TWI835011B - A method and apparatus for simulating the acting track of an object - Google Patents

Abstract

A method for simulating the acting track of an object include capturing a plurality frame of images from a video; determining a plurality of outline coordinates of a first tracking object according to a first feature of the first tracking object in each image; identifying a first frame according to the outline coordinates of the first tracking object; generating a vector according to the relative relationship of the reference coordinate and the center coordinate of the first frame, and stimulating the acting track of the target object according to the vector, wherein the target object related to the first tracking object , or directly; and quantifying the acting track, and drawing to the corresponding coordinates of a data visualization draw.

Description

Method and device for simulating the trajectory of objects

The present invention relates to image tracking technology; specifically, the present invention relates to a method and device for simulating the trajectory of a specific object.

With the development of artificial intelligence, image recognition and object detection technologies have also been greatly improved (such as the object detection algorithm YOLO (You Only Look Once)), and are used in various applications such as autonomous driving, smart medical care, and face recognition. in the field. Through machine learning technology, AI can continuously learn to identify the characteristics of specific objects through training sets, thereby quickly and accurately capturing target objects in different images, and continuously tracking the target objects as they move.

At present, object detection technology can already achieve object tracking technology such as identifying vehicles on the road, classifying vehicle types, and continuously tracking the vehicle during its movement. However, when the shape of the target to be tracked is not fixed (such as air flow, water column, etc.), because the AI cannot identify the shape of the target, learn its characteristics through the training set, or even sense its existence from the image, existing objects Tracking technology is difficult to apply to track such targets.

One object of the present invention is to provide a method and device for simulating the trajectory of an object, which can track a target object and then simulate the trajectory or range of its action (such as water jets or airflow, etc.).

One object of the present invention is to provide a method or device for simulating the trajectory of an object, which can easily estimate the range of the image that has been affected by the object and present it to the user.

The method of simulating the trajectory of the object includes capturing a plurality of frames of images from the video; determining the plurality of outline coordinates of the first tracking object based on the first characteristics of the first tracking object in each image; and identifying the first tracking object based on the outline coordinates. Frame; generate a vector based on the relative relationship between the reference coordinates and the coordinates of the center point of the first frame, and simulate the action trajectory of the target object based on the vector, where the target object is associated with the first tracking object; and digitize the action trajectory , and plotted to the corresponding coordinates of a data visualization diagram. Through this method, it is possible to simulate its trajectory (such as the direction and range of the air shot) by tracking tangible objects (such as gloves, spray guns, etc.), and then estimate the impact of invisible substances (such as air, water, etc.) Distribution range. And the user can also easily understand the scope of the fixed area that has been affected (for example, cleaned) and the scope that has yet to be affected based on the data visualization diagram.

The device for simulating the trajectory of an object includes at least one memory; and at least one processor coupled to the memory, the processor being configured to: determine positioning according to the first characteristic of the first tracking object in each of the images. Find the plurality of outline coordinates of the first tracking object; locate the first frame shape according to the outline coordinates of the first tracking object, wherein the first frame shape surrounds the outline coordinates of the first tracking object; according to the The relative relationship between the reference coordinates and the coordinates of the center point toward the first frame generates a vector, and the trajectory of the target object is simulated according to the vector, wherein the target object can be the same as the first tracking object or directly or indirectly connected to the Associated with the first tracking object; and digitizing the action trajectory and drawing it to the corresponding coordinates of a data visualization diagram. With this device, you can track Trace tangible objects (such as gloves, spray guns, etc.) to simulate their trajectory (such as the direction and range of the ejected air), and then estimate the distribution range of intangible substances (such as air, water, etc.). And the user can also easily understand the scope of the fixed area that has been affected (for example, cleaned) and the scope that has yet to be affected based on the data visualization diagram.

