METHOD FOR DETERMINING A STATE OF A TAILGATE OF A PICKUP VEHICLE BY ANALYSIS OF A REGION OF INTEREST, COMPUTER PROGRAM PRODUCT, ELECTRONIC COMPUTING DEVICE AS WELL AS CAMERA SYSTEM

Abstract

A method for determining a state of a tailgate (9) of a pickup vehicle (1) by a camera system (2), includes capturing a cargo bed (5) of the pickup vehicle (1) in an image (8) captured by a camera (3), determining a first region of interest (ROI1) in the captured image (8) by an electronic computing device (4) of the camera system (2), wherein the first region of interest (ROI1) includes the potential tailgate (9) at least partially, determining a parameter optically characterizing the tailgate (9) depending on the captured image (8), analyzing the first region of interest (ROI1) with regard to the characterizing parameter, and determining a closed state of the tailgate (9) as state or determining an opened state of the tailgate (9) as state depending on the analysis of the first region of interest (ROI1). A computer program product, an electronic computing device (4) as well as a camera system (2) are also disclosed.

Claims

1. A method for determining a state of a tailgate of a pickup vehicle by a camera system of the platform vehicle the method comprising: capturing a cargo bed of the pickup vehicle in an image captured by a camera of the camera system; determining a first region of interest in the captured image by an electronic computing device of the camera system, wherein the first region of interest includes the potential tailgate at least partially; determining a parameter optically characterizing the tailgate depending on the captured image; analyzing the first region of interest with regard to the characterizing parameter; and determining a closed state of the tailgate as state or determining an opened state of the tailgate as state depending on the analysis of the first region of interest.

2. The method according to claim 1, wherein least one region of interest is determined depending on a histogram backprojection by the electronic computing device.

3. The method according to claim 1, wherein a second region of interest is determined, which at least in certain areas includes the cargo bed, and the characterizing parameter is determined depending on the second region of interest by the electronic computing device.

4. The method according to claim 3, wherein of the first region of interest a first histogram is generated, and of the second region of interest a second histogram is generated, and as a characterizing parameter a pixel value in the histograms is determined and by comparison of the first histogram with the second histogram as analysis based on a contrast comparison the state of the tailgate is determined by the electronic computing device.

5. The method according to claim 3, wherein in the case of a predetermined number of non-zero pixel values of the pixel values in the first region of interest the closed state is determined, and in the case of a predetermined number of zero pixel values of the pixel values in the first region of interest the opened state is determined by the electronic computing device.

6. The method according to claim 1, wherein in addition the first region of interest is analyzed by a Hough transformation and depending on this analysis the state of the tailgate is determined by the electronic computing device.

7. The method according to claim 1, wherein the first region of interest is determined depending on at least one predetermined size indication of the cargo bed by the electronic computing device.

8. The method according to claim 1, wherein the first region of interest depending on at least one extrinsic parameter of the camera is determined by the electronic computing device.

9. The method according to claim 1, wherein the first region of interest is subdivided into a multitude of subregions and for a respective subregion a respective motion vector of the subregion as characterizing parameter is determined and depending on the multitude of determined motion vectors for the multitude of the subregions the state of the tailgate is determined by the electronic computing device.

10. The method according to claim 9, wherein by analysis of an optical flow within the respective subregion the respective motion vector is determined by the electronic computing device.

11. The method according to claim 1, wherein on the basis of a contrast comparison in the image and based on a motion vector analysis in the image the state of the tailgate is determined by the electronic computing device.

12. A computer program product with program means, which are stored in a computer-readable medium to perform the method for determining a state of a tailgate according to claim 1, when the computer program product is executed on a processor of an electronic computing device.

13. An electronic computing device with a computer program product according to claim 12.

14. A camera system for a pickup vehicle comprising: at least one camera; and an electronic computing device with a computer program product with program means, which are stored in a computer-readable medium, wherein the camera system is configured for performing a method according to claim 1.

Description

[0043] These show in:

[0044] FIG. 1 a schematic perspective view of an embodiment of a pickup vehicle comprising an embodiment of a camera system;

[0045] FIG. 2 a schematic perspective view of a taken image;

[0046] FIG. 3 a further schematic view of a taken image; and

[0047] FIG. 4 yet a further schematic view of a taken image.

