WO2003019234A1 - Verfahren zur erkennung und verfolgung von objekten - Google Patents
Verfahren zur erkennung und verfolgung von objekten Download PDFInfo
- Publication number
- WO2003019234A1 WO2003019234A1 PCT/EP2002/009362 EP0209362W WO03019234A1 WO 2003019234 A1 WO2003019234 A1 WO 2003019234A1 EP 0209362 W EP0209362 W EP 0209362W WO 03019234 A1 WO03019234 A1 WO 03019234A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- box
- pixels
- object box
- objects
- segment
- Prior art date
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
- G01S7/48—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00
- G01S7/4802—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S17/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/66—Radar-tracking systems; Analogous systems
- G01S13/72—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar
- G01S13/723—Radar-tracking systems; Analogous systems for two-dimensional tracking, e.g. combination of angle and range tracking, track-while-scan radar by using numerical data
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S17/00—Systems using the reflection or reradiation of electromagnetic waves other than radio waves, e.g. lidar systems
- G01S17/88—Lidar systems specially adapted for specific applications
- G01S17/93—Lidar systems specially adapted for specific applications for anti-collision purposes
- G01S17/931—Lidar systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/58—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
- G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
- G01S13/88—Radar or analogous systems specially adapted for specific applications
- G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
- G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
- G01S2013/9327—Sensor installation details
- G01S2013/93271—Sensor installation details in the front of the vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/08—Detecting or categorising vehicles
Definitions
- the present invention relates to a method for the detection and tracking of objects on the basis of images of objects in a field of view of the sensor that are detected by a sensor for detecting electromagnetic radiation, in particular a laser scanner, and which contain depth-resolved images, in which the detection and Tracking is carried out using models associated with the objects for the temporal development of dynamic state variables, each of which describes a temporally variable state of the object and comprises dynamic state variables for determining the position of the respective object, and in the case of which objects are assigned an assignment of pixels of a current image for at least one object.
- a generic method according to the invention is characterized in that at least one object is assigned as a static property an object box that represents an expansion of the object in the image plane, has a specific shape and size, and corresponds in shape and size to the object represented by the object , and that the dynamic state variables for determining the position of the object comprise the coordinates of a reference point of the object box, via which the position of the object can be determined.
- a depth-resolved image of a sensor is understood to be a set of image points detected when the visual range of the sensor is scanned, to which points or, depending on the resolution of the sensor, also regions of an object detected by the sensor, the image points corresponding to the position of the associated object points in at least two dimensions, both of which are not perpendicular to the viewing direction of the sensor.
- the pixels can furthermore contain data about further, in particular optical, properties of the object points, for example their reflectivity.
- This data is referred to below as the optical properties of the pixels.
- sensors for electromagnetic radiation for detecting such deeply resolved images are known. These can preferably be optoelectronic sensors which offer good spatial resolution and are therefore preferred for the method according to the invention.
- systems with stereo video cameras can be used which have a device for converting the raw data recorded by the cameras into depth-resolved images.
- laser scanners are preferably used which, during a scan, scan a visual range with at least one pulsed radiation beam, which sweeps over a predetermined angular range and detect, mostly diffuse, reflected radiation pulses of the radiation beam from a point or area of an object.
- the transit time of the emitted, reflected and detected radiation pulses is recorded for distance measurement.
- the raw data recorded in this way for a pixel can then contain as coordinates the angle at which the reflection was recorded and the distance from the object point determined from the transit time of the radiation pulses.
- the radiation can in particular be visible or infrared light.
- Dynamic state variables are understood to mean variables which each describe a state of the object which changes over time and which at least approximately corresponds to a state of the object which is also variable over time and which is represented by the object. These dynamic state variables therefore do not include variables that may be changeable over time within the scope of the method, but whose variability is not based on a corresponding variability of the object, but rather only on a change due to detection, in particular a change in the representation of the object , Since objects should be tracked, the dynamic state variables include in particular dynamic state variables for determining the position of the respective object.
- the detection and tracking of objects takes place on the basis of models with which the temporal development of the dynamic state variables can be described at least approximately, whereby the type and complexity of the models inter alia. according to the object detection and tracking process and the type of typically occurring objects to be tracked.
- the pixels of the image are assigned to at least one object.
- an object box is assigned to at least one object as a static property, which represents an expansion of the object in the image plane and which has a specific shape and size.
- the object box thus represents a geometric structure in the level of the depth-resolved image with a certain shape and size.
- the object box thus essentially describes the outline of the object represented by the object in the level of the depth-resolution image, which is not variable or assumed to be variable in time ,
- the correspondence can only be very rough, in particular the shape of the object box can represent a very extensive simplification of the shape of the object, for example a simple geometric basic shape.
- the object box itself therefore represents neither a dynamic state nor a dynamic state variable in the sense mentioned above.
- the dynamic state variables further comprise the coordinates of a reference point of the object box, via which the position of the object box and thus of the object can be determined. This enables object tracking via the dynamic state variables for the coordinates of the reference point.
- the reference point of the object box can be any point, the position of which is fixed relative to the object box for the duration of the method.
- the geometric center of gravity of the object box can be provided as the reference point.
- a major advantage of this method compared to other methods is that the position of the object in a current image can be determined within wide limits regardless of the number of pixels assigned to the object, provided that enough pixels are available to determine the position of the object box , If instead, for example, the geometric center of gravity of the pixels assigned to the object were used as the position of the object, rapid fluctuations in the view of the object could result in large fluctuations or jumps in the measured position, which would result in comparatively large fluctuations in the predicted or estimated position of the object.
- the object box or the object represented by the object is assigned an axis - net, and that the dynamic state variables comprise an orientation variable, via which the orientation of the axis and thus the object box to a predetermined reference axis can be determined.