100: Step chart

101: Steps

103: Steps

105: Steps

107: Steps

108: Steps

109: Steps

201:Image

202:Region of Interest (ROI)

203:First tracking object

204: Partial image

205: First frame shape

207:Reference coordinates

303:Reflection

305: Error identification box

307: Correct identification frame

403:Coordinates of the center point of the first frame

405:Vector

407a:Coordinates

407b:Coordinates

407c:Coordinates

500:Data visualization chart

503: Reference indicator

503a: Instant Coverage

503b:Trajectory indicator

700:Device

710:Memory

720: Processor

721:Image capture unit

722:Image recognition module

723:Computing unit

725: Data conversion unit

730:Display unit

740:Image source

T: action trajectory

R: range

A: Expansion angle

V _N :variable

FIG. 1 is a step sequence diagram of a method for simulating the trajectory of an object according to an embodiment of the present invention.

FIG. 2A is a schematic diagram of capturing an image according to an embodiment of the present invention.

FIG. 2B is a schematic diagram of a captured image including a region of interest (ROI) according to an embodiment of the present invention.

Figure 3 is a schematic diagram of an image and its recognition results according to another embodiment of the present invention.

FIG. 4 is a schematic diagram of the trajectory of a simulated object according to an embodiment of the present invention.

FIG. 5A is a schematic diagram of action trajectory values according to an embodiment of the present invention.

FIG. 5B is a schematic diagram of action trajectory values according to another embodiment of the present invention.

FIG. 5C is a schematic diagram of a data visualization diagram according to an embodiment of the present invention.

FIG. 6 is a schematic diagram of the trajectory of a simulated object when multiple target objects are included according to another embodiment of the present invention.

Figure 7 is a schematic diagram of a device for simulating the trajectory of an object according to yet another embodiment of the present invention.

As used herein below, "about," "approximately," or "substantially" includes the stated value and an average within an acceptable range of deviations from the particular value as determined by one of ordinary skill in the art, taking into account the measurements in question and The specific amount of error associated with a measurement (i.e., the limitations of the measurement system). For example, "about" can mean within one or more standard deviations of the stated value, or within ±30%, ±20%, ±10%, ±5%. Furthermore, the terms "about", "approximately" or "substantially" used in this article can be used to select a more acceptable deviation range or standard deviation based on optical properties, etching properties or other properties, and one standard deviation may not apply to all properties. .

It will be understood that, although the terms "first," "second," "third," etc. may be used herein to describe various elements, components, regions, layers and/or sections, these elements, components, regions, and/or parts shall not be limited by these terms. These terms are only used to distinguish one element, component, region, layer or section from another element, component, region, layer or section. Thus, a "first element", "component", "region", "layer" or "section" discussed below could be termed a second element, component, region, layer or section without departing from the teachings herein.

Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be construed to have meanings consistent with their meanings in the context of the relevant technology and the present invention, and are not to be construed as idealistic or excessive Formal meaning, unless expressly defined as such herein.

The present invention relates to how to simulate the trajectory of a target object (such as a spray gun) to evaluate the scope of action of unfixed or invisible substances (such as air, water). For example, equipment such as water jets or air spray guns may be used to clean factory equipment. Although current image recognition technology can identify tangible objects (such as spray guns), it is difficult to identify the objects ejected by the spray guns. The distribution range of water column or air flow. Therefore, the present invention aims to provide a trajectory of air flow or water column ejected from a spray gun, thereby assessing the cleaning status of the equipment and quickly distinguishing between cleaned and uncleaned equipment areas. The present invention will be described in detail below through various embodiments. It should be understood that each embodiment is only illustrative to enable those of ordinary skill in the art to understand the content of the present invention, without departing from the spirit of the present invention. , the following exemplary embodiments may be modified, and this disclosure is intended to cover such modifications.

