are calculated according to the width W and height h of the image, and then the grid division is performed, where the width W of the image The height h can be calculated from the collected images, or can be preset in advance; or, the width W1 and height h1 of each grid area are preset, and grid division can be performed at this time; the above is only an example, not This is limited. Among them, X is an integer greater than 0. When the image size is constant, the larger X is, the smaller the grid area is.

Among them, because the shooting background of each image is the same (even if the background is slightly different due to jitter during shooting, the background can be aligned through calibration), therefore, when mapped to the same coordinate system, each frame The total area occupied by the image is the same, that is, each frame of the image is divided into grids according to a preset method, and has the same number of grid areas, and the coordinate positions of the corresponding grid areas are the same; The grid regions of the frame image are encoded in the same order (for example, from the first row to the last row, and sequentially encoded from left to right), and the same grid region in each frame image is the corresponding raster. Grid area.

Step 103 can be implemented in the following manner.

First, in accordance with the order of arrangement, each frame of the image is sequentially motion-recognized, and the images containing the moving subject are identified, and the images containing the moving subject are also arranged in sequence according to the acquisition time; wherein, motion recognition can be used in the prior art The method of extracting and matching based on feature points will not be described in detail here. Suppose that among the 60 frames of images collected, only the 6th frame to the 45th frame contains moving subjects, while the 1st frame to the 5th frame images and the 46th frame to the 60th frame images do not contain moving subjects. , Then only the 6th frame to the 45th frame image can be saved, and the 1st frame to the 5th frame image, the 46th frame image to the 60th frame image can be discarded (ie deleted).

Secondly, for images containing moving subjects, determine the grid area occupied by moving subjects in each image, and select several frames of images from these images containing moving subjects. The moving subjects in any two frames of several frames of images Does not overlap. Specifically, for example, each grid region can be numbered; then select a frame of image first, and determine the number of the grid region occupied by the moving subject in the selected image. The selected image is considered to meet the preset conditions Image, as in the above example, you can select the sixth frame image, and confirm that the number of the grid area occupied by the moving subject in the sixth frame image is 1, 2, where the sixth frame image is considered to meet the preset conditions Images; then, according to the order of the images, the grid areas occupied by the moving subjects in each image are separately judged. As in the above example, the images from the 7th frame to the 45th frame are sequentially filtered, assuming the 7th frame If the number of grid areas occupied by moving subjects in the image is 3 or 4, it is determined that the grid area occupied by moving subjects in the seventh frame image does not overlap with the grid area occupied by moving subjects in the sixth frame image (number 3, The grid area of 4 and the grid areas of numbers 1 and 2 do not overlap), so the image of the seventh frame is considered to meet the preset conditions; the number of grid areas occupied by moving subjects in the image of the eighth frame is 4, 6, 7, it is determined that the 8th frame of the image The grid area occupied by the moving subject overlaps with the grid area occupied by the moving subject in the 7th frame image (the 8th frame image and the 7th frame image both occupy the grid area of number 4), therefore, the 8th frame image is not The image that meets the preset conditions can be discarded; the number of the grid area occupied by the moving subject in the 9th frame image is 7, 9, 10, then the grid area occupied by the moving subject in the 9th frame image and the first The grid areas occupied by moving subjects in the 6-frame image and the 7th frame image do not overlap (the grid areas numbered 7, 9, and 10 and the grid areas numbered 1 to 4 do not overlap), so the 9th frame image can be It is considered as an image that satisfies the preset conditions; and so on, until the screening of the 45th frame image is completed. Therefore, the grid areas occupied by the moving subjects in any two frames of the selected frames of images do not overlap. Among them, the selected image can be set as needed, any frame of the image containing the moving subject can be used as the selected image, for example, the first frame of the image as the selected image (such as the 6th in the above example) Frame image).

