RU2749941C2 - Universal way to photograph traffic violations - Google Patents

Abstract

FIELD: transportation.

SUBSTANCE: invention relates to methods for determining the excess of the speed limit of vehicles. The proposed method for photographing traffic violations is that the speed of the vehicle, its movement and the time spent on movement are controlled according to the readings of its own sensors received from the transmitting module placed on the vehicle, and the change in its position relative to other vehicles and road elements is determined by its photographs. To present the results, the photo fixations are carried out by means of six pairs of photographs taken at precisely known times, necessary to ensure reliability of the evidence; the pair of the photographs formed by the combination of the four real photographs are used to determine the speed of the vehicle in the time interval between them, the speed of the vehicle is determined by the values of the displacements of the beginning and end of the body elements that have an extension along the trajectory of movement, in units of the length of these elements in each photograph in a pair; the determination of the exact amount of movement of the vehicle on the ground is carried out after its identification, clarification of its overall dimensions and the dimensions of the body elements selected as reference; preliminary selection of vehicles suspected of violating the speed limit is performed using wheels as reference body elements, reducing their many sizes to three standard sizes; the vehicle must be equipped with a transmitting module that transmits a standardized signal containing at least the VIN of the vehicle (VIN of the vehicle) and the readings of its speed sensor, and the traffic control complexes have devices that receive these signals.

EFFECT: claimed method improves the accuracy of determining the permissible vehicle speed.

3 cl, 6 dwg

Description

3.2.4.1. The technical field to which the invention relates.

The invention relates to the field of application of means and methods for monitoring compliance with traffic rules (SDA). The claimed method for determining the speed of a vehicle (TS) is not a method of direct measurement of speed and the technical means that use it do not need appropriate metrological certification except in terms of the accuracy of measuring time intervals. The measurement of speed is made on the basis of photographic data of vehicles suspected of violating the speed limit by any means of length comparison, or by devices of the vehicle itself, the readings of which are broadcast (if available) by Wi-Fi-, Bluetooth- or other beacon installed on the vehicle.

Classes of invention:

3.2.4.2. State of the art.

Currently, there is a wide range of means and methods for monitoring compliance with traffic rules, which are based on both direct measurement methods and indirect methods. In addition to monitoring the position of the vehicle on the road, the most important of all parameters is the monitoring of compliance with the speed limit. As is known, there is no direct way of measuring speed, by definition, and, in all likelihood, one should distinguish between direct ones arising from the definition of the term speed and indirect ones. Direct methods of measuring the speed of a vehicle are based on measuring the distance traveled by the vehicle and measuring the time during which it has traveled this distance. This also includes the measurement method based on the mathematical approximation of the vehicle displacement function from time and the subsequent finding of the time derivative of this function. The ratio of the displacement to the time interval for which it occurred gives the value of the average speed for a given segment of the trajectory. Most indirect measurement methods are based on the measurement of the return time or change in the frequency of the electromagnetic wave reflected from the vehicle (radars, lidars). Significant disadvantages of the known methods are the presence of hidden processes in the production of measurements, a completely erroneous approach in providing information about the errors of devices based on them and the errors of the methods themselves, which often remain outside the scope of consideration of the errors of devices, incorrect information about the confidence intervals and confidence probabilities for the measured values.

The objective of the claimed invention is to improve the accuracy and reliability of measurements, to provide visibility and accessibility for a wide range of involved persons to understand the photographic recording materials and the possibility for them to directly recheck the measurement results, to eliminate calibration processes when measuring the vehicle speed.

The prior art of the claimed invention is well described in the corresponding section of the description of the invention-analogue RU 2559418 C2 (priority from October 15, 2013 authors I.F. Komlik (RU). Simonchik K.K. (RU), Bondar D.B. ( Barskiy I.V. (RU) "A method for determining the position and speed of a vehicle on the road and a complex for its implementation" [5] and together with the description of the analogous invention itself forms the state of the art of the claimed invention. is given below.

Known "Method for determining the speed and coordinates of vehicles with their subsequent identification and automatic registration of violations of traffic rules and a device for its implementation" ("Arrow"), RF patent RU 2382416 C2, publ. 20.02.2010. (see the source [1]) This method of measuring coordinates and speed involves measuring the coordinates and speed of vehicles (TC) using a pulsed radar and a video camera. The range and speed of the vehicle are determined by the radar. At the same time, the coordinates and speeds of the same vehicles in the frame are calculated from the vehicle image received from the video camera, after which the obtained independent data streams about all vehicles that are currently on the selected section of the roadway are compared. To obtain metrologically reliable data on the speed and range to the vehicle, the radar data is used, and the data received from the video camera is taken as the reliable data on the azimuth of the same vehicle. As a result of the comparison, each vehicle is provided with further support until the state registration plate is recognized, after which a video camera for recognizing the state registration plate forms a frame of the vehicle image with the recognized state registration plate, speed and other necessary information.

The disadvantage of this method is the need to integrate data from a video camera and radar to ensure the operation of the entire system. If video data is not delivered for any reason or is of poor quality (for example, in difficult weather conditions or in conditions of insufficient visibility), the reliability of the system drops to the point of complete inoperability. The disadvantages of this method include the fact that the determination of the speed (and location) of the vehicle and its identification (license plate recognition) occur at different points in time. At the same time, the determination of the vehicle coordinates to accompany it to the recognition zone occurs at a considerable distance (up to 1000 m), where license plate recognition is impossible. The disadvantage is that the method is implemented only with the help of permanently operating two independent systems (radar and video camera), which leads to a significant increase in the size and power consumption of the system. In addition, the radar used to implement the method (due to the used physical principle) does not provide information about the azimuth of the target, and the video camera, as follows from the description of the patent, does not provide metrologically reliable data on the speed (due to the low accuracy of determining the position of the vehicle from the video image at large distances), thus creating the possibility of incorrect tracking of the vehicle into the recognition zone.

The known "Method for determining the speed of a vehicle", RF patent RU 2442218 C1 (publ. 10.02.2012), (see source [2]) using the results of measuring the distance to the car, obtained at different times by analyzing the license plate image obtained from a camcorder. The range to the vehicle is calculated by changing the size of the license plate image obtained on the matrix of the video camera at different times, taking into account the correction of distortions in the proportions of the license plate plate. For each of the moments of fixing the license plate, the distance to the vehicle is determined, and the speed is calculated by changing the distance in a fixed time.

The disadvantage of this method is the lack of accounting for errors associated with a change in the angle of observation of the car when it moves in the control area. In addition, the limited scope of the method is also a disadvantage. It is applicable only for systems located at a low height above the road, and provided that the angle (2-4 degrees) between the direction of movement of the vehicle and the axis of the video camera is small. It is these conditions that provide a small angular measurement error, however, they also make the application of the method unsuitable in cases where the system is designed for automatic control of heavy traffic flows. For such streams, measurement at a low angle to the direction of travel will inevitably lead to mutual shading of vehicles, which implies the need to install the camera at a significant (up to 10 m) height.

Known "Method for determining the speed of a vehicle", RF patent RU 2419884 C1 (publ. 05/27/2011), (see source [3]). According to the proposed method, video frames are recorded with the image of the plate of the state registration plate of the vehicle when it is moved in the control zone of the video camera, and the parameters of the location of the video camera relative to the road plane are additionally measured. The coordinates of the points of the corners and the center of the GRZ plate are measured in the video frame coordinate system at the beginning and at the end of the control zone and the direction of movement of the GRZ plate in the road coordinate system is determined by the points of the center of the GRZ plate at the beginning and at the end of the control zone. Next, the amount of movement of the center point of the GRZ plate is determined in a plane parallel to the plane of the road located at the height of the movement of the center point of the GRZ plate. According to the indicated amount of movement and the time during which this movement was made, the vehicle speed is determined.

The preliminary measurement of the parameters of the location of the video camera relative to the plane of the road is a laborious task, during which it is necessary to take into account, as follows from the description of the patent, a significant number of parameters associated with both the position and parameters of the camera and lens, and with the parameters of the road in the control zone (height, angles of rotation, roll, tilt, focal length of the lens, resolution and dimensions of the matrix in the video camera, angles of tilt and roll of the road, etc.). The need to solve it significantly complicates the process of installing a system that implements the described method. Reducing the number of such parameters is possible only when the video camera is installed directly above the traffic lane (or with a small offset from its axis), which significantly reduces the scope of the proposed method, since it requires the construction of special engineering structures above the road.

The disadvantages also include the possibility of uncontrolled random displacement of the camera (when the fastening is loosened due to vibrations, soil settlement, etc.) during operation, leading to a decrease in the measurement accuracy.

