WO2009087536A2 - Method and system for detecting moving objects - Google Patents

Method and system for detecting moving objects Download PDF

Info

Publication number
WO2009087536A2
WO2009087536A2 PCT/IB2008/055566 IB2008055566W WO2009087536A2 WO 2009087536 A2 WO2009087536 A2 WO 2009087536A2 IB 2008055566 W IB2008055566 W IB 2008055566W WO 2009087536 A2 WO2009087536 A2 WO 2009087536A2
Authority
WO
WIPO (PCT)
Prior art keywords
vehicles
passage
detecting
route according
homologous
Prior art date
Application number
PCT/IB2008/055566
Other languages
Spanish (es)
French (fr)
Other versions
WO2009087536A3 (en
Inventor
Ferran Lisa Mingo
Original Assignee
Imagsa Technologies S.A.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Imagsa Technologies S.A. filed Critical Imagsa Technologies S.A.
Publication of WO2009087536A2 publication Critical patent/WO2009087536A2/en
Publication of WO2009087536A3 publication Critical patent/WO2009087536A3/en

Links

Classifications

    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/04Detecting movement of traffic to be counted or controlled using optical or ultrasonic detectors
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00Image analysis
    • G06T7/70Determining position or orientation of objects or cameras
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06VIMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00Scenes; Scene-specific elements
    • G06V20/50Context or environment of the image
    • G06V20/52Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G06V20/54Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/0104Measuring and analyzing of parameters relative to traffic conditions
    • GPHYSICS
    • G08SIGNALLING
    • G08GTRAFFIC CONTROL SYSTEMS
    • G08G1/00Traffic control systems for road vehicles
    • G08G1/01Detecting movement of traffic to be counted or controlled
    • G08G1/017Detecting movement of traffic to be counted or controlled identifying vehicles
    • G08G1/0175Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules

Definitions

  • the main object of the present invention is a method and a device that detect the passage of vehicles through a road.
  • Another object of the present invention is a method and a device that also determine the length, height and speed of the vehicles.
  • Advanced traffic control systems use artificial vision systems that study different traffic conditions in a scene by analyzing digital images.
  • the capture of these images generally occurs with two systems, one that detects the passage of a vehicle and another that captures at least one digital image of the scene.
  • Non-intrusive systems based on ultrasound and microwave have also been proposed. This type of systems usually presents problems related to lack of precision or excessive dependence on the environmental conditions on the road.
  • optical detectors have been created that interpret the changes in the scene as indicators of the passage of a vehicle, but their use is very limited because they are excessively sensitive to changes in the ambient lighting (caused by the passage of clouds, by the lights and shadows produced by other cars, by elements external to the road, etc.), and because they are very sensitive to vibrations, which makes it impossible to install them on traffic lights, lampposts, bridges or porches.
  • the present invention relates to a method and a device that solve the above problem in a single device that detects the passage of the vehicle and takes high resolution images for later identification.
  • the detection of the vehicle solves the deficiencies of other systems of detection by artificial vision, since it uses an algorithm that analyzes the height of the objects, which allows it to discard the brightness and shadows on the road that would confuse other methods, and performs a filtering of the images.
  • the method is also totally insensitive to vibrations, since it works with static images ("snapshots") instead of video, which facilitates its installation in any element available on public roads (street lamps, traffic lights, bridges, porches, etc.). ).
  • Another important advantage of the method is that it has been designed to work at a very high speed, which makes it possible to detect vehicles regardless of its speed, and to be easily integrated into a specific application integrated circuit (ASIC) or other similar device, which allows its large-scale manufacturing and very low cost.
  • ASIC application integrated circuit
  • the method used for the detection of vehicles on the road is based on the estimation of their height by means of a system that contains two or more means of image acquisition. Comparing the acquired images, discrepancies between homologous pixels of the homologous images are detected, deducing the existence of objects with non-zero height.
  • road refers to any type of road, road, road, etc. through which vehicles whose characteristics you want to know pass.
  • vehicle is not limited only to cars, since the present procedure and system are capable of detecting the passage of any non-zero height object, such as motorcycles, animals, people on foot, etc.
  • a "point” corresponds to a part of a vehicle, road, etc. in the real world, as opposed to the points of which the images are composed, which we will refer to as pixels.
  • the "value of a pixel” refers to a brightness value of said pixel.
  • a method of detecting the passage of vehicles by a route comprising the following two phases is provided:
  • This first phase includes the following operations:
  • two "homologous images” are two images of the same area of the track from two horizontally separated points of view a distance corresponding to the distance that separates the image acquisition means.
  • the image acquisition means obtain images of the pathway upon receiving the radiation in it reflected, for example the visible natural radiation.
  • b) Generate a model of the path by analyzing the correspondences between the pixels of the simultaneous homologous images acquired in the previous step by some technique of correspondence analysis between images, for example using fixed reference points of the path.
  • This model is a correspondence function ⁇ : R 2 -> R 2 that gives us for each pixel of the image of one of the cameras, for example the left in case they are horizontally separated, the position of its homologous pixel in Ia image of the other camera, that is to say on the right in case they are separated horizontally.
  • the Euclidean distance between the position (X 11 V 1 ) of a pixel in the left image and its counterpart (Xd, and d) in the right image gives us by trigonometric methods the distance between the means of image acquisition and the path in that Place of the scene.
  • phase 2 The procedure described in phase 2 is carried out continuously, simultaneous simultaneous images are acquired consecutively at very high speed, and deducting from them the number of vehicles that travel along the road during a certain period of time.
  • the operations of the phase 2 are carried out a minimum of K times per second, K being a parameter that depends, according to the following formula, on the maximum speed at which the vehicles on the road (Vmax), its minimum length (Lmm) and the longitudinal distance of the road that is covered by the analysis region (L r ⁇ g ).
  • phase 2 The accuracy of vehicle speed and length estimates is greater the higher the frequency of image acquisition.
  • the operations of phase 2 are performed 50 or more times per second. Thanks to the repetitive implementation of the previous procedure, we unequivocally detect the entry and exit of an object in the analysis region, which allows us to capture a complete image of the vehicle at the most appropriate time for our application. For example, we can take a picture of the back of the vehicle to analyze its registration.
  • the method further comprises the operation of emitting in the direction of the pathway, by means of a radiation emitting means, a radiation suitable for the operation of the image acquisition means.
  • the radiation emitting medium could emit visible, infrared, ultrasound, radar, microwave, etc.
  • the adaptation of the power of the emitted radiation and the sensitivity of the sensors is carried out dynamically in each capture by means of a feedback scheme.
  • Said scheme takes as an initial data an estimate of the ambient luminosity calculated based on the evolution of the average of the pixel values in the analysis region, in the moments in which there is no vehicle on the track.
  • the method of detecting the passage of vehicles through a track also includes the operation of determining the height of the detected vehicle. This function is evident from the operation described above, since it is possible to determine the height of a point by basic trigonometry.
  • the method of detecting the passage of vehicles through a road comprises also the operation of determining the length of the detected vehicle. For this, in each pair of acquired images, several regions of simultaneous homologous analysis are analyzed in the form of a band perpendicular to the pathway. From the information related to the height of the homologous pixels located in each of the pairs of band-shaped analysis regions, the vehicle length is deduced, since the longitudinal distance between the analysis regions is known.
  • the method of detecting the passage of vehicles through a track also includes the operation of determining the speed of the detected vehicle. Since the images are taken sequentially at a known frequency, it can be determined how long the vehicle remains within one of the analysis regions. And since, as explained above, we can know the length of the vehicle, the vehicle speed can be deduced.
  • the method of detecting the passage of vehicles through a road also includes the operations of taking the date and time of the detection, measuring the geographical location of the detection equipment, and encrypting and communicating the results obtained through a means of communication.
  • the number of vehicles that have traveled on a road and their average speed could be sent to a traffic control center, for example, daily, along with the identification of the Telec point at which the measurement was made and the date and hour.
  • a traffic control center for example, daily
  • Several similar equipment could also be interconnected to manage the detection of vehicles on several roads of the same road (for example on multi-lane roads), detecting the situations in which a vehicle is changing lanes to avoid two vehicles being counted in place of one. For security reasons, all shipments are made in encrypted form.
  • a system for detecting the passage of vehicles through a track is described, characterized in that it comprises the following devices:
  • the image acquisition means can be digital cameras, digital video cameras, infrared cameras, ultrasonic detectors, radar detectors, etc.
  • the distance between the left and right infrared cameras (3, 4) is within the range of 20 cm to 100 cm.
  • a processing means connected to the image acquisition means (3, 4), which subtracts, for each pixel of one of the simultaneous homologous images, the value of the homologous pixel in the other simultaneous homologous image and determines, depending on of the number of discrepancies, if a detection occurs.
  • the processing means has the ability to encrypt the data for sending to, for example, a traffic control center.
  • the processing means could be any device capable of carrying out the calculations necessary to determine the presence of vehicles on the road, although according to preferred embodiments of the invention it could be a microcontroller, a computer, an ASIC, a DSP, an FPGA, etc.
  • a radiation emitting medium can be added that emits radiation in the direction of the pathway, thus avoiding dependence on the natural visible light.
  • the radiation could be of any type, as long as it is reflected in the vehicles that travel along the road and can be detected by the means of acquiring images.
  • the radiation emitting means can emit visible, infrared, ultrasound, radar, microwave, etc. radiation. However, visible radiation is not advisable, as it could disturb the drivers of vehicles that travel along the road.
  • a means can be added to determine the date and time at which the detection of the vehicle occurs and a means to identify the geographical position of the equipment, which could be, for example, a GPS positioning system
  • the system may comprise a communication means for communicating the results obtained, for example to a traffic control center.
  • the communication medium can be of any type capable of transmitting the results obtained although, according to preferred embodiments, it could use the Internet, Ethernet, USB, RS232, RS485, Bluetooth, IEEE802.Hb / g, Zigbee, radio, telephone line, UMTS, GSM, infrared, etc.
  • Figure 1. Shows a scheme of the vehicle detection system according to the present invention.
  • Figure 2. Shows an example of the type of images acquired by one of the cameras of the example of the present invention.
  • Figures 3 and 4.- Shows a pair of simultaneous homologous images captured by a pair of cameras of the vehicle detection system according to the present invention.
  • Figures 5 and 6.- Shows the pair of simultaneous homologous images of Figure 3, road in which three regions of analysis are seen in the form of a band perpendicular to the road.
  • Figures 7 and 8.- Shows the pair of simultaneous homologous images of the previous figure taken at the moment of passing a vehicle, from which a detail of one of the analysis regions with the shape of a band perpendicular to the road has been extracted.
  • Figure 9. Shows a plan view of the location of the system of the present invention on a bridge that crosses the road whose traffic volume is to be known.
  • Figure 10. Shows a perspective view of the location of the system according to the invention above the road.
  • Figure 1 1 shows a plan view of the regions detected by the pair of cameras of the system according to the present invention.
  • Figure 12.- Shows an elevation of the vehicle detection system of Figure 1 1.
  • the system (1) for detecting the passage of vehicles (7) by a road is composed of one or more pairs of image acquisition means that take pictures of the scene of the road or public road from Two different positions horizontally.
  • the system includes a radiation emitting medium that emits radiation with the power necessary to reach the vehicles (7) at the working distance of the system.
  • the radiation emitting medium is an infrared lamp (2)
  • the image acquisition means are a pair of infrared cameras (3, 4).
  • Figure 2 shows the type of images acquired by one of the infrared cameras (3, 4), at the moment of passing a vehicle (7). The use of infrared radiation avoids the danger of disturbing or dazzling the conductors during the night.
  • Figures 3 and 4 show two images acquired by the infrared cameras (3, 4), the figure 3 being the image taken by a left infrared camera (3), and Figure 4 the image taken by a right infrared camera (4) .
  • the slight difference existing between the positions of the horizontal signaling lines of the road (8, 9) is observed, caused by the difference in position of the infrared cameras (3, 4), which in this example is 40 cm.
  • the processing means is a microcontroller (5), which is connected to the infrared cameras (3, 4), to the infrared lamp (2), to a real time clock (66) and to a GPS positioning system (67) .
  • the microcontroller (5) controls the capture of images by infrared cameras (3, 4) and controls the emission level of the infrared lamp (2) depending on the conditions of the scene.
  • the microcontroller (5) has the capacity to analyze the images acquired by infrared cameras (3, 4), either by receiving them directly or after passing through a digitizing element (for example, digital analog converter).
  • the microcontroller (5) also has the ability to communicate the results of its analysis through a communications medium (6), in this example an Ethernet port (6).
  • the results that are transmitted through the Ethernet port (6) are encrypted and combined with the date and time measured with the real time clock (66) and the coordinates of its geographical location measured with the GPS positioning system (67).
  • the microcontroller (5) can periodically send, for example with hourly frequency, the number and the average speed of the vehicles (7) that have passed through that section of the road network.
  • the system (1) which is integrated in a compact manner in a housing, is installed on a porch that crosses the road whose traffic you want to measure perpendicularly, as shown in Figure 9.
  • the system starts up an initialization process corresponding to the phase 1 described above, and which consists of capturing a pair of homologous images of the road, figures 3 and 4, and in calculating its correspondence ⁇ function as the cross correlation between both images, based on the regions of interest defined by the user.
  • This function gives us for each pixel of the region in the left homologous image the position of its homologous pixel in the right homologous image.
  • the horizontal signaling lines (8, 9) of the track could be taken.
  • the detection process corresponding to phase 2 consists, in this For example, in simultaneously capturing three analysis regions (12, 13, 14) in the form of a band perpendicular to the road every 10 milliseconds (corresponding to a frequency of 100 images / second) by each of the infrared cameras (3, 4) , and in determining the number of discrepancies in each region of analysis (12, 13, 14) in the form of a band, as shown in Figures 5 and 6.
  • each pixel in each region of analysis is subtracted ( 12, 13, 14) of the right homologous image of the value of its homologous pixel in the left homologous image, according to the correspondence of positions ⁇ established in the initialization process, and the absolute value of the result is compared with a reference threshold t predefined If the value of a pixel and its counterpart are not similar enough, a discrepancy is considered to have occurred.
  • the value of a pixel could be, for example, a brightness value or the percentage of black of the pixel.
  • an analysis of the region in question is carried out in the homologous images left and right to verify if the lines (8, 9) of the horizontal road marking can be detected in any of them. If the lines (8, 9) cannot be detected, it is concluded that the image is saturated by the presence of a vehicle (7) of high height (for example, a truck or coach), so it is also concluded that there is a vehicle (7) in the analysis region. In other words, it is a vehicle of such a large size that it covers all the image acquired by the cameras. In case that the horizontal signaling lines (8, 9) of the road can be detected, it is concluded that there is no vehicle (7) in the analysis region.
  • a vehicle (7) of high height for example, a truck or coach
  • the analysis process combines different results to extract additional information about the vehicles (7) that circulate around the scene:
  • the height of the vehicles (7) is estimated, for example, by calculating the cross correlation between the analysis regions (12, 13, 14) in the left and right homologous images, and estimating the height by trigonometric methods simple.
  • the distance between an object (15) and the cameras (3, 4) can be obtained by triangulation based on the difference between its position in the left image and its position in the right image, value that is estimated by cross correlation. Having that distance and knowing the height h the height of the object (15) is calculated.

