Classifications

• • 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
• • G—PHYSICS
  • G01—MEASURING; TESTING
  • G01W—METEOROLOGY
  • G01W1/00—Meteorology
• • 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
  • G06V20/584—Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads of vehicle lights or traffic lights

Definitions

• the present invention relates to a method for detecting a weather condition in an environment of a vehicle, to a control unit for detecting a weather condition in an environment of a vehicle and to a corresponding computer program product.
• DE 10 2007 062 258 A1 describes an optical rain sensor device for a motor vehicle.
• the present invention proposes a method for detecting a weather condition in an environment of a vehicle, a control device for detecting a weather condition in an environment of a vehicle and a corresponding computer program product according to the main claims.
• Advantageous embodiments emerge from the respective subclaims and the following description.
• the invention is based on the recognition that in the dark a field of vision of a driver of a vehicle has different characteristics when different weather conditions such as rain, snow, fog, whirling dust or dry weather prevail.
• different weather conditions such as rain, snow, fog, whirling dust or dry weather prevail.
• Image information to be mapped when the driver's field of view is illuminated by at least one headlight, the weather conditions can trigger a weather-typical, optical behavior as a characteristic feature.
• the light can be refracted by the prevailing weather conditions and thus trigger a weather-typical refraction behavior or scattering behavior as a characteristic feature.
• an evaluation of the characteristic features of safety-relevant systems in the vehicle can provide important parameters which can help to ensure greater safety.
• comfort systems can alter and adjust control parameters through knowledge of the weather condition, for example to avoid misinterpreting other information.
• optical detection systems can adapt image processing to the current weather conditions in order to improve object recognition and evaluation in different weather conditions.
• the present invention provides a method for detecting a weather condition in an environment of a vehicle, the method comprising the steps of: Determining at least one imaging feature in image information, wherein the image information represents an image of at least a portion of the environment and the imaging feature represents a lighting effect on a reflective or self-luminous object in the region of the environment; and
• Determining weather information for characterizing the weather condition based on a comparison between the mapping feature and at least one expected value.
• an atmospheric phenomenon can be understood.
• a weather condition such as fog, rain, fallen rain, snowfall, fallen snow, or even lack thereof may be.
• the weather condition may be a consequence of an atmospheric phenomenon.
• wind, dust or sand can stir and blow.
• An environment of a vehicle may be an environment within sight of the vehicle.
• the environment may be an area in front of the vehicle that can be illuminated directly or indirectly by headlights of the vehicle.
• an imaging feature can be used to reduce a picture quality of a picture caused by at least one physical effect
• a contour of a road sign can be blurred.
• a lamp may have a "yard or a halo.
• precipitation or condensed air moisture in a headlight cone can be visible as an imaging feature.
• water crystals or water droplets can throw back incoming light to a high degree in the direction of the headlights.
• Image information can be understood to mean an image file, for example a raster graphic.
• a raster graphic rasterformig arranged pixels can each represent a pixel.
• a pixel may have a brightness value and / or a color value.
• a light effect for example, a refraction or a light diffraction or a light reflection or a light scattering on an object and / or a beam path from the headlight and to the object and / or a beam path from the object to an objective
• a reflective or self-luminous object for example, a reflector, a street sign, a lantern, a headlight, a
• Weather information may be information resulting from the current weather condition.
• the weather information may be information about a visibility or a road condition.
• An expected value may be taken to mean a stored value that has been recorded, for example, in the past as a past image feature in a similar object and weather conditions. The expected value may thus be a brightness distribution, a shape, an extension or an arrangement of a region of the image or a combination thereof. By a comparison, a largest match can be found and the weather condition associated with the selected expectation value can be provided as weather information.
• the light effect may represent at least one light-dark gradient, and in the
• a scattering characteristic and / or a refractive index of the light-dark gradient can be determined from the imaging feature, wherein the scattering characteristic represents a light scatter in an image area around the reflective or self-illuminating object, and the refractive index converts a refraction of light in the image area represents the reflective or self-luminous object.
