WO2010099847A1

WO2010099847A1 - Method and device for determining visibility range for a vehicle

Info

Publication number: WO2010099847A1
Application number: PCT/EP2010/000316
Authority: WO
Inventors: Thanh-Binh To; Wojciech Waclaw Derendarz; Henry Türmer
Original assignee: Volkswagen Aktiengesellschaft
Priority date: 2009-03-05
Filing date: 2010-01-20
Publication date: 2010-09-10
Also published as: DE102009011866A1

Abstract

The present invention relates to a method for determining a visibility range for a vehicle (10). In the method, an image information of an environment of the vehicle (10) is captured by means of an image capture device (2, 3). Those image segments comprising a contrast greater than a predetermined contrast threshold are determined from the image information. Distances (d1-d4) between each image capture device (2, 3) and locations (12-15) in the environment corresponding to the image segments are determined for each image segment so determined. The greatest of the determined distances (d1-d4) is used as the dimension for the visibility range.

Description

description

Method and apparatus for determining a range of visibility for a vehicle

The present invention relates to a method and a device for determining a visibility for a vehicle, which can be used, for example, to control fog lamps or an automatic cruise control system of the vehicle.

A detection of off-vehicle conditions which limit the driver's visibility, such as Fog, smoke, heavy rain, heavy snowfall can contribute significantly to improving traffic safety. On the one hand, driver assistance systems, such as an automatic cruise control system, an automatically controlled rear fog light or automatically controlled fog lights are automatically adjusted depending on the specific visibility, thereby relieving the driver and increasing driving safety is achieved. In addition, the driver can be issued a warning, for example in the event of sudden fog, so that the driver can adjust his speed and an accident and especially a heavy pile-up can be prevented.

Previous sensors for visibility measurement, which are used outside of vehicles, so-called scatter or transmission meters. However, these special visibility detection devices require a large free area for proper measurement. In addition, they are unwieldy and expensive to buy and also bad to integrate in a vehicle.

DE 10 2008 003 948 A1 discloses a device that detects the presence of a component that interferes with a view of space that is present in front of a vehicle. The ingredient results from a natural appearance in the room and may include, for example, fog. The apparatus includes means for acquiring an image of the space, means for extracting a reference image, means for estimating the brightness values, means for detecting an obstacle, means for masking, means for calculating reliability, means for calculating a Total brightness intensity and means for determining. The means for detecting the mapping of the space is used to detect the image seen from the vehicle and composed of a group of pixels having brightness values. The means for extracting the reference image is used to extract the reference image from the map generated by the means for detecting the image is detected. The reference image includes a portion of the map of the space located a predetermined distance away from the vehicle. The means for estimating the brightness values is used to estimate the brightness values of the groups of bump elements in the reference image. The obstacle estimating means is used to detect the obstacle from the reference image to obtain a contour of the obstacle in the reference image. The means for masking is applied to mask a region including the image of the obstacle in the reference image in accordance with the contour of the obstacle detected by the obstacle detection block to produce a masked reference image. The means for calculating the reliability is applied to calculate the reliability of determining the existence of the component based on the area which is masked in generating the masked reference image. The means for calculating the total brightness intensity is applied to calculate the total brightness intensity as a function of the brightness values of the groups of pixels in the masked reference image to achieve the total brightness intensity. The means for determining is applied to determine the existence of the component in accordance with the total intensity intensity calculated by the total brightness calculation block and the reliability of determining the existence of the component calculated by the reliability calculation block. Although the method disclosed in DE 10 2008 003 948 A1 provides a reliable determination as to whether fog is present, it requires complex and complex image analyzes which include an object determination. In addition, the device requires various sensors, such as a camera and a distance radar, whereby the device is relatively expensive.

EP 1 498 721 A1 discloses a device for detecting fog for a vehicle. The device comprises a camera having a plurality of pixels arranged in array form, the camera capturing a part of its surroundings, in particular the area in front of a vehicle. The device further comprises an image evaluation unit for evaluating an image captured by the camera and for this purpose compares the brightness value of each pixel of the camera with a presettable minimum brightness value. A fog indicating output signal is output when none of the pixels has a brightness value smaller than the minimum brightness value. However, this device is not suitable for determining a visibility. In addition, the device is highly dependent on the intensity of the environment, which may compromise the reliability of fog detection.

DE 695 04 761 T2 discloses a method for detecting the presence of fog or smoke in the vicinity of a motor vehicle. The procedure will require at least At least one image of a spatial area is taken, the contrast of at least part of the image is analyzed, and a signal for detecting fog or smoke is sent as a function of this analysis. For the analyzed image part, a characteristic of a number of white pixels and a characteristic of a number of black pixels are determined, and this value is compared with thresholds. If one of these values of pixel counts is greater than its threshold, it is assumed that there is no fog. However, the disclosed method does not provide a way to determine a current visibility.