Abstract

Method for detecting the passage of vehicles (7) on a road, characterised in that it comprises carrying out repetitively at last the following operations: acquiring, at least, a pair of simultaneous equivalent images of the road; subtracting, for each pixel of one of the simultaneous equivalent images, the value of the equivalent pixel in the other simultaneous equivalent image, locating its position by the Ω correspondence function; comparing the values obtained in the previous operation with a preset threshold value, obtaining the number of discrepancies, if a detection is produced, wherein the Ω correspondence function, that relates the position of the pixels of an equivalent image with the position of the pixels of the other equivalent image, is calculated in a previous starting operation using fixed elements of the road as reference points.

Description

PROΠFΓHMIFNTO Y sisTFMA ΠF ΠFTFΠΠIÓN ΠF OR.IFTOS FM PROΠFΓHMIFNTO AND sisTFMA ΠF ΠFTFΠΠIÓN ΠF OR.IFTOS FM
MDVlMlFMTnMDVlMlFMTn
Π F S Π R I P Π I Ó NΠ F S Π R I P Π I Ó N
OBJETO DE LA INVENCIÓNOBJECT OF THE INVENTION
El objeto principal de Ia presente invención es un procedimiento y un dispositivo que detectan el paso de vehículos por una vía. Otro objeto de Ia presente invención es un procedimiento y un dispositivo que además determinan Ia longitud, altura y velocidad de los vehículos.The main object of the present invention is a method and a device that detect the passage of vehicles through a road. Another object of the present invention is a method and a device that also determine the length, height and speed of the vehicles.
ANTECEDENTES DE LA INVENCIÓNBACKGROUND OF THE INVENTION
El análisis y control de tráfico rodado es una necesidad cada vez mayor en las sociedades avanzadas, en las que los problemas de movilidad suponen un gran coste económico y en las que los accidentes de tráfico suponen un enorme coste en vidas humanas.The analysis and control of road traffic is a growing need in advanced societies, in which mobility problems involve a great economic cost and in which traffic accidents are a huge cost in human lives.
Los sistemas avanzados de control de tráfico utilizan sistemas de visión artificial que estudian diferentes condiciones del tráfico en una escena analizando imágenes digitales. La captura de esas imágenes se produce en general con dos sistemas, uno que detecta el paso de un vehículo y otro que captura al menos una imagen digital de Ia escena.Advanced traffic control systems use artificial vision systems that study different traffic conditions in a scene by analyzing digital images. The capture of these images generally occurs with two systems, one that detects the passage of a vehicle and another that captures at least one digital image of the scene.
El hecho de tener dos elementos en muchos casos supone una gran complejidad de instalación y mantenimiento. Además, los elementos que detectan el paso de los vehículos suelen ser costosos, ya que se trata de elementos de mucha precisión, como por ejemplo radares, o de elementos intrusivos, cuya instalación requiere Ia ejecución de obras en Ia calzada conThe fact of having two elements in many cases implies a great complexity of installation and maintenance. In addition, the elements that detect the passage of vehicles are usually expensive, since they are very precise elements, such as radars, or intrusive elements, whose installation requires the execution of works on the road with
Ia consiguiente interrupción del tráfico. También se han propuesto sistemas no intrusivos basados en ultrasonidos y en microondas. Este tipo de sistemas suele presentar problemas relacionados con falta de precisión o excesiva dependencia de las condiciones de entorno en Ia carretera. Como alternativa a estos elementos se han creado detectores ópticos que interpretan los cambios en Ia escena como indicadores del paso de un vehículo, pero su uso queda muy limitado al ser excesivamente sensibles a cambios en Ia iluminación ambiental (provocados por el paso de nubes, por las luces y las sombras producidas por otros coches, por elementos externos a Ia carretera, etc.), y por ser muy sensibles a las vibraciones, Io que imposibilita su instalación sobre semáforos, farolas, puentes o pórticos.The subsequent traffic interruption. Non-intrusive systems based on ultrasound and microwave have also been proposed. This type of systems usually presents problems related to lack of precision or excessive dependence on the environmental conditions on the road. As an alternative to these elements, optical detectors have been created that interpret the changes in the scene as indicators of the passage of a vehicle, but their use is very limited because they are excessively sensitive to changes in the ambient lighting (caused by the passage of clouds, by the lights and shadows produced by other cars, by elements external to the road, etc.), and because they are very sensitive to vibrations, which makes it impossible to install them on traffic lights, lampposts, bridges or porches.
DESCRIPCIÓN DE LA INVENCIÓNDESCRIPTION OF THE INVENTION
La presente invención se refiere a un procedimiento y un dispositivo que resuelven Ia problemática anterior en un único aparato que detecta el paso del vehículo y toma imágenes de alta resolución para su posterior identificación. La detección del vehículo soluciona las deficiencias de otros sistemas de detección por visión artificial, ya que utiliza un algoritmo que analiza Ia altura de los objetos, Io que Ie permite descartar los brillos y las sombras en Ia calzada que confundirían a otros métodos, y realiza un filtrado de las imágenes.The present invention relates to a method and a device that solve the above problem in a single device that detects the passage of the vehicle and takes high resolution images for later identification. The detection of the vehicle solves the deficiencies of other systems of detection by artificial vision, since it uses an algorithm that analyzes the height of the objects, which allows it to discard the brightness and shadows on the road that would confuse other methods, and performs a filtering of the images.
El método también es totalmente insensible a las vibraciones, ya que trabaja con imágenes estáticas ("instantáneas") en lugar de vídeo, Io que facilita su instalación en cualquier elemento disponible en Ia vía pública (farolas, semáforos, puentes, pórticos, etc.).The method is also totally insensitive to vibrations, since it works with static images ("snapshots") instead of video, which facilitates its installation in any element available on public roads (street lamps, traffic lights, bridges, porches, etc.). ).
Otra ventaja importante del método es que ha sido concebido para trabajar a muy alta velocidad, Io cual posibilita Ia detección de vehículos independientemente de su velocidad, y para ser fácilmente integrado en un circuito integrado de aplicación específica (ASIC) u otro dispositivo análogo, Io que posibilita su fabricación a gran escala y muy bajo coste.Another important advantage of the method is that it has been designed to work at a very high speed, which makes it possible to detect vehicles regardless of its speed, and to be easily integrated into a specific application integrated circuit (ASIC) or other similar device, which allows its large-scale manufacturing and very low cost.
El método utilizado para Ia detección de vehículos en Ia vía se basa en Ia estimación de Ia altura de éstos mediante un sistema que contiene dos o más medios de adquisición de imágenes. Comparando las imágenes adquiridas se detectan discrepancias entre píxeles homólogos de las imágenes homologas, deduciéndose Ia existencia de objetos con altura no nula.The method used for the detection of vehicles on the road is based on the estimation of their height by means of a system that contains two or more means of image acquisition. Comparing the acquired images, discrepancies between homologous pixels of the homologous images are detected, deducing the existence of objects with non-zero height.
En el presente documento, se entenderá que "vía" hace referencia a cualquier tipo de camino, calzada, carretera, etc. por Ia que pasen vehículos cuyas características se desea conocer. De igual modo, el término "vehículo" no se limita sólo a los coches, ya que el presente procedimiento y sistema son capaces de detectar el paso de cualquier objeto de altura no nula, como motocicletas, animales, personas a pie, etc.In this document, it will be understood that "road" refers to any type of road, road, road, etc. through which vehicles whose characteristics you want to know pass. Similarly, the term "vehicle" is not limited only to cars, since the present procedure and system are capable of detecting the passage of any non-zero height object, such as motorcycles, animals, people on foot, etc.
Además, diremos que un "punto" corresponde a una parte de un vehículo, carretera, etc. en el mundo real, en contraposición con los puntos de los que están compuestas las imágenes, a los que haremos referencia como píxeles. Además, se entenderá que el "valor de un píxel" hace referencia a un valor de luminosidad de dicho píxel.In addition, we will say that a "point" corresponds to a part of a vehicle, road, etc. in the real world, as opposed to the points of which the images are composed, which we will refer to as pixels. In addition, it will be understood that the "value of a pixel" refers to a brightness value of said pixel.
Por tanto, de acuerdo con un aspecto de Ia presente invención, se proporciona un procedimiento de detección del paso de vehículos por una vía que comprende las dos fases siguientes:Therefore, in accordance with one aspect of the present invention, a method of detecting the passage of vehicles by a route comprising the following two phases is provided:
Fase 1 :Phase 1 :
Se realiza por Io menos una vez como inicialización del sistema y consiste en analizar, al menos en un par de imágenes homologas simultáneas, Ia posición de Ia carretera, y establecer una referencia de Ia distancia a Ia que se encuentra Ia calzada para cada píxel de Ia imagen. Esta primera fase comprende las siguientes operaciones:It is done at least once as system initialization and it consists of analyzing, at least in a pair of simultaneous homologous images, the position of the road, and establishing a reference of the distance at which the roadway is for each pixel of the image. This first phase includes the following operations:
a) Adquirir, al menos, un par de imágenes homologas simultáneas de Ia vía sin vehículos mediante dos medios de adquisición de imágenes, preferentemente separados horizontalmente. Diremos que dos "imágenes homologas" son dos imágenes de Ia misma zona de Ia vía desde dos puntos de vista separados en dirección horizontal una distancia correspondiente a Ia distancia que separa los medios de adquisición de imágenes. Los medios de adquisición de imágenes obtienen imágenes de Ia vía al recibir Ia radiación en ella reflejada, por ejemplo Ia radiación natural visible.a) Acquire at least one pair of simultaneous homologous images of the path without vehicles by means of two image acquisition means, preferably horizontally separated. We will say that two "homologous images" are two images of the same area of the track from two horizontally separated points of view a distance corresponding to the distance that separates the image acquisition means. The image acquisition means obtain images of the pathway upon receiving the radiation in it reflected, for example the visible natural radiation.
b) Generar un modelo de Ia vía mediante el análisis de las correspondencias entre los píxeles de las imágenes homologas simultáneas adquiridas en el paso anterior mediante alguna técnica de análisis de correspondencias entre imágenes, por ejemplo utilizando para ello puntos de referencia fijos de Ia vía. Este modelo es una función de correspondencia Ω: R2->R2 que nos da para cada píxel de Ia imagen de una de las cámaras, por ejemplo Ia izquierda en caso de que estén separadas horizontalmente, Ia posición de su píxel homólogo en Ia imagen de Ia otra cámara, es decir Ia de Ia derecha en caso de que estén separadas horizontalmente. La distancia euclidea entre Ia posición (X11V1) de un píxel en Ia imagen izquierda y su homólogo (Xd,yd) en Ia imagen derecha nos da por métodos trigonométricos Ia distancia entre el medio de adquisición de imágenes y Ia vía en ese lugar de Ia escena.b) Generate a model of the path by analyzing the correspondences between the pixels of the simultaneous homologous images acquired in the previous step by some technique of correspondence analysis between images, for example using fixed reference points of the path. This model is a correspondence function Ω: R 2 -> R 2 that gives us for each pixel of the image of one of the cameras, for example the left in case they are horizontally separated, the position of its homologous pixel in Ia image of the other camera, that is to say on the right in case they are separated horizontally. The Euclidean distance between the position (X 11 V 1 ) of a pixel in the left image and its counterpart (Xd, and d) in the right image gives us by trigonometric methods the distance between the means of image acquisition and the path in that Place of the scene.
Fase 7-Phase 7-
Se realiza de forma repetitiva y consiste en detectar objetos que alteran la situación establecida en el modelo obtenido en Ia primera fase. En esta segunda fase se realizan al menos las siguientes operaciones:It is done repetitively and consists of detecting objects that they alter the situation established in the model obtained in the first phase. In this second phase at least the following operations are carried out:
a) Adquirir, al menos, un par de imágenes homologas simultáneas de Ia vía mediante los dos medios de adquisición de imágenes separados y detectar píxeles en las imágenes homologas que no coincidan con el modelo establecido en Ia fase anterior, porque Ia distancia a Ia carretera en su posición sea diferente a Ia establecida. Dicho de otro modo, mientras no exista ningún vehículo en Ia vía el valor de un píxel será aproximadamente igual al de su homólogo, ya que ambos píxeles corresponderán al mismo punto, normalmente de Ia vía, y por Io tanto ambos tendrán Ia misma luminosidad. En cambio, cuando pase un vehículo de altura no nula, en algún momento se producirá una situación en que un píxel corresponderá a un punto de Ia vía, mientras que su píxel homólogo corresponderá un punto del vehículo. En esta situación, Ia resta de los valores de este par de píxeles homólogos dará como resultado un valor superior a un valor umbral establecido, produciéndose una discrepancia. Por tanto, calculando Ia resta entre los valores de los píxeles de las dos imágenes homologas, desplazando previamente los valores de los píxeles en una de ellas un número de píxeles proporcional al modelo Ω, y comparando los resultados con el valor umbral establecido, obtenemos el número de discrepancias. En el caso de que el número de discrepancias en esta operación supere un número máximo predefinido, concluimos que hay un objeto en Ia vía que no estaba presente en el momento de Ia generación del modelo (fase 1 ). De acuerdo con una realización preferida de Ia invención, para realizar este proceso a muy alta velocidad centramos nuestro análisis en regiones de las imágenes homologas con forma de banda perpendicular a Ia vía.a) Acquire at least one pair of simultaneous homologous images of the path by means of the two means of acquiring separate images and detect pixels in the homologous images that do not coincide with the model established in the previous phase, because the distance to the road in its position it is different from the one established. In other words, as long as there is no vehicle on the track, the value of a pixel will be approximately equal to that of its counterpart, since both pixels will correspond to the same point, usually on the track, and therefore both will have the same luminosity. On the other hand, when a vehicle of non-zero height passes, at some point there will be a situation in which a pixel will correspond to a point of the track, while its homologous pixel will correspond to a point of the vehicle. In this situation, the subtraction of the values of this pair of homologous pixels will result in a value greater than an established threshold value, producing a discrepancy. Therefore, calculating the subtraction between the pixel values of the two homologous images, previously displacing the pixel values in one of them a number of pixels proportional to the Ω model, and comparing the results with the established threshold value, we obtain the number of discrepancies In the event that the number of discrepancies in this operation exceeds a predefined maximum number, we conclude that there is an object in the path that was not present at the time of the generation of the model (phase 1). In accordance with a preferred embodiment of the invention, to perform this process at a very high speed, we focus our analysis on regions of the homologous images in the form of a band perpendicular to the pathway.
b) En el caso de que el número de discrepancias detectadas en Ia operación anterior no sea superior al número máximo predefinido, podemos, o bien concluir que no hay ningún objeto en Ia vía que no estaba presente en el momento de Ia generación del modelo de Ia vía (fase 1 ), o bien que existe un objeto con una altura excesivamente alta que produce Ia saturación de Ia imagen (por ejemplo un camión o autocar). Para distinguir entre estas dos posibilidades se verifica si al menos una de las regiones de análisis de las dos imágenes homologas contiene alguna característica conocida presente en Ia vía. En el caso de no detectar esta característica en alguna de las dos imágenes concluimos que hay un objeto en Ia vía que no estaba presente en el momento de Ia generación del modelo (fase 1 )b) In the event that the number of discrepancies detected in the previous operation does not exceed the maximum predefined number, we can either conclude that there is no object in the path that was not present in the moment of the generation of the model of the track (phase 1), or that there is an object with an excessively high height that produces the saturation of the image (for example a truck or coach). To distinguish between these two possibilities, it is verified if at least one of the analysis regions of the two homologous images contains some known characteristic present in the pathway. In the case of not detecting this characteristic in any of the two images we conclude that there is an object in the path that was not present at the time of the generation of the model (phase 1)
El procedimiento descrito en Ia fase 2 se realiza de modo continuo, adquiriéndose imágenes homologas simultáneas de modo consecutivo a muy alta velocidad, y deduciéndose de ellas el número de vehículos que transitan por Ia vía durante un determinado período de tiempo. De acuerdo con una realización preferida de Ia presente invención, se llevan a cabo las operaciones de Ia fase 2 un mínimo de K veces por segundo, siendo K un parámetro que depende, según Ia siguiente fórmula, de Ia velocidad máxima a Ia que circulan los vehículos por Ia carretera (Vmax), su longitud mínima (Lmm) y Ia distancia longitudinal de Ia vía que queda cubierta por Ia región de análisis (LrΘg). Si no tenemos en cuenta Ia altura de los vehículos podemos aproximar este parámetro con Ia siguiente fórmula (por ejemplo, para una motocicleta de 1 ,5 m de largo, una velocidad máxima de 300 Km/hora y una región de detección de 10 cm necesitamos una frecuencia de análisis mínima de 52 imágenes por segundo):The procedure described in phase 2 is carried out continuously, simultaneous simultaneous images are acquired consecutively at very high speed, and deducting from them the number of vehicles that travel along the road during a certain period of time. In accordance with a preferred embodiment of the present invention, the operations of the phase 2 are carried out a minimum of K times per second, K being a parameter that depends, according to the following formula, on the maximum speed at which the vehicles on the road (Vmax), its minimum length (Lmm) and the longitudinal distance of the road that is covered by the analysis region (L rΘg ). If we do not take into account the height of the vehicles we can approximate this parameter with the following formula (for example, for a motorcycle of 1.5 m long, a maximum speed of 300 km / hour and a detection region of 10 cm we need a minimum analysis frequency of 52 images per second):
K > Vχasκ K> Vχasκ
V min reg )V min reg)
La precisión de las estimaciones de velocidad y longitud de los vehículos es mayor cuanto mayor Ia frecuencia de adquisición de imágenes. De acuerdo con una realización preferida de Ia invención, las operaciones de la fase 2 se realizan 50 o más veces por segundo. Gracias a Ia realización repetitiva del anterior procedimiento detectamos de forma inequívoca Ia entrada y Ia salida de un objeto en Ia región de análisis, Io que nos permite capturar una imagen completa del vehículo en el momento más adecuado para nuestra aplicación. Por ejemplo, podemos tomar una imagen de Ia parte posterior del vehículo para analizar su matrícula.The accuracy of vehicle speed and length estimates is greater the higher the frequency of image acquisition. In accordance with a preferred embodiment of the invention, the operations of phase 2 are performed 50 or more times per second. Thanks to the repetitive implementation of the previous procedure, we unequivocally detect the entry and exit of an object in the analysis region, which allows us to capture a complete image of the vehicle at the most appropriate time for our application. For example, we can take a picture of the back of the vehicle to analyze its registration.