- the reference axis can in particular be a fixed axis with respect to the sensor used. In this way, the orientation of an object can also be easily tracked largely independently of the number of pixels assigned to the object in a current image. It is particularly important here that large fluctuations or jumps in the position of the reference point of the object box, in particular the center of gravity, can be avoided if the orientation of the object box or the object is represented as a rotation about the reference point.
- a search area in the current image is preferably defined as a function of the predicted values of the dynamic state variables determining the object position, in which search is carried out for pixels for the object and its shape and size from at least one property of the object box depends. At least for the dynamic state variables describing the position of the object, new values are then determined using the position of the pixels assigned to this object in the current image. In particular, values of the dynamic state variables from previous iterations can be used.
- the search area can be selected to be larger in accordance with the object box, so that the search area for the reference point, which can be smaller than the extension of the object box, does not have to be artificially expanded unnecessarily.
- a Kaiman filter can be used with particular preference in which the search area is then determined via predicted values of measured variables and predicted variances or covariances of the measured variables, the measured variables comprising the dynamic state variables which determine the position and, if appropriate, the orientation of the object box, or of depend clearly on this.
- the search area can result from first determining a search area for the reference point, preferably the center of gravity of the object box or the object, and then guiding the reference point in the object box with the object box along the edge of the search area for the reference point, the Search area for the reference point together with the area covered by the object box determines the search area.
- the algorithm for determining the search area of the reference point can be designed independently of the shape and size of the object box, so that a particularly straightforward method results.
- segments are formed for the detection of objects from the pixels of the current image, each comprising a single pixel or a set of all pixels, of which at least two each correspond to at least one segmentation criterion as belonging to the segment are defined, and which do not have any common pixels, that for at least one object, depending on the predicted values in the state variable determining the object or object box position, a search area in the current image is defined by searching for pixels to be assigned to the object for the object, and the position and / or orientation of the object box is determined such that at least one segment that is in the search area also lies in the object box.
- This segmentation criterion can here and in the following in particular be a criterion for a maximum permissible distance and / or a maximum permissible difference of an optical property, in particular the reflectivity.
- an inclination of a surface determined from pulse widths and / or heights can also be used.
- the object box is positioned in a variant of the invention in such a way that a segment in the search area in the object box.
- the position and / or orientation of the object box can be determined in such a way that a number of segments in the search area may also lie in the object box. This can in particular facilitate the assignment of several segments to one object.
- the position and / or orientation of the object box is determined as a function of the distances between pixels of segments lying in the object box and in the search area and the edge of the object box.
- the distance can in particular be the distance of the image point from the edge of the object box along a perpendicular to a tangent to the edge of the object box.
- the position and / or orientation of the object box can preferably be minimized by minimizing the distances between pixels of segments lying in the object box and in the search area and the edge of the object box.
- the distances can be weighted differently, for example depending on the direction, distance or quality of a pixel.
- the assignment of segments to an object box or the corresponding object can take place in particular as a function of the average distance of the pixels of a segment from the edge of the object box.
- an assignment only takes place if the average distance exceeds a predetermined threshold value.
- the average distance used for the assignment is based on an extension of the object box, in the case of a rectangle, for example, the length or width.
- At least one object box has an edge strip of a predetermined width running along its circumference within the object box, and for object segments to be assigned the object box is positioned such that the number of pixels in the edge strip is maximized.
- object classes to be assigned to objects are defined, for each of which an object box with an object class-specific shape and size is defined, and that at least one object during or after its formation, one object class and one for the object class is assigned to a specific object box.
- the object class can preferably be determined in that it corresponds to objects with certain geometric properties, that is to say in particular certain shapes and sizes.
- the shapes can only be very abstract and therefore simply specified and the sizes lie within a specified value range.
- object classes to be assigned to objects are defined that correspond to objects with certain shapes and sizes, that the shape and / or size of the object box can be parameterized by at least one static status parameter corresponding to the shapes and sizes of the objects corresponding to the object class, and that at least one object is assigned an object class and an object box according to the object class-specific shape and a predetermined initial value of the static state parameter during or after its formation.
- the object classes can preferably further serve to define the models used for object tracking, as described in the applicant's German patent application filed on the same day as the present application with the attorney's file number S7880 and the title "Method for recognizing and tracking objects” is the content of which is hereby incorporated by reference into the description.
- segments are formed before the object formation, each of which comprises a single pixel or a set of all pixels, at least two of which are defined as belonging to the segment in accordance with at least one segmentation criterion, and which have no common pixels with one another that an object is created on the basis of at least one segment and that the object has an object class and one of the object class is assigned to the corresponding object box that completely accommodates the segment.
- segments facilitates both the selection of a suitable object class and the determination of the position and, if necessary, orientation of the object box.
- the position and, if appropriate, orientation can be determined using the developments of the method according to the invention described above.
- the initial value can in particular be determined in such a way that the extent of the corresponding object box corresponds to the most frequent or average extent of objects corresponding to the object class.
- the shape and / or size of the object box can be parameterized by at least one static status parameter, and that segments are formed from the pixels of the image, each comprising a single pixel or a set of all pixels, of which at least two each are defined as belonging to the segment in accordance with at least one segmentation criterion and which do not have any common pixels.
- segments are formed from the pixels of the image, each comprising a single pixel or a set of all pixels, of which at least two each are defined as belonging to the segment in accordance with at least one segmentation criterion and which do not have any common pixels.
- the adaptation is carried out by changing the value of at least one corresponding static status parameter. Detection of a deviation of the extent of the object from the corresponding extent of the object box can preferably take place in that the boundaries of the object are reliably detected. This can be done by Corners are detected or, after a partial obscuring of the ends of the contour of the object visible in the image by another object lying between the sensor and the object, via the detected size of the object.