Alternatively, step 103 can also be implemented in the following manner. According to the arrangement order, perform motion recognition on each frame in sequence and when recognizing the image containing the moving subject, determine the grid area occupied by the moving subject in the image, and determine the grid occupied by the moving subject in the image Whether the grid area overlaps with the grid area occupied by moving subjects in other images. As in the above example, motion recognition is performed on the first frame image to the fifth frame image in sequence, and no moving subject is identified, then the first frame image to the fifth frame image are discarded; the sixth frame image is subjected to motion recognition and Identified the moving subject, and determined that the grid area occupied by the moving subject is numbered 1, 2. Since the sixth frame image is the first to identify the moving subject, the sixth frame image is considered to meet the pre- Set the conditional first image, perform motion recognition on the seventh frame image and identify the moving subject, and determine the number of the grid area occupied by the moving subject as 3, 4, then determine the moving subject in the seventh frame image The occupied grid area does not overlap with the grid area occupied by the moving subject in the sixth frame image (the grid areas numbered 3 and 4 and the grid areas numbered 1 and 2 do not overlap), so the seventh frame image is considered to be satisfactory Images with preset conditions; the recognition method of the 8th frame image to the 45th frame image is similar to the above example, and will not be repeated here; the moving subjects are not recognized in the 46th frame image to the 60th frame image, then the Images from frame 46 to frame 60 are discarded.

It should be emphasized that, as mentioned in this embodiment and subsequent embodiments, the moving subjects in the image refer to the pictures of the moving subjects in the image.

In step 104, if the moving subjects in the image are separated from the background image, the moving subjects separated in each frame image are combined on a background image, if the moving subjects in the image are not separated from the background image Coming directly, if there is jitter during overshooting, there may be some deviation between the background images of each frame of the image, you can calibrate the deviation before synthesizing). The synthesis methods here are similar to the prior art, and will not be repeated here.

The second embodiment of the present invention relates to a motion track synthesis method. The second embodiment is substantially the same as the first embodiment, and the main difference is that in the second embodiment of the present invention, motion recognition will be performed on the currently collected image, and the image containing the moving subject is selected, and it is repeated multiple times in a row When no moving subject is identified in the collected image, the collection is stopped.

4 is a flowchart of a method for synthesizing a motion track according to a second embodiment of the present invention; wherein steps 202-204 are substantially the same as steps 102-104 in FIG. 2 and will not be repeated here, the difference is that , Step 201, acquiring multiple frames of images, including the following sub-steps:

Substep 2011, acquiring images according to a preset frequency;

Sub-step 2012, perform motion recognition on the currently collected image and judge whether the recognition is successful; if yes, enter sub-step 2013, if not, enter sub-step 2014;

Sub-step 2013, add the currently acquired image to the first image sequence set; return to sub-step 2011.

Sub-step 2014, discard the currently acquired image; then enter sub-step 2015.

Substep 2015, accumulate the number of frames of consecutively discarded images, and determine whether the preset number of frames K is reached; if it is, stop the acquisition and go to step 202; if not, return to substep 2011. Where K is the preset number of frames and L is greater than 1

In this embodiment, a first image sequence set, such as a sequence P, may be preset in the electronic device. The electronic device performs motion recognition on the currently collected image in real time. If a moving subject is recognized from the currently collected image, indicating that the recognition is successful, the currently collected image is added to the first image sequence set P, which is regarded as the first image sequence set An element Pi of P, i is an integer greater than or equal to 1, i represents the sequence number of the image in the first image sequence set P; if no moving subject is identified from the currently collected image, indicating that the recognition failed, then the The currently acquired image is discarded. Since the images added to the first image sequence set are added in the order of acquisition, the images in the first image sequence set are arranged according to the acquisition time. For example, the sequence P = {P1, P2, P3, ..., Pm} indicates that m frames of images have been acquired.