De acuerdo con otra realización preferida de Ia invención, el procedimiento comprende además Ia operación de emitir en dirección a Ia vía, mediante un medio emisor de radiación, una radiación adecuada para el funcionamiento de los medios de adquisición de imágenes. Así, el medio emisor de radiación podría emitir radiación de tipo visible, infrarroja, ultrasonidos, radar, microondas, etcAccording to another preferred embodiment of the invention, the method further comprises the operation of emitting in the direction of the pathway, by means of a radiation emitting means, a radiation suitable for the operation of the image acquisition means. Thus, the radiation emitting medium could emit visible, infrared, ultrasound, radar, microwave, etc.
La adaptación de Ia potencia de Ia radiación emitida y de Ia sensibilidad de los sensores se realiza de forma dinámica en cada captura mediante un esquema realimentado. Dicho esquema toma como dato de partida una estimación de Ia luminosidad ambiental calculada en base a Ia evolución de Ia media de los valores de los píxeles en Ia región de análisis, en los instantes en que no hay un vehículo en Ia vía.The adaptation of the power of the emitted radiation and the sensitivity of the sensors is carried out dynamically in each capture by means of a feedback scheme. Said scheme takes as an initial data an estimate of the ambient luminosity calculated based on the evolution of the average of the pixel values in the analysis region, in the moments in which there is no vehicle on the track.
De acuerdo con realizaciones preferidas de Ia invención, el procedimiento de detección del paso de vehículos por una vía comprende además Ia operación de determinar Ia altura del vehículo detectado. Esta función se desprende de modo evidente del funcionamiento descrito anteriormente, ya que es posible determinar Ia altura de un punto mediante trigonometría básica.In accordance with preferred embodiments of the invention, the method of detecting the passage of vehicles through a track also includes the operation of determining the height of the detected vehicle. This function is evident from the operation described above, since it is possible to determine the height of a point by basic trigonometry.
Además, de acuerdo con otra realización preferida de Ia invención, el procedimiento de detección del paso de vehículos por una vía comprende además Ia operación de determinar Ia longitud del vehículo detectado. Para ello se analizan, en cada par de imágenes adquiridas, varias regiones de análisis homologas simultáneas con forma de banda perpendicular a Ia vía. De Ia información relativa a Ia altura de los píxeles homólogos localizados en cada uno de los pares de regiones de análisis con forma de banda se deduce Ia longitud del vehículo, ya que se conoce Ia distancia en longitudinal entre las regiones de análisis.In addition, according to another preferred embodiment of the invention, the method of detecting the passage of vehicles through a road comprises also the operation of determining the length of the detected vehicle. For this, in each pair of acquired images, several regions of simultaneous homologous analysis are analyzed in the form of a band perpendicular to the pathway. From the information related to the height of the homologous pixels located in each of the pairs of band-shaped analysis regions, the vehicle length is deduced, since the longitudinal distance between the analysis regions is known.
De acuerdo con una realización preferida más de Ia presente invención, el procedimiento de detección del paso de vehículos por una vía comprende además Ia operación de determinar Ia velocidad del vehículo detectado. Puesto que las imágenes se toman de manera secuencial a una frecuencia conocida, se puede determinar cuánto tiempo permanece el vehículo dentro de una de las regiones de análisis. Y puesto que, como se ha explicado anteriormente, podemos conocer Ia longitud del vehículo, se puede deducir Ia velocidad del vehículo.In accordance with a further preferred embodiment of the present invention, the method of detecting the passage of vehicles through a track also includes the operation of determining the speed of the detected vehicle. Since the images are taken sequentially at a known frequency, it can be determined how long the vehicle remains within one of the analysis regions. And since, as explained above, we can know the length of the vehicle, the vehicle speed can be deduced.
De acuerdo con otras realizaciones preferidas de Ia invención, el procedimiento de detección del paso de vehículos por una vía comprende además las operaciones de tomar Ia fecha y Ia hora de Ia detección, medir Ia localización geográfica del equipo de detección, y encriptar y comunicar los resultados obtenidos mediante un medio de comunicación. Así, se podría enviar el número de vehículos que ha transitado por una carretera y su velocidad media a un centro de control de tráfico, por ejemplo, diariamente, junto con Ia identificación del punto kilométrico en que se ha realizado Ia medida y Ia fecha y hora. También se podría interconectar varios equipos similares para gestionar Ia detección de vehículos en varias vías de una misma calzada (por ejemplo en carreteras de varios carriles), detectando las situaciones en las que un vehículo está cambiando de carril para evitar que sean contabilizados dos vehículos en lugar de uno. Por motivos de seguridad, todos los envíos se realizan en forma encriptada. De acuerdo con un segundo aspecto de Ia invención, se describe un sistema de detección del paso de vehículos por una vía, caracterizado porque comprende los siguientes dispositivos:In accordance with other preferred embodiments of the invention, the method of detecting the passage of vehicles through a road also includes the operations of taking the date and time of the detection, measuring the geographical location of the detection equipment, and encrypting and communicating the results obtained through a means of communication. Thus, the number of vehicles that have traveled on a road and their average speed could be sent to a traffic control center, for example, daily, along with the identification of the kilometric point at which the measurement was made and the date and hour. Several similar equipment could also be interconnected to manage the detection of vehicles on several roads of the same road (for example on multi-lane roads), detecting the situations in which a vehicle is changing lanes to avoid two vehicles being counted in place of one. For security reasons, all shipments are made in encrypted form. In accordance with a second aspect of the invention, a system for detecting the passage of vehicles through a track is described, characterized in that it comprises the following devices:
a) Dos medios de adquisición de imágenes que adquieren dos imágenes homologas simultáneas de un tramo de Ia vía. De acuerdo con realizaciones preferidas de Ia invención, los medios de adquisición de imágenes pueden ser cámaras digitales, cámaras de vídeo digitales, cámaras infrarrojas, detectores de ultrasonidos, detectores de radar, etc. Además, de acuerdo con otra realización preferente de Ia presente invención, Ia distancia entre las cámaras infrarrojas izquierda y derecha (3, 4) está dentro del rango de 20 cm a 100 cm.a) Two image acquisition means that acquire two simultaneous homologous images of a section of the track. In accordance with preferred embodiments of the invention, the image acquisition means can be digital cameras, digital video cameras, infrared cameras, ultrasonic detectors, radar detectors, etc. In addition, according to another preferred embodiment of the present invention, the distance between the left and right infrared cameras (3, 4) is within the range of 20 cm to 100 cm.
b) Un medio de procesamiento conectado a los medios de adquisición de imágenes (3, 4), que resta, para cada píxel de una de las imágenes homologas simultáneas, el valor del píxel homólogo en Ia otra imagen homologa simultánea y determina, en función del número de discrepancias, si se produce una detección. Además, el medio de procesamiento tiene Ia capacidad de encriptar los datos para su envío a, por ejemplo, un centro de control de tráfico. El medio de procesamiento podría ser cualquier dispositivo capaz de efectuar los cálculos necesarios para determinar Ia presencia de vehículos sobre Ia vía, aunque de acuerdo con realizaciones preferentes de Ia invención podría ser un microcontrolador, un ordenador, un ASIC, un DSP, una FPGA, etc.b) A processing means connected to the image acquisition means (3, 4), which subtracts, for each pixel of one of the simultaneous homologous images, the value of the homologous pixel in the other simultaneous homologous image and determines, depending on of the number of discrepancies, if a detection occurs. In addition, the processing means has the ability to encrypt the data for sending to, for example, a traffic control center. The processing means could be any device capable of carrying out the calculations necessary to determine the presence of vehicles on the road, although according to preferred embodiments of the invention it could be a microcontroller, a computer, an ASIC, a DSP, an FPGA, etc.
Además, de acuerdo con realizaciones preferidas de Ia invención se puede añadir un medio emisor de radiación que emita radiación en dirección a Ia vía, evitando así Ia dependencia de Ia luz visible natural. La radiación podría ser de cualquier tipo, siempre que se refleje en los vehículos que transitan por Ia vía y pueda ser detectada por los medios de adquisición de imágenes. De acuerdo con realizaciones preferentes de Ia invención, el medio emisor de radiación puede emitir radiación de tipo visible, infrarroja, ultrasonidos, radar, microondas, etc. Sin embargo, no es aconsejable Ia radiación visible, ya que podría molestar a los conductores de los vehículos que transiten por Ia vía.In addition, according to preferred embodiments of the invention, a radiation emitting medium can be added that emits radiation in the direction of the pathway, thus avoiding dependence on the natural visible light. The radiation could be of any type, as long as it is reflected in the vehicles that travel along the road and can be detected by the means of acquiring images. In accordance with preferred embodiments of the invention, the radiation emitting means can emit visible, infrared, ultrasound, radar, microwave, etc. radiation. However, visible radiation is not advisable, as it could disturb the drivers of vehicles that travel along the road.
De acuerdo con realizaciones preferidas de Ia invención se puede añadir un medio para determinar Ia fecha y hora en que se produce Ia detección del vehículo y un medio para identificar Ia posición geográfica del equipo, que podría ser, por ejemplo, un sistema de posicionamiento GPSIn accordance with preferred embodiments of the invention, a means can be added to determine the date and time at which the detection of the vehicle occurs and a means to identify the geographical position of the equipment, which could be, for example, a GPS positioning system