In addition, the implementation of the method in measuring equipment requires a qualified confirmation by the state metrological service of the correctness of determining the location parameters of each video camera (and its periodic confirmation directly at the installation site), since the inaccuracy of determining these parameters has a catastrophic effect on the error in speed measurements. The disadvantages of the method also include the use of the time interval between the first and the last frame for calculating the speed without taking into account the dynamics of the speed change during the movement of the vehicle in the field of view.

Known complex video recording and measuring the speed and coordinates of vehicles (see source [4]), Utility model RU 113398 U1, publ. 02/10/2012), containing a radar, a video unit and a processing unit connected to them, located in the casing, the radar is configured to simultaneously measure the speed, range, azimuth and dimensions of the vehicle, and contains a transmitter, at least two receivers and a digital module signal processing. The disadvantage of this complex is the need for significant computing resources, the inability to check the accuracy of measuring the speed and the correct identification of the target by an alternative method. The disadvantages also include the lack of redundancy in the system, which can lead to short-term uncontrolled failures in the measurement and / or identification of targets. Also, the disadvantage of the complex is its low secrecy associated with constant (albeit insignificant in power) microwave radiation, which makes it possible to detect it using radar detectors.

The closest analogue.

The closest analogs of the proposed method are the methods described in inventions RU 2559418 C2 (priority from October 15, 2013 authors I.F. Komlik (RU), K.K. Simonchik (RU), D.V.Bondar (RU), Barsky IV (RU) "A method for determining the position and speed of a vehicle on the road and a complex for its implementation" [5] and RU 2557667 C2 (priority from December 12, 2013) Mustafin R. G. (RU) // Method for determining the speed of a vehicle //

The invention-analogue RU 2559418 C2 (Barskiy I.V. et al.) Proposes a method for determining the position and speed of a vehicle on the road, in which video frames with the image of the plate of the state registration plate (GRZ) of the vehicle are obtained by means of a video camera when it is moved in the control zone video cameras, determine the coordinates of the points of the corners of the GRZ plate in the coordinate system of the photosensitive matrix, convert the coordinates of the points of the corners of the GRZ plate from the coordinate system of the photosensitive matrix to the coordinate system of the road, taking into account the parameters of the location of the video camera relative to the road plane, bind the coordinate grid in the plane of the road to the coordinate grid in the plane photosensitive matrix, recognize the symbols and format of the license plate by means of optical character recognition, determine the coordinates of the corners and the center of the license plate on the image, based on the obtained coordinates of the image plate on the matrix of the video camera restore the coordinates of the license plate on the road at the given points in time, using the grids, the vehicle speed is calculated from the displacement of the GRZ coordinates, the vehicle speed is corrected, taking into account the peculiarities of the GRZ location.

Determination of the amount of vehicle displacement based on frame-by-frame analysis of the GRZ image is used in the Strelka and Azimuth measuring complexes. If we consider the GRZ as an obligatory part of the vehicle body, then these complexes can be considered as analogs of the claimed invention.

The difference between the invention-analogue from the above is that it contains a stage at which data on the position of the vehicle on the road is obtained from the radar, at the stage of binding the coordinate grid in the plane of the road to the coordinate grid in the plane of the photosensitive matrix, the parameters of the perspective projection are determined using the data on the position and speed of the vehicle, received simultaneously from the radar and from the video camera, and on the basis of these parameters determine the speed of the vehicle by the displacement of the coordinates of the license plate. The refinement of the parameters of the perspective projection is preferably carried out periodically at predetermined time intervals, and in cases where the magnitude of the measurement error goes beyond the specified one, the radar is automatically turned on to further refine the parameters of the perspective projection. As the results of measurements of the speed and position of the vehicle, it is preferable to use data from both a radar and a video camera as a result of image analysis. The radar is designed with the ability to simultaneously measure the speed, range, azimuth and dimensions of the vehicle and contains a transmitter, at least two receivers and a digital signal processing module. The field of view of the radar is preferably coordinated with the recognition area of the license plate provided by the video camera. This method allows you to significantly reduce the cost of periodic manual calibration, as well as significantly reduce the energy and computing resources required for the operation of the system, due to the short-term connection of the radar only when correction is necessary. The obtained technical result is achieved due to the fact that the calibration of the proposed measurement method is carried out in a fully automatic mode based on data from the radar and from the video camera. Calibration provides automatic determination of perspective projection parameters for subsequent measurement of vehicle speed and position only from its image on a video camera without using a radar.

In the invention-analogue RU 2557667 C2 (Mustafin R.G.), it is proposed to install active or passive radio tags RFID (radio frequency identification) on the vehicle, embedding information about the vehicle into the reflected radio signal, with the help of which the vehicle speed is measured.

Fundamental disadvantages of existing methods for measuring vehicle speed.

It should be noted that in analogous inventions there is the possibility of operator or third-party interference in the measurement process by shifting the optical axis of the video camera or the direction of the radar beam after the process of their calibration. This does not allow considering this method as an automatic way of measuring speed.

The closest analogue uses a combined method for measuring vehicle speed. At the initial stage, this method uses microwave measurements of the range and Doppler speed of the vehicle, the data array of which is compared with the coordinates and speed of movement of images on the video camera matrix from frame to frame. Based on this mapping, the coordinate system referenced to the road plane is brought into line with the matrix coordinate system. After this step, the need for a microwave part, according to the description of the present invention, allegedly disappears, and all measurements of the vehicle speed are carried out on the basis of the analysis of the vehicle images on the matrix of the video camera. It should be noted that the accuracy of the comparison of the coordinate system (virtual coordinate grid) on the road with the coordinate system on the matrix is determined by the accuracy of measurements performed by the microwave unit, the size of the road element corresponding to one pixel, the accuracy of measuring the time and the nature of the frame formation (sequence of recording pixels) of the recording in time. The root-mean-square error in measuring the vehicle speed by the value of the Doppler shift is determined by the root-mean-square error in measuring the Doppler frequency, which in turn is determined by the duration of the measuring pulse and the error in measuring the angle between the direction of the radius vector from the emitter to the vehicle and the direction of the vehicle speed. With a probe pulse duration of 25 milliseconds, the Doppler frequency measurement error is 40 Hz. This gives an error in measuring the radial speed (velocity components along the radius vector ~ 1 km / h. To obtain the true vehicle speed, the radial speed must be multiplied by the inverse cosine of the angle between the vehicle speed direction and the radius vector on the vehicle. which reflects the signal received by the detector is impossible, then the angular size of the vehicle from the location of the video camera, taking into account the degree of uncertainty of the distance to the vehicle, is (a / r ~ 1/20, where a is the size of the rectangular projection of the vehicle onto the matrix plane, r is the distance to the vehicle , about 2-3 degrees When the device is installed near the road, the base angle is 5-10 degrees, which gives a root-mean-square error of the order of (2/10) 20% of the measured vehicle speed.

It should be emphasized that in order to comply with the principle of the presumption of innocence, the person brought to administrative responsibility must, firstly, have the technical ability to control this parameter, and secondly, the accuracy of measuring the critical parameter (in this case, the vehicle speed), on the basis of which the administrative responsibility, must ensure that the measured value of the critical parameter is in the confidence interval, excluding the values of the parameter for which administrative liability is not carried out, (note: the event "hitting the true speed of the vehicle outside the confidence interval of the measured speeds" is incredible event). In the case of a normal (Gaussian) distribution of measurement errors, the confidence interval with the width of the root-mean-square error (in the standard notation V ± ΔV) corresponds to the confidence probability w ₁ = 0.683; double confidence interval V ± 2ΔV w ₂ = 0.955; triple confidence interval V ± 3ΔV w ₃ = 0.997. This is generally known information from the theory of measurement errors of physical quantities (see sources [14, 15]).

The confidence probability that the result of a measurement used as an accusatory one falls within the established confidence interval should be such that the probability of wrongly prosecuting on the basis of the measurement data during the use of this measurement method is close to zero. If we take the total amount of fines 50 million fines per year, 20 years - the duration of the method, then, taking the probability that in this set of fines there will be 1 wrongly prosecuted equal to 0.1, we get the probability that the result of measuring the vehicle speed will be released beyond the confidence interval, should be less (or at least on the order of) 10 ^-10 . This probability corresponds to a confidence interval approximately equal to 10 root-mean-square errors, 10ΔV. Thus, with the root-mean-square error declared in the invention, an analogue of RU 2559418 C2 ΔV ~ 3 m / s (~ 10 km / h), the confidence interval corresponding to the reliable hit of the measured speed in this interval is about 100 km / h. This makes the use of this invention and the device based on it senseless from a legal point of view, i.e. in terms of guaranteeing compliance with the presumption of innocence. The same applies to all existing meters used for vehicle control, for which rms errors of ± 2-3 km / h are declared (or 2-3% at speeds over 100 km / h). For them, the confidence interval that guarantees compliance with the presumption of innocence is at least 20-30 km / h. It is no coincidence that it is close to the speed limit allowed by Art. 12.9 of the Administrative Code of the Russian Federation. Those. the legislator noticed such a problem and was resolved in a somewhat awkward way. Thus, the lack of clarity in the presentation of measurement results and the lack of accuracy of instruments used to monitor traffic, forces to abandon the principle of the presumption of innocence or violate traffic safety, making the applied restrictions meaningless. The claimed invention is directed to resolve this conflict.