- a static state parameter parameterizing the object box is adapted so that the segment is completely in the adapted object box is located, the assignment preferably taking place only if the object box for the object meets at least one predetermined consistency condition.
- Object classes to be assigned to objects are particularly preferably defined, for each of which an object box with an object class-specific shape and a range of permissible values are defined for at least one static state parameter that parameterizes the object box. Furthermore, at least one consistency condition is provided for at least one object class with respect to a static state parameter that parameterizes the expansion of the object box.
- the object classes are defined via minimum and maximum extents of objects to which the objects of the object class correspond, these minimum and maximum extents can be used for the consistency check. For example, a segment with a length of 10 m should not be used to adapt an object box Lead object of the object class passenger cars. In order to avoid such problems, the consistency condition is preferably checked.
- the shape of the object box is a rectangle, since at least most vehicles and also people can be represented by rectangles at least in rough approximation.
- rectangles are particularly easy to handle in the operations of the method.
- the object box can particularly preferably be parameterized by a static state parameter for its length and one for its width.
- the object classes can in particular be defined by value ranges for the width and length of the objects that are to be represented by objects of the object class.
- the corresponding object boxes then have a shape as a rectangle, the size of which is parameterized by the static state parameters for the width and length, the values of the static state parameters having to lie in value ranges which correspond to those of the object class definition.
- segments are formed for the detection of objects, each comprising a single pixel or a set of all pixels, of which at least two are defined as belonging to the segment in accordance with at least one segmentation criterion and which do not have any common pixels, and that a segment-object assignment for at least one object is assigned a quality upon detection, which is determined as a function of the maximum possible number of pixels in the object box and the number of pixels actually found in the object box.
- the maximum possible number of pixels can be determined on the basis of the position of the object box relative to the sensor and the resolving power of the sensor at the given object box position, or theoretically by variation over all possible positions of the object box relative to the sensor.
- At least one object box has an edge strip of a predetermined width running along its circumference within the object box, and a quality of the assignment of pixels depending on the pixels located in the edge strip and the total number of pixels located in the object box is determined.
- the ratio of the two values can be used for this.
- a quality can be determined very quickly, which indicates how exactly the pixels, which should correspond to the contour of the object, are indicated along the contour of the object box. are arranged.
- a segmentation can also particularly preferably be carried out or used here, resulting in a quality of the segment-object assignment.
- a masking detection is carried out, in which it is determined whether a part of the object or the object represented by it from the direction of view of the sensor through another object or another through this other object object shown is covered, and that when a cover is detected, the pixels missing due to the cover are taken into account when calculating the quality. This can be done either by reconstructing the corresponding pixels and including them in the calculation, or by reducing the maximum number of pixels by the number of hidden pixels while determining the quality. This also makes it easy to see whether the extent of an object represented by the object corresponds to the extent of the object box or not even in the case of concealment.
- segments are formed for the detection of objects, each comprising a single image point or a set of all image points, of which at least two are defined as belonging to the segment in accordance with at least one segmentation criterion, and none of them have common pixels, and that a segment-object assignment for at least one object is assigned a quality upon detection which depends on the minimum distances between pixels located in the object box and the edge of the object box and on the number of pixels in the object box lying points is determined.
- the normalized mean can be particularly preferred simple or quadratic distance can be used to determine a quality value. In this way, it can in particular be easily determined to what extent the pixels assigned to an object also lie on the edge of the object box or in the vicinity thereof. This allows, for example, actually substantially rectangular objects such as. B. cars from other objects such. B. differentiate bushes, since the latter have segments with pixels, which may also be in large numbers within the object box.
- the distances can be particularly preferably related to an expansion of the object box, so that comparable quality values result for both small and large object boxes.
- An object quality which is determined as a function of the quality of at least the last segment-object assignment, can particularly preferably be assigned to the object as a property. A subsequent application can use this object property to recognize the reliability of an object determination.
- the object quality can preferably be determined as a moving, optionally weighted mean value of the qualities determined for several segment-object assignments.
- Another object of the invention is a computer program with program code means to carry out the inventive method when the program is executed on a computer.
- the invention also relates to a computer program product with program code means on a computer-readable data carrier are stored in order to carry out the method according to the invention when the computer program product is executed on a computer.
- a computer is understood here to mean any data processing device with which the method can be carried out.
- these can have digital signal processors and / or microprocessors with which the method is carried out in whole or in part.
- Another object of the invention is a method for recognizing and tracking objects in a field of view of a sensor for detecting electromagnetic radiation, preferably an optoelectronic sensor, in particular a laser scanner, in which images of at least part of its field of vision that are successively depth-resolved in time are recorded, and in which the captured images are processed with the inventive method for recognizing and tracking objects.
- a sensor for detecting electromagnetic radiation preferably an optoelectronic sensor, in particular a laser scanner, in which images of at least part of its field of vision that are successively depth-resolved in time are recorded, and in which the captured images are processed with the inventive method for recognizing and tracking objects.
- the invention relates to a device for recognizing and tracking objects with at least one sensor designed to detect depth-resolved images of a monitoring area for detecting electromagnetic radiation, preferably an optoelectronic sensor, in particular a laser scanner, and with a data processing device designed to carry out the method according to the invention is.
- FIG. 1 is a schematic plan view of a scene with a vehicle on which a laser scanner is attached and two vehicles in the field of view of the laser scanner,
- FIG. 2 shows a flowchart of a method for recognizing and tracking objects according to a preferred embodiment of the invention
- FIG. 3 shows the scene in FIG. 1 at a later point in time
- FIG. 5 shows a schematic top view of a scene with a vehicle to which a laser scanner is attached and two vehicles in the field of view of the laser scanner, and
- Fig. 6 shows the scene in Fig. 5 at a later time.