The electronic device also accumulates the number of consecutively discarded images and determines whether the number of consecutively discarded images reaches the preset number of frames K. If the number of consecutively discarded images reaches the preset number of frames K, the No moving subject is identified in the collected K frame images, it is considered that no moving subject has been detected, and the acquisition is stopped at this time. Among them, the preset number of frames K can be set according to actual experience or actual needs. For example, K is 10, and if no moving subjects are identified in the first to tenth frames of the collected image, the number of frames of consecutively discarded images is 10, which reaches the preset number of frames K, so it is considered to be moving If the subject never enters the shooting screen, stop the acquisition (this time the trajectory synthesis ends); or, from the first frame image to the 45th frame of the collected images, the moving subject is recognized, and from the 46th frame image, it has been continuous After 10 frames were collected, no moving subjects were identified in the collected images, that is, the 46th frame to the 56th frames of the collected images did not identify the moving subject. It is considered that the moving subject has run out of the shooting screen and the collection is stopped. ; And take the first frame image to the 45th frame image as the acquired multi-frame image. Assuming that moving subjects are recognized in the collected first frame image to the 30th frame image, and no moving subjects are recognized in the collected 31st frame image to the 35th frame image, the collected 36th frame image to the 50th frame image All of the moving subjects are recognized, then the 31st frame image to the 35th frame image are discarded, but because the discarded image has only 5 frames, which is less than the preset frame number K, so the collection will not stop, And record the 36th frame image to the 50th frame image acquired subsequently, and use the 1st frame image to the 30th frame image, the 36th frame image to the 50th frame image as the acquired image, that is, a total of 45 frame images are acquired.

In this embodiment, in the synthesis of motion trajectories, the method of motion recognition is used to preliminarily screen the collected images, whether to continue to collect images or to stop collecting images, is determined by whether the moving subject is located in the shooting screen, rather than collected as in the prior art A preset fixed frame number of images; that is, in this embodiment, if it is detected that no moving subject is identified in the continuously acquired K frame images, then even if the number of currently acquired images containing moving subjects does not reach the preset At a fixed number of frames, the collection will stop, because even if the collection continues, because the moving subject is no longer in the shooting screen, the collected image has no practical use for the synthesis of the movement track; therefore, this embodiment is more in line with the synthesis of the movement track when acquiring the image The actual demand for the image has improved the effectiveness of image acquisition, which can improve the synthesis efficiency of the motion track.

The third embodiment of the present invention relates to a motion track synthesis method. The third embodiment is substantially the same as the second embodiment, and the main difference is that in the third embodiment of the present invention, a specific way of dividing a plurality of grid regions is provided.

FIG. 5 is a flowchart of the motion track synthesis method of the third embodiment.

Steps

301 and 304 are substantially the same as

steps

201 and 204 in FIG. 4 and will not be repeated here. The difference is that

Step 302, dividing the image into multiple grid regions according to a preset method, including the following sub-steps:

Sub-step 3021, select a frame of images containing moving subjects from the multi-frame images, and determine the target frame area of the moving subjects in the selected image; wherein, the moving subjects in the selected image are located in the target frame area.

Since the third embodiment is based on the improvement of the second embodiment, the multi-frame images acquired here are all images containing moving subjects and are arranged in order according to the acquisition order; that is, the first image sequence set P in the second embodiment Multi-frame images in. The selected image may be any frame image in the first image sequence set P; for example, the first frame image P1 in the first image sequence set P may be selected, and the target of the moving subject in the selected image P1 may be determined The frame area R1, or, one image in the middle of the first image sequence set P can be selected; or, the selected image can also be the motion in the first few images in the first image sequence set P The image with the largest subject, for example, compare the size of the moving subject in the first three frames P1, P2, P3 of the first image sequence set P, and select the larger frame of the moving subject, if the second frame of the image The medium moving subject is the largest, then the selected image is the second frame image P2 in the first image sequence set P. It should be noted that the selected image is not limited to the above example, the designer can set it according to actual needs, for example, it can also calculate the average of the size of the moving subject in the obtained multi-frame images or several of them, and then average The image with the moving subject near the value is selected as the image. For the same moving subject, the closer to the camera when moving, the larger the moving subject in the collected image, and the farther away from the camera when moving, the smaller the moving subject in the collected image.

In this embodiment, as shown in FIG. 6, the target frame area R1 is a rectangular frame area, and the target frame area R1 may be the area where the smallest circumscribed rectangular frame of the moving subject is located, thereby ensuring that the moving subject is located in the target frame area R1. Alternatively, the target frame area R1 may be similar to the area of the smallest circumscribed rectangular frame, but the area is larger than the area of the smallest circumscribed rectangular frame, that is, after the smallest circumscribed rectangular frame of the moving subject is identified, the smallest circumscribed rectangular frame is enlarged , The target frame area R1 is slightly larger than R1min, but it is best not to exceed 1/4 of the area of the smallest circumscribed rectangular frame; that is, the area ratio of the target frame area and the area of the smallest circumscribed rectangular frame is greater than 1 and less than or equal to 5 / 4.