A method of simulating an object's trajectory according to an embodiment of the present invention is described with reference to FIG. 1 and FIG. 2A. At step 101, the method of simulating an object's trajectory includes capturing a plurality of frames of images 201 from a video. The video records the scene of the target object (such as a spray gun) and the possible range of the target object (such as the equipment to be cleaned). In a preferred embodiment, the plurality of frame images 201 may be continuously captured images 201. In other embodiments, the images 201 may also be captured at intervals. For example, images 201 are captured from the video every 0.5 seconds within a period of time. The time for capturing the images 201 can also have a longer (for example, every second) or shorter (for example, every 0.3 seconds) interval according to the needs. The difference in length may affect the accuracy of the trajectory of the simulated object. For example, when the interval time is longer, the timing difference between the plurality of frame images 201 is larger, and the simulated trajectory also has larger discontinuities, so it is less accurate.

Continuing to describe this embodiment with reference to FIG. 2B , capturing multiple frame images 201 from the video may further include capturing only images of a region of interest (ROI) 202 (area shown with diagonal lines) defined by the user. . For example, when the user wants to know the current cleaning status of a certain equipment, he or she can capture only the image 201 covering the part of the equipment in the video. The method of capturing the ROI 202 may further include performing morphological processing on the captured ROI 202, such as performing expansion processing on the captured ROI 202 to increase the area of the ROI 202. Through this processing, the possibility of being missed when the object to be identified is too close to the edge of ROI 202 or slightly beyond the edge can be avoided. In addition, it is also possible The image 201 is processed using logical operations to obtain the ROI 202 of a specific shape. For example, a portion of the image 204 can be subtracted from the image 201 to obtain a hollow frame-shaped ROI 202).

It should be noted that the image 201 can be a file in any format (such as a JPEG file, etc.), and image processing (such as dimensionality reduction) can be performed on the image 201 to improve the efficiency of image processing and reduce required performance. Or the image 201 is converted from the RGB model to the HSV model to facilitate subsequent image recognition using color, and the user can adjust the color more intuitively when setting color-related features. The foregoing format description and image processing of the image 201 are all exemplary descriptions. The image 201 can be in any other format or process as long as it does not conflict with the content of the present invention.

The method of simulating the trajectory of an object according to one embodiment of the present invention will be continued to be described with reference to FIG. 2A . At step 103, the method includes determining the outline coordinates of the first tracking object 203 in the image 201 based on the first feature. The first tracking object 203 may be a physical object such as a glove, and may have a specific first characteristic (for example, red). According to the first feature, the first tracking object 203 can be identified from the image 201 (for example, through image color filtering technology), and then the plural outline coordinates of the location of the first tracking object 203 can be obtained. In different embodiments, the first feature can also be set by the user (for example, the user can set the first feature to different colors such as blue, green, etc.). At step 105, the method can identify a first frame 205 surrounding the outline coordinates according to the outline coordinates (that is, the range surrounded by the first frame 205 can cover all outline coordinates). It should be noted that in this embodiment, the first frame 205 has a minimum area that can surround the outline coordinates. However, in different embodiments, the first frame 205 can also have different area ranges (for example, it can have The present invention does not set limits around twice the area of these contour coordinate ranges, etc.).

In another embodiment of the present invention, a method for simulating the trajectory of an object may include performing morphological image processing (such as corrosion, expansion, etc.) on the identified outline coordinates of the first tracking object 203 to eliminate noise and restore the identified object. Broken outlines, etc., make the outline coordinates closer to the true shape of the first tracking object 203. Thereafter, the area of the outline coordinates can be calculated and compared with the estimated area of the first tracking object 203 (that is, the area of the first tracking object 203 presented in the image 201), thereby eliminating the error range (for example, ±5%). ) outside the contour coordinates. Similarly, the area of the first frame 205 can be calculated and compared with the estimated area of the first frame 205 (ie, the area of the frame surrounding the first tracking object 203 presented in the image 201) to find out the error. The first frame shape 205 outside the range (for example, ±5%) is excluded, and the contour coordinates corresponding to the first frame shape 205 are excluded. Referring to FIG. 3 specifically, the left side of FIG. 3 is an example diagram of the image 201, and the right side is a schematic diagram of the recognition result of the image 201. When the scene in the image 201 has smooth objects such as metal, the first tracking object 203 may produce a reflection 303 on the smooth object, and the image 201 may separate the first tracking object 203 and the reflection 303 after recognition. At the same time, it is recognized as the first tracking object 203 (the wrong recognition box 305 in the right picture shows the wrong recognition result of mistakenly recognizing the reflection 303 as the first tracking object 203, and the correct recognition box 307 shows the correct recognition of the first tracking object 203 the correct identification result), which in turn affects the identification result. Since the reflection of an object (as shown in the error recognition frame 305) usually does not have the same complete shape as the original object (as shown in the correct recognition frame 307), by excluding the outline coordinates whose area is outside the error range, this embodiment can avoid The opportunity to misjudge the reflection 303 of the first tracking object 203 as the first tracking object 203 .

