A method for mapping three-dimensional motion trajectory in video

Abstract

The present invention discloses a method for mapping a three-dimensional motion trajectory in a video, comprising the following steps: decomposing the video frame by frame and sorting it according to a time axis; selecting a starting frame, capturing and selecting a target therein, sending the obtained visual recognition information to a processing terminal, and obtaining an RGB grayscale value model of the target; performing a color gamut distribution analysis on the RGB grayscale value model, calculating the relative area ratios of different regions in the target, and archiving the model for subsequent invocation; capturing each frame one by one through a visual recognition camera, comparing the frame with the archive to locate the target, and then drawing a two-dimensional motion trajectory of the target; comparing the identified target with the target identified in the starting frame, calculating the ratio of the overall area change, thereby calculating the target's travel route on the Z axis, and merging the target with the two-dimensional image to draw the target's three-dimensional motion trajectory. The present invention has the advantages that the sampling and calculation processes are greatly simplified, and the required peripherals are simpler.

Description

Technical Field

[0001] The present invention relates to the field of visual mapping, and in particular to a method for mapping three-dimensional motion trajectories in a video. Background Art

[0002] With the explosive growth of video data, it has become an urgent problem to analyze and browse videos quickly and accurately, which is particularly prominent in video surveillance applications. Key frame extraction as a feasible solution has attracted more and more attention. The evaluation of key frames mainly depends on whether they can fully and accurately reproduce the main events in the original video, and minimize redundancy while ensuring comprehensive extraction. The key frame extraction algorithms widely used at present are usually based on the underlying feature analysis of the video, and the key frame extraction is based on the changes in the content features (color, shape, motion, etc.) of a single frame or a small number of frames. However, due to the lack of complete feature analysis in the time dimension, it is difficult to grasp the number of key frames to be extracted and determine the position of the key frames as a whole. It is easy to be disturbed by environmental changes, target posture changes, target occlusion, etc., resulting in the omission of moving targets, which in turn leads to the failure to extract the real key frames. There is a difference between the extraction results and the real semantics of the video, and it cannot fully and accurately reflect the real semantics of the video. In other words, the key frame extraction results do not conform to the visual perception of the human eye.

[0003] On the basis of not losing the semantic information of the video, it is of great research significance and practical demand to extract key frames quickly and accurately. The spatiotemporal motion trajectory of the video as a key frame extraction criterion provides an effective solution for this. The spatiotemporal motion trajectory of the target can accurately reflect the change of the target's motion state, and has a wide range of applications both in civilian and military fields.

[0004] However, a simple method to track the motion trajectory of a target in a video is to two-dimensionalize the trajectory of the target and depict it on a plane. If it is necessary to map the three-dimensional motion trajectory of the target through a two-dimensional video, the optical flow method is used in the prior art. Although the optical flow method is feasible for extracting the spatiotemporal motion trajectory of the video, the amount of calculation is large and it is difficult to meet the real-time requirements of the system. Therefore, there is an urgent need for an efficient method for mapping three-dimensional motion trajectories in videos. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method for mapping three-dimensional motion trajectories in videos in view of the deficiencies in the prior art.

[0006] Technical solution: The method for mapping three-dimensional motion trajectories in a video described in the present invention comprises the following steps:

[0007] S1, decompose the video frame by frame and sort it according to the timeline;

[0008] S2, selecting the starting frame of the video, capturing and selecting the target with the visual recognition camera, sending the obtained visual recognition information to the processing terminal, and obtaining the RGB grayscale value model of the target;

[0009] S3, performing color gamut distribution analysis on the RGB gray value model obtained in S2, calculating the relative area ratios of different regions in the target, and archiving the RGB gray value model of the target and the relative area ratios of different regions for subsequent use;

[0010] S4, capturing each frame decomposed from the video one by one through a visual recognition camera, comparing it with the archived RGB grayscale value model of the target and the relative area ratio of different regions to locate the target, and then draw a two-dimensional motion trajectory of the target;

[0011] S5. Compare the target identified in S4 with the target identified in the starting frame, calculate the ratio of the overall area change, and thus calculate the target's travel route on the Z axis. Combined with the two-dimensional target obtained in S4, the three-dimensional motion trajectory of the target can be drawn.

[0012] Preferably, in S2, in addition to the starting frame, several frames are randomly selected from the frames decomposed from the video for visual recognition camera capture and the target is selected, the extracted frames are visually captured and analyzed to calculate the RGB grayscale value model of the target at different angles in each frame and the relative area ratio of different regions, and the measured data are composited to draw a three-dimensional model of the target.

[0013] Preferably, the two-dimensional motion trajectory of the target obtained in S4 is marked with frame marks and time axis marks.

[0014] Preferably, the target's travel route on the Z axis acquired in S5 is marked with frame marks and time axis marks.

[0015] Preferably, in S5, composite drawing of the travel route on the Z axis and the two-dimensional motion trajectory is achieved through composite positioning of the frame mark and the time axis mark.

[0016] Compared with the prior art, the present invention has the following beneficial effects: a visual recognition camera is used to capture and select a target in advance, and the obtained visual recognition information is sent to a processing terminal to obtain an RGB grayscale value model of the target and the relative area ratios of different regions in the target. The surveying and mapping process in the complete video is converted from point-by-point calculation to capture measurement and comparison, the sampling and calculation process is greatly simplified, and the required peripherals are simpler, so real-time surveying and mapping can be achieved, and the equipment cost is greatly reduced. DETAILED DESCRIPTION

[0017] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] In the present invention, unless otherwise clearly specified and limited, the terms "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integrated connection; it can be a mechanical connection, an electrical connection, or a communication connection; it can be a direct connection, or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0019] The technical solution of the present invention is described in detail with specific embodiments below. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described in detail in some embodiments.