3.2.4.3. The essence of the invention.

As already noted, the vehicle speed is measured in the coordinate system associated with the measuring device. For this reason, in all cases, the speed is measured not by the vehicle itself, but by measuring the speed of the projection of the vehicle. In this case, the most difficult and unsolved problem is taking into account the spatial location of the vehicle speed vector. This problem is the main obstacle to considering the used speed meters as mobile (not stationary) automatic meters. In the initial versions of photo-radars, to solve the problem of the uncertainty of the angle between the radar beam (which itself has a divergence of 4 degrees large enough for the required accuracy), it was proposed to install the photo-radar at a certain angle to the road markings on its straight sections. But this condition, necessary for the measurement accuracy, was violated both by the operators of photo-radars, setting them not in their regular positions, and not on straight sections, and by the "yaw" of the vehicle's speed within the traffic lane or when changing lanes to another lane. Thus, at the moment, the existing photo-radars to measure the speed of vehicles use hidden operations, through which the influence of third-party factors, including operators, on the measurement result is possible and which cannot be controlled in a “manual” mode by any interested person. The proposed invention removes these questions and makes the procedure for measuring the vehicle speed completely transparent, automatic and accessible for control by a person held liable on the basis of the measurement data, and the cost of devices using exclusively the claimed measurement method is hundreds of times cheaper. Of the existing means, the most close to solving this problem are the Kordon-2M, Avtodoria, Kordon-Temp complexes, which present two photographs of the vehicle for viewing (see, [6, 17, 18], Figures 1, 2.) but, unfortunately, in no way not related to each other.

The claimed determination method is based not on a stationary coordinate system associated with a photo-radar, but on a local coordinate system associated with the vehicle itself, included in the video camera frame. Generally speaking, at any turns of a car, the threshold line between the wheels is at least in the same vertical plane with the vector of the speed of the rear wheel of the car. Those. this line can be taken as the coordinate axis, and the midpoint on this line as a reference point. As a rule, this line coincides with the velocity vector and its length can be used as a scale along the named coordinate axis. The length of the named line (like any other line on the vehicle) is known from the vehicle outline drawing and can also be measured on the vehicle itself. Subsequent frames, captured at precisely known moments in time, which are automatically indicated on the frame with an accuracy of no worse than 0.0001 sec, will show the displacement of this line at other moments in time. Thus, when comparing several frames, we have the longitudinal displacement of the characteristic line in units of the length of this line, which can be easily converted into speed. With a frame size of 2500x3000 pixels (7.5 Mpc) and with the size of the photographed scene, the width is approximately two lengths of a car, i.e. about 6-8 m, we get the length per pixel, 2 mm. With an uncertainty of the time interval of 0.0001 s against a background of 0.05 s, we obtain an uncertainty of the speed of 40 mm / s, or about 0.15 km / h. And the reliable value of the speed falls within the confidence interval of ± 1.5 km / h, which is commensurate with the accuracy of maintaining the speed determined by the division value on the car's speedometer (division value - 10 km / h). The standard accuracy of the dial gauge is about half the scale division, but not less than 1/4. Therefore, generally speaking, the requirement for the driver to maintain the engine power and, accordingly, the vehicle speed with an accuracy significantly different from this value is absurd. Thus, it is possible to blame the driver for deliberate violation of the speed limit only if the deviation of the measured speed from the established traffic rules exceeds 3-5 km / h. At the same time, the root-mean-square error of measuring the speed by a measuring device should in this case be no more than 0.5 km / h. The estimates obtained indicate that the claimed method for measuring the vehicle speed meets this requirement. For the purpose of measuring speed, any line on the vehicle body in the direction of travel can be used, the length of which can be determined from the dimensional drawing or on the vehicle itself. For the currently used speed measurement methods, in the case of presenting their results in a pair of sequential photographs with an indication of the time interval with an accuracy of 0.0001 sec, the proposed method can be an addition that allows you to check the measurements and guarantee their reliability.

At the same time, for the purpose of instantaneous assessment of the vehicle speed and its possible falling into the interval of forbidden speeds during automatic fixing, it is required to have a universal standard of length located in the registration field of the video camera, since the determination of the overall dimensions is possible only after identification of the vehicle brand associated with its registration number and real the image of the fixed vehicle.

Such a standard is the vehicle wheel, which has a standard range of standard sizes: R12, R13, R14, R15… R20. In the analogous invention, the width of the license plate is partially used as a standard (reference) of length. However, in the claimed method, this standard (reference) cannot be used, since the dimensions of the license plate in the direction of travel are negligible and the determination accuracy associated with it is unsatisfactory. Thus, for a preliminary assessment and selection of vehicles suspected of violating the speed limit, the wheel size is used as the length standard and the vehicle displacement in the video camera frame is measured in units of the wheel diameter along the direction of displacement. If it is impossible to assign the size of a wheel to one or another standard size by the angular size of its vertical diameter, 2-3 established standard sizes are used based on the size of the area in the frame occupied by the car. For example, all cars in the frame are classified as R14, all trucks are R18. In this case, the error in estimating the speed will be about 10% due to the wrong standard size, which is quite enough for the purposes of preliminary selection of vehicles suspected of violating the speed limit. In addition, in the setting mode, the operator can enter the standard size of the wheels of the photographed car and further assignment of the standard sizes will occur automatically according to the angular size of the wheel. These actions of the operator cannot be considered as an influence on the measurement result, since serve only for the preliminary selection of suspected vehicles and to reduce the amount of further work of the operator to identify vehicles that have really violated the speed limit. The final determination of the vehicle speed is made using a cascade of 4 photographs, which, in fact, make it possible to carry out only one body element up to 6 measurements by the suspects themselves and to make an independent assessment of the parameters of the error probability distribution in this series of measurements.

The use of this method for determining the conditional speed of a vehicle in a smartphone with a navigator does not require its registration as a device that photo-fixes a suspected vehicle violation, as a measuring instrument, since the speed is actually measured by conventional means of measuring length and time. Thus, in order to use a smartphone with a navigator as a means of fixing a vehicle suspected of violating traffic rules, it is necessary to ensure that the smartphone receives a cascade of photos with the geographic coordinates of the smartphone and the relative time on each photo, with an accuracy of at least 0.0001 seconds, as well as the immediate sending of the received data to electronic storage.

The possibility of using linear estimates in the claimed method.

Obviously, due to the laws of trigonometry and the peculiarities of perspective projection, various parts, elements of the vehicle body structure (TC) are displayed in the photograph nonlinearly, i.e. on a different scale, as well as sections of the trajectory along which these elements move. However, it is also obvious that in the case of coincidence at the moments of taking photographs of a point on the trajectory of movement of a body element corresponding to the end of the element with a point corresponding to its beginning in an earlier photograph, this fact, recorded in a pair of photographs, will mean that the movement of the vehicle is equal paths between these points of the body element. This fact does not depend on the quality of the lens and the distortion it introduces and can be used to calibrate any pair of photographs. Considering that in a perspective projection straight lines on the ground are converted into straight lines on the matrix (photographic film), it is possible to match any pair of points on the real body of the vehicle, thereby determining the real distance between them, to a pair of points in the photographs. In this case, the accuracy of determining the vehicle speed will be determined only by the resolution of the matrix (photographic film). The process of obtaining stroboscopic photographs cannot be considered as a measurement of speed, but only as a process of fixing the movement of a vehicle, its trajectory and the trajectory of movement of its body elements. Already after photographing the nature of its movement, the vehicle speed is measured using conventional length measuring instruments, taking into account the knowledge of the exact time intervals when the photographs were taken in the cascade. The confidence interval characterizing the accuracy of such a measurement should be taken as the root-mean-square error of the instruments with which drivers control the speed of the vehicles they drive. In the standard sense, this is half the division value of the vehicle speedometer. The division price is 10 km / h. Thus, the confidence interval, on the basis of which it is possible to judge whether the driver is guilty or not guilty of violating the speed limit, is ± 5 km / h. The confidence level of the results of such a measurement (i.e., with a confidence interval of ± 5 km / h), based on the principle of the presumption of innocence, should be such that the appearance of an erroneously imposed fine in the totality of ever imposed fines would be an incredible event. An event with a probability of at least 0.1 can be considered such an incredible event. With the number of penalties imposed, for example, in 2016, about 50 million, and assuming the duration of these measurements at least 20 years, we find that the probability of the measurement result going beyond the confidence interval of ± 5 km / h should be 10 ^-10 . Based on the normal distribution of measurement errors, this means that the root-mean-square error of the measurement method and the instrument making the measurements should be no worse than ± 0.5 km / h. To date, there are no instruments that guarantee such an RMS error. The proposed method for determining the vehicle speed is intended to fill this gap. The existing vehicle speed meters, having real (and not declared by the manufacturer) root-mean-square errors of 2-3% of the measured speed, provide a confidence interval that guarantees compliance with the principle of the presumption of innocence of 20-30 km / h. This is very difficult to meet the safety requirements. More precisely, it does not satisfy them.