The method of simulating the trajectory of an object according to one embodiment of the present invention will be continued to be described with reference to FIGS. 2 and 4 . At step 107 , the method includes generating a vector 405 based on the relative relationship between the reference coordinate 207 and the center point coordinate 403 of the first frame 205 . In this embodiment, the reference coordinate 207 is a preset fixed coordinate (for example, it can be the center point coordinate of the image), and the vector 405 is directed from the reference coordinate 207 toward It is generated by connecting a line to the center point coordinate 403 of the first frame 205. The target object may be the first tracking object 203 (for example, the same spray gun), or when the target object is difficult to identify and track (for example, the features are less obvious or obscured), the target object may be the same as the first tracking object 203 Directly or indirectly connected objects (for example, the first tracking object 203 is a glove, and the target object is a spray gun held in the glove). With this configuration, even when the target object is difficult to identify and track (for example, when the cleaning staff holds the spray gun in their hands, the spray gun itself is small and blocked by the cleaning staff's hands), it can still be directly or indirectly connected to it through identification and tracking. The first tracking object 203 achieves the purpose of tracking the target object.

It should be noted that although the reference coordinates 207 are fixed coordinates set in advance in this embodiment, in different embodiments, the second tracking object can also be used to determine the center point coordinates 207 . The second tracking object includes a second feature (for example, blue), and the second feature is different from the first feature to avoid confusion during the identification process. The second characteristic of the second tracking object can also be set by the user. According to the second feature, the second tracking object can be identified from the image 201, a plurality of outline coordinates of the second tracking object can be obtained, and a second frame surrounding the outline coordinates can be generated. The center point of the second frame shape can be used as the reference coordinate 207, and a vector 405 can be generated by connecting the reference coordinate 207 toward the center coordinate 403 of the first frame shape 205.

The method of simulating the trajectory of an object in this embodiment will be described below with reference to FIG. 4 . Step 108 includes simulating the action trajectory T of the target object according to the vector 405 . In this embodiment, the simulated target object's action trajectory T not only includes vector 405 but also includes the action trajectory parameters of the target object, and the action trajectory parameters can be set by the user. The vector 405 is associated with the basic action direction of the target object, and the action trajectory parameter is associated with the range of the action trajectory T. For example, assuming that the action trajectory T is a fan-shaped range, the action trajectory parameters may include the range R of the target object (for example, the straight-line distance that the air or water sprayed by the spray gun can reach, that is, the radius of the fan), expansion Angle A (i.e. fan The central angle of the circle), etc. The vector 405 can initially simulate the action direction of the target object. Combining the action trajectory parameters (range R, dispersion angle A, etc.) can make the simulation of the target object's action trajectory more accurate. It should be noted that although the action trajectory T is a sector-shaped range in this embodiment, the action trajectory T can also be simulated as other shapes (such as triangles, etc.) in different embodiments, and the present invention does not limit this. .

