FR2892822A1 - DEVICE AND METHOD FOR MEASURING THE SPEED OF A MOBILE ELEMENT IN A TERMINAL ENVIRONMENT - Google Patents

Abstract

The invention relates to a method and a device for measuring the speed of at least one moving element in a bounded environment (3). The device comprises: - video capture means (1) whose capture field covers said environment (3), - calibration means (2) matching the granularity of the captured images and the actual distance in said environment ( 3), - means for locating a distinctive sign of said movable element.According to the invention, the device comprises means for identifying the position in said environment (3) of said distinctive sign on said video captured at a first moment and at a time. second time, said calibration means (2) being able to convert said evolution of said distinctive sign on said video between the first instant and the second instant in a distance traveled (d1) in said environment (3), said device comprises means calculating the speed of said element over said distance traveled between the first instant and the second instant.

Description

Device and method for velocity measurement of a movable element in a

bounded environment.

  The invention relates to a device and a method for measuring the speed of a moving element in a bounded environment.

  Speed measurement is information used in many areas and especially on roads or during sports competitions such as for swimming.

  Often the speed measurement is accompanied by the installation of sensors on the moving device whose speed is to be measured. However, in the applications mentioned above, it is difficult to equip sportsmen or vehicles with a sensor. For the sportsmen, the presence of a sensor can be troublesome according to the movements carried out or dangerous in case of falls, during the matches of football or rugby for example. The equipment of the vehicles does not have the same problems but is difficult to envisage in the case where the speed measurements of the vehicles are made by surprise during speed checks, the vehicles being then obviously not equipped with sensors specific to these vehicles. purposes.

  The speed measurement of an athlete can be used for informative purposes during the retransmission of the sporting event for the spectators or television viewers but also for pedagogical purposes to improve the performances.

  The invention proposes a speed measurement of a mobile element operating in a bounded environment that does not require measuring means on the mobile element. To this end, the invention relates to a device for measuring the speed of at least one mobile element in a bounded environment comprising: video capture means whose capture field covers the environment, calibration means implementing match the granularity of the captured images and the actual distance in the environment, - means of locating a distinctive sign of the movable element. According to the invention, the device comprises means for identifying the position in the environment of the distinctive sign on the captured video at a first instant and at a second instant, the calibration means being able to convert the evolution of the distinctive sign. on the video between the first instant and the second instant in a distance traveled in the environment, the device comprises means for calculating the speed of the element over the distance traveled between the first instant and the second instant.

  According to a preferred embodiment, the device comprises means for recovering the dimensions of the bounded environment and the distance between the environment and the video capture means.

  According to a preferred embodiment, the calibration means map the granularity of the captured images and the actual distance according to the intrinsic parameters of the capture means, and the position of the capture means with respect to the absolute reference of the environment.

  According to a preferred embodiment, the capture means comprise a fixed camera placed above the environment.

  According to a preferred embodiment, the device comprises means for determining the direction in which the mobile element is moving, means for identifying the position of the distinctive sign seeking the position on the video as a function of the direction.

  Advantageously, the distinctive element is the centroid of an area of the movable element. Preferably, the movable member is a swimmer and the environment is a pool.

  Advantageously, the device comprises means for identifying the different water lines of the pool.

  According to another embodiment, the mobile element is a car and the environment is a road.

  According to another aspect, the invention relates to a method for measuring the speed of at least one mobile element in a bounded environment comprising: a video capture step whose capture field covers said environment; in correspondence the granularity of the captured images and the real distance in said environment, - a step of locating a distinctive sign of said movable element. According to the invention the method comprises a step of identifying the position in the environment of the distinctive sign on the captured video at a first instant and at a second instant, said calibration step converting said evolution of said distinctive sign on said video between the first instant and the second instant in a distance traveled in said environment, said method comprising a step of calculating the speed of said element over said distance traveled between the first instant and second instant.

  The invention will be better understood and illustrated by means of examples of advantageous embodiments and implementations, in no way limiting, with reference to the appended figures in which:

  FIG. 1 represents an exemplary device according to a preferred embodiment of the invention, FIG. 2 represents an example of an environment according to the preferred embodiment of the invention, FIG. preferred embodiment of a method according to the invention; FIG. 4 represents another aspect of the invention.