Thus, we state that the confidence interval of measurement, on the basis of which a person is brought to administrative responsibility, is determined by the accuracy of the instruments by which this person has the ability to control the measured value, and the confidence probability is determined by the probability of violation of the principle of the presumption of innocence in the entire set of measurements by this method. Having formulated the requirement for the inaccuracy of the proposed method as ± 0.5 km / h (~ ± 0.5% of 100 km / h), let us estimate the possibility of its satisfaction.

If we take as a standard the shooting of 4 consecutive photographs (see Figure 3.) with an interval of 0.02 seconds between them, then at a vehicle speed of 250 km / h in a time of 0.06 seconds, the vehicle will shift by ~ 4.2 m (approximately the length of the car body). at a distance of 35 m from the vehicle, all manipulations with measurements in the photograph will take place in the range of angles (with a margin) ± 5 ° (angular aperture of a photo or video camera), which is characterized by a nonlinearity of about 0.1%. That is, assuming that the distances in the photo are proportional to the distances on the ground, we will be mistaken by no more than 0.2%. In addition, the coincidence in the photo of the beginning and end of any characteristic size (body element) of the vehicle eliminates this error as well. If necessary, it is also possible to carry out a rigorous calculation of the displacements taking into account the nonlinearities of the transmission of distances from the terrain to the photograph. At the same time, it should be noted that it does not matter which section of the road control is carried out - straight-line or with a turn. All these moments are recorded in the photo and can be taken into account in the event of a dispute in the calculation of the vehicle speed.

If in the theorem of sines with respect to the side of a triangle - the sine of the opposite angle instead of the sine of the angle we take the angle itself (in radians), then for angles ± 15 ° {(Sin0.26-0.26) /0.26 ~ 0.011} the difference will be about 1%, in the range ± 10 ° the difference will be {(Sin0.17-0.17) /0.17 ~ 0.0048} less than 0.5%, and in the range of ± 5 ° - {(Sin0.09-0.09) /0.09 ~ 0.0013} about 0.1%. The range of ± 10 ° corresponds to photographing a passenger car "on board" from a distance of 15 m. The best shooting is at an angle of 45 ± 15 ° to the direction of travel. At 30 ° between the axis of the scene and the direction of the vehicle speed, the projection of the vehicle side and its displacement onto the matrix plane is halved, which halves the accuracy of measuring the vehicle speed from a photograph as compared to a photograph taken at an angle of 90 ° to the direction of travel. At 60 ° between the axis of the scene and the direction of the vehicle speed, the conditions for recognizing the license plate worsen, since the ratio of the plate width to its height decreases by a factor of 2. Nevertheless, the working range of shooting angles can be extended up to 3 times worsening of the named conditions relative to the best ones. At the boundaries of the 45 ± 25 ° range, either accuracy (at an angle of 20 ° to the trajectory, used for mobile photo-radars of the KRIS type) or recognizability (at an angle of 20 ° from the normal to the trajectory), in comparison with the 45 ° direction, degrades in 2.1 times. The limiting angle at which the GRZ can theoretically be different is estimated from the following considerations. With the relief height of the figures on the GRZ 2 mm, their width 50 mm, and the gap between them is 12-15 mm, in the photo the white gaps are completely obscured by black numbers at a sliding angle of about 10 °, even for a perfectly flat plate of the GRZ.

Let us consider in more detail the proposed method for determining the vehicle speed from photographs.

Conditions for obtaining photographs. When entering a vehicle into a controlled field, a photo or video camera can take four consecutive shots (see Figure 3) with an interval of, for example, 0.02 sec (50 frames per sec.) Or 0.04 sec (25 frames per sec.), At each snapshot displays the time of its execution with an accuracy not worse than 0.0001 sec. This corresponds to an accuracy of measuring the time interval between successive photographs of 0.5% in the first case and 0.25% in the second; or by the extreme photos in the series, respectively 0.2% and 0.1%.

Requirements for photographs: Photos are taken at an angle of 45 ° ± 25 ° from a distance of at least 15 m to the vehicle, an absolutely stationary (controlled by the immovability of the background in all photographs) camera with a resolution of at least 2000 pixels for the length of the vehicle or a recognizable element of its external body structure. For extended trucks, obviously we can talk about its cabin and adjacent elements. The photographs, in principle, can be combined into one stroboscopic image, similar to that shown in Figure 4a.

So, let's take one of 6 possible paired combinations of the vehicle location on the photo and on the road (see the explanatory figures B-D in Figure 3). Let us choose in the photographs of this combination as a reference element of the external structure of the vehicle body, the size of which in the direction of displacement can be known from the dimensional drawing, or measured directly on the car (for example, a sill, side door, rear wheel). Obviously, all 4 points on the terrain, corresponding to the selected points in two combined photographs, lie (see Figures 4, 5) on one straight line T'-T '' in Figure 5-b or tt (T-M) in Figure 4 - trajectories of the selected points of the body element. It is also obvious that this line and the optical center of the lens of a photo or video camera (denoted by the letter O) lie in the same plane. This plane, crossing the plane of the camera matrix, forms a straight line pp, which is a projection of the T'-T '' (tt) line. In the same plane tt-O-pp there is a line O'-O '' passing through the optical center O of the lens, and perpendicular to the line pp on the matrix. We consider this line as the optical axis of the lens, although in the general case it is not. This "optical" axis O'-O '' intersects the trajectory line at an angle α ₀ . Let's make a reservation right away that this angle is within the limits of α ₀ = π / 4 ± π / 8. The angle between the axis O'-O '' and the line pp is denoted as ϕ ₀ (Fig. 5-b, in Fig. 4 the designation is omitted). As mentioned above, ϕ ₀ = π / 2 and in the standard case cannot deviate from this value more than beyond the angular aperture of the camera.

Let us designate the position of the extreme (limiting) points of the body element on the trajectory of its movement on the ground by points A _n , the corresponding points on the matrix (photographic film) of the camera as a _n . The points of intersection of the line O'-O '' with the lines tt and pp will be taken as the origin of coordinates for counting the position, respectively, of the points A _n on the trajectory and the corresponding points a _n on the matrix. The angle between the line O'-O '' and the line A _n -a _{n is} denoted as β _n- Similarly to the accepted designations for the angles formed by the line O'-O '', the angle at the vertex (point) A _{n is} denoted as Ф _n , and the angle at the vertex (point on the matrix) a _n , as ϕ _n. We also denote the segments (and their lengths) of the O'-O '' line from the optical center O to the trajectory as F ₀ , and to the matrix as ƒ. Similarly, the segment (side of the triangle) OA _n , as F _n , segment Oa _n , as ƒ _n . The angle between the lines tt and pp will be denoted as θ. Then, for the angles shown in Figure 4 of the triangles, the relations are obvious

Ф _n + ϕ _n = π-θ, Ф _n = α ₀ ⁺ β _n ϕ _n = ϕ-α _n and previously indicated α ₀ = π / 4 ± π / 8 ϕ ₀ = π / 2 ± π / 20.

In the accepted notation, the coordinates of the point A _n on the trajectory through the angle α _{n can} be written as

To simplify the right-hand side of the expression, we make the substitution A ^* _n = A _n / F ₀ sin (α ₀ ) and get

We express the cotangent of the sum of the angles in terms of the tangents of the partial angles, as

ctg (α ₀ + β _n ) = (1-tgα ₀ ⋅tgβ _n ) / (tgα ₀ + tgβ _n )

(G.B.Dwight, Tables of integrals and other mathematical formulas, M. Nauka, 1977, p. 74, formula 405.03 ctg (A-B) = (1 + tgA tgB) / (tgA-tgB)

Similarly, in the accepted notation, the coordinates of the projection of the point An on the matrix through the angle β _{n can} be written as

We replace in (3) the value of the angle β _n obtained from (4) as

We also replace the coordinates of the vanishing point S a _s = ƒ ₀ ⋅tg (β _s ), where the angle between the reference axis O''O 'and the direction to the vanishing point α _s = -α ₀

Substituting the expressions for the angles (5) and (5 ') into expression (3) and taking into account that the tangent of the arctangent of the argument is equal to the argument, we obtain the expression:

А ^* _n = -ƒ ₀ / a _s - [1+ (a _s / ƒ ₀ ) (a _n / ƒ ₀ )] / [- (a _s / ƒ ₀ ) + (a _n / ƒ ₀ )], which after simplification it looks like this (the figure shows the situation when a _s <0, i.e. it is negative)

We make a check of this obtained expression for linking the coordinates on the trajectory with the coordinates in the photograph.