The method of simulating the trajectory of an object according to one embodiment of the present invention will be described below based on FIG. 4 , FIG. 5A , FIG. 5B and FIG. 5C . Step 109 includes digitizing the action trajectory T and drawing it to the corresponding coordinates of the data visualization diagram 500 . Referring to FIG. 5A , the action trajectory T of the target object can be simulated according to the foregoing content, wherein digitizing the action trajectory T may include, for example, giving the same numerical value to each coordinate of the action trajectory T distribution in the same frame image 201 (i.e., the sector range). Each point of the distribution has the same value), for example, the coordinates 407a, 407b and 407c can be given the same value (for example, 10). After multiple frames of images 201 undergo the same processing, since the range of the distribution of the active trajectory T in each frame of image 201 may be different, the variable V _n representing the accumulated value of each point coordinate will accumulate with the timing of each frame of image 201 . The coordinate value of this point in this frame. Therefore, corresponding to the distribution of the action trajectories T in the plural frame image 201, the variable V _n at each coordinate point can accumulate values of different sizes (for example, in Figure 5A, the overlapping part of the two action trajectories T has a variable V _n with a value of 20), Then, the degree to which each coordinate point is affected by the target object can be displayed through the data visualization diagram 500 .

Referring to FIG. 5B , in different embodiments, different coordinate points of the distribution of the action trajectory T in the image 201 can also be given values of different sizes. For example, when a water column or airflow is ejected from the nozzle of a spray gun, it will be affected by air resistance and gravity, and the density of the distribution will decrease as the distance increases. Therefore, the coordinates closer to the target object can be given a larger value (for example, 40), and the coordinates further away from the target object can be given decreasing values (that is, coordinate 407a can be given a value greater than coordinate 407b. The coordinate 407b can be given a value greater than the coordinate 407c). After the same processing is performed on the plurality of images 201, each coordinate point will also accumulate its own value, thereby simulating that, for example, when the target object is a water jet spray gun, the extent to which the area is affected by the water jet will also decrease as the distance increases. It should be noted that the above-mentioned numerical settings are all exemplary, and other different numerical settings can be made without departing from the spirit of the present invention, and the present invention does not set limits.

The data visualization diagram 500 refers to a way of graphically presenting numerical values through colors, blocks, lines, etc. Referring to Figure 5C, in this embodiment, the value of each coordinate point can be presented in the form of a heat map according to the level of the numerical value. (That is, the coordinate points that are affected by the target object to a higher degree are closer to red on the heat map, and the coordinate points that are affected to a lower degree are closer to purple. It should be noted that although in this embodiment, the heat map is used However, in different embodiments, the numerical values may also be presented in other types of data visualization diagrams 500 (such as contour diagrams, etc.), and the present invention does not limit this.

Referring to FIG. 5C , the method of simulating the trajectory of an object according to an embodiment of the present invention will be continued to be described. In this embodiment, the data visualization diagram 500 may only present the region of ROI 202 . For example, when the equipment that the user wants to clean presents a frame-shaped area in the image, the data visualization diagram 500 can only present the color distribution of the frame-shaped area to facilitate the user to quickly understand the cleaning status of the equipment. In addition, the data visualization diagram 500 may further include reference indicators 503, which may include real-time coverage 503a and trajectory indicators 503b of the action trajectory T. Among them, the area covered by the action trajectory T on the data visualization map 500 (that is, acted upon by the target object) is calculated and divided by the total area of the image 201 or ROI 202 to obtain the real-time coverage of the action trajectory T. 503a; Calculate the total value of the action trajectory T digitized on the data visualization diagram 500, and divide it by the total area of the image 201 or ROI 202 to obtain the trajectory index 503b of the action trajectory T. Since the real-time coverage 503a and the trajectory indicator 503b will change with time series and are related to the distribution of the action trajectory T, so The user can use these reference indicators 503 to determine the real-time status of the target object. It should be noted that the foregoing description of the reference index 503 is only illustrative. In different embodiments, the reference index 503 may also include other reference indexes 503 that can provide reference information related to the action trajectory T, such as Cumulative action time, etc.