Finalmente, de acuerdo con otra realización preferida más, el sistema puede comprender un medio de comunicación para comunicar los resultados obtenidos, por ejemplo a un centro de control de tráfico. El medio de comunicación puede ser de cualquier tipo capaz de transmitir los resultados obtenidos aunque, de acuerdo con realizaciones preferidas, podría utilizar Internet, Ethernet, USB, RS232, RS485, Bluetooth, IEEE802.Hb/g, Zigbee, radio, línea telefónica, UMTS, GSM, infrarrojos, etc.Finally, according to yet another preferred embodiment, the system may comprise a communication means for communicating the results obtained, for example to a traffic control center. The communication medium can be of any type capable of transmitting the results obtained although, according to preferred embodiments, it could use the Internet, Ethernet, USB, RS232, RS485, Bluetooth, IEEE802.Hb / g, Zigbee, radio, telephone line, UMTS, GSM, infrared, etc.
DESCRIPCIÓN DE LOS DIBUJOSDESCRIPTION OF THE DRAWINGS
Para complementar Ia descripción que se está realizando y con objeto de ayudar a una mejor comprensión de las características de Ia invención, de acuerdo con un ejemplo preferente de realización práctica de Ia misma, se acompaña como parte integrante de dicha descripción, un juego de dibujos en donde con carácter ilustrativo y no limitativo, se ha representado Io siguiente:To complement the description that is being made and in order to help a better understanding of the characteristics of the invention, according to a preferred example of practical implementation thereof, a set of drawings is attached as an integral part of said description. where, for the purposes of illustration and not limitation, the following has been represented:
Figura 1.- Muestra un esquema del sistema de detección de vehículos de acuerdo con Ia presente invención. Figura 2.- Muestra un ejemplo del tipo de imágenes que adquiere una de las cámaras del ejemplo de Ia presente invención.Figure 1.- Shows a scheme of the vehicle detection system according to the present invention. Figure 2.- Shows an example of the type of images acquired by one of the cameras of the example of the present invention.
Figuras 3 y 4.- Muestra un par de imágenes homologas simultáneas capturada por un par de cámaras del sistema de detección de vehículos de acuerdo con Ia presente invención.Figures 3 and 4.- Shows a pair of simultaneous homologous images captured by a pair of cameras of the vehicle detection system according to the present invention.
Figuras 5 y 6.- Muestra el par de imágenes homologas simultáneas de la figura 3, carretera en el que se aprecian tres regiones de análisis con forma de banda perpendicular a Ia carretera.Figures 5 and 6.- Shows the pair of simultaneous homologous images of Figure 3, road in which three regions of analysis are seen in the form of a band perpendicular to the road.
Figuras 7 y 8.- Muestra el par de imágenes homologas simultáneas de Ia figura anterior tomadas en el momento de paso de un vehículo, de donde se ha extraído un detalle de una de las regiones de análisis con forma de banda perpendicular a Ia carretera.Figures 7 and 8.- Shows the pair of simultaneous homologous images of the previous figure taken at the moment of passing a vehicle, from which a detail of one of the analysis regions with the shape of a band perpendicular to the road has been extracted.
Figura 9.- Muestra una vista en planta de Ia ubicación del sistema de Ia presente invención en un puente que cruza Ia carretera cuyo volumen de tráfico se desea conocer.Figure 9.- Shows a plan view of the location of the system of the present invention on a bridge that crosses the road whose traffic volume is to be known.
Figura 10.- Muestra una vista en perspectiva de Ia ubicación del sistema de acuerdo con Ia invención por encima de Ia carretera.Figure 10.- Shows a perspective view of the location of the system according to the invention above the road.
Figura 1 1 .- .Muestra una vista en planta de las regiones detectadas por el par de cámaras del sistema de acuerdo con Ia presente invención.Figure 1 1 .- It shows a plan view of the regions detected by the pair of cameras of the system according to the present invention.
Figura 12.- Muestra un alzado del sistema de detección de vehículos de Ia figura 1 1 .Figure 12.- Shows an elevation of the vehicle detection system of Figure 1 1.
REALIZACIÓN PREFERENTE DE LA INVENCIÓN De acuerdo con Ia figura 1 , el sistema (1 ) de detección del paso de vehículos (7) por una vía se compone de uno o más pares de medios de adquisición de imágenes que toman imágenes de Ia escena de Ia carretera o vía pública desde dos posiciones diferentes en horizontal. Para garantizar que existe luz visible suficiente veinticuatro horas al día, el sistema incluye un medio emisor de radiación que emite radiación con Ia potencia necesaria para alcanzar los vehículos (7) a Ia distancia de trabajo del sistema. En este ejemplo, el medio emisor de radiación es una lámpara infrarroja (2), y los medios de adquisición de imágenes son un par de cámaras infrarrojas (3, 4). En Ia figura 2 se muestra el tipo de imágenes que adquiere una de las cámaras infrarrojas (3, 4), en el momento de paso de un vehículo (7). El uso de radiación infrarroja evita el peligro de molestar o deslumbrar a los conductores durante Ia noche.PREFERRED EMBODIMENT OF THE INVENTION According to Figure 1, the system (1) for detecting the passage of vehicles (7) by a road is composed of one or more pairs of image acquisition means that take pictures of the scene of the road or public road from Two different positions horizontally. To ensure that there is sufficient visible light twenty-four hours a day, the system includes a radiation emitting medium that emits radiation with the power necessary to reach the vehicles (7) at the working distance of the system. In this example, the radiation emitting medium is an infrared lamp (2), and the image acquisition means are a pair of infrared cameras (3, 4). Figure 2 shows the type of images acquired by one of the infrared cameras (3, 4), at the moment of passing a vehicle (7). The use of infrared radiation avoids the danger of disturbing or dazzling the conductors during the night.
Las figuras 3 y 4 muestran dos imágenes adquiridas por las cámaras infrarrojas (3, 4), siendo Ia figura 3 Ia imagen tomada por una cámara infrarroja izquierda (3), y Ia figura 4 Ia imagen tomada por una cámara infrarroja derecha (4). Se observa Ia ligera diferencia existente entre las posiciones de las líneas de señalización horizontal de Ia carretera (8, 9), causada por Ia diferencia de posición de las cámaras infrarrojas (3, 4), que en este ejemplo es de 40 cm.Figures 3 and 4 show two images acquired by the infrared cameras (3, 4), the figure 3 being the image taken by a left infrared camera (3), and Figure 4 the image taken by a right infrared camera (4) . The slight difference existing between the positions of the horizontal signaling lines of the road (8, 9) is observed, caused by the difference in position of the infrared cameras (3, 4), which in this example is 40 cm.
El medio de procesamiento es un microcontrolador (5), que está conectado a las cámaras infrarrojas (3, 4), a Ia lámpara infrarroja (2), a un reloj de tiempo real (66) y a un sistema de posicionamiento GPS (67). El microcontrolador (5) controla Ia captura de imágenes por las cámaras infrarrojas (3, 4) y controla el nivel de emisión de Ia lámpara infrarroja (2) en función de las condiciones de Ia escena.The processing means is a microcontroller (5), which is connected to the infrared cameras (3, 4), to the infrared lamp (2), to a real time clock (66) and to a GPS positioning system (67) . The microcontroller (5) controls the capture of images by infrared cameras (3, 4) and controls the emission level of the infrared lamp (2) depending on the conditions of the scene.
El microcontrolador (5) tiene capacidad para analizar las imágenes adquiridas por las cámaras infrarrojas (3, 4), bien sea recibiéndolas directamente o después de pasar por un elemento digitalizador (por ejemplo, conversor analógico digital). El microcontrolador (5) también dispone de capacidad para comunicar los resultados de sus análisis a través de un medio de comunicaciones (6), en este ejemplo un puerto Ethernet (6). Los resultados que se transmiten por el puerto Ethernet (6) van encriptados y combinados con Ia fecha y hora medidas con el reloj de tiempo real (66) y las coordenadas de su localización geográfica medidas con el sistema de posicionamiento GPS (67).The microcontroller (5) has the capacity to analyze the images acquired by infrared cameras (3, 4), either by receiving them directly or after passing through a digitizing element (for example, digital analog converter). The microcontroller (5) also has the ability to communicate the results of its analysis through a communications medium (6), in this example an Ethernet port (6). The results that are transmitted through the Ethernet port (6) are encrypted and combined with the date and time measured with the real time clock (66) and the coordinates of its geographical location measured with the GPS positioning system (67).
Así, el microcontrolador (5) puede enviar periódicamente, por ejemplo con frecuencia horaria, el número y Ia velocidad media de los vehículos (7) que han pasado por ese tramo de Ia red viaria.Thus, the microcontroller (5) can periodically send, for example with hourly frequency, the number and the average speed of the vehicles (7) that have passed through that section of the road network.
El sistema (1 ), que está integrado de forma compacta en una carcasa, se instala sobre un pórtico que cruza en perpendicular Ia carretera cuyo tráfico se desea medir, como se observa en Ia figura 9.The system (1), which is integrated in a compact manner in a housing, is installed on a porch that crosses the road whose traffic you want to measure perpendicularly, as shown in Figure 9.
En este ejemplo, se realiza en primer lugar y como mínimo una vez cuando se pone en marcha el sistema un proceso de inicialización correspondiente a Ia fase 1 descrita anteriormente, y que consiste en capturar un par de imágenes homologas de Ia carretera, figuras 3 y 4, y en calcular su función Ω de correspondencia como Ia correlación cruzada entre ambas imágenes, tomando como base las regiones de interés definidas por el usuario. Esta función nos da para cada píxel de Ia región en Ia imagen homologa izquierda Ia posición de su píxel homólogo en Ia imagen homologa derecha. Como puntos de referencia fijos para el cálculo de Ia función a se podrían tomar, en este ejemplo, las líneas (8, 9) de señalización horizontal de Ia vía.In this example, it is carried out first and at least once when the system starts up an initialization process corresponding to the phase 1 described above, and which consists of capturing a pair of homologous images of the road, figures 3 and 4, and in calculating its correspondence Ω function as the cross correlation between both images, based on the regions of interest defined by the user. This function gives us for each pixel of the region in the left homologous image the position of its homologous pixel in the right homologous image. As fixed reference points for the calculation of the function a, in this example, the horizontal signaling lines (8, 9) of the track could be taken.