DE 102 19 788 C1 discloses a method and a device for vision measurement with image sensor systems which consist of at least two image sensors which record substantially the same scene. Objects are detected from the image sensor signals, their distance to the image sensor system is calculated, the object contrast is determined and the visibility is determined. The method requires reliable object detection and is based on the assumption that the two detected objects have the same contrast, that is, that the contrast differences observed in the image are due to differences in separation only.

DE 103 03 046 A1 discloses a method and a device for the quantitative estimation of the visibility in vehicles. Image data from the environment are recorded and evaluated to estimate the visibility. In this case, at least one region of a captured image is evaluated to form a characteristic measure, to this measure by means of a reference that sets the measure in relation to a sight, determines a visibility and uses the determined visibility for output and / or control of vehicle components. The evaluation of the image data is preferably carried out on the basis of a measure of the sharpness or the contrast of the image. Additionally or alternatively, the environment of the vehicle is illuminated, detects the backscattered light and formed an intensity-dependent measure of the detected backscattered light to determine the visibility.

Object of the present invention is to provide a simple and inexpensive method and a corresponding device for determining a visibility for a vehicle, in particular in a visual obstruction by fog, smoke and the like.

According to the present invention, this object is achieved by a method for determining a visibility for a vehicle according to claim 1 and a device for determining a visibility for a vehicle according to claim 11. The dependent claims define preferred and advantageous embodiments of the invention. In accordance with the present invention, a method for determining a range of visibility for a vehicle is provided. The method comprises capturing image information of an environment of the vehicle by means of an image capture device. From the image information, image portions are determined which have a contrast that is greater than a predetermined contrast threshold value. For the particular image sections, a distance between the image capture device and the location in the environment corresponding to the image section is determined in each case. Finally, the largest of the specific distances is determined as a measure of the visibility.

The image capture device may be e.g. a stereo camera, so that the distances can be determined by means of a stereo method from the image information. The image information may, for example, have a multiplicity of image points arranged, for example, in array form, and an image section may comprise two adjacent image points of the image information, such that a contrast of the image section comprises the contrast of the two neighboring pixels. Thus it is possible in a simple way not only to detect fog, but also to determine a visibility in fog, smoke, heavy rain or snowfall and the like. By using the stereo camera, which can also be used, for example, for other functions of the vehicle, e.g. a roadway or driving tube detection, is a cost-effective detection of the image information available. In addition, the method does not require a transmitting unit, which illuminates, for example, an area in the vicinity of the vehicle, whereby the cost of implementing the method can be kept low. In addition, since no transmitting unit is necessary, i.e., the method is based on passive sensing, interference with oncoming traffic is excluded. Since the contrast is determined directly by comparing adjacent pixels, no reference data is necessary. In addition, no object detection is necessary since the method operates on a pixel level and not on an object level. As a result, no high computing power is required, whereby the process is quick and inexpensive to implement. In addition, the contrast calculation is independent of the exposure control of the camera and thus the visibility estimate of the exposure control of the camera is independent. As a result, the method can be reliably used in a wide variety of weather conditions and at different times of the day.

According to one embodiment, the contrast may be determined using a logarithmic image processing model. Such models are also known as logarithmic image processing models, so-called LIP models, and offer independence from the exposure control of the camera. Investigations have shown that a con- threshold value in the range of 3% to 10%, in particular a contrast threshold value with a value of 5% are suitable for classifying an object as being visible.

According to a further embodiment, the method comprises filtering the determined visibility with a statistical method, for example filtering with a Kalman filter. This can increase the robustness of the system.

According to one embodiment, depending on the particular visibility, a vehicle device, such as a vehicle. a fog lamp, a fog lamp, an automatic cruise control system or a warning device for the driver of the vehicle, driven. By automatically controlling the fog lamps and the automatic setting of the cruise control system, the driver can be relieved. By the warning device, the driver can be notified in good time to a dangerous situation with visual impairment, whereby the traffic safety can be increased.

According to the present invention, there is further provided an apparatus for determining a visibility of a vehicle. The apparatus includes an image capture device for acquiring image information of an environment of the vehicle and a processing unit coupled to the image capture device. The processing unit determines those image portions having a contrast which is greater than a predetermined contrast threshold from the ^"image information. Furthermore, determines the processing unit for each of the predetermined image portions respectively, a distance between the image capture device and those points in the surroundings of the vehicle, which the From the distances thus determined to the locations in the environment corresponding to the particular image sections, the greatest distance is determined as a measure of the visibility.

A device configured as described above can be used to carry out the methods described above and therefore also includes the advantages described above in connection with the methods.

The present invention will be explained below with reference to the accompanying drawings.