[0048] In the figures same elements or elements having the same function are equipped with the same reference signs.

[0049] FIG. 1 in a schematic perspective view shows an embodiment of a pickup vehicle 1 comprising an embodiment of a camera system 2. The pickup vehicle 1 is in particular a motor vehicle. The camera system 2 comprises at least one camera 3 as well as an electronic computing device 4. The camera 3 is in particular configured as central high-mounted braking lamp camera. In other words, it is in particular envisaged that by means of the camera 3 a cargo bed 5 of the pickup vehicle 1 can be captured. The camera 3 thus is oriented in the direction of a tailgate 6 of the pickup vehicle 1 so that a rear space 7 of the pickup vehicle 1 can be captured.

[0050] The camera 3 in particular is a central high-mounted camera 3 in the region of the third brake light of the pickup vehicle 1. This camera 3 is also referred to as Center High Mount Stop Light Camera (CHMSL Camera). By means of the camera 3 it is in particular facilitated that a cargo bed 5 of the motor vehicle can be captured. The pickup vehicle 1 can in particular also be referred to as pickup vehicle. The camera 3 is in particular already mounted on the motor vehicle and for instance configured for observation of a trailer arranged on the pickup. Additionally now the state of a tailgate 9 can be captured by the method according to the invention.

[0051] FIG. 2 in a schematic view shows an embodiment of an image 8 taken by the camera 3.

[0052] The image 8 shows in particular the cargo bed 5.

[0053] In the method for determining a state of the tailgate 9 of the pickup vehicle 1 by the camera system 2 a capturing of the cargo bed 5 of the pickup vehicle 1 is effected by means of the camera 3 of the camera system 2. A determining of a first region of interest ROI 1 in the captured image 8 by means of the electronic computing device 4 of the camera system 2 is effected, wherein the first region of interest ROI1 includes the potential tailgate 9 at least partially. A determining of a parameter opticallycharacterizing the tailgate 9 depending on the captured image 8 is effected. An analyzing of the first region of interest ROI1 with regard to the characterizing parameter is effected and a determining of a closed state of the tailgate 9 as state or a determining of an opened state of the tailgate 9 as state depending on the analysis of the first region of interest ROI1 is effected.

[0054] In what is shown in FIG. 2 it can here in particular be seen that the first region of interest ROI1 is determined depending on a histogram backprojection. In particular FIG. 2 shows the histogram backprojection.

[0055] For determining the state of the tailgate 9 it is in particular envisaged that a second region of interest ROI2 is determined, which at least partially includes the cargo bed 5, and the characterizing parameter is determined depending on the second region of interest ROI2.

[0056] In the following embodiment the second region of interest ROI2 is in particular a front part of the cargo bed 5. In particular the second region of interest ROI2 is a front edge 10 of the cargo bed 5. FIG. 2 shows in particular that by a first region of interest ROI1 a first histogram is generated and by the second region of interest ROI2 a second histogram is 15 generated and as characterizing parameter a pixel value in the histograms is determined and by comparison of the first histogram with the second histogram as analysis based on a contrast comparison the state of the tailgate 9 is determined.

[0057] Further, it is in particular envisaged that in the case of a predetermined number of non-zero pixel values the pixel values of the closed state is determined and in the case of a predetermined number of zero pixel values of the pixel values in the opened state is determined.

[0058] In other words FIG. 2 shows in particular that the pickup vehicle 1 is in a static state. For this purpose in a first state the capturing of the cargo bed 5 within the image 8 is effected.

[0059] In particular the capturing of the cargo bed 5 is effected by means of histogram backprojection. Then the determining of the first region of interest ROI1 and the determining of the second region of interest ROI2 is effected. In particular herein size indications of the pickup vehicle 1, in particular the cargo bed 5, can be used in order to determine the region of interest ROI1, ROI2 accordingly. These size indications are in particular predetermined and can thus be utilized by the electronic computing device 4 for analysis.

[0060] In particular it is envisaged that in the case of a predetermined number of non-zero pixel values of the pixel values the closed state is determined and in the case of a predetermined number of zero pixel values of the pixel values in the opened state is determined. In particular in the case of a large number of non-zero pixel values within the first region of interest ROI1 it can be determined that the tailgate 9 is closed, whereas in the case of a large number of zero pixel values it can be concluded that the tailgate 9 is opened.