El proceso de detección correspondiente a Ia fase 2 consiste, en este ejemplo, en capturar simultáneamente tres regiones de análisis (12, 13, 14) con forma de banda perpendicular a Ia carretera cada 10 milisegundos (correspondiente a una frecuencia de 100 imágenes/segundo) mediante cada una de las cámaras infrarrojas (3, 4), y en determinar el número de discrepancias en cada región de análisis (12, 13, 14) con forma de banda, como se muestra en las figuras 5 y 6. Para ello se resta el valor de cada píxel en cada región de análisis (12, 13, 14) de Ia imagen homologa derecha del valor de su píxel homólogo en Ia imagen homologa izquierda, según Ia correspondencia de posiciones Ω establecida en el proceso de inicialización, y se compara el valor absoluto del resultado con un umbral t de referencia predefinido. Si el valor de un píxel y su homólogo no son Io suficientemente parecidos se considera que se ha producido una discrepancia. El valor de un píxel podría ser, por ejemplo, un valor de luminosidad o el porcentaje de negro del píxel.The detection process corresponding to phase 2 consists, in this For example, in simultaneously capturing three analysis regions (12, 13, 14) in the form of a band perpendicular to the road every 10 milliseconds (corresponding to a frequency of 100 images / second) by each of the infrared cameras (3, 4) , and in determining the number of discrepancies in each region of analysis (12, 13, 14) in the form of a band, as shown in Figures 5 and 6. To do this, the value of each pixel in each region of analysis is subtracted ( 12, 13, 14) of the right homologous image of the value of its homologous pixel in the left homologous image, according to the correspondence of positions Ω established in the initialization process, and the absolute value of the result is compared with a reference threshold t predefined If the value of a pixel and its counterpart are not similar enough, a discrepancy is considered to have occurred. The value of a pixel could be, for example, a brightness value or the percentage of black of the pixel.
Así, cuando no pasa ningún vehículo por Ia vía (Figuras 5 y 6), el píxel A y el píxel A', situados en Ia banda central 13, corresponderán a un mismo punto situado sobre Ia calzada, y por Io tanto Ia diferencia entre los valores de esos píxeles será menor que el umbral t de referencia. Sin embargo, si está pasando un vehículo por Ia vía (figuras 7 y 8), el píxel A corresponderá a un punto sobre Ia calzada, mientras que el píxel A' corresponderá a un punto del vehículo. Por este motivo, Ia resta de sus valores producirá como resultado un valor mayor que el umbral t de referencia, dando lugar a una discrepancia.Thus, when no vehicle passes through the road (Figures 5 and 6), the pixel A and the pixel A ', located in the central band 13, will correspond to the same point located on the roadway, and therefore the difference between the values of those pixels will be less than the reference threshold t. However, if a vehicle is passing by the route (Figures 7 and 8), the pixel A will correspond to a point on the road, while the pixel A 'will correspond to a point of the vehicle. For this reason, the subtraction of its values will result in a value greater than the reference threshold t, resulting in a discrepancy.
Sl el número de discrepancias en una región de detección en forma de banda es mayor que un umbral n (idealmente n=1 ) se concluye que existe un vehículo (7) en dicha región de análisis.Sl the number of discrepancies in a band-shaped detection region is greater than a threshold n (ideally n = 1) it is concluded that there is a vehicle (7) in said analysis region.
En caso de que el número de discrepancias sea menor que el umbral n, se realiza un análisis de Ia región en cuestión en las imágenes homologas izquierda y derecha para verificar si se pueden detectar en alguna de ellas las líneas (8, 9) de Ia señalización horizontal de Ia calzada. Si no se pueden detectar las líneas (8, 9), se concluye que Ia imagen está saturada por Ia presencia de un vehículo (7) de mucha altura (por ejemplo, un camión o autocar), por Io que también se concluye que hay un vehículo (7) en Ia región de análisis. Dicho de otro modo, se trata de un vehículo de tamaño tan grande que cubre toda Ia imagen adquirida por las cámaras. En caso de que se puedan detectar las líneas (8, 9) de señalización horizontal de Ia calzada se concluye que no existe ningún vehículo (7) en Ia región de análisis.If the number of discrepancies is less than the threshold n, an analysis of the region in question is carried out in the homologous images left and right to verify if the lines (8, 9) of the horizontal road marking can be detected in any of them. If the lines (8, 9) cannot be detected, it is concluded that the image is saturated by the presence of a vehicle (7) of high height (for example, a truck or coach), so it is also concluded that there is a vehicle (7) in the analysis region. In other words, it is a vehicle of such a large size that it covers all the image acquired by the cameras. In case that the horizontal signaling lines (8, 9) of the road can be detected, it is concluded that there is no vehicle (7) in the analysis region.
El tamaño de estas regiones de análisis (12, 13, 14), así como Ia calidad de las imágenes (número de píxeles), es suficientemente grande para asegurar que recogen un porcentaje amplio del carril por el que pasan los vehículos (7), pero es Io suficientemente pequeño para que Ia carga computacional permita una frecuencia de adquisición de imágenes suficientemente alta.The size of these analysis regions (12, 13, 14), as well as the quality of the images (number of pixels), is large enough to ensure that they collect a large percentage of the lane through which vehicles (7) pass, but it is small enough for the computational load to allow a sufficiently high image acquisition frequency.
Sobre estas regiones de análisis (12, 13, 14) se extraen características que permiten acentuar Ia presencia del vehículo (7), por ejemplo pasando un filtro pasa altos que acentúa los contornos de Ia imagen.On these analysis regions (12, 13, 14) characteristics are extracted that allow accentuating the presence of the vehicle (7), for example by passing a high pass filter that accentuates the contours of the image.
Esta detección es completamente insensible a las sombras y cualquier otro elemento sobre Ia superficie de Ia carretera (charcos, manchas, etc.). En las figuras 10, 11 y 12 se muestra una vista en perspectiva, una planta y un alzado de Ia posición de las cámaras infrarrojasThis detection is completely insensitive to shadows and any other element on the surface of the road (puddles, spots, etc.). In figures 10, 11 and 12 a perspective view, a plan and an elevation of the position of the infrared cameras are shown
(3, 4).(3. 4).
El proceso de análisis combina diferentes resultados para extraer información adicional sobre los vehículos (7) que circulan por Ia escena:The analysis process combines different results to extract additional information about the vehicles (7) that circulate around the scene:
- Analizando las imágenes formadas por los píxeles clasificados como no pertenecientes a Ia calzada se estima Ia altura de los vehículos (7), por ejemplo, calculando Ia correlación cruzada entre las regiones de análisis (12, 13, 14) en las imágenes homologas izquierda y derecha, y estimando Ia altura por métodos trigonométricos sencillos.- Analyzing the images formed by the pixels classified as not belonging to the roadway, the height of the vehicles (7) is estimated, for example, by calculating the cross correlation between the analysis regions (12, 13, 14) in the left and right homologous images, and estimating the height by trigonometric methods simple.
Como muestra las figuras 11 y 12, Ia distancia entre un objeto (15) y las cámaras (3, 4) se puede obtener por triangulación partiendo de Ia diferencia que existe entre su posición en Ia imagen izquierda y su posición en Ia imagen derecha, valor que se estima mediante Ia correlación cruzada. Teniendo esa distancia y conociendo Ia altura h se calcula Ia altura del objeto (15).As shown in Figures 11 and 12, the distance between an object (15) and the cameras (3, 4) can be obtained by triangulation based on the difference between its position in the left image and its position in the right image, value that is estimated by cross correlation. Having that distance and knowing the height h the height of the object (15) is calculated.
- Comparando Ia evolución temporal de los resultados de los análisis anteriores en diferentes regiones de análisis (12, 13, 14) se estima Ia velocidad a Ia que circulan los vehículos (7).- Comparing the temporal evolution of the results of the previous analyzes in different analysis regions (12, 13, 14), the speed at which the vehicles circulate (7) is estimated.
Los mismos métodos trigonométricos que permiten calcular Ia distancia entre un objeto (15) y las cámaras (3, 4), permiten determinar Ia distancia entre las regiones de análisis (12, 13, 14). Si conocemos Ia distancia longitudinal entre dos regiones (12, 13, 14), y conocemos el instante temporal en el que se produce Ia entrada de un vehículo (7) a cada una, podemos estimar su velocidad de paso.The same trigonometric methods that allow to calculate the distance between an object (15) and the cameras (3, 4), allow to determine the distance between the analysis regions (12, 13, 14). If we know the longitudinal distance between two regions (12, 13, 14), and we know the time at which the entry of a vehicle (7) occurs to each one, we can estimate its speed of passage.
- Combinando Ia estimación de Ia velocidad a Ia que circulan los vehículos (7) con Ia medida del tiempo que tardan en atravesar una región de análisis (12, 13, 14) se estima Ia longitud de los vehículos (7).- Combining the estimation of the speed at which the vehicles circulate (7) with the measurement of the time it takes to cross an analysis region (12, 13, 14), the length of the vehicles (7) is estimated.
Finalmente, aunque en este ejemplo se ha descrito una realización preferida de Ia invención, el experto en Ia materia podrá concebir otras realizaciones sin por ello salir del alcance de las reivindicaciones adjuntas. Finally, although in this example a preferred embodiment of the invention has been described, the person skilled in the art will be able to conceive other embodiments without thereby going beyond the scope of the appended claims.