Fig. 1 shows a schematic representation of a device for determining a visibility for a vehicle according to the present invention. Fig. 2 shows a vehicle having a device for determining a visibility according to the present invention on a roadway.

Fig. 1 shows a device 1 for determining a visibility for a vehicle. The device 1 comprises an image capture device 2, 3 for acquiring image information of an environment of the vehicle. The image capture device 2, 3 in the present embodiment comprises two cameras 2, 3, which are arranged as a stereo camera arrangement on the vehicle. The cameras 2, 3 are preferably arranged on a front side of the vehicle such that they can receive an area in the direction of travel in front of the vehicle. The device 1 further comprises a processing unit 4, which is coupled to the stereo camera arrangement 2, 3. The processing unit 4 comprises, for example, a microprocessor computing unit which is suitable for processing image information. The stereo camera arrangement 2, 3 supplies, for example, image information of the surroundings of the vehicle in the form of pixels which are arranged in a matrix arrangement, a so-called array arrangement. Each pixel includes a brightness value and possibly a color information of a corresponding location of the surroundings of the vehicle.

As shown in FIG. 1, the processing unit 4 includes a 3D reconstruction unit and a contrast calculation unit.

In the contrast calculation unit, the contrast values between in each case two adjacent pixels are calculated and those positions of the camera image are determined at which the contrast is greater than, for example, 5%. The contrast formula of the Logarithmic Image Processing Model, the so-called LIP model, is used to ensure independence from the exposure control of the camera. Visibility was defined by the World Metereological Organization as the maximum distance at which a black object has a contrast of at least 5% on clear skies. Therefore, in the present embodiment, a contrast threshold of 5% was used. The contrast C is defined by the Weber contrast definition as follows:

Where L _{0 is} the luminance of the object and I_ _{B is} the luminance of the immediately adjacent background. The LIP model establishes a consistency with Weber's definition of contrast and the LIP model can be used as a definition for the term "visibility." The contrast between two adjacent pixels is calculated in the LIP model M] f {x, y) -f (x ', y) C ^MM M-min (f (x, y), f (x ^> , y ^< )) ^W

Here, M is the maximum possible gray value of a pixel in the image, and f (x, y) and f (x ', y') are the gray values of two adjacent pixels defined in the LIP model with the positions (x, y) and (* ', /). In this model, the inverted brightness values are used as gray values. Due to the previously described contrast formula of the LIP model, the contrast determination is independent of an exposure control of the stereo camera arrangement 2, 3. In the contrast calculation unit of the processing unit 4, all sections or regions of the image information are determined in which the contrast for each two adjacent pixels according to equation (2) is greater than 5%.

In the 3D reconstruction unit of the processing unit 4, a distance determination is carried out by means of a stereo method for the image section or image regions of the image information which have a contrast of more than 5%. The distances thus determined are supplied to a vision estimation unit of the processing unit 4. In the vision estimation unit, the largest of the determined distances is determined and used as a measure of the visibility. In order to increase the robustness of the system, the specific visibility can be filtered by statistical methods, for example with a Kalman filter.

Depending on the specific visibility, further devices of the vehicle, such as a fog light 5, a cruise control system 6 or a warning device 7 can then be activated. The fog lamp 5 may for example comprise both fog lamps and a rear fog lamp of the vehicle. Depending on the particular visibility then the fog lights and / or the rear fog light can be switched on and off individually and automatically. Thereby, the driver of the vehicle is relieved of the operation of corresponding operating devices of the vehicle and it is avoided that the driver of the vehicle forgets to switch the fog lamps on or off when fog comes up or there is no more fog. In vehicles with an automatic cruise control system 6, for example, a maximum speed to which the vehicle can be adjusted, depending on the specific visibility can be limited. This automatically reduces the speed of the vehicle in the onset of fog, which can increase road safety. Finally, a warning device 7, such as an electronic display in a dashboard of the vehicle, which indicates a current visibility in meters, or an audible warning signal, can alert the driver of the vehicle to reduced visibility, thereby increasing traffic safety. FIG. 2 shows a vehicle 10 equipped with a device 1 for determining a visibility for the vehicle 10. The device 1 comprises, as described in connection with FIG. 1, the stereo camera arrangement 2, 3 and a processing unit not shown in FIG. 2. The vehicle 10 is located on a carriageway 11 and moves in the illustration of Fig. 2 from right to left. The stereo camera arrangement 2, 3 is arranged on the vehicle 10 in such a way that it detects a surrounding area of the vehicle 10 in the direction of travel in front of the vehicle 10. In the detection range of the stereo camera arrangement 2, 3 there is inter alia a traffic sign 12 on the right-hand lane edge of the lane 11, an oncoming vehicle 13, a tree 14 and a vehicle 15 traveling in front.