Sub-step 3022, the selected image is divided into a plurality of grid areas according to the size of the target frame area; wherein the target frame area is located in one of the grid areas.

In this embodiment, the target frame region R1 is used as a basic cell to divide the image into grids. For example, for the width W and height H of an image, the coordinate position of one of the corner points of the target frame area R1 can be used as the starting point, and the width w1 and height h1 of the target frame area R1 can be used as standard Divided by the interval w1, the height H of the entire image is divided into grid regions every h1; preferably, for example, the coordinate position of the upper corner of the target frame region R1 is used as the starting point. Therefore, the target frame area is located in one of the grid areas; in this embodiment, the target frame area is a grid unit, but it is not limited to this. It should be noted that due to the limitation of the image width W and height H, the grid area at the edge of the image may be smaller than the target frame area.

Step 303, select several frames of images that satisfy the preset conditions from the multi-frame images; where the preset conditions include that the image contains moving subjects and the grid area occupied by the moving subjects in any two frames of images does not overlap, including the following sub step:

Sub-step 3031, add the selected image to the second image sequence set, and mark the grid area where the selected image is located as occupied.

Among them, the second image sequence set Q can be preset in the electronic device to save the images that meet the preset conditions; the selected image is used as the basis for comparing other images, so the selected image is considered to meet the preset The conditional image is added to the second image sequence set Q, and the grid area where the selected image is located is marked as occupied, where the grid area where the selected image is located is the target frame area R1. In this embodiment, the grid area that has been occupied by the moving subject is marked to facilitate subsequent comparison.

In sub-step 3032, for each frame of the multi-frame images except the selected image, the screening is performed in the order of arrangement.

As shown in Figure 7, the screening methods include,

Sub-step 3032-1, judging whether the grid area where the moving subject in the image is marked as occupied; if not, enter sub-step 3032-2; if yes, enter sub-step 3032-3;

Sub-step 3032-2, add the image to the second image sequence set, and mark the grid area where the moving subject in the image is occupied as occupied.

Substep 3032-3, discard the image.

For each frame of the first image sequence set P except the selected image, first determine the grid area where the moving subject is located, and then determine whether the grid area where the moving subject is located in the image is marked as occupied. In this embodiment, you can first determine the target frame area of the moving subject in the image, and then determine whether the grid area where the target frame area of the moving subject in the image is marked as occupied; if it is not marked as occupied, then If the image is considered to meet the preset condition, the second image sequence set Q is added; if it is marked as occupied, the image is deemed not to meet the preset condition and the image is discarded. Among them, the number of grid areas where the target frame area of the moving subject is located may be one or more. If the number of grid areas where the target frame area of the moving subject is located is multiple, as long as one grid area is marked as If it is occupied, it is considered that the image does not meet the preset conditions.

For example, to set the second image sequence set Q, if the image selected from the first image sequence set P = {P1, P2, P3, ..., Pm} is P1, then let Q1 = P1 and set P1 The grid area where it is located is marked as occupied; P2 ~ Pm in the first image sequence set P = {P1, P2, P3, ..., Pm} are sequentially filtered, and the moving subject in the currently filtered image is first determined For the target frame area, when the target frame area is not marked as occupied, the image is added to the second image sequence set Q, and if it is marked as occupied, the image is discarded. Specifically, you can number the grid area, use the number to indicate the grid area where the moving subject in the image is located, determine the number of the grid area where the moving subject in the image is located, and determine whether the grid area indicated by the number has been marked For occupation, if it has been marked as occupied, it means that the image does not meet the preset conditions. For example, if the number of the grid area where P1 is located is 3, the number 3 grid area is marked as occupied; the number of the grid area where P2 is located is 5 because the number 5 grid area is not marked as occupied , Then add P2 to the second image sequence set Q, that is, let Q2 = P2, and mark the grid area numbered 5 as occupied; the grid area where P3 is located is numbered 6, 7, due to the number 6, 7 If the grid area is not marked as occupied, P3 is added to the second image sequence set Q, that is, Q3 = P3, and at the same time, the grid areas numbered 6 and 7 are marked as occupied; the grid area number where P4 is located Is 7, 8, because the grid area numbered 7 is marked as occupied, so discard P4; the grid area where P5 is located is numbered 8, because the grid area numbered 8 is not marked as occupied, then P5 Add the second image sequence set Q, let Q4 = P5, and mark the grid area number 8 as occupied; and so on, until the screening of Pm is finished, the second image sequence set Q = {Q1, Q2, Q3 ..., Qs}, which means that s that meet the preset conditions are selected from the m frames of the first image sequence set P Image, wherein, s is an integer less than or equal to m.