- a laser scanner 12 is arranged on the front of a vehicle 10, the viewing area 14 of which comprises the area lying in front of the vehicle in the direction of travel.
- the viewing area 14 there are a truck 16 and a two-wheeler 18, which are represented by the rectangles with narrow dotted lines.
- the laser scanner 12 scans its viewing area 14 with a rotating, pulsed radiation beam 20 and detects the position of object points, for example the object points 22 and 24 on the truck 16 or the two-wheeler 18, at which a pulse of the radiation beam was reflected, on the basis of the angle at which the radiation pulse was sent or received and the distance to the laser scanner 12, which results from the transit time of the radiation pulse from the radiation to the reception after reflection.
- a depth-resolved image is thus acquired, which comprises the image points acquired during the scanning, which contain at least the coordinates of the object points 22 and 24 at which the radiation beam 18 was reflected.
- the pixels are represented in the figures by the item points.
- the complete images are updated at a time interval T.
- the images captured by the laser scanner 12 are output, possibly after a correction of the coordinates of the image points, to a data processing device 26 in the vehicle 10, in which the images are further processed.
- the data processing device 26 not shown in the figures, has a processor and a memory as well as an interface for transmitting the data from the laser scanner. Furthermore, an interface is provided for outputting data to other data processing devices in the vehicle 10, which, however, are also not shown in the figures.
- an object recognition and tracking method is carried out in the data processing device 26, in which a corresponding program is stored and executed.
- a Kaiman filter is used for object detection and tracking, the center of gravity coordinates of the object being used as dynamic state variables in this exemplary embodiment. These coordinates of the center of gravity are also used at the same time as measured values in the Kaiman filtering. The process flow is roughly shown in the flow chart in FIG. 2.
- step S10 After a new depth-resolved image has been captured by the laser scanner 12, the image is read in in step S10.
- pixels of the image are combined into segments in step S12.
- a pixel belongs to a segment if it is at a distance from at least one other pixel of the segment that is smaller than a predetermined maximum distance.
- the formation of segments and in particular the choice of distance criteria can be made using a Euclidean distance, at which the differences between the coordinates of the image point in the direction of travel direction may be weighted less than transverse to the travel direction, or two distances in each of the coordinate directions of a given one Coordinate system, for example a Cartesian coordinate system or a polar coordinate system.
- step S14 depending on predicted values of the dynamic state variables, that is, when using the Kaiman filter more precisely, on predicted values of the corresponding measured variables and, for example, predicted variances and covariances of the measured variables, a search range for each object is defined, in the search for pixels or segments are sought which are to be assigned to the object to be recognized.
- step S 16 segments and objects are then assigned. This assignment results in new values of the measured variables and in particular new values of the coordinates of the center of gravity of the object.
- step S 18 new estimated values for the dynamic state variables of the detected, existing objects are then calculated as a function of previous values of the dynamic state variables, the new values of the measured variables and further variables, such as, for example, variances and covariances of the dynamic state variables and the measured variables.
- step S20 If objects could not be assigned to all segments during object recognition, new objects are formed from unassigned segments in step S20.
- the properties of these objects and in particular initial values of their dynamic state variables result at least in part from the properties of the segments on which the objects are based.
- the data relating to the objects, in particular their number and center of gravity coordinates, are then calculated and stored in step S22, if not already done, or output to subsequent applications.
- the method then continues after the acquisition of a new depth-resolved image by the laser scanner 12 with step S10.
- FIG. 1 and 3 show, among other things, the result of an object detection and tracking, in which the center of gravity of the object is formed by the center of gravity of the segment assigned to the object.
- part of the truck 16 is covered by the two-wheeler 18, so that there are only a few pixels along the long side of the truck 16.
- segments 28 and 30 result in FIG. 1, which are each represented by the picture elements 22 and 24, which are connected to each other by solid lines.
- Fig. 3 in which the same vehicles are shown at a later point in time, the two-wheeler 18 is now so far in front of the truck 16 that its entire long side is visible to the laser scanner 12, so that there is now a substantially larger number of objects - or pixels 32 for the truck 16 results, while only two pixels 34 are detected for the two-wheeler.
- the pixels 32 now result in a significantly larger segment 36 which can be assigned to the truck 16.
- the positions of the center of gravity are the positions identified by points 42 and 44, respectively.
- the truck 16 in the image in FIG. 3 has practically not moved relative to the vehicle 10 compared to the image in FIG. 1, as shown in FIG. 3 there is a comparatively large jump in the measured coordinates of the object, that doesn't correspond to reality.
- three object classes are provided in the example, which are distinguished by the dimensions of the objects corresponding to the objects in the object classes.
- an object class for two-wheelers which corresponds to objects which, represented by a rectangle, have a length between 1.5 m and 2.5 m and a width between 0.3 m and 1.2 m.
- an object class for passenger cars is provided which corresponds to objects which, represented by a rectangle, have a length between 3 m and 6 m and a width between 1.5 m and 2 m.
- an object class for trucks is provided which corresponds to objects which, represented by a rectangle, have a length between 6 m and 15 m and a width between 2 m and 3 m.
- Objects of these object classes are each assigned object boxes, which are also rectangular and whose extent can be parameterized for the width and length of the rectangles by means of static status parameters.
- the values of the static status parameters are selected so that the extension of the object boxes is maximum. This means, for example, that the object box 46 shown in FIG. 1 for a truck is three meters wide and 15 meters long.
- the coordinates of the center of gravity of the object box which is represented by the point 48 for the object box 46 in FIGS. 1 and 3, are now selected as dynamic state variables in the method according to the invention.
- the focus of the object box is also seen as the position of the object in the method.