According to the present invention, image information of the surroundings of the vehicle 10 is detected by the stereo camera device 2, 3 and provided to the processing unit 4. The processing unit 4 analyzes the image information on a pixel basis to determine areas or pixel pairs which exceed a predetermined contrast threshold of, for example, 5%. For this purpose, as described in connection with FIG. 1, an LIP model is used which ensures a contrast determination independently of the exposure control of the camera. Due to fog not shown in FIG. 2, which is located between the vehicle 10 and the further objects 12-15, the processing device 4 determines contrast values, for example for pixels which correspond to the traffic sign 12 and for pixels which correspond to the oncoming vehicle 13 of more than 5%. For pixels which correspond to the tree 14 or the preceding vehicle 15, contrast values of less than 5% are determined in the present visual situation. For the areas or image sections of the image information which have a contrast of more than 5%, the processing unit 4 then uses the stereo method and the two cameras 2, 3 to determine distances to the locations in the surroundings of the vehicle which correspond to the image sections or Pixels with more than 5% contrast. In the present case, therefore, the distances di and d _{4 are} determined. On the other hand, the distances d ₂ and d ₃ are not determined since the pixels of the tree 14 and the preceding vehicle 15 are not sufficiently contrasted (ie, less than 5%). Now the maximum of the certain distances d ^ d _{4 is} determined. In the present example, d _{4 is} the largest of the specific distances. Thus, as a measure of the visibility for the vehicle 10, the distance d _{4 is} determined. Depending on this distance, as described in connection with FIG. 1, rear fog lights, an automatic cruise control system or warning devices for the driver can be controlled.

In the example shown in FIG. 2, the view improves such that also pixels that represent the preceding vehicle 15 have a contrast value of more than 5%. sen, so the distance d _{3 is determined} as the measure of the visibility and driven the devices of the vehicle accordingly.

Although a contrast threshold of 5% was used to determine the range of vision in the preceding description, other contrast thresholds may be used, for example in a range of 3-10%. Moreover, other image processing models than the previously described LIP model can be used to determine a contrast between adjacent pixels or in image portions of the image information. In addition, the stereo camera arrangement 2, 3 and the processing unit 4 connected thereto can also be used for other functions, so-called driver assistance functions of the vehicle, such as an obstacle object detection and a roadway and driving route detection.

LIST OF REFERENCE NUMBERS

Device Image acquisition device Image acquisition device Processing unit Fog light Cruise control Warning device Vehicle Lane Traffic sign oncoming vehicle, tree driving vehicle

Claims

claims

A method of determining a range of vision for a vehicle, comprising:

Acquiring image information of an environment of the vehicle (10) by means of an image acquisition device (2, 3),

Determining image portions of the image information having a contrast greater than a predetermined contrast threshold,

Determining distances (drd ₄ ) between each of the image capture device (2, 3) and locations (12-14) in the environment corresponding to the image sections, and

Determining the largest of the determined distances (d _! -D ₄ ) as a measure of the visibility.

2. The method according to claim 1, characterized in that the image capture device (2, 3) comprises a stereo camera, and that the distances (drd ₄ ) are determined by means of a stereo method from the image information.

3. The method of claim 1 or 2, characterized in that an image section comprises two adjacent pixels of the image information and a contrast of the image portion comprises the contrast of the two adjacent pixels.

4. The method according to any one of the preceding claims, characterized in that the contrast is determined by means of a logarithmic image processing model.

5. The method according to any one of the preceding claims, characterized in that the contrast threshold value comprises a value in the range of 3% -10%.

6. The method according to any one of the preceding claims, characterized in that the contrast threshold value comprises a value of 5%.

7. The method according to any one of the preceding claims, characterized in that the method comprises a filtering of the determined visibility with a statistical method.

8. The method according to claim 7, characterized in that the filtering comprises filtering with a Kalman filter.

9. The method according to any one of the preceding claims, characterized in that method comprises a driving of a vehicle device (5-7) in dependence of the determined visibility.

10. The method according to claim 9, characterized in that the Fahrzeugvoπϊchtung a rear fog lamp (5), a fog light (5), an automatic cruise control system (6) and / or a warning device (7) for the driver of the vehicle (10).

11. A device for determining a range of vision for a vehicle, comprising:

- An image detection device (2, 3) for detecting an image information of an environment of the vehicle (10), and

- A processing unit (4) which is coupled to the image capture device (2, 3), wherein the processing unit (4) is configured:

To determine image portions of the image information having a contrast greater than a predetermined contrast threshold,

To determine distances (dτd ₄ ) between each of the image capture devices (2, 3) and sites (12-15) in the environment corresponding to the particular image sections, and

- To determine the largest of the certain distances (d ^ d ^ as a measure of the visibility.

12. The device according to claim 11, characterized in that the device (1) for carrying out the method according to one of claims 2-10 is configured.