1. By definition, when approaching the vanishing point on the projection (photograph) (ie a _n → a _s - the coordinate of the corresponding point on the real trajectory should go to infinity. A ^* _n → ∞, because [a _s -a _n ] → 0, - the denominator in expression (6) tends to zero, as it is in expression (6).

2. When the projection coordinate tends to the matrix a _n → + ∞ (see Figure 4), the coordinate on the real trajectory comes to point E. We transform expression (6) by dividing the numerator and denominator by a _n . As a _n → + ∞, all terms containing a _n in the denominator tend to zero.

A ^* _n = -ƒ / a _s - [(ƒ / a _n ) + (a _s / ƒ)] / [(a _s / a _n ) - 1] = - ƒ / a _s -a _s / ƒ. This value corresponds to a point on the real trajectory, into which the line drawn through the optical center falls parallel to the projection line of the trajectory on the matrix, which therefore corresponds to the infinite value of the coordinate of the point on the matrix (photograph). The remaining terms ƒ ₀ / a _s and a _s / ƒ ₀ correspond to the segments O''D and DE and their sum is the coordinate of the point E.

3. The situation when the reference axis (optical axis) is perpendicular to the trajectory corresponds to the fact that a _s → -∞. In this case, proceeding in the same way as in item 2, we obtain the only term a _n / ƒ on the right-hand side. This corresponds to a simple proportion A _n / F ₀ ⋅sin (α ₀ ) = a _n / ƒ for similar in this case triangles A _n -A ₀ -O and a _n -a ₀ -O.

Now, using expression (6), we can obtain an expression for the lengths M of the trajectory segments (threshold, wheelbase or other elements of the car's appearance) in two positions (far and near) through their coordinates on the projection. Once again, for definiteness, we point out that these segments correspond, for example, to the threshold of the car or the segment between the axles of the wheels (wheelbase).

Let us denote in the far (English far) position this value M = M _ƒ = A ₂ -A ₁ , and in the near (English close) position M = M _c = A ₄ -A ₃

In both expressions, the same term in square brackets is the constant W = D ₀ (ƒ ₀ ² + a _s ² ) / ƒ ₀ - the scale parameter characterizing the central projection. In curly brackets, we have a fraction, in the numerator of which are the values m _ƒ = ( a ₂ - a ₁ ) and m _c = ( a ₄ - a ₃ ), which are the lengths of the named segments on the projection (photograph). The denominator of this fraction contains values of the type d ₁ - ( a _s - a ₁ ), d ₂ = ( a _s - a ₂ ), d ₃ = ( a ₅ - a _h ), d ₄ = ( a _s - a ₄ ) , in the sense of being the distance in the photograph between the vanishing point and the corresponding ends of the named segments (threshold, wheelbase) in the corresponding positions of the car (near or far). Thus, the problem with the inaccuracy of determining the origin and the procedure for finding the origin is completely eliminated. (There remains a problem associated with the inaccuracy of determining the vanishing point S of the projections of the trajectories of the points of the car in the photograph, which is solved by varying the values of d _1,2,3,4 until the same values of the calculation results are obtained by the four equivalent formulas below.)

In the new notation, expressions (7) and (7 ') will be written as follows:

The scaling parameter W of these expressions is represented in two ways:

Each of these expressions for the scale parameter can be substituted into expressions for moving S, which we can write in a similar way to M. Those. we can express the real displacements of the beginning and end of the segment associated with the threshold of the car or its wheelbase, as S = S _b = A ₃ -A ₁ ; S = S _e = A ₄ -A ₂ (subscripts around S: b - begin; e - end). Recall that. Having obtained a numerical expression of the value of S through the known M and dividing it by the time interval between the moments of obtaining photographs of the near and far positions of the car (or a combining stroboscopic photograph), we get the speed of the car. From expression (6) the connection between the coordinates of points on the real trajectory and the coordinates of points on its projection, in a similar way to expressions (7), we obtain:

where s _b = ( a ₃ - a ₁ ) and s _e = ( a ₄ - a ₂ ) are the displacements, respectively, of the beginning and end of the segment (threshold, wheelbase, etc.) measured in the photograph (in mm, pixels or other conventional units)

When substituting in (10) and (10 ') the expressions for W from (9) and (9' '), we obtain four equivalent expressions connecting S with M through a scale factor formed from the quantities measured in the photograph:

As a confirmation of the correctness of these expressions, we note that if in the photographs the end of the segment in the near position coincides with its beginning in the far position (i.e. points А ₂ , A ₃ coincide, and the distances d ₂ = d ₃ ), then s _b = m _ƒ , as _e = m _c , and it turns out S = M.

The speed of movement of the vehicle (TS) in the time interval between the moments of taking photos is defined as the ratio of the movement of the vehicle in units of length of any selected element of the vehicle body, which has an extension along its trajectory and the difference between the named points in time. The true speed of the vehicle can then be determined by multiplying the obtained speed value in units of the length of the body element by its metric length obtained from the outline drawing or by measuring it directly. Obviously, for a couple of photographs there is a large set of vehicle body elements, in which the most preferable is the choice of an element, the length of which along its trajectory is equal to or close to the value of vehicle displacement. An important point is in the case of a mechanical shutter or camera shutter, taking into account the non-simultaneous illumination of the matrix cells, due to the peculiarities of the movement of the shutter edge.

и подставляем его в выражения (9) и (9'). Получаем систему двух уравнений с двумя неизвестными, из которой определяем расстояние от оптического центра объектива фотокамеры О до траектории tt автомобиля D₀ и условное фокусное расстояние ƒ₀ объектива, которое определяется масштабом (размером) фотографии, используемой для измерений расстояний. Дополнительно к этому мы можем определить в каждой зафиксированной позиции автомобиля угол a _n между осью автомобиля и оптической осью объектива фотокамеры по отношению длины b и ширины а автомобиля или его элементов в той же протяженности и их реальных размеров А и В (tgα_n=(a/b)/(A/B)) либо по отношению максимального D и минимального d диаметров искаженных изображений каждого колеса. (sin α_n=d/D).To build a more complete picture of the relative position of a moving car and a camera, in principle, it is possible to determine the values of D ₀ (D ₀ = F ₀ ⋅sin (α ₀ )) and ƒ, after which it is possible to determine all the angles and sides of triangles in the plane of construction. To determine D ₀ and ƒ we take expressions for

and substitute it into expressions (9) and (9 '). We obtain a system of two equations with two unknowns, from which we determine the distance from the optical center of the camera lens O to the trajectory tt of the car D ₀ and the conditional focal length ƒ _{0 of the} lens, which is determined by the scale (size) of the photograph used for measuring distances. In addition to this, we can determine in each fixed position of the car the angle a _n between the axis of the car and the optical axis of the camera lens in relation to the length b and width a of the car or its elements in the same length and their real dimensions A and B (tgα _n = ( a / b) / (A / B)) or the ratio of the maximum D and minimum d diameters of the distorted images of each wheel. (sin α _n = d / D).

As noted above, in the case of a sufficiently small angular aperture of the camera, linear estimates are admissible, although the coordinate of a point on the selected trajectory of a car body element is nonlinearly related to the coordinate of its projection on the camera matrix. However, for each trajectory there is a set of constants F ₀ , ƒ ₀ , α ₀ , which must be determined when calibrating the device. This is the reason for the insufficient measurement accuracy and causes the complexity of the process of measuring the vehicle speed by any method, which in turn causes the complexity of the device and its cost.

The complexity of the device and the processes used for its operation is also the reason that these processes are difficult to understand by persons brought to responsibility, and cannot be monitored by them for the observance of their rights. The claimed method is simple to understand, does not contain hidden processes that affect the measurement result, and this result can be easily controlled by any person using the proposed method.

All known methods of measuring the vehicle speed are based on measuring the vehicle's position at certain points in time in the coordinate system of the device (observer). In this case, various types of invariants are used, which provide a comparison of measured values with standards. This is done either by measuring the time points of the vehicle passing certain boundaries (complexes Avtodoria, Azimuth, invariant - distance traveled), or by determining the speed and distance to the vehicle using location methods (complexes Kordon, KRIS, invariant - speed light, orientation), or by determining the distance to the state. registration plate used as an invariant, placed on the vehicle (Complex "Azimut", analogous invention).