In this embodiment, a specific way of dividing the grid area is proposed, that is, the target frame area of the moving subject in a frame of image is selected as the basic unit for division, wherein the moving subject in the selected image is located in the target frame area ; The size of the grid area divided in this way is appropriate, so that the overlap judgment can be performed more quickly; because relative to the size of the moving subject, if the grid area is too small, the moving subject in each image will occupy If there are multiple grid areas, the amount of work spent in overlapping judgment will be greater (because it is checked whether each grid area is marked as occupied). If the grid area is too large, the selected conditions meet the preset conditions The distance between moving subjects in the image will be relatively large, making the distance between two adjacent moving subjects in the last synthesized motion trajectory larger, affecting the synthesis effect.

It should be noted that the third embodiment may also be an improvement on the basis of the first embodiment; when the third embodiment improves on the basis of the first embodiment, since step 301 does not use motion recognition The collected images are filtered, so there may be images that do not contain moving subjects in the acquired multi-frame images; in step 302, the selected image must contain moving subjects. Therefore, the selected image can be many The first frame of the frame image contains the image of the moving subject.

The fourth embodiment of the present invention relates to a motion track synthesis method. The fourth embodiment is substantially the same as the third embodiment, and the main difference is that in the fourth embodiment of the present invention, the image of the moving subject suddenly being too large can be eliminated to eliminate external interference and improve the synthetic effect of the motion track.

8 is a flowchart of the screening method in the fourth embodiment. FIG. 8 is an improvement based on FIG. 7. Before sub-step 3032-1, it also includes sub-step 3032-0 to determine the moving subject in the image. Whether the target frame areas are all located in one of the grid areas; if yes, go to sub-step 3032-1, if no, go to sub-step 3032-3.

Specifically, due to instability during shooting, the camera may suddenly move closer to or away from the moving subject. At this time, the moving subject in the image will suddenly become larger, that is, the target frame area of the moving subject in the image will exceed a grid area. (For example, occupy two or more grid areas at the same time). Among them, when setting the target frame area of the moving subject, the target frame area can be set to be larger than the area where the smallest circumscribed rectangular frame of the moving subject is located, as described in sub-step 3021 of the third embodiment, the target The area ratio of the frame area to the area of the smallest circumscribed rectangular frame can be greater than 1 and less than or equal to 5/4, so that a normal variable margin is left. The normal variable margin refers to the moving subject during the movement The amount of change in the size of moving subjects in the image when normally approaching or away from the camera.

In this embodiment, when selecting images that meet the preset conditions, the judgment on the number of occupied grid areas is increased, and moving subjects whose moving subjects are suddenly too large can be eliminated, that is, shooting due to unstable shooting can be avoided as much as possible The less accurate images participate in the trajectory synthesis, thereby improving the synthesis effect and authenticity of the motion trajectory.

The fifth embodiment of the present invention relates to a motion track synthesis method. The fifth embodiment is almost the same as the second embodiment, the main difference is that: in the fifth embodiment of the present invention, images with excessively large moving subjects can be eliminated, and when continuous multi-frame images have the phenomenon of excessively large moving subjects , Indicating that the shooting distance is likely to be too close, and a prompt message is generated.