- the assignment between the segment 28 or 36 and the object box 46 representing the corresponding object takes place in that the object box is arranged around the center of gravity for a given fixed extent by means of translations and rotation such that the distances between the pixels forming the segment and the edge of the object box are minimized.
- This treatment of the object also results in an improved position of the search area in which segments for an object are searched.
- 3 schematically shows a search area 50 for the center of gravity of the object assigned to the truck 16, i. H. the object box 46.
- the position of the search area 50 results from the predicted position of the center of gravity of the object box and the extent of the search area for the center of gravity from predicted variances and covariances of these measured variables.
- the actual search area for segments in the inventive method according to this embodiment results from the fact that a rectangle corresponding to the object box is guided with its center of gravity along the edge of the search area 50 and the area swept by the rectangle together with the search area 50 for the focus as a search area defined for segments becomes.
- the search area 50 for the center of gravity can also be elliptical in another embodiment of the invention, for example when choosing a different model for the truck.
- the search area 52 for the two-wheeler 18, that is to say the center of gravity of the object box 54 corresponding to this is shown schematically in FIG. 3.
- the two-wheeler 18 or the segment 55 formed from the pixels 34 is located in the search area 50 of the object assigned to the truck 16.
- the predetermined extent that is to say the shape and size of the object box 46 and the determination of the position by minimizing the distances of the segment pixels to be assigned to the object and lying within the object box 46 from the edge of the object box, make it possible to assign the segment 55 to that of the truck 16 representing object is not given, because of the number of pixels used related average distance of the pixels from the edge of the object box would not be minimized.
- This distance can be assigned to the object as a quality indicator and output as an object property with the object data in step S22, so that the quality of the assignment can be assessed by the following applications.
- the distance is used to only allow an assignment of a segment to an object if a certain threshold value is undershot. For example, in a car whose outline corresponds to a rectangle to a very good approximation, it is required that the mean square distance of the pixels of a segment from the edge of the object box must not exceed a certain maximum value. This makes it possible, for example, to distinguish between a large, similarly shaped, diffuse object, such as a tree, in which the pixels are typically not arranged in a row, but rather in a cluster.
- the size of the object box after an assignment during or after the object formation is significantly larger than the object, the extent of which is to be represented by the object box. If necessary, the size of the object box can then be adapted in the course of the method. This is shown in Figures 5 and 6.
- step S12 the segment 70 was formed from the pixels 68, which could not be assigned to any object in steps S14 and S16.
- a new object is then formed from segment 70 in step S20.
- the length and width of a delimiting rectangle 72 that most closely surrounds the segment 70 are determined. This width and length is then compared with the value ranges specified for the definition of the object classes.
- the object class for trucks and thus the object class-specific object box 74 are assigned to the object corresponding to the truck 66, the extent of which is parameterized by the two static state parameters object box length and object box width.
- the maximum values for both the object box length and the object box width are: permitted according to the definition of the object class, used here as length 15 m and width 3 m.
- the resulting object box 74 is thus significantly longer than the truck 66.
- the coordinates of the center of gravity of the object box 74 are used as the dynamic state variables of the truck 66.
- the truck 66 is no longer covered by the two-wheeler 18, so that the captured image points 76, which correspond to object points on the truck 66, cover a coherent angular range 78 with respect to the laser scanner 12.
- the length of the object box does not correspond to the actual length of the truck 66 and can therefore be adapted.
- the length of the new, adapted object box 84 is therefore selected for adaptation so that a pixel theoretically lying on the edge of the original object box 74 is indicated by the angle adjacent to the angular region 78 counterclockwise, indicated by the cross 80, just no longer in the object box.
- This takes into account the limited resolving power of the laser scanner 12 at the given distance and the given angle.
- this adaptation takes place at the end of step S 16. First of all, it is checked whether the detected contour of the object box 74 could be covered at its ends by another object, for example the two-wheeler 18. This could be the case if the pixels adjacent to the pixels located at the ends of the detected contour are at a smaller distance from the sensor or laser scanner 12.
- the object box is adapted by selecting at least one static status parameter that parameterizes the object box, for example the width and / or length, so that at least all pixels of the segments assigned to the object must be in the new object box, but these must be located with a specified maximum tolerance are tightly enclosed.
- the static state parameters at the respective angle and the respective distance are preferably chosen to be larger in accordance with the resolving power of the laser scanner 12, as described above.