Claims

R F I V I N ΓH Π Δ Π I O N F S RFIVIN ΓH Π Δ Π IONFS
1. Procedimiento de detección del paso de vehículos (7) por una vía, caracterizado porque comprende efectuar de modo repetitivo al menos las siguientes operaciones:1. Procedure for detecting the passage of vehicles (7) in one way, characterized in that it comprises performing at least repetitively the following operations:
adquirir, al menos, un par de imágenes homologas simultáneas de Ia vía mediante, al menos, dos medios de adquisición de imágenes (3, 4) separados una determinada distancia;acquiring, at least, a pair of simultaneous homologous images of the path by means of at least two image acquisition means (3, 4) separated by a certain distance;
restar, mediante un medio de procesamiento (5), para cada píxel de una de las imágenes homologas simultáneas, el valor del píxel homólogo en Ia otra imagen homologa simultánea localizando su posición mediante Ia función de correspondencia Ω;subtract, by means of a processing means (5), for each pixel of one of the simultaneous homologous images, the value of the homologous pixel in the other simultaneous homologous image locating its position by means of the Ω correspondence function;
comparar, mediante el medio de procesamiento (5), los valores obtenidos en Ia operación anterior con un valor umbral predeterminado, obteniéndose el número de discrepancias;compare, by means of the processing means (5), the values obtained in the previous operation with a predetermined threshold value, obtaining the number of discrepancies;
determinar, mediante el medio de procesamiento (5), en función del número de discrepancias, si se produce una detección,determine, by means of the processing means (5), based on the number of discrepancies, if a detection occurs,
donde Ia función correspondencia Ω, que relaciona Ia posición de los píxeles de una imagen homologa con Ia posición de los píxeles de Ia otra imagen homologa, se calcula en una operación previa de inicialización mediante alguna técnica de análisis de correspondencias entre imágenes utilizando elementos fijos de Ia vía como puntos de referencia.where the Ω correspondence function, which relates the position of the pixels of a homologous image with the position of the pixels of the other homologous image, is calculated in a previous initialization operation by some technique of correspondence analysis between images using fixed elements of The path as reference points.
2. Procedimiento de detección del paso de vehículos por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque Ia operación de restar para cada píxel el valor de su píxel homólogo se realiza en regiones de análisis con forma de banda (12, 13, 14) perpendicular a Ia vía.2. Procedure for detecting the passage of vehicles through a route according to any of the preceding claims, characterized in that the operation of subtracting for each pixel the value of its homologous pixel It is performed in regions of band-shaped analysis (12, 13, 14) perpendicular to the pathway.
3. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con Ia reivindicación anterior, caracterizado porque las operaciones anteriores se realizan 50 o más veces por segundo.3. Procedure for detecting the passage of vehicles (7) by a route according to the preceding claim, characterized in that the above operations are performed 50 or more times per second.
4. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además Ia operación de emitir en dirección a Ia vía, mediante un medio emisor de radiación (2), una radiación adecuada para el funcionamiento de los medios de adquisición de imágenes (3, 4).4. Method for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it also comprises the operation of emitting in the direction of the route, by means of a radiation emitting means (2), a radiation suitable for the operation of the image acquisition means (3, 4).
5. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además Ia operación de determinar Ia hora en que se realiza una detección.5. Procedure for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it also comprises the operation of determining the time at which a detection is performed.
6. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además determinar el lugar geográfico que el que se realiza una detección.6. Method for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it further comprises determining the geographical location that the detection is performed.
7. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además Ia operación de comunicar los resultados obtenidos mediante un medio de comunicación (6).7. Procedure for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it further comprises the operation of communicating the results obtained by means of a communication means (6).
8. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende Ia operación de encriptar los resultados previamente a su envío mediante el medio de comunicación (6).8. Procedure for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it comprises the operation of encrypting the results prior to its sending by means of communication (6).
9. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además Ia operación de determinar Ia altura del vehículo (7) detectado.9. Method for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it further comprises the operation of determining the height of the vehicle (7) detected.
10. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además Ia operación de determinar Ia longitud del vehículo (7) detectado.10. Procedure for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it further comprises the operation of determining the length of the vehicle (7) detected.
11. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque comprende además Ia operación de determinar Ia velocidad del vehículo (7) detectado.11. Method for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that it further comprises the operation of determining the speed of the vehicle (7) detected.
12. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado porque se utilizan las líneas de señalización horizontal de Ia vía como puntos fijos de referencia para el cálculo de Ia función a.12. Method for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that the horizontal signaling lines of the route are used as fixed reference points for the calculation of the function a.
13. Procedimiento de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones anteriores, caracterizado por el hecho de que dichos dos medios de adquisición de imágenes (3,4) están separados una determinada distancia horizontalmente.13. Method for detecting the passage of vehicles (7) by a route according to any of the preceding claims, characterized in that said two image acquisition means (3,4) are separated a certain distance horizontally.
14. Sistema (1 ) de detección del paso de vehículos (7) por una vía, caracterizado porque comprende los siguientes dispositivos:14. System (1) for detecting the passage of vehicles (7) in one way, characterized in that it comprises the following devices:
dos medios de adquisición de imágenes (3, 4), que adquieren dos imágenes homologas simultáneas;two means of image acquisition (3, 4), which acquire two simultaneous homologous images;
un medio de procesamiento (5) conectado a los medios de adquisición de imágenes (3, 4), que resta, para cada píxel de una de las imágenes homologas simultáneas, el valor del píxel homólogo en Ia otra imagen homologa simultánea y determina, en función del número de discrepancias, si se produce una detección.a processing means (5) connected to the image acquisition means (3, 4), which subtracts, for each pixel of one of the simultaneous homologous images, the value of the homologous pixel in the other simultaneous homologous image and determines, in depending on the number of discrepancies, if a detection occurs.
15. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con Ia reivindicación 14, donde los medios de adquisición de imágenes (3, 4) están separados horizontalmente una distancia entre 20 cm y 100 cm.15. System (1) for detecting the passage of vehicles (7) by a route according to claim 14, wherein the image acquisition means (3, 4) are horizontally separated by a distance between 20 cm and 100 cm.
16. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14 ó 15, caracterizado porque además comprende un medio emisor de radiación (2), que emite en dirección a Ia vía radiación una radiación adecuada para los medios de adquisición de imágenes (3, 4).16. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14 or 15, characterized in that it also comprises a radiation emitting means (2), which emits in the direction of the radiation path adequate radiation for the image acquisition means (3, 4).
17. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-15, caracterizado porque comprende además un sistema (67) de posicionamiento GPS para determinar Ia posición geográfica donde se produce una detección.17. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-15, characterized in that it further comprises a GPS positioning system (67) for determining the geographical position where a detection.
18. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-16, caracterizado porque comprende además un medio (66) para determinar Ia fecha y Ia hora en que se produce una detección.18. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-16, characterized in that it also comprises a means (66) for determining the date and time at which a detection.
19. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-17, caracterizado porque comprende además un medio de comunicación (6) para comunicar los resultados obtenidos.19. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-17, characterized in that It also includes a means of communication (6) to communicate the results obtained.
20. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-18, caracterizado porque los medios de adquisición de imágenes (3, 4) se eligen de Ia siguiente lista: cámaras digitales, cámaras de vídeo digitales, cámaras infrarrojas, detectores de ultrasonidos y detectores de radar.20. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-18, characterized in that the image acquisition means (3, 4) are chosen from the following list: cameras digital, digital video cameras, infrared cameras, ultrasonic detectors and radar detectors.
21. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-19, caracterizado porque el medio emisor de radiación emite radiación de un tipo de Ia siguiente lista: luz visible, luz infrarroja, ultrasonidos y radar.21. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-19, characterized in that the radiation emitting means emits radiation of a type from the following list: visible light, light infrared, ultrasound and radar.
22. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-20, caracterizado porque el medio de procesamiento (5) se elige de entre Ia siguiente lista: un microcontrolador, un ordenador, un ASIC, un DSP y una FPGA.22. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-20, characterized in that the processing means (5) is chosen from the following list: a microcontroller, a computer, an ASIC, a DSP and an FPGA.
23. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14-21 , caracterizado porque el medio de comunicación (5) se comunica mediante al menos uno de los modos de Ia siguiente lista: Internet, Ethernet, USB, RS232, RS485, Bluetooth, IEEE802.Hb/g, Zigbee, radio, línea telefónica, UMTS, GSM o infrarrojos.23. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14-21, characterized in that the communication means (5) is communicated by at least one of the modes of the following list: Internet, Ethernet, USB, RS232, RS485, Bluetooth, IEEE802.Hb / g, Zigbee, radio, telephone line, UMTS, GSM or infrared.
24. Sistema (1 ) de detección del paso de vehículos (7) por una vía de acuerdo con cualquiera de las reivindicaciones 14 a 23, caracterizado por el hecho de que el medio de procesamiento (5) está configurado para:24. System (1) for detecting the passage of vehicles (7) by a route according to any of claims 14 to 23, characterized in that the processing means (5) is configured to:
- comparar, mediante el medio de procesamiento (5), los valores obtenidos en Ia operación de resta con un valor umbral predeterminado, obteniéndose el número de discrepancias;- compare, by means of the processing means (5), the values obtained in the subtraction operation with a predetermined threshold value, obtaining the number of discrepancies;
- determinar, mediante el medio de procesamiento (5), en función del número de discrepancias, si se produce una detección,- determine, by means of the processing means (5), based on the number of discrepancies, if a detection occurs,
Y para calcular, en una operación previa de inicialización, mediante alguna técnica de análisis de correspondencias entre imágenes utilizando elementos fijos de Ia vía como puntos de referencia, Ia función correspondencia Ω, que relaciona Ia posición de los píxeles de una imagen homologa con Ia posición de los píxeles de Ia otra imagen homologa. And to calculate, in a previous initialization operation, by some technique of correspondence analysis between images using fixed elements of the pathway as reference points, the Ω correspondence function, which relates the position of the pixels of a homologous image with the position of the pixels of the other homologous image.
PCT/IB2008/055566 2007-12-31 2008-12-29 Method and system for detecting moving objects WO2009087536A2 (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
ES200703498A ES2330499B1 (en) 2007-12-31 2007-12-31 PROCEDURE AND SYSTEM FOR DETECTION OF MOVING OBJECTS.
ESP200703498 2007-12-31