  The represented modules are functional units, which may or may not correspond to physically distinguishable units. For example, these modules or some of them may be grouped into a single component, or be functionalities of the same software. On the other hand, some modules may be composed of separate physical entities.

  FIG. 1 represents a first embodiment in which a swimmer operates in a swimming pool 3 and one wishes to know his instantaneous speed. A camera 1 is at the top of the basin. The camera is centered on the pool so that its field of action covers the whole pool. Its optical axis is centered on the pool. This camera is connected to a processing device 2 via an ethernet-type communication network if it is a camera equipped with an IP interface. This communication may also be different in other embodiments. The pool-centric camera is able to capture the video of different swimmers moving in the pool. The camera is therefore placed at a sufficient height so as to cover the entire pool The treatment device 2 knows the characteristics of the pool and in particular its width and length. The processing device 2 calibrates the camera 1 to establish a correspondence between the distance measurements in the pelvis 3 and the distance measurements in the captured images. The calibration step consists of estimating the extrinsic and intrinsic parameters of the camera.

  The extrinsic parameters express the 3D position of the camera relative to the absolute reference of the scene. They are decomposed into a matrix of rotation and a matrix of translation: - R is a matrix of rotation between the absolute reference of the scene and the local coordinate system of the camera, - t represents the vector of translation between the origin of the absolute reference of the scene and the local landmark of the camera.

  The intrinsic parameters contain information about the internal data of the camera. Assuming a pure perspective projection model without nonlinear distortions, the intrinsic parameters are five: the ratio between the sizes of the pixels in x and y and the focal length, the angle between the axes of the reference image, and the coordinates of the projection of the optical center in the image.

  To estimate the intrinsic and extrinsic parameters of the camera, metric data from the scene can be used (eg pool size). These metric data related to the position of the 3D points are expressed in the absolute reference linked to the scene. This mark may be chosen arbitrarily, it may advantageously be chosen so that the center of the mark is an angle of the pool, the X and Y axes corresponding to the sides of the pool (length and width) and the Z axis being taken so that the mark thus formed evening orthonormed direct (ie Z perpendicular to the body of water). An initial or preliminary manual matching step between points in the image and known 3D points in this marker then allows the calculation of the intrinsic and extrinsic parameters of the camera. Knowing the intrinsic and extrinsic parameters, it is possible to know the position in the image of any 3D point whose coordinates are known in absolute reference. Conversely, it is not possible to know the 3D position of a point of the image but only the equation of the 3D line on which this point is located (for a point of the image plane p, the set of points 3D belonging to the line passing through the image point p and the optical center is projected in p). This equation is as follows: faX + bY + cZ = d eX + fY + gZ = h Where the parameters a, b, c, d, e, f, g, h are known. We must therefore add here a hypothesis on the prior knowledge of the 3D surface on which the object evolves. In most cases, the planar hypothesis is sufficient (plan of water, football field ...) It is however possible, although complex to implement, to consider surfaces not planes, the main difficulty residing in the estimation of the 3D parametric equations of the surface. Having positioned the absolute reference at an angle of the pool with the Z axis perpendicular to the plane, the equation of the body of water exclaims simply as Z = O. If we consider a point p of the image knowing the extrinsic and intrinsic parameters we can calculate the 3D coordinates of a line on which is the corresponding 3D point in the scene. By now adding the hypothesis of the membership of this point in the water body (valid hypothesis in the case of a swimmer moving in the body of water), the 3D point is determined by the intersection of the plane of water. water (equation Z = 0) with the 3D projection line. Taking the equation of the projection line and introducing the equation Z = O, we find a system of two equations with two unknowns X and Y that can be solved in order to evaluate in the absolute reference of the scene the X and Y coordinates of the 3D point corresponding to the point p of the image. In the following, we will call pos (t) the vector of the X and Y coordinates associated with the point p in the absolute reference of the scene.

  The calibration step is also optionally followed by a step of detecting the water lines in the pool, so as to then restrict the search area of a swimmer during the speed estimation. The smaller the search area, the faster the speed calculation time and the real time, which is important in the case where the calculated speed represents an instantaneous speed to be displayed in real time, for example 25 times per second.