[0020] Embodiment 1: A method for mapping a three-dimensional motion trajectory in a video, comprising the following steps:

[0021] S1, decompose the video frame by frame and sort it according to the timeline;

[0022] S2, selecting the starting frame of the video, capturing and selecting the target with the visual recognition camera, sending the obtained visual recognition information to the processing terminal, and obtaining the RGB grayscale value model of the target;

[0023] S3, performing color gamut distribution analysis on the RGB gray value model obtained in S2, calculating the relative area ratios of different regions in the target, and archiving the RGB gray value model of the target and the relative area ratios of different regions for subsequent use;

[0024] S4, capturing each frame decomposed from the video one by one through a visual recognition camera, comparing it with the archived RGB grayscale value model of the target and the relative area ratio of different regions to locate the target, and then draw a two-dimensional motion trajectory of the target;

[0025] S5. Compare the target identified in S4 with the target identified in the starting frame, calculate the ratio of the overall area change, and thus calculate the target's travel route on the Z axis. Combined with the two-dimensional target obtained in S4, the three-dimensional motion trajectory of the target can be drawn.

[0026] Embodiment 2: A method for mapping a three-dimensional motion trajectory in a video, comprising the following steps:

[0027] S1, decompose the video frame by frame and sort it according to the timeline;

[0028] S2, select the starting frame of the video, capture it with a visual recognition camera and select the target, send the obtained visual recognition information to the processing terminal, obtain the RGB gray value model of the target, in addition to the starting frame, randomly select several frames from the frames decomposed from the video to capture them with a visual recognition camera and select the target, visually capture and analyze the extracted frames to calculate the RGB gray value model of the target at different angles in each frame;

[0029] S3, performing color gamut distribution analysis on the RGB gray value model obtained in S2, calculating the relative area ratios of different regions in the target, archiving the RGB gray value model of the target and the relative area ratios of different regions for subsequent calls, visually capturing and analyzing the extracted frames to measure the RGB gray value model of the target at different angles in each frame and the relative area ratios of different regions, and drawing a three-dimensional model of the target after compounding the measured data;

[0030] S4, capturing each frame decomposed from the video one by one through a visual recognition camera, comparing it with the archived RGB grayscale value model of the target and the relative area ratio of different regions to locate the target, and then drawing a two-dimensional motion trajectory of the target, the obtained two-dimensional motion trajectory of the target is marked with a frame mark and a time axis mark;

[0031] S5. Compare the target identified in S4 with the target identified in the starting frame, calculate the ratio of the overall area change, and thus calculate the target's route on the Z axis. The obtained target's route on the Z axis is marked with frame marks and time axis marks. The composite positioning of the frame mark and the time axis mark is used to realize the composite drawing of the route on the Z axis and the two-dimensional motion trajectory, so as to draw the three-dimensional motion trajectory of the target.

[0032] The advantage of this technical solution is that the visual recognition camera captures and selects the target in advance, and sends the obtained visual recognition information to the processing terminal to obtain the RGB grayscale value model of the target and the relative area ratio of different regions in the target. The surveying and mapping process in the complete video is transformed from point-by-point calculation to snapshot measurement and comparison, the sampling and calculation process is greatly simplified, and the required peripherals are simpler, which can realize real-time surveying and mapping and greatly reduce the equipment cost.

[0033] In the present invention, unless otherwise clearly specified and limited, the first feature being "above" or "below" the second feature may mean that the first feature is in direct contact with the second feature, or the first feature and the second feature are in indirect contact through an intermediate medium. Moreover, the first feature being "above", "above" and "above" the second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. The first feature being "below", "below" and "below" the second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature. In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example" or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example.

[0034] Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradiction.

[0035] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for mapping three-dimensional motion trajectories in a video, characterized in that: The following steps are involved: S1, decompose the video frame by frame and sort it according to the timeline; S2, selecting the starting frame of the video, capturing and selecting the target with the visual recognition camera, sending the obtained visual recognition information to the processing terminal, and obtaining the RGB grayscale value model of the target; S3, performing color gamut distribution analysis on the RGB gray value model obtained in S2, calculating the relative area ratios of different regions in the target, and archiving the RGB gray value model of the target and the relative area ratios of different regions for subsequent use; S4, capturing each frame decomposed from the video one by one through a visual recognition camera, comparing it with the archived RGB grayscale value model of the target and the relative area ratio of different regions to locate the target, and then draw a two-dimensional motion trajectory of the target; S5. Compare the target identified in S4 with the target identified in the starting frame, calculate the ratio of the overall area change, and thus calculate the target's travel route on the Z axis. Combined with the two-dimensional target obtained in S4, the three-dimensional motion trajectory of the target can be drawn.

2. The method for mapping three-dimensional motion trajectories in a video according to claim 1, characterized in that: In S2, in addition to the starting frame, several frames need to be randomly selected from the frames decomposed from the video for visual recognition camera capture and target selection. The extracted frames are visually captured and analyzed to calculate the RGB grayscale value model of the target at different angles in each frame and the relative area ratio of different regions. The measured data are composited to draw a three-dimensional model of the target.

3. The method for mapping three-dimensional motion trajectories in a video according to claim 1, characterized in that: The two-dimensional motion trajectory of the target obtained in S4 is marked with frame marks and time axis marks.

4. The method for mapping three-dimensional motion trajectories in a video according to claim 1, characterized in that: The target's travel path on the Z axis obtained in S5 is marked with frame marks and time axis marks.

5. A method for mapping three-dimensional motion trajectories in a video according to claim 3 or 4, characterized in that: In S5, the composite drawing of the travel route on the Z axis and the two-dimensional motion trajectory is achieved through the composite positioning of the frame mark and the time axis mark.