9 is a flowchart of a motion trajectory synthesis method according to a fifth embodiment of the present invention; wherein, steps 602 to 604 are substantially the same as steps 202 to 204 in FIG. 4 and will not be repeated here. Step 601 contains sub Step 6011 to sub-step 6019, where sub-step 6011 to sub-step 6012, sub-step 6014, sub-step 6018, sub-step 6019 are substantially the same as sub-step 2011 to sub-step 2015 in FIG. 4, and will not be repeated here, The difference is that it also includes:

Substep 6013, determine whether the moving subject in the currently collected image meets the preset size condition; wherein, the size condition includes that the width of the moving subject is less than half the width of the currently collected image, and the height of the moving subject is less than the currently collected image the height of. When the judgment result of substep 6013 is yes, enter substep 6014; when the judgment result of substep 6013 is no, enter substep 6015, and then enter substep 6016.

Substep 6015, discard the currently acquired image.

Sub-step 6016, accumulating the number of frames of consecutively discarded images, and determining whether the preset number of frames L is reached; where L is the preset number of frames and L is greater than 1. If the judgment result is yes, then enter sub-step 6017, and then return to sub-step 601, if the judgment result is no, then directly return to sub-step 6011.

Substep 6017, generating a prompt message indicating that the moving subject is too large.

If the judgment result of sub-step 6013 is negative, it means that the moving subject in the image is too large, that is, the moving subject is too close to the camera. Keep track, please stay away. "

In substep 6016, the preset frame number L may be set according to actual needs, and the preset frame number L and the preset frame number K in substep 6019 may be the same or different.

Among them, sub-step 6015 and sub-step 6016 are similar to sub-step 6018 and sub-step 6019, respectively, but the difference is that the sub-step 6015 discards an image with an excessively large moving subject, while the sub-step 6018 discards that there is no moving subject. The sub-step 6016 accumulates the number of continuously captured images of moving subjects that are too large, while the sub-step 6019 accumulates the number of frames of consecutively captured images that do not recognize a moving subject.

Therefore, in this embodiment, images with excessively large moving subjects can be eliminated, and the photographer can be promptly reminded that the shooting distance is too close, which is beneficial for collecting images containing moving subjects of a relatively large size for the synthesis of a moving trajectory, which can improve the movement The synthetic effect of the track.

The steps of the above methods are divided only for clarity, and can be combined into one step or split into some steps and decomposed into multiple steps when implemented, as long as they include the same logical relationship, they are all within the scope of protection of this patent ; Add irrelevant modifications to the algorithm or process or introduce irrelevant design, but do not change the core design of the algorithm and process are within the scope of protection of the patent.

The sixth embodiment of the present invention relates to an electronic device, as shown in FIG. 10, including:

At least one processor 10; and,

A memory 11 communicatively connected to the at least one processor;

A camera 12 connected to the at least one processor and the memory;

The camera 12 is used to collect images; the memory 11 stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor 10 to enable the at least one processor 10 Able to perform the above method embodiments.

Among them, the memory 11 and the processor 10 are connected by a bus. The bus may include any number of interconnected buses and bridges. The bus connects one or more processors 10 and various circuits of the memory 11 together. The bus can also connect various other circuits such as peripheral devices, voltage regulators, and power management circuits, etc., which are well known in the art, and therefore, they will not be further described herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices on the transmission medium. The data processed by the processor 10 is transmitted on the wireless medium through the antenna. Further, the antenna also receives the data and transmits the data to the processor 10.

The processor 10 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. The memory 11 may be used to store data used by the processor 10 when performing operations.

The seventh embodiment of the present invention relates to a computer-readable storage medium that stores a computer program. When the computer program is executed by the processor, the above method embodiments are implemented.

Compared with the prior art, the embodiment of the present invention divides the image into a plurality of grid areas according to a preset method; and selects a plurality of frames of images satisfying preset conditions from multiple frames of images for synthesis, where the preset conditions include, The image contains moving subjects and the grid areas occupied by the moving subjects in any two frames of images do not overlap. Therefore, the grid method is used to select the image for synthesis, which is not affected by the movement speed of the moving subject, the direction of movement, and the change in the size of the moving subject. it is good.