The undoubted advantage of these devices is the use of a universal invariant located on the vehicle itself. At the same time, their significant drawback, which reduces the measurement accuracy, is the use as an invariant of the GRZ plate, which has not significant dimensions in general, and negligible dimensions in the direction of movement of the vehicle, which does not allow using it as a direct measurement standard. In the claimed invention, it is proposed to use body elements as a standard of movement, having an extension in the direction of movement of the vehicle and the dimensions of which are known either from the dimensional drawings, or can be measured directly on the vehicle, or are standardized (wheels, GRZ). Wheels, disks, state. The registration plate in the claimed invention is used as a reference for the preliminary selection of offenders and photo-fixation of persons suspected of committing an offense of the vehicle. Further clarification, which guarantees the observance of the principle of the presumption of innocence, is made on the basis of an even simpler method justified below, using the dimensions of the vehicle body elements.

Let us denote, as before, the length (length) of the vehicle body element, known from the dimensional drawing or direct measurement, as M *, or as M * ₁ and M * _2, its dimensions in the road coordinate system corresponding to the first and second photographs (see. Figure 5, 3d). It's obvious that

and M * ₁ = M * 2 = M * / 2.

Similarly, we denote the displacements of the corresponding points A ₁ and A ₂ as

and S * ₁ = S * ₂ = S * / 2.

Similar values on the camera matrix will be:

m ₁ = a ₁ - a ₂ , a m ₂ = a ₄ - a ₃ , s ₁ = a ₃ - a ₁ and s ₂ = a ₄ - a ₂ . Moreover, m ₁ ≠ m ₂ , as ₁ ≠ s ₂ and it is obvious that none of the values m ₁ , m ₂ , under strict consideration, can be applied as a standard for measuring the values s ₁ and s ₂ .

However, if we construct (compose) the values S * = S * ₁ + S * ₂ and s = s ₁ + s ₂ from the displacements, M * = M * ₁ + M * ₂ and m = m ₁ + m ₂ from the lengths of the element body of the vehicle, then the values M * and m can be used as standards of high accuracy.

Summing up expressions like (11) and highlighting the values of M *, m, S * and s in the resulting expression, we obtain that the ratio S * / M * practically does not differ from the ratio s / m.

It should be emphasized that in the case, for any S and M, if S = M, then the photo will also show s ₁ = m ₁ ≠ m ₂ = s ₂ , but s ₁ + s ₂ = s = m = m ₁ + m ₂ . Those. the correction factor between S * / M * and s / m is exactly 1.

This means that the s / m ratio obtained from the photographs in the matrix coordinate system is exactly equal to the S / M ratio in the road coordinate system. Determination of this ratio is not associated with the use of certified measuring instruments and can be performed both from a photograph of sufficient size using a measuring ruler and by counting pixels that fit on the corresponding sections of the trajectory of a vehicle body element. With a body element length of about 2000 mm and a photo resolution of 1 pixel per 1 mm for the interval between photos (frames in a video stream) 0.04 sec (0.02 sec), we obtain an error in measuring the speed (in the absence of an error in measuring the time interval) of about 25 mm / sec (50 mm / s) or 0.09 km / h (0.18 km / h). Assuming that the measurement of the time interval and the indication in the photograph of the moment of its receipt with an accuracy of 10 ^-4 sec, we have an additional error in the value of the speed of about 0.25% (0.5%) of the measured value of the speed. This error must be added with the error associated with measuring the displacement (according to the well-known rule [14, 15] of squares of errors) and for a speed of 100 km / h it will be 0.27 km / h (0.53 km / h).

To ensure compliance with the presumption of innocence in the measurement, a confidence interval of ~ 10 root-mean-square errors is required, i.e. 2.7 km / h (5.3 km / h). Considering that the driver can control the vehicle speed with an accuracy of about 5 km / h, the ratio of these error values seems to be very reasonable. At the same time, it can be seen that for shooting 25 frames / sec, the camera resolution can be reduced by 2 times. Considering that the angle α ₀ between the trajectory and the optical axis of the camera is about 45 °, the resolution should be increased 1.5 times. Those. when the size of the picture being shot is 6 m long, 1 pixel must correspond to 1.5 mm or a matrix of 2500 × 4000 pixels (10 Mpc) is required. The cameras of modern smartphones have such a resolution. This means that a smartphone may well cope with the task of controlling the speed limit, transferring the costs of producing the corresponding equipment to a completely different category - almost a thousand times cheaper.

The method does not require automatic execution of measurements without the participation of an operator in order to comply with the principle of the presumption of innocence. Subject to accurate measurement and indication of the time intervals between the moments of obtaining photographs (frames), this method of determining the vehicle speed is understandable to everyone and its result can be independently checked by any person with secondary education. Automation is necessary only in order to facilitate the work on the selection of episodes of offenses and to prevent corruption actions of the service personnel, but not in any way in order to justify the violation of the principle of the presumption of innocence. In the automatic version, the processing of cascades of photographs occurs in two stages. At the first stage, the selection of vehicles suspected of violating the speed limit takes place, based on the data obtained by existing automatic control means, or on the basis of determining the relative displacement of the universal elements of the appearance of any vehicle, which are wheels, wheel disks (of several standard sizes, the displacement is determined in units of diameter and the direction of the trajectory with respect to the optical axis of the camera), the plate of the GRZ (the angular velocity is determined by the angle of rotation of the GRZ and the distance in units of the size of its length).

Thus, according to claim 1 of the claims, we conclude that according to the "Universal method of photo-fixing traffic violations" (see figure 3) traffic violations are recorded by 4 consecutive photographs with a resolution of ~ 1 mm of roadway space per 1 pixel of the photo or video matrix - cameras performed at regular intervals depending on the nature of the offense being recorded (by the location of the vehicle, by the vehicle maneuver, by the vehicle speed). In the case of detecting violations of the speed limit, photographs are taken by a stationary camera at short intervals, preferably at 45 ° to the direction of movement of the vehicle. The primary reason for entering the episode into the database of suspects in committing an offense under Art. 12.9 of the Code of Administrative Offenses of the Russian Federation is the displacement of the vehicle position in the frame in units of the vehicle wheel diameter more than the specified displacement, in accordance with the speed limits in the controlled area. Each of the four photographs in the cascade indicates: with an accuracy of 1-5 m, the geographical coordinates of the camera that took the photograph; one of 8 lens directions (north, northeast, east, southeast, south, southwest, west, northwest); exact relative time of execution of photographs with an accuracy of 0.0001 sec; date; speed limitation in the controlled area. In the future, the type of vehicle is determined by the type in the photograph and the registration data (according to the license plate, VIN), the dimensions of the elements of its body are determined based on the type of vehicle, which are taken as standards. For each paired combination of photos from 4 in the cascade and the data of the vehicle outline drawing, the vehicle displacement is calculated in units of the length of the selected body element (it is preferable to select an element whose length is close to or equal to the vehicle displacement in the frame). For this, the length of the body element is measured between points on its trajectory on each of the photographs in a pair and summed up. Thus, we get the value m = m ₁ + m ₂ . From the photographs we measure the displacement of the beginning of the element s ₁ and its end s ₂ . We also sum up and get the value s = s ₁ + s ₂ . Taking the ratio s / m, we obtain the displacement of the vehicle in units of the length of its body element. Multiplying this ratio by the length of the element, we obtain the displacement of the vehicle for the time between the photos S, Calculating the time interval between the photos, as Δt = t ₂ -t _{1, we} calculate the speed of the vehicle as V = S / Δt. And so for each of the 6 pairs of photos. We calculate the arithmetic mean, determine the deviation of each of the 6 values and from them calculate the root-mean-square error ΔV. To comply with the principle of the presumption of innocence, this value must be less than 1/10 of the difference between the arithmetic mean of the speed and the value of the speed limit, i.e. ΔV <(VV _limit ) / 10 or, in other words, the value of the speed limit should not fall within the interval 10 times the root mean square error about the average value of the measured speed. Otherwise, there is a possibility that the actual value of the measured speed is less than the speed limit and the principle of the presumption of innocence is violated. The values of the arithmetic mean of the speeds determined from the photographs and the mean square error from their totality, together with the general calculation formulas, are indicated in the decision on bringing to administrative responsibility. Within the framework of the current legislation, it seems permissible to refer the verification of the values of the quantities to the interested party - the person involved. The stated algorithm is simpler and more accurate than the calculations and photo processing procedures used in the methods and complexes mentioned in the "Prior Art" section. The segment of the trajectory of the body element on which it lies, determined by two points, always lies in the same plane with the projection center (optical center of the lens) and the projection of this segment onto the camera matrix, which greatly simplifies the calculations. This does not raise doubts about the possibility of implementing the invention in this part.