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Multimedia (AREA)
- Computer Networks & Wireless Communication (AREA)
- Theoretical Computer Science (AREA)
- Electromagnetism (AREA)
- Image Analysis (AREA)
- Optical Radar Systems And Details Thereof (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Burglar Alarm Systems (AREA)
Abstract
Description
Claims
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP02796260A EP1419402B1 (de) | 2001-08-22 | 2002-08-21 | Verfahren zur erkennung und verfolgung von objekten |
DE50206591T DE50206591D1 (de) | 2001-08-22 | 2002-08-21 | Verfahren zur erkennung und verfolgung von objekten |
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
DE10141055.7 | 2001-08-22 | ||
DE10141055A DE10141055B4 (de) | 2001-08-22 | 2001-08-22 | Verfahren zur Bestimmung von Bewegungsinformationen |
DE10148068.7 | 2001-09-28 | ||
DE10148068A DE10148068A1 (de) | 2001-09-28 | 2001-09-28 | Verfahren zur Erkennung und Verfolgung von Objekten |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2003019234A1 true WO2003019234A1 (de) | 2003-03-06 |
Family
ID=26009978
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/EP2002/009362 WO2003019234A1 (de) | 2001-08-22 | 2002-08-21 | Verfahren zur erkennung und verfolgung von objekten |
Country Status (4)
Country | Link |
---|---|
EP (1) | EP1419402B1 (de) |
AT (1) | ATE324597T1 (de) |
DE (1) | DE50206591D1 (de) |
WO (1) | WO2003019234A1 (de) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2006034886A1 (de) * | 2004-09-28 | 2006-04-06 | Robert Bosch Gmbh | Radarsensor und verfahren zur abstands- und geschwindigkeitsregelung |
DE102007013023A1 (de) | 2007-03-19 | 2008-09-25 | Ibeo Automobile Sensor Gmbh | Probabilistische Rasterkarte |
DE102007038364A1 (de) | 2007-08-10 | 2009-02-12 | Robot Visual Systems Gmbh | Verfahren zur Geschwindigkeitsmessung von Fahrzeugen mittels Laserscanner |
DE102009007055A1 (de) | 2009-02-02 | 2010-08-05 | Robot Visual Systems Gmbh | Verfahren zur Messung der Geschwindigkeit eines Fahrzeuges und sichtbaren Zuordnung in einer Dokumentation |
US7880643B2 (en) | 2001-12-19 | 2011-02-01 | Logobject Ag | Method and device for following objects, particularly for traffic monitoring |
US10916026B2 (en) | 2018-05-16 | 2021-02-09 | Toyota Research Institute, Inc. | Systems and methods of determining stereo depth of an object using object class information |
USRE48490E1 (en) | 2006-07-13 | 2021-03-30 | Velodyne Lidar Usa, Inc. | High definition LiDAR system |
US10983218B2 (en) | 2016-06-01 | 2021-04-20 | Velodyne Lidar Usa, Inc. | Multiple pixel scanning LIDAR |
US11073617B2 (en) | 2016-03-19 | 2021-07-27 | Velodyne Lidar Usa, Inc. | Integrated illumination and detection for LIDAR based 3-D imaging |
US11082010B2 (en) | 2018-11-06 | 2021-08-03 | Velodyne Lidar Usa, Inc. | Systems and methods for TIA base current detection and compensation |
US11137480B2 (en) | 2016-01-31 | 2021-10-05 | Velodyne Lidar Usa, Inc. | Multiple pulse, LIDAR based 3-D imaging |
US11703569B2 (en) | 2017-05-08 | 2023-07-18 | Velodyne Lidar Usa, Inc. | LIDAR data acquisition and control |
US11796648B2 (en) | 2018-09-18 | 2023-10-24 | Velodyne Lidar Usa, Inc. | Multi-channel lidar illumination driver |
US11808891B2 (en) | 2017-03-31 | 2023-11-07 | Velodyne Lidar Usa, Inc. | Integrated LIDAR illumination power control |
US11885958B2 (en) | 2019-01-07 | 2024-01-30 | Velodyne Lidar Usa, Inc. | Systems and methods for a dual axis resonant scanning mirror |
US11933967B2 (en) | 2019-08-22 | 2024-03-19 | Red Creamery, LLC | Distally actuated scanning mirror |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0564858A2 (de) * | 1992-04-06 | 1993-10-13 | Siemens Aktiengesellschaft | Verfahren zur Auflösung von Clustern bewegter Segmente |
US5291207A (en) * | 1991-12-27 | 1994-03-01 | Honda Giken Kogyo Kabushiki Kaisha | Method for estimating position of objective obstacle for vehicle |
US5475494A (en) * | 1992-12-22 | 1995-12-12 | Mitsubishi Denki Kabushiki Kaisha | Driving environment surveillance apparatus |
US5638164A (en) * | 1994-08-09 | 1997-06-10 | Development Authority, Rafael State of Israel--Ministry of Defense Armament | Apparatus and method for laser imaging |
EP0973121A2 (de) * | 1998-07-14 | 2000-01-19 | DaimlerChrysler AG | Bildverarbeitungsverfahren und Vorrichtungen zur Erkennung von Objekten im Verkehr |
EP1020736A2 (de) * | 1999-01-13 | 2000-07-19 | Honda Giken Kogyo Kabushiki Kaisha | Radargerät |
US6125191A (en) * | 1995-05-22 | 2000-09-26 | Honda Giken Kogyo Kabushiki Kaisha | Object detecting system for vehicle |
-
2002
- 2002-08-21 AT AT02796260T patent/ATE324597T1/de not_active IP Right Cessation
- 2002-08-21 DE DE50206591T patent/DE50206591D1/de not_active Expired - Lifetime
- 2002-08-21 EP EP02796260A patent/EP1419402B1/de not_active Expired - Lifetime
- 2002-08-21 WO PCT/EP2002/009362 patent/WO2003019234A1/de not_active Application Discontinuation