  The detection of water lines is done by a gradient-based image processing, either by the detection of straight contours, straight lines, or by a technique based on the contrast of water lines, black color, using tools of mathematical morphology, by rectilinear filtering or by using the top hat, tool of known mathematical morphology which makes it possible to extract the fine and contrasted details of an image.

  Once the segmentation of the pond into water lines has been performed, the treatment device 2 restricts the search area of a swimmer to its water line.

  The processing device 2 thus receives the images of the camera 1 in real time. It therefore analyzes at every moment the video data in order to match the distance d2 on the video and the distance dl actually traveled by the swimmer. These measurements can be made very closely to measure the instantaneous speed. The processing device 2 first locates a distinctive swimmer sign from which it analyzes the speed. This distinctive sign can be detected before the start of swimming when the swimmer waits for the start of the race on the starting block. Each swimmer is therefore marked by this distinctive sign and the distance traveled by the swimmer is the distance traveled by this distinctive sign. For example, you can choose the swimmer's swimming cap. This makes it possible to distinguish between different swimmers.

  The distinctive sign is preferably an element of the swimmer which remains close to the surface of the water and parallel thereto. This distinctive sign is manually initialized by a human machine interface on the processing device. The operator then simply clicks on the distinctive sign and the processing device 2 analyzes this distinctive sign so as to extract the characteristics necessary for its recognition thereafter. Among the necessary characteristics are the color, the texture, the shape.

  In other embodiments, the distinctive sign may be recognized automatically by the treatment device which then selects a distinctive sign for each swimmer in each water line. The evolution of the distinctive sign from one image to another is also conceivable by object tracking methods in a sequence of images known to those skilled in the art. When the processing device 2 has obtained the distance d1 from the distance d2 and the calibration data, it calculates the speed of the swimmer at each desired instant t by the following formula: Instantaneous speed = (Pos (t + dt) Pos (t)) / dt With (Pos (t + dt) - Pos (t)) representing the distance d2 traversed between the instants t and t + dt, the swimmer moving in a plane and not in a 3D surface.

  FIG. 2 represents a search zone in which the treatment device 2 will search for the swimmer. The swimmer whose speed is to be measured is in the L4 water line. The basin has 7 water lines, from L1 to L7. From the first measurements made, the processing device 2 analyzes the direction of movement of the swimmer. This advantageously allows it to reduce the search area of the distinctive sign in the image and thus save time for calculating the instantaneous speed.

  When the swimmer reaches the end of the line, the search zone situated upstream of the swimmer's movement no longer contains the distinctive sign and the search zone as shown in FIG. 2 is not exploitable. The search area is then expanded in both directions, forward and backward of the swimmer. This search area may be limited to the line in which the swimmer moves. Similarly, when the swimmer arrives at the end of the line, either the race is over, or the swimmer will disappear under water time to turn around. The processing device can not therefore locate the distinctive sign at this time. It will therefore analyze successively all the images without being able to generate instantaneous speed until the distinctive sign is again identifiable. The processing device 2 displays the instantaneous speed in real time. Thus all the spectators in the pool can visualize the value of this speed. The processing device 2 also transmits the instantaneous speed with the video as the associated metadata. In this way, during the broadcast of the video, the instantaneous speed thus measured can be displayed simultaneously with the images. This transmission of speed is done through metadata. The video data is transmitted live with the associated speed as a meta data to broadcasting means, for example television. The associated video data and metadata are also recorded on a hard drive or other permanent storage medium for later use, for example by swimmers and coaches during the course viewing to evaluate and understand their performance.

  FIG. 3 represents a method according to the invention, implemented by the processing device 2 as well as by the camera 1. During a first step E1, a video representative of the bounded environment in which the swimmer evolves is captured . This video makes it possible to calibrate the camera during a step E2 and thus to match a distance d1 in the pool and a distance d2 on the video as indicated above. During a step E3, the manipulator indicates on an image of the video the distinctive sign that distinguishes the mobile element on the video. During a step E4, this mobile element is characterized so as to locate it in the following images. This movable element is characterized by the peculiarities of the movable element as indicated above. Then, during a step E5, it is possible to start the capture of the mobile element when it is desired to start the speed measurements of the mobile element. During a step E6, the speed calculations are started, for example when the race has started. During a step E7, from the first speed calculations, the calculated speed is displayed and / or transmitted to a remote device to be recorded or broadcast.