Referring to FIG. 6 , in different embodiments, the image 201 of the same frame may include a plurality of first tracking objects 203 with the same first characteristics (for example, red), wherein the first tracking objects 203 can be identified from the image 201 based on the first characteristics. these first tracking objects 203, and then respectively obtain the outline coordinates of plural groups of these first tracking objects 203 (the outline coordinates of the same group correspond to the same first tracking object 203), and generate a set of outline coordinates around the first tracking objects 203 A plurality of first frame shapes 205 (the outline coordinates of the same group correspond to one first frame shape 205). In this embodiment, complex vectors 405 can be generated from the reference coordinates 207 to the center point coordinates of the first frames 205, and complex action trajectories T can be generated by these vectors 405, thereby simulating multiple targets at the same time. The status of the object's function. Specifically, when monitoring the cleaning status of equipment, there may be situations where multiple cleaning tasks are performed at the same time (for example, multiple cleaning personnel hold spray guns to spray water jets at the same time). With this configuration, the conditions of multiple spray guns can be simulated at the same time. Act on the track T to understand the immediate cleaning status of the equipment.

Referring to FIG. 7 , a device 700 for simulating the trajectory of an object according to another embodiment of the present invention is described, in which the device includes a memory 710 , a processor 720 coupled to the memory, and a display unit 730 . The memory 710 stores a plurality of instructions. When executing these instructions, the processor 720 is configured to execute the method of simulating the trajectory of an object in any of the foregoing embodiments. Specifically, the processor 720 includes an image capture unit 721, an image recognition module 722, a calculation unit 723 and a data conversion unit 725. The capture unit 721 captures a plurality of frames of images 201 from the image source 740 (ie, the video containing the target object and its scope scene), and passes them to the image recognition module 722 . The image recognition module 722 determines the outline coordinates of the first tracking object 203 from the captured image 201 and identifies the first frame 205 . Thereafter, the image recognition module 722 passes the recognition result to the calculation unit 723, and the calculation unit 723 generates a vector 405 based on the relative relationship between the first frame 205 identified by the image recognition module 722 and the reference coordinates 207. By using the action trajectory parameters and the vector 405, the calculation unit 723 can simulate the action trajectory T of the target object in each frame of image 201, and digitize the action trajectory T. Among them, the variable V _N representing the accumulated value of each coordinate point can be stored in the register. Along with the timing of the plurality of frame images 201 , the variable V _N continues to accumulate the result of the digitized action trajectory T by the calculation unit 723 . And the data conversion unit 725 will image the value of the variable V _N representing each coordinate point (such as expressing the value with a color), so that each coordinate point has its own image characteristics (such as color), and transmit the signal to the display unit 730, to present it to the user for viewing.

100…step diagram 101…steps 103…steps 105…steps 107…steps 108…steps 109…steps

Claims (24)