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5291207A (en) * | 1991-12-27 | 1994-03-01 | Honda Giken Kogyo Kabushiki Kaisha | Method for estimating position of objective obstacle for vehicle |
EP0564858A2 (de) * | 1992-04-06 | 1993-10-13 | Siemens Aktiengesellschaft | Verfahren zur Auflösung von Clustern bewegter Segmente |
US5475494A (en) * | 1992-12-22 | 1995-12-12 | Mitsubishi Denki Kabushiki Kaisha | Driving environment surveillance apparatus |
US5638164A (en) * | 1994-08-09 | 1997-06-10 | Development Authority, Rafael State of Israel--Ministry of Defense Armament | Apparatus and method for laser imaging |
US6125191A (en) * | 1995-05-22 | 2000-09-26 | Honda Giken Kogyo Kabushiki Kaisha | Object detecting system for vehicle |
EP0973121A2 (de) * | 1998-07-14 | 2000-01-19 | DaimlerChrysler AG | Bildverarbeitungsverfahren und Vorrichtungen zur Erkennung von Objekten im Verkehr |
EP1020736A2 (de) * | 1999-01-13 | 2000-07-19 | Honda Giken Kogyo Kabushiki Kaisha | Radargerät |
Non-Patent Citations (1)
Title |
---|
VERLY J G ET AL: "MODEL-BASED AUTOMATIC TARGET RECOGNITION (ATR) SYSTEM FOR FORWARDLOOKING GROUNDBASED AND AIRBORNE IMAGING LASER RADARS (LADAR)", PROCEEDINGS OF THE IEEE, IEEE. NEW YORK, US, vol. 84, no. 2, 1 February 1996 (1996-02-01), pages 126 - 163, XP000555861, ISSN: 0018-9219 * |
Cited By (32)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7880643B2 (en) | 2001-12-19 | 2011-02-01 | Logobject Ag | Method and device for following objects, particularly for traffic monitoring |
US7554485B2 (en) | 2004-09-28 | 2009-06-30 | Robert Bosch Gmbh | Radar sensor and method for distance and cruise control |
WO2006034886A1 (de) * | 2004-09-28 | 2006-04-06 | Robert Bosch Gmbh | Radarsensor und verfahren zur abstands- und geschwindigkeitsregelung |
USRE48490E1 (en) | 2006-07-13 | 2021-03-30 | Velodyne Lidar Usa, Inc. | High definition LiDAR system |
USRE48688E1 (en) | 2006-07-13 | 2021-08-17 | Velodyne Lidar Usa, Inc. | High definition LiDAR system |
USRE48666E1 (en) | 2006-07-13 | 2021-08-03 | Velodyne Lidar Usa, Inc. | High definition LiDAR system |
USRE48504E1 (en) | 2006-07-13 | 2021-04-06 | Velodyne Lidar Usa, Inc. | High definition LiDAR system |
USRE48503E1 (en) | 2006-07-13 | 2021-04-06 | Velodyne Lidar Usa, Inc. | High definition LiDAR system |
USRE48491E1 (en) | 2006-07-13 | 2021-03-30 | Velodyne Lidar Usa, Inc. | High definition lidar system |
DE102007013023B4 (de) * | 2007-03-19 | 2017-10-05 | Sick Ag | Probabilistische Rasterkarte |
DE102007013023A1 (de) | 2007-03-19 | 2008-09-25 | Ibeo Automobile Sensor Gmbh | Probabilistische Rasterkarte |
EP2026103A1 (de) | 2007-08-10 | 2009-02-18 | ROBOT Visual Systems GmbH | Verfahren zur Geschwindigkeitsmessung von Fahrzeugen mittels Laserscanner |
DE102007038364A1 (de) | 2007-08-10 | 2009-02-12 | Robot Visual Systems Gmbh | Verfahren zur Geschwindigkeitsmessung von Fahrzeugen mittels Laserscanner |
EP2221640A2 (de) | 2009-02-02 | 2010-08-25 | JENOPTIK Robot GmbH | Verfahren zur Messung der Geschwindigkeit eines Fahrzeuges und sichtbaren Zuordnung in einer Dokumentation |
DE102009007055A1 (de) | 2009-02-02 | 2010-08-05 | Robot Visual Systems Gmbh | Verfahren zur Messung der Geschwindigkeit eines Fahrzeuges und sichtbaren Zuordnung in einer Dokumentation |
US11137480B2 (en) | 2016-01-31 | 2021-10-05 | Velodyne Lidar Usa, Inc. | Multiple pulse, LIDAR based 3-D imaging |
US11822012B2 (en) | 2016-01-31 | 2023-11-21 | Velodyne Lidar Usa, Inc. | Multiple pulse, LIDAR based 3-D imaging |
US11698443B2 (en) | 2016-01-31 | 2023-07-11 | Velodyne Lidar Usa, Inc. | Multiple pulse, lidar based 3-D imaging |
US11550036B2 (en) | 2016-01-31 | 2023-01-10 | Velodyne Lidar Usa, Inc. | Multiple pulse, LIDAR based 3-D imaging |
US11073617B2 (en) | 2016-03-19 | 2021-07-27 | Velodyne Lidar Usa, Inc. | Integrated illumination and detection for LIDAR based 3-D imaging |
US11550056B2 (en) | 2016-06-01 | 2023-01-10 | Velodyne Lidar Usa, Inc. | Multiple pixel scanning lidar |
US11561305B2 (en) | 2016-06-01 | 2023-01-24 | Velodyne Lidar Usa, Inc. | Multiple pixel scanning LIDAR |
US11808854B2 (en) | 2016-06-01 | 2023-11-07 | Velodyne Lidar Usa, Inc. | Multiple pixel scanning LIDAR |
US10983218B2 (en) | 2016-06-01 | 2021-04-20 | Velodyne Lidar Usa, Inc. | Multiple pixel scanning LIDAR |
US11874377B2 (en) | 2016-06-01 | 2024-01-16 | Velodyne Lidar Usa, Inc. | Multiple pixel scanning LIDAR |
US11808891B2 (en) | 2017-03-31 | 2023-11-07 | Velodyne Lidar Usa, Inc. | Integrated LIDAR illumination power control |
US11703569B2 (en) | 2017-05-08 | 2023-07-18 | Velodyne Lidar Usa, Inc. | LIDAR data acquisition and control |
US10916026B2 (en) | 2018-05-16 | 2021-02-09 | Toyota Research Institute, Inc. | Systems and methods of determining stereo depth of an object using object class information |