The second part of the invention is the idea that information about the vehicle speed can also be obtained not only knowing its overall characteristics and being in the associated coordinate system, but also using the information circulating in its systems and devices. In this case, simply by fixing in a specific place the information received from its own sensors. At present, in modern cars, information about speed, mileage, engine temperature and engine speed, fuel level, and time in digital form is displayed on the dashboard and on-board computer from sensors. WiFi and Bluetooth data transmission technologies are widespread. It is not difficult to create a specialized standard for the presentation of this data for traffic monitoring purposes. From the point of view of pollution of the city by radio wave noise from numerous cars, it is preferable to use directional infrared radiation for data transmission from a car. All that is needed for this: manufacturers of cars and on-board computers should start producing dashboards and on-board computers equipped with modules that transmit the values of a number of parameters of vehicle sensors via standard communication channels, for example, Wi-Fi, and manufacturers of devices that control vehicle movement should provide their devices are modules that receive these signals from the monitoring area. Obviously, the signal must be in a standardized form and its package must contain information about the VIN of the vehicle (as well as the driver's license number of the driver who controls it). In the invention-analogue RU 2557667 C2 (see source [7]) it was proposed to equip cars with passive and active RFID tags for their identification by radio signal. The present invention proposes to expand the functions of such devices for transmitting data from a vehicle (TS) in terms of the amount of information transmitted (data about the vehicle, its devices, driver) and operating frequencies (microwave, IR, visible light) and replace the measurement of the vehicle speed with an external device receiving information from the sensors of the vehicle itself by means of modules (TM) that are serially installed on the vehicle, transmitting data that the driver of the vehicle sees on the monitoring devices. In this case, the question of the relationship between the measurement error and the principle of the presumption of innocence practically disappears. In addition, control over the vehicle can be carried out by devices comparable in complexity, size and price to inexpensive smartphones. And also their functions can be fully taken over by existing smartphones capable of connecting to the named communication channels. In a more sophisticated design, these broadcasting modules (TM), which are serially installed on the vehicle and their on-board computers, could receive messages related to this vehicle from the traffic control system and from other road users. The inventive method can be applied in all existing complexes of photo and video fixation, by presenting paired photographs and displaying the exact coordinates and exact (up to 0.0001 sec) time of obtaining photographs in the information attached to the photograph.

3.2.4.4. List of figures, drawings and other materials.

Figure 1. An example of the presentation of evidence in the Cordon-2M complex

Figure 2. An example of the presentation of evidence in the Cordon-Temp complex

Figure 3. Scheme of presentation of evidentiary material in the claimed invention.

Figure 4. Scheme of construction and transformation of the coordinate system on the road and the matrix of a digital or analog photo or video camera.

Figure 5. a) Diagram of finding the vanishing point S. b) Diagram of the location of the vehicle (TC) and the device that records the event of the passage of the vehicle

3.2.4.5. Information confirming the possibility of carrying out the invention.

Section 3.2.4.3 specifies the requirements for the resolution of photographs and the accuracy of measuring time intervals. These requirements are met by modern devices (Avtodoria, Azimut, etc.), but these devices do not use all the possibilities inherent in the photographs they receive. Since all measurements can be made within the capabilities of a smartphone, the implementation of the proposed method reduces the cost of the technical result by about 1000 times, increases the accuracy by 10 times and, most importantly, is characterized by absolute observance of the principle of the presumption of innocence and the absence of instrumental processes hidden from the suspect in the offense.

In an extended version of the claimed invention, it is assumed instead of an RFID tag on a vehicle during its production (or when installing an on-board computer on it), for example, to install an IR beacon (LED on the windshield in front of the interior rear-view mirror or other options with the installation of two- four LEDs, high-frequency modulation of the light of running lights, headlights) constantly broadcasting a coded signal containing the VIN of the vehicle and the readings of its sensors (minimum - speedometer readings in km / h, mileage in meters, time in tenths of seconds). (This kind of technology is currently used everywhere, in every home - take, for example, a TV remote control, air conditioner, WiFi in computer networks, Bluetooth. That is, it is inexpensive and massive in the implementation of technology.) These data are entered into the message about the offense. In the future, when using a driver's license with a microchip, for example, as an ignition key, the data on the driver's license can be entered in the report of the offense. In this case, the need for Art. 26.1 of the Code of Administrative Offenses of the Russian Federation, bringing to administrative responsibility not the person who committed the offense, but the owner of the vehicle. In general, the claimed invention, moving away from direct measurement of speed with ever more sophisticated instruments, makes it possible to reduce the cost of the process of obtaining legally significant information about the vehicle's behavior by almost a thousand times and to increase its accuracy and reliability tenfold.

10.7.4.4. Brief Description of Drawings

Figure 1. An example of the presentation of evidence in the Cordon-2M complex. The figure shows two frames with a car made by the Cordon-2M complex. You can clearly see the displacement of the car in the frame. If the information accompanying the frames would indicate the time for each frame with an accuracy of 0.0001 sec, then by measuring the length of the threshold and its displacement along itself with an ordinary ruler with an accuracy of 0.1 mm, the person brought to administrative responsibility would be able to verify the correctness of the grounds for bringing , or prove the opposite.

Figure 2. An example of the presentation of evidence in the Cordon-Temp complex. The figure shows two frames with a car made by the Cordon-Temp complex: at the entrance to the controlled area and at the exit from it. The execution time for each photo is indicated with an accuracy of 0.001 sec. The distance between the control points is indicated, which the involved person, however, cannot measure. If there was a second photo for each episode, the involved person could make sure that the speed values indicated in the information accompanying the frames are correct. The example also shows that it is possible to represent the data on the time interval between photographs with an accuracy of 0.0001 s, since the times indicated in the photo with an accuracy of 0.001 s refer to the time synchronized with the global one.

Figure 3. Scheme of presentation of evidentiary material in the claimed invention. Inset 3-a presents 4 consecutive photographs associated with fixing traffic violations. In the case of a violation of the speed limit, photographs are taken for a short period of time while the vehicle passes the controlled area. In the event of other traffic violations, the size of the controlled area may be larger and cover the entire area of the vehicle's maneuver. Accordingly, the intervals between frames increase. Tab 3-b contains 6 paired combinations of photographs that can be used to determine vehicle displacement. In the 3-c tab, an example of a dimensional drawing of a vehicle is presented, which can be used to measure the speed of a vehicle. The 3D tab shows examples of determining the displacement s and the length of the reference body element in the photographs. In this case, the movement of the car in units of the length of the body element is determined in a simplified form, as S / M = (s ₁ + s ₂ ) / (m ₁ + m ₂ ), the speed as V = S / (t ₂ -t ₁ ), where t ₂ , t ₁ - moments of time for the corresponding positions of the car. The square of the relative error in measuring the speed is defined as the sum of the squares of the relative errors in determining the time and determining the displacement, associated mainly with the resolution of the camera, and not with the accuracy of manufacturing the body element.

Figure 4. a) Scheme of transformation of the coordinate system on the road and the matrix of a digital or analog photo or video camera.

Т-М (or tt) - a line containing a segment of the trajectory of a vehicle body element (TC) taken as a standard of length. Point O '' has a coordinate A ₀ = 0, point T _n - coordinate A _n , the positive direction of the coordinate axis on the trajectory is shown by an arrow. O'-O '' is the optical axis intersecting the trajectory line and passing through the optical center of the lens O (projection center). These two lines lie in a plane intersecting the camera matrix along the pp line and containing the projection a _{n of the} point A _n on the trajectory of the body element. The O'-O '' axis is close to the normal to the plane of the matrix and intersects the trajectory line at an angle of 45 ° ± 25 °. Based on this construction and the similarity of right-angled triangles O'O''M, OO'D, OED, MDM ', OSM' of the triangles shown in the diagram, the relationship of coordinates on the trajectory and coordinates on its photographic projection is calculated. Details in the text of the description of the invention.

Figure 5. b) Scheme of finding the vanishing point S. Similarly, the lines can be drawn through any identical points on the car body, a) The layout of the vehicle (TC) and the device that records the event of the passage of the vehicle The figure schematically shows the overall drawing of the vehicle (TC ) and the position of the vehicle along the trajectory along the T'-T '' line. Two positions characterized by the coordinates of the beginning and end of the body element (here a threshold) A ₁ , A ₂ and A ₃ , A ₄ on the trajectory, a ₁ , a ₂ and a ₃ , a ₄ on the matrix correspond to the positions fixed by a fixed camera in two photographs from each combination of Figure 3. From each photograph, the length of the body element along the trajectory of movement is determined and the average between them is taken (value m). By comparing the position of the beginning and end of the body element, its average displacement s is determined. From the ratio s / m multiplied by the length of the body element and divided by the time interval between the photographs, the vehicle speed is found. On the outline diagram, the vertical and "horizontal" diameters of the wheel are marked with different letters to emphasize that in the photograph these are different values that allow you to determine the angle between the direction of movement and the direction of shooting. Details in the test description.