A method for simulating the trajectory of an object, including: capturing an image frame from a video; determining the plurality of contour coordinates of a first tracking object in the image based on a first feature of the first tracking object; These outline coordinates identify a first frame shape, where the position and size of the first frame shape are associated with the outline coordinates; a vector is generated according to the relative relationship between a reference coordinate and the coordinates of the center point of the first frame shape, and according to The vector simulates an action trajectory of a target object, wherein the target object is related to the first tracking object; and a value is given to each coordinate of the action trajectory and plotted to the corresponding coordinates of a data visualization diagram. The method of claim 1, wherein the target object can be the same as the first tracking object or directly or indirectly connected to the first tracking object. The method described in claim 1, wherein the image capturing step includes setting an area to be monitored so that only the area to be monitored is captured when capturing the videos; the step of drawing to the data visualization map includes only Generate a visualization of the data within the area to be monitored. The method of claim 3, wherein capturing the image frame may further include expanding the area to be monitored to intercept an expanded area to be monitored. The method of claim 1, wherein determining the contour coordinates of the first tracking object in the image includes dilating and eroding the contour of the first tracking object. The method of claim 1, wherein the step of simulating the action trajectory includes simulating the action trajectory according to an action trajectory parameter and the vector. The method of claim 1, further comprising: calculating an outline area based on the outline coordinates; comparing the outline area with an actual area of the first tracking object to generate a comparison result; and eliminating an error range based on the comparison result. In addition, these outline coordinates. The method as described in claim 1, further comprising: calculating an area of a first frame of the first frame; comparing the area of the first frame with an estimated area of the first frame to generate a comparison result; and excluding the first frame shape outside the error range according to the comparison result, and further excluding the contour coordinates corresponding to the first frame shape outside the error range. The method of claim 1, wherein the step of drawing to the data visualization diagram further includes: associating the action trajectory with at least one reference indicator; and merging the reference indicator in the data visualization diagram. The method of claim 1, further comprising: determining the center point coordinates of a second tracking object, wherein the second tracking object has a second characteristic different from the first characteristic of the first tracking object; and setting The coordinates of the center point of the second tracking object are the reference coordinates. The method of claim 1, wherein the step of assigning a numerical value to each coordinate in the action trajectory includes setting the same or different values according to different position coordinates of the action trajectory. The method of claim 1, wherein the dimension of the image frame can be reduced or converted from the RGB color model to the HSV color model. A device for simulating the action trajectory of an object, including: at least one memory on which a plurality of instructions are stored; and at least one processor coupled to the memory. When executing the instructions, the processor is configured to: from a An image frame is captured in the video; in the image, a plurality of outline coordinates of the first tracking object are determined according to a first characteristic of the first tracking object; and a first frame shape is identified according to the outline coordinates, wherein the first The position and size of the frame are associated with the outline coordinates; a vector is generated based on the relative relationship between a reference coordinate and the coordinates of the center point of the first frame, and an action trajectory of a target object is simulated according to the vector, wherein the target The object is associated with the first tracking object; and each coordinate in the action trajectory is given a value and plotted to the corresponding coordinate in a data visualization diagram. The device of claim 13, wherein the target object can be the same as the first tracking object or directly or indirectly connected to the first tracking object. The device of claim 13, wherein the processor is configured to include setting an area to be monitored in the image capture, so that only the area to be monitored is captured when capturing the videos; in the image capture, The data visualization diagram includes only generating the data visualization diagram within the area to be monitored. The device of claim 15, wherein capturing the image frame may further include expanding the area to be monitored to intercept an expanded area to be monitored. The device of claim 13, wherein determining the contour coordinates of the first tracking object in the image includes expanding and eroding the contour of the first tracking object. The device of claim 13, wherein the processor is configured to simulate the action trajectory including simulating the action trajectory according to an action trajectory parameter and the vector. The device of claim 13, the processor is further configured to: calculate an outline area based on the outline coordinates; compare the outline area with an actual area of the first tracking object to generate a comparison result; and generate a comparison result based on the outline coordinates. The comparison results exclude those contour coordinates outside the error range. The device of claim 13, the processor is further configured to: calculate an area of a first frame of the first frame; compare the area of the first frame with an estimated area of the first frame to determine Generate a comparison result; and exclude the first frame shape outside the error range based on the comparison result, and further exclude the contour coordinates corresponding to the first frame shape outside the error range. The device of claim 13, wherein the processor is further configured to: associate the action trajectory with at least one reference index; and merge the reference index into the data visualization graph. The device of claim 13, wherein the processor is further configured to: Determine the center point coordinates of a second tracking object, wherein the second tracking object has a second characteristic different from the first characteristic of the first tracking object; and use the center point coordinates of the second tracking object as a reference coordinates. The device of claim 13, wherein when a value is given to each coordinate in the action trajectory, the processor can be configured to set the same or different values according to different position coordinates of the action trajectory. The device of claim 13, wherein the dimension of the image frame can be reduced or converted from the RGB color model to the HSV color model.