That is, those skilled in the art can understand that all or part of the steps in the method of the foregoing embodiments can be completed by instructing relevant hardware through a program, which is stored in a storage medium and includes several instructions to make a device ( It may be a single chip microcomputer, a chip, etc.) or a processor to execute all or part of the steps of the methods described in the embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code .

Those of ordinary skill in the art can understand that the above-described embodiments are specific examples for implementing the present invention, and in actual applications, various changes can be made in form and details without departing from the spirit and range.

Claims

A motion trajectory synthesis method, which is characterized by including:

Acquiring multiple frames of images; wherein the multiple frames of images are arranged according to the acquisition time;

Divide the multi-frame image into multiple grid areas according to a preset manner;

Selecting from the multi-frame images several frames of images that meet preset conditions; wherein, the preset conditions include that the several frames of images all contain moving subjects and the grid area occupied by the moving subjects in any two frames of images Do not overlap

The motion trajectory of the moving subject is synthesized according to the selected frames of images.
The method for synthesizing a motion track according to claim 1, wherein the acquiring multiple frames of images includes:

Acquire images according to preset frequency;

Perform motion recognition on the currently collected image, and if a moving subject is identified from the currently collected image, add the currently collected image to the first image sequence set; wherein, the first image sequence set contains the acquired The multi-frame image;

If the moving subject is not recognized from the continuously collected K frame images, the collection is stopped; where K is the preset number of frames and K is greater than 1.
The method for synthesizing a motion track according to claim 1, wherein the dividing the multi-frame image into a plurality of grid regions according to a preset manner includes:

Selecting an image containing a moving subject from the multi-frame image, and determining a target frame area of the moving subject in the selected image; wherein, the moving subject in the selected image is located in the target frame In the area

The selected image is divided into a plurality of grid areas according to the size of the target frame area; wherein the target frame area is located in one of the grid areas.
The method for synthesizing a motion track according to claim 3, wherein the selecting of a plurality of frames of images satisfying preset conditions from the multi-frame images includes:

Adding the selected image to the second image sequence set, and marking the grid area where the selected image is located as occupied;

For each frame of the multi-frame image except the selected image, the screening is performed according to the order of sorting. The screening method includes, when it is determined that the grid area where the moving subject in the image is located is not marked as When occupied, add the image to the second image sequence set, and mark the grid area where the moving subject in the image is occupied as occupied;

Wherein, the second image sequence set includes the several frames of images satisfying preset conditions.
The method for synthesizing a motion track according to claim 4, wherein the grid area where the moving subject in the image is located is not marked as occupied, including:

Determine the target frame area of the moving subject in the image;

It is determined that the grid area where the target frame area of the moving subject in the image is not marked as occupied.
The method for synthesizing a motion track according to claim 4, wherein before determining that the grid area where the target frame area of the moving subject in the image is located is not marked as occupied, further comprising:

It is determined that the target frame areas of the moving subject in the image are all located in one of the grid areas.
The motion trajectory synthesis method according to claim 2, wherein after identifying the moving subject from the currently collected image and before adding the currently collected image to the first image sequence set, include:

It is determined that the moving subject in the currently collected image satisfies a preset size condition, where the size condition includes that the width of the moving subject is less than half the width of the currently collected image, and the The height is less than the height of the currently collected image;

If none of the moving subjects in the continuously collected L-frame images meets the size condition, a prompt message indicating that the moving subjects are too large is generated; where L is a preset number of frames and L is greater than 1.
The trajectory synthesis method according to any one of claims 3 to 7, wherein the target frame area is an area where the smallest circumscribed rectangular frame of the moving subject is located; or, the target frame area and all The area of the smallest circumscribed rectangle is similar in shape, and the area ratio is greater than 1 and less than or equal to 5/4.
An electronic device, characterized in that it includes:

At least one processor; and,

A memory in communication connection with the at least one processor;

A camera connected to the at least one processor and the memory;

The camera is used to collect images; the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor, so that the at least one processor can execute The method for synthesizing a motion track according to any one of claims 1 to 8.
A computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the motion trajectory synthesis method according to any one of claims 1 to 8 is implemented.