List of literature describing the prior art of the invention

1. Osipov C.K. (RU), Malinkin (RU) A.Yu., Method for determining the speed and coordinates of vehicles with their subsequent identification and automatic registration of traffic violations and a device for its implementation "(" Strelka "), invention RU 2382416 C2, priority 20 March 2008, publ. 20.02.2010, https://patents.google.com/patent/RU2382416C2/en?oq=RU+2382416+C2

2. Zarubin Yu.L. (RU), Ubozhenko H.B. (RU), Stukalov D.A. (RU), Vovk M.A. (RU), Aistov A.A. (RU), "Method for determining the speed of a vehicle", Invention RU 2442218 C1 (priority 03 December 2010 publ. 02/10/2012), Patent holders: Limited Liability Company "Recognition Technologies" (RU) https://patents.google. com / patent / RU2442218C1 / en? oq = RU + 2442218 + C1

3. Ubozhenko N.V. (RU), Aistov A.A. (RU), Zarubin Yu.L. (RU), Stukalov D.A. (RU), Vovk M.A. (RU) "Method for determining the speed of a vehicle", RF patent RU 2419884 C1 (publ. 05/27/2011), http://www.findpatent.ru/patent/241/2419884.html Patent owners RU 2419884: Limited Liability Company "Recognition Technologies" (RU)

4. Averkiev M.V., Evdokimov L.A., Mezhuev R.P., Komlik I.F., Markin A.A., Bondar D.V., Bodnar V.M., Prigorovsky V.M., Barskiy IV, Useful model "Complex of video recording and measurement of the speed of movement and coordinates of vehicles", priority October 19, 2011 RU 113398 U1, publ. 10.02.2012), https://patents.google.com/patent/RU113398U1/ru

5. Komlik I.F. (RU), Simonchik K.K. (RU), Bondar D.V. (RU), Barsky I.V. (RU) "Method for determining the position and speed of a vehicle on the road and a complex for its implementation", Invention RU 2559418 C2 (priority October 15, 2013), ttps: //patents.google.com/patent/RU2559418C2/en? oq = RU + 2559418C2 +

6. Burlutskiy A.S. (RU), Ilyina Yu.V. (RU) Method for determining the average speed of a vehicle on a road section containing curved segments // Invention RU 2589802, http://www.findpatent.ru/patent/258/2589802.html Avtodoria "(RU)

7. Mustafin Ramil Gamilovich // Method for determining the speed of a vehicle // Invention RU 2557667 C2, priority from December 12, 2013 // http://www.findpatent.ru/patent/255/2557667.html state budgetary educational institution of higher professional education "Kazan State Power Engineering University" (FGBOU VPO "KSPEU") (RU)

8. Klukas Alan (GB), SMITH Brian David Vincent (GB), Method and system for measuring vehicle speed // Invention RU 2543947 // http://www.findpatent.ru/patent/254/2543947.html Patent holders RU 2543947: ZM INNOVATIVE PROPERTIES COMPANY (US)

9. Ubozhenko N.V. (RU), Zarubin Yu.L. (RU), Vovk M.A. (RU) Method for determining the distance from the speed meter camera to the vehicle (options) // Invention RU 2470376 // http://www.findpatent.ru/patent/247/2470376.html Recognition Technologies "(RU)

10. Matsur I.Yu. (RU), Method for detecting, identifying and determining the speed of a vehicle and a device for its implementation _ // Invention 2422909 // http://www.findpatent.ru/patent/242/2422909.html. Holders of the patent RU 2422909: Matsur Igor Yurievich (RU)

11. Kislitsyn V.O. (RU), Artemov N.V. (RU), V.A. Kalinin (RU), Shubarev B.A. (RU), Dikarev V.I. (RU), Method for detecting, identifying and determining the speed of a vehicle. // Invention RU 2571148, http://www.findpatent.ru/patent/257/2571148.html Patent holders RU 2571148: Open Joint Stock Company Avangard (RU)

12. Novikov V.P. (RU), Karyukhin D.A. (RU), Serga E.V. (RU), Anisimov V.I. (RU), Shchukin V.V. (RU), Shishkin K.E. (RU), Mamakov P.V. (RU), Method for determining the speed of a vehicle // Invention RU 2579645 http://www.findpatent.ru/patent/257/2579645.html, Owners of patent RU 2579645: Limited Liability Company "New Technologies" (RU),

13. Petrichkovich Ya.Ya. (RU), Khamukhin A.V. (RU), Method and system for determining the speed of a vehicle // Invention RU 2592712, http://www.findpatent.ru/patent/259/2592712.html

14. Agekyan Tateos Artemyevich, Fundamentals of the theory of errors for astronomers and physicists, M. 1968, 148 pp., Publishing house "Nauka", Main edition of physical and mathematical literature.

15. Seidel Alexander Natanovich. Errors in measuring physical quantities / A.N. Seidel; Physics and technology in-t them. A.F. Ioffe of the USSR Academy of Sciences. - L .: Nauka, 1974 .-- 108 p.

16. Description of the measuring complex "Azimut" (Russia) http://tbdd.ru/node/78

17. Description of the measuring complex "Cordon M2" (Russia) http://www.simicon.ru/rus/product/gun/cordon_m2.html

18. Description of the measuring complex "Cordon Temp" (Russia) http://www.simicon.ru/rus/download/cordon_m/cordon_temp_brochur_2017_gost.pdf

19. Description of the measuring complex "MultaRadar CD moving" (Germany) http://www.unikart.su/multaradar-cd-moving.html

20. Description of the measuring complex "KRIS" (Russia) http://www.simicon.spb.ru/rus/download/KRIS-C.pdf

Claims (3)

1. A method of photographing traffic violations, which consists in the fact that the speed of the vehicle (vehicle), its movement and the time spent on movement are controlled according to the indications of its own sensors received from the broadcasting module placed on the vehicle, and the change in its position relative to other vehicles and road elements are determined by its photographs, and the identification of the vehicle is carried out according to the image of its state registration plate (GRZ) and / or by the signal received from the broadcasting module placed on the vehicle, characterized in that, to present the results of photographic recording, they are carried out by means of six pairs of photographs taken at precisely known points in time necessary to ensure the reliability of the evidence; pairs of photographs formed by a combination of four real photographs are used to determine the speed of the vehicle in the time interval between them, the speed of the vehicle is determined by the values of the displacements of the beginning and end of the body elements, which have a length along the trajectory of movement, in units of the length of these elements in each photograph in a pair; the determination of the exact amount of movement of the vehicle on the ground is carried out after its identification, clarification of its overall dimensions and the dimensions of the body elements selected as reference; the amount S of the car's displacement is determined from each pairwise combination of four photographs as S = M (s _b + s _e ) / (m _c + m _f ) and, more precisely, as S = M (s _b / m _f )) / (d ₂ / d ₃ ) = M (s _e / m _f )) / (d ₁ / d ₄ ) = M (s _b / m _c )) / (d ₄ / d ₁ ) = M (s _e / m _c ) ) / (d ₃ / d ₂ ) by four similar formulas, where M is the real length of any selected body element in the direction of movement, s _b and s _e are the displacement of the beginning and end of the body element, measured in arbitrary units in the photographs, m _c and m _f - the lengths of the selected body element in the near and far positions, measured in the same units in the photographs; d _1,2,3,4 - distances from the beginning and end of the body element in two positions to the vanishing point of the trajectories of the body elements, measured in the same arbitrary units in the photographs; the reliability of the violation of the speed limit is established from a comparison of 1/10 of the speed excess and the root-mean-square error, determined from the six obtained speed values; preliminary selection of vehicles suspected of violating the speed limit is performed using wheels as reference elements of the body, reducing their many sizes to three standard sizes; the vehicle must be equipped with a broadcasting module that transmits a standardized signal containing at least the VIN of the vehicle (VIN of the vehicle) and the readings of its speed sensor, and the traffic control complexes have devices that receive these signals. 2. The method according to claim 1, characterized in that, in addition to the VIN of the vehicle and the readings of its speed sensor, the translating module includes the mileage and the machine time in the signal it generates; information about the license plate of the vehicle and about the person managing it; to determine the movement and speed of the vehicle, information from its sensors is used, transmitted by the transmitting module of the vehicle. 3. The method according to claim 2, characterized in that the transmitting module of the vehicle is configured to receive information from the traffic management system and from other vehicles.

Images