• the scattering characteristic value and / or the refractive index value can be compared with at least one expected value for the scattering characteristic value and / or the refractive index value.
• a bright-dark gradient can be understood to mean a brightness gradient from one bright point to a neighboring darker point.
• a light-dark gradient between a white pixel to a black pixel is large. Between a gray pixel and a dark gray pixel, the light-dark gradient is small. For example, drops of water in the air may break light. Thus, the light-dark gradient at an edge between, for example, a light source and a dark background becomes smaller than in clear air. From a driver's view of the vehicle, e.g. Vehicle headlight of another vehicle in fog a so-called halo. The halo gets darker and darker. Therefore, the light-dark gradient is small here. For example, a scattering characteristic may represent a size of the halo. The denser the fog, the bigger the
• a halo can also be displayed on a retro- reflective object, such as a reflector can be arranged.
• a refractive index may represent a decomposition of the light in the halo into spectral components.
• a large refractive index may indicate ice crystals in the air.
• An expected value for the scattering characteristic may be a stored empirical value which represents a measure of a particular scatter.
• An expectation value for the refractive index may be a stored empirical value representing a measure of a particular refraction.
• the light effect may also represent at least one characteristic of a light spot, which is caused at least by a light beam emitted by a headlight of the vehicle.
• a position of the light spot can be determined from the imaging feature, and in the step of determining further the position can be compared with an expected value for the position.
• a spot of light can be understood to mean a region of high brightness that can be caused by a beam of a spotlight.
• the light spot can image a contour of a light beam from the headlight.
• the light spot can be projected onto a street.
• light spots can also "float” in front of the vehicle as a result of a reflection of particles.
• a “floating" light spot can have a different shape than a projected light spot.
• the "floating" light spot may be perceived at a position other than the projected light spot, a characteristic may represent a shape and location of the light spot, for example, a position may reflect whether the light spot is projected on the street and the street Particles may be, for example, dust, water droplets or ice crystals
• An expectation value for the position may be an empirical value representing an expected position of the light spot in a given weather condition In particular, fog can be detected particularly easily and safely.
• Width of the light spot can be determined, the height being an apparent vertical Extension of the light spot represents and the width represents an apparent horizontal extension of the light spot. Further, in the step of determining, the height may be compared with an expectation value for the height and the width compared with an expectation value for the width. Under a height and a width of the light spot, an apparent dimension of the
• the height and / or width may represent a countable number of pixels in an alignment along and / or across the vehicle.
• An expectation value for the height and / or the width may be an empirical value representing an expected height and / or width of the light spot in a given weather condition.
• the light effect may represent at least one ground brightness
• a light distribution of the ground brightness may be determined.
• the light distribution may be compared with an expected value for the light distribution. Under a floor brightness one of the
• Vehicle can be understood from perceptible brightness of the road.
• a light distribution can represent a perceptible distribution of light and dark spots on the street. For example, only a small part of the emitted light can be reflected back to the vehicle when the road is wet. A remnant of the emitted light may be reflected off the vehicle under total reflection from the wet roadway. For example, in the case of snow-covered roads, a large part of the emitted light can be diffusely reflected back to the vehicle over a large area. On dry roads a medium part of the emitted light can be reflected back to the vehicle.
• the relationships may be reversed, since in the case of rain, the light reflected away from the oncoming vehicle may strike the own vehicle as total reflection, corresponding to a direct headlight beam. In snow, the light can be diffusely reflected over a large area.
• An expected value for the light distribution can be an experience value representing an expected light distribution of the ground brightness in a certain weather condition. This can be a road condition in front of the vehicle are recorded as a particularly safety-relevant parameter, and corrective interventions of safety and comfort systems are adapted, and the driver of the vehicle is alerted.
• the present invention the
• a lateral brightness value of the brightness can be determined from the imaging feature and, in the step of determining, the side area brightness value can also be compared with an expected value for the side area brightness value.