  Figure 4 shows another aspect of the invention in which the camera is movable. The camera is slaved to the speed of the swimmer. The camera is equipped with speed measuring equipment. In this case, the instantaneous speed is given by the speed of the camera and simply calculated by usual speed calculation devices. The speed measurement is thus transmitted to the processing device 2.

  In other embodiments, the distinguishing sign is the centroid of an image recognizable area. This advantageously makes it possible to obtain real value displacements providing more precise instantaneous speed measurements.

  In other embodiments, it is possible to measure the speed of several swimmers simultaneously. For this, each swimmer is identified by the treatment device 2 in his water line. The treatment device identifies for each swimmer a distinctive sign, for example the bathing cap. The lines of water being usually separated by floats whose color is detectable, the search device can easily isolate each swimmer. In other embodiments, it is possible to have several cameras above the pool to cover the entire pool if a single camera is not sufficient. The cameras can therefore be placed at different angles to cover the pool. The mapping between the distances d1 and d2 is then done by combining the parameters of the different cameras and their position relative to the pool.

  Another preferred embodiment relates to measuring the speed of a vehicle on a road. This can also be the case during a grand prix formula 1 for example or during speed control by the authorities on the roads.

  In this case, a video camera is placed above the road and observes the movement of the vehicle in an area whose extent is bounded, for example a road section of 300m. The camera can be hung in height, for example hanging from a bridge.

  As in the previous embodiment described in FIGS. 1 to 4, the camera is connected to a processing device that calibrates the camera and makes it possible to establish a correspondence between the distance measured on the video and the actual distance traveled by the camera. vehicle. The operation described for FIGS. 1 to 4 is also valid for this embodiment. The distinctive sign of the vehicle is for example the center of gravity of the vehicle.

Claims (10)

claims   A device for measuring the speed of at least one moving element in a bounded environment (3) comprising: video capture means (1) whose capture field covers said environment (3); calibration means (2) mapping the granularity of the captured images to the actual distance in said environment (3), - means for locating a distinctive sign of said movable element, characterized in that it comprises means for identifying the position in said environment (3) of said distinctive sign on said captured video at a first instant and at a second instant, said calibration means (2) being able to convert said evolution of said distinctive sign on said video between the first instant and the second instant in a distance traveled (dl) in said environment (3), said device comprises means for calculating the speed of said element over said distance traveled between the first instant and second instant.   2. Device according to claim 1 characterized in that it comprises means for recovering the dimensions of said environment (3) and the distance between said environment (3) and said video capture means (1).   3. Device according to claim 2 characterized in that said calibration means (2) match the granularity of the captured images and the actual distance (dl) according to the intrinsic parameters of said capture means (1), and the position of said capture means with respect to the absolute reference of said environment (3).   4. Device according to one of the preceding claims characterized in that said capture means (1) comprise a fixed camera placed above said environment (3).   5. Device according to one of the preceding claims characterized in that it comprises means for determining the direction in which said moving element evolves, said means for identifying said position of said distinctive sign seeking said position on said video according to said direction.   6. Device according to one of the preceding claims characterized in that said distinctive element is the center of gravity of a zone of said movable element.   7. Device according to one of the preceding claims characterized in that said movable member is a swimmer and said environment (3) is a pool.   8. Device according to the preceding claim characterized in that it comprises means for identifying the different water lines of said pool.   9. Device according to one of claims 1 to 6 characterized in that said movable element is a car and said environment is a road. 25   A method for measuring the speed of at least one moving element in a bounded environment comprising: a video capture step (E1) whose capture field covers said environment; a calibration step (E2) setting corresponding to the granularity of the captured images and the actual distance in said environment, - a step (E3) of locating a distinctive sign of said movable element, characterized in that it comprises a step (E4) of identification of the position in said environment of said distinctive sign on said captured video at a first instant and at a second instant, said calibration step converting said evolution of said distinctive sign on said video between the first instant and the second instant into a distance traveled in said environment, said method comprising a step (E6) of calculating the speed of said element over said distance traveled between the first instant and second instant. 10