US11796648B2 (en) | 2018-09-18 | 2023-10-24 | Velodyne Lidar Usa, Inc. | Multi-channel lidar illumination driver |
US11082010B2 (en) | 2018-11-06 | 2021-08-03 | Velodyne Lidar Usa, Inc. | Systems and methods for TIA base current detection and compensation |
US11885958B2 (en) | 2019-01-07 | 2024-01-30 | Velodyne Lidar Usa, Inc. | Systems and methods for a dual axis resonant scanning mirror |
US11933967B2 (en) | 2019-08-22 | 2024-03-19 | Red Creamery, LLC | Distally actuated scanning mirror |
Also Published As
Publication number | Publication date |
---|---|
EP1419402A1 (de) | 2004-05-19 |
DE50206591D1 (de) | 2006-06-01 |
EP1419402B1 (de) | 2006-04-26 |
ATE324597T1 (de) | 2006-05-15 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
EP1298454B1 (de) | Verfahren zur Erkennung und Verfolgung von Objekten | |
DE112012002885B4 (de) | Verfahren und Vorrichtung zur bildbasierten Fahrzeugerfassung und Entfernungsmessung | |
EP1267178B1 (de) | Verfahren zur Verarbeitung eines tiefenaufgelösten Bildes | |
DE102006057552B4 (de) | System und Verfahren zur Messung des Abstands eines vorausfahrenden Fahrzeugs | |
EP1419402B1 (de) | Verfahren zur erkennung und verfolgung von objekten | |
EP1589484B1 (de) | Verfahren zur Erkennung und/oder Verfolgung von Objekten | |
EP1306690A2 (de) | Verfahren zur Erkennung und Verfolgung von Objekten | |
WO2004055548A1 (de) | Verfahren zur erkennung und verfolgung von objekten | |
WO2007033870A1 (de) | Verfahren und fahrerassistenzsystem zur sensorbasierten anfahrtsteuerung eines kraftfahrzeugs | |
DE102014117102B4 (de) | Spurwechselwarnsystem und Verfahren zum Steuern des Spurwechselwarnsystems | |
WO2003015053A1 (de) | Verfahren zur bestimmung einer modellfahrbahn | |
EP1475764A2 (de) | Verfahren und Vorrichtung zur Ermittlung der Wahrscheinlichkeit für eine Kollision eines Fahrzeugs mit einem Gegenstand | |
WO2004111943A1 (de) | Verfahren und vorrichtung zur objektortung für kraftfahrzeuge | |
EP1531342B1 (de) | Verfahren zur Erkennung von Fussgängern | |
DE102006036402A1 (de) | System und Verfahren zur Messung der Entfernung eines vorausfahrenden Fahrzeugs | |
EP1298012B1 (de) | Verfahren zur Erkennung und Verfolgung von Objekten | |
DE10148070A1 (de) | Verfahren zur Erkennung und Verfolgung von Objekten | |
EP3663881B1 (de) | Verfahren zur steuerung eines autonomen fahrzeugs auf der grundlage von geschätzten bewegungsvektoren | |
DE10148062A1 (de) | Verfahren zur Verarbeitung eines tiefenaufgelösten Bildes | |
DE102006037600B4 (de) | Verfahren zur auflösungsabhängigen Darstellung der Umgebung eines Kraftfahrzeugs | |
EP1291674A2 (de) | Verfahren zur Erkennung und Verfolgung von Objekten | |
DE102008050456B4 (de) | Verfahren und Vorrichtung zur Fahrspurerkennung | |
DE102006007550A1 (de) | Vorrichtung und Verfahren zur Erkennung einer Fahrbahnmarkierung für ein Kraftfahrzeug | |
DE102018121158A1 (de) | Verfahren zum Erfassen von Bodenabtastpunkten und Fahrerunterstützungssystem, das dafür konfiguriert ist, ein derartiges Verfahren auszuführen | |
DE10148068A1 (de) | Verfahren zur Erkennung und Verfolgung von Objekten |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AK | Designated states |
Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BY BZ CA CH CN CO CR CU CZ DE DM DZ EC EE ES FI GB GD GE GH HR HU ID IL IN IS JP KE KG KP KR LC LK LR LS LT LU LV MA MD MG MN MW MX MZ NO NZ OM PH PL PT RU SD SE SG SI SK SL TJ TM TN TR TZ UA UG US UZ VC VN YU ZA ZM Kind code of ref document: A1 Designated state(s): AE AG AL AM AT AU AZ BA BB BG BR BY BZ CA CH CN CO CR CU CZ DE DK DM DZ EC EE ES FI GB GD GE GH GM HR HU ID IL IN IS JP KE KG KP KR KZ LC LK LR LS LT LU LV MA MD MG MK MN MW MX MZ NO NZ OM PH PL PT RO RU SD SE SG SI SK SL TJ TM TN TR TT TZ UA UG US UZ VC VN YU ZA ZM ZW |
|
AL | Designated countries for regional patents |
Kind code of ref document: A1 Designated state(s): GH GM KE LS MW MZ SD SL SZ TZ UG ZM ZW AM AZ BY KG KZ MD RU TJ TM AT BE BG CH CY CZ DE DK EE ES FI FR GB GR IE IT LU MC NL PT SE SK TR BF BJ CF CG CI CM GA GN GQ GW ML MR NE SN TD TG Kind code of ref document: A1 Designated state(s): GH GM KE LS MW MZ SD SL SZ UG ZM ZW AM AZ BY KG KZ RU TJ TM AT BE BG CH CY CZ DK EE ES FI FR GB GR IE IT LU MC PT SE SK TR BF BJ CF CG CI GA GN GQ GW ML MR NE SN TD TG |
|
121 | Ep: the epo has been informed by wipo that ep was designated in this application | ||
WWE | Wipo information: entry into national phase |
Ref document number: 2002796260 Country of ref document: EP |
|
WWP | Wipo information: published in national office |
Ref document number: 2002796260 Country of ref document: EP |
|
REG | Reference to national code |
Ref country code: DE Ref legal event code: 8642 |
|
WWG | Wipo information: grant in national office |
Ref document number: 2002796260 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: JP |
|
WWW | Wipo information: withdrawn in national office |
Country of ref document: JP |