Publications (2)

Publication Number Publication Date
WO2009087536A2 true WO2009087536A2 (en) 2009-07-16
WO2009087536A3 WO2009087536A3 (en) 2009-09-03

Family

ID=40761915

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/IB2008/055566 WO2009087536A2 (en) 2007-12-31 2008-12-29 Method and system for detecting moving objects

Country Status (2)

Country Link
ES (1) ES2330499B1 (en)
WO (1) WO2009087536A2 (en)

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN102368356A (en) * 2011-10-24 2012-03-07 姜廷顺 Monitoring system and method used for freeway ramp mouth
CN102394012A (en) * 2011-06-21 2012-03-28 陆化普 Monitoring system having preset bit function for vehicle crossing over markings and method thereof
US8242476B2 (en) 2005-12-19 2012-08-14 Leddartech Inc. LED object detection system and method combining complete reflection traces from individual narrow field-of-view channels
US8436748B2 (en) 2007-06-18 2013-05-07 Leddartech Inc. Lighting system with traffic management capabilities
US8600656B2 (en) 2007-06-18 2013-12-03 Leddartech Inc. Lighting system with driver assistance capabilities
US8723689B2 (en) 2007-12-21 2014-05-13 Leddartech Inc. Parking management system and method using lighting system
US8842182B2 (en) 2009-12-22 2014-09-23 Leddartech Inc. Active 3D monitoring system for traffic detection
US8908159B2 (en) 2011-05-11 2014-12-09 Leddartech Inc. Multiple-field-of-view scannerless optical rangefinder in high ambient background light
US9235988B2 (en) 2012-03-02 2016-01-12 Leddartech Inc. System and method for multipurpose traffic detection and characterization
US9378640B2 (en) 2011-06-17 2016-06-28 Leddartech Inc. System and method for traffic side detection and characterization
US10488492B2 (en) 2014-09-09 2019-11-26 Leddarttech Inc. Discretization of detection zone
USRE49342E1 (en) 2007-12-21 2022-12-20 Leddartech Inc. Distance detection method and system

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112489447A (en) * 2020-11-24 2021-03-12 杭州海康威视数字技术股份有限公司 Method, device and system for detecting vehicle running speed and electronic equipment

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19517028A1 (en) * 1995-05-10 1996-11-14 Bosch Gmbh Robert Process for creating an object mask for video surveillance systems
EP1074958A1 (en) * 1999-07-23 2001-02-07 Matsushita Electric Industrial Co., Ltd. Traffic congestion measuring method and device and applications thereof
US6198987B1 (en) * 1997-04-18 2001-03-06 Samsung Electronics Co., Ltd. Method and a multi-functional apparatus for determining the class of a vehicle
DE10022218A1 (en) * 2000-05-04 2001-11-08 Oezkan Mustafa Imaging sensor for road junction control compares detected image with stored image to improve traffic flow
EP1395059A2 (en) * 2002-08-30 2004-03-03 Fuji Jukogyo Kabushiki Kaisha Intruding-object detection apparatus
US20040125207A1 (en) * 2002-08-01 2004-07-01 Anurag Mittal Robust stereo-driven video-based surveillance

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP4003623B2 (en) * 2002-11-19 2007-11-07 住友電気工業株式会社 Image processing system using a pivotable surveillance camera
JP4537794B2 (en) * 2004-08-02 2010-09-08 株式会社東芝 Image processing apparatus and image processing program
NL1031867C1 (en) * 2005-07-08 2007-01-09 Everhardus Fransiscu Weijdeven Method for determining vehicle data.
JP4973008B2 (en) * 2006-05-26 2012-07-11 富士通株式会社 Vehicle discrimination device and program thereof

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
DE19517028A1 (en) * 1995-05-10 1996-11-14 Bosch Gmbh Robert Process for creating an object mask for video surveillance systems
US6198987B1 (en) * 1997-04-18 2001-03-06 Samsung Electronics Co., Ltd. Method and a multi-functional apparatus for determining the class of a vehicle
EP1074958A1 (en) * 1999-07-23 2001-02-07 Matsushita Electric Industrial Co., Ltd. Traffic congestion measuring method and device and applications thereof
DE10022218A1 (en) * 2000-05-04 2001-11-08 Oezkan Mustafa Imaging sensor for road junction control compares detected image with stored image to improve traffic flow
US20040125207A1 (en) * 2002-08-01 2004-07-01 Anurag Mittal Robust stereo-driven video-based surveillance
EP1395059A2 (en) * 2002-08-30 2004-03-03 Fuji Jukogyo Kabushiki Kaisha Intruding-object detection apparatus

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8242476B2 (en) 2005-12-19 2012-08-14 Leddartech Inc. LED object detection system and method combining complete reflection traces from individual narrow field-of-view channels
US8436748B2 (en) 2007-06-18 2013-05-07 Leddartech Inc. Lighting system with traffic management capabilities
US8600656B2 (en) 2007-06-18 2013-12-03 Leddartech Inc. Lighting system with driver assistance capabilities
US8723689B2 (en) 2007-12-21 2014-05-13 Leddartech Inc. Parking management system and method using lighting system
USRE49342E1 (en) 2007-12-21 2022-12-20 Leddartech Inc. Distance detection method and system
US8842182B2 (en) 2009-12-22 2014-09-23 Leddartech Inc. Active 3D monitoring system for traffic detection
US8908159B2 (en) 2011-05-11 2014-12-09 Leddartech Inc. Multiple-field-of-view scannerless optical rangefinder in high ambient background light
USRE47134E1 (en) 2011-05-11 2018-11-20 Leddartech Inc. Multiple-field-of-view scannerless optical rangefinder in high ambient background light
USRE48763E1 (en) 2011-05-11 2021-10-05 Leddartech Inc. Multiple-field-of-view scannerless optical rangefinder in high ambient background light
US9378640B2 (en) 2011-06-17 2016-06-28 Leddartech Inc. System and method for traffic side detection and characterization
CN102394012A (en) * 2011-06-21 2012-03-28 陆化普 Monitoring system having preset bit function for vehicle crossing over markings and method thereof
CN102368356A (en) * 2011-10-24 2012-03-07 姜廷顺 Monitoring system and method used for freeway ramp mouth
US9235988B2 (en) 2012-03-02 2016-01-12 Leddartech Inc. System and method for multipurpose traffic detection and characterization
USRE48914E1 (en) 2012-03-02 2022-02-01 Leddartech Inc. System and method for multipurpose traffic detection and characterization
US10488492B2 (en) 2014-09-09 2019-11-26 Leddarttech Inc. Discretization of detection zone

Also Published As

Publication number Publication date
ES2330499B1 (en) 2010-09-21
WO2009087536A3 (en) 2009-09-03
ES2330499A1 (en) 2009-12-10

Similar Documents

Publication Publication Date Title
ES2330499B1 (en) PROCEDURE AND SYSTEM FOR DETECTION OF MOVING OBJECTS.
ES2352300T3 (en) SYSTEM TO DETECT VEHICLES.
US11753781B2 (en) Apparatus, system, and method for smart roadway stud control and signaling
ES2724353T3 (en) System and methods for the support of autonomous vehicles through environmental perception and sensor calibration and verification
US20200124430A1 (en) Detecting types of travel corridors on which personal mobility vehicles travel
CA2998166C (en) System and method for vehicle detection
US20140313504A1 (en) Procedure and device for the measurement of dynamic luminance and retroreflection of road markings and signals and the obtention of the shape, position and dimensions of the same
ES2359852T3 (en) METHOD AND SYSTEM FOR THE MAPPING OF REACH SENSOR DATA ON IMAGE SENSOR DATA.
US20140112537A1 (en) Systems and methods for intelligent monitoring of thoroughfares using thermal imaging
JP6456838B2 (en) Street device health monitoring
JP2011027594A (en) Map data verification system
JP5734521B2 (en) In-vehicle device and center device
CN110648528A (en) Wisdom highway management system
ES2645504T3 (en) Determination of at least one characteristic of a vehicle
KR101287585B1 (en) Crossroad.crosswalk traffic control system
EP3058508A1 (en) Method and system for determining a reflection property of a scene
KR101475453B1 (en) Jaywalking detection system
ES2743605T3 (en) Information sharing system that uses hygrometry and road temperature sensors
JP3781281B2 (en) Method and apparatus for measuring route facilities
KR101917297B1 (en) System and method for collecting traffic information
CN110462698A (en) The method of lighting system and control lighting system
KR102588554B1 (en) Automotive sensor integration module
KR102148826B1 (en) Speed-sensitive Road transversal person Warning Device
CN115968489A (en) Roadside detection and alarm system and method
ES2561912B1 (en) Procedure for monitoring and safety management in crosswalks and device for its realization

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 08869504

Country of ref document: EP

Kind code of ref document: A2

122 Ep: pct application non-entry in european phase

Ref document number: 08869504

Country of ref document: EP

Kind code of ref document: A2