• a ground-level side area can be understood as a lane of the road adjacent to one's own traffic lane.
• a ground-level side area can be understood as a side strip of the street or a banquet of the street.
• the near-bottom side region may have, for example, soiling.
• a page range brightness value may represent a brightness reflected from the page area. As a result, it can be detected, for example, whether the roadway is dirty or snow-covered in addition to the own lane.
• An expectation value for the side-panel brightness value may be an empirical value representing an expected side-range brightness value of the near-side-brightness in a given weather condition.
• an additional imaging feature may be determined, wherein the additional image information represents an additional image of at least the portion of the environment at an additional time, and the additional imaging feature provides an additional lighting effect represents the additional time at the reflecting or self-illuminating object in the part of the environment.
• the weather information may further be determined based on an additional comparison between the additional mapping feature and the expected value or at least one additional expected value, wherein the weather information is further determined based on a change between the comparison and the additional comparison.
• Estimation of the currently prevailing weather condition is based on a determination of characteristic features for both reflective and self-luminous light objects.
• This can e.g. Headlamps and lights of foreign traffic, reflectors illuminated by their own headlights, as well as an illuminated area of the street directly in front of the vehicle.
• the characteristics of these light objects such as maximum brightness, gray scale gradient, color,
• Fig. 3 shows an image of a driver's field of view in dry weather.
• the driver's field of view is partially illuminated by light from headlights of a vehicle.
• a straight course of a road 300, each with a directional lane is so far in front of the vehicle to detect until the light is no longer sufficient
• the self-luminous objects 402 represent two headlights 402 of an oncoming vehicle.
• the self-luminous objects 402 are shown as two equally large and spaced approximately the same height from each other arranged very bright image areas.
• the oncoming vehicle is not recognizable.
• the light of the two headlights 402 is on windshield wiper streaks on one
• the light of the two headlights 402 is also scattered on the windshield wiper streaks on a windshield of the ego vehicle.
• the self-luminous objects 402 also have lightsabers 406.
• a right roadside 408 of the road 300 is snow-covered and reflects a portion of the light of the headlights back to the diffuser
• FIG. 3 is a street 300 in front of a courtesy vehicle illuminated by headlights of the ego vehicle.
• the road 300 has a broken center mark 304.
• a right roadside 408 is snow-covered, throwing back diffused light from the headlights of the ego vehicle.
• the headlights of the ego vehicle illuminate individual snowflakes 902 in the air in front of the vehicle.
• the snowflakes 902 appear in the picture as bright white dots.
• the ego vehicle is another vehicle opposite.
• the headlights 402 of the other vehicle illuminate the snow track in front of the other vehicle very bright. Therefore, this part of the driver's field of view is overexposed in Figure 10 and appears as an amorphous white area 1002. Due to the characteristic distribution of light on the road 300 in front of the vehicle on a snowy road or also because of the overexposed surface 1002 in front of the oncoming vehicle, the method presented here can determine weather information "snow on the road". By evaluating a number of snowflakes 902 in the spotlight, it is possible to deduce a density of the snowfall, and thus a weather information "probable visibility".
• the beam of the headlights hits the road surface. Due to road bumps, the measured height, as well as the measured width of the light spot has a fluctuation range. Since the road surface reflects only light in the center of the headlight beam with high intensity back to the vehicle, the light spot appears narrow. In light fog, particles in the air in front of the vehicle are already illuminated by the beam of the headlights and make it visible. Depending on the density of the fog, the light spot appears at different heights in front of the vehicle. Since the fog also reflects light of lesser intensity from side areas of the headlight beam, the light spot appears wider in light fog than the light spot in a clear atmosphere.

Applications Claiming Priority (2)

Publications (1)

Family

ID=46125441

Family Applications (1)

Country Status (6)

Families Citing this family (28)

Family Cites Families (30)

• 2012
  • 2012-05-15 EP EP12722129.9A patent/EP2721554A2/de not_active Withdrawn
  • 2012-05-15 CN CN201280078252.5A patent/CN105593875B/zh active Active
  • 2012-05-15 JP JP2014515114A patent/JP5851597B2/ja active Active
  • 2012-05-15 US US14/126,885 patent/US9946937B2/en active Active
  • 2012-05-15 WO PCT/EP2012/059033 patent/WO2012171739A2/de active Application Filing
  • 2012-06-12 DE DE102012209810A patent/DE102012209810A1/de active Pending

Non-Patent Citations (1)

Also Published As

Similar Documents

Legal Events