[0061] FIG. 3 in a schematic perspective view shows the image 8. In particular FIG. 3 shows an enlarged view of the tailgate 9 from an inside 11. In particular FIG. 3 shows that additionally the first region of interest ROI1 is analyzed by a Hough transformation 12 and depending on this analysis the state of the tailgate 9 is determined. In particular this has its background in the fact that, should for instance due to a low contrast comparison, since for instance the difference of pickup vehicle 1 from an environment is not significant, because for instance the tailgate 9 is opened and the motor vehicle moves on a dark ground, by the Hough transformation 12 it can additionally be verified in which state the tailgate 9 is. This has its background in particular in the fact that by the Hough transformation 12 in the first region of interest ROI1 a corresponding line can be detected when the tailgate 9 is in the closed state. The line cannot be detected when the tailgate 9 is in an opened state. Yet again additionally it may be envisaged that based on an edge recognition it can be verified whether the tailgate 9 is opened or closed. For this purpose it is in particular envisaged that in the closed state of the tailgate 9 merely a top edge 12 of the tailgate 9 can be captured. In the opened state also the two lateral edges 13 can be captured, whereby it can be reliably determined whether the tailgate 9 is in an opened state or in a closed state.

[0062] FIG. 4 in a schematic perspective view shows a further image 8. It can in particular be envisaged that the first region of interest ROI1 is determined depending on at least one predetermined size indication of the cargo bed 5. Further it can in particular be envisaged that the first region of interest ROI1 is determined depending on at least one extrinsic parameter of the camera 3. In particular the at least one extrinsic parameter is predetermined. In particular for the method according to the invention it is necessary that the nominal position of the camera 3 is known. The extrinsic parameters are in particular one rotation parameter and one translation parameter of the camera 3. Should for instance the camera 3 have an incorrect orientation, this can be compensated for by corresponding corrections in the image 8. By a corresponding correction of the incorrect orientation of the camera 3 the opened or the closed state, respectively, of the tailgate 9 can be determined in an improved way.

[0063] Further FIG. 4 shows in particular that the first region of interest ROI1 is subdivided into a multitude of subregions 14 and for a respective subregion 14 a respective motion vector 15 of the subregion 14 as characterizing parameter is determined and depending on the multitude of the determined motion vectors 15 for the multitude of the subregions 14 the state of the tailgate 9 is determined. In particular in FIG. 4 the pickup vehicle 1 is in a dynamic state, in particular a moved state. By the camera 3 in particular a multitude of images 8 is taken so that the motion vectors 15 can be reliably captured. The analysis for the respective motion vectors 15 is effected in particular on the basis of an optical flow within the respective subregions 14 within the multitude of the taken images 8.

[0064] For instance for this purpose it may be envisaged that the first region of interest ROI1 is subdivided into the subregions 14, wherein these for instance form a 8 by 8 or 16 by 16-matrix. When the pickup vehicle 1 is in motion, the optical flow is determined, wherein this is performed for instance by the Sum of Squared Differences (SSD) or by the Sum of Absolute Differences (SAD). It can then be concluded that the tailgate 9 is closed, if the movement vectors 15 are nearly 0, since in particular the camera 3 and the tailgate 9 are fixed on the pickup vehicle 1 so that no relative movement of the camera 3 towards the closed tailgate 9 can be determined. On the other hand, if the movement vector 15 is captured as being large, in particular larger than a predetermined threshold value, it can be concluded that the tailgate 9 is in the opened state, since in particular the road is captured and not the tailgate 9 as such.

[0065] In particular it may be envisaged that on the basis of a contrast comparison in the image 8 and based on a movement vector analysis in the image 8 the state of the tailgate 9 is determined. In other words it can in particular be envisaged that both by the contrast comparison as it is in particular represented in FIG. 2, as well as by the motion vector analysis, as it is in particular represented in FIG. 4, additionally the state of the tailgate 9 can be determined. In particular this can then for instance be performed based on the formula:

Criterion.sub.tailgate=(weight.sub.static*Output.sub.static)+(vehicleMotion) *(weight.sub.dynamic*Output.sub.dynamic);

[0066] The invention also relates to a computer program product with program code means, which are stored in a computer-readable medium, in order to perform the method for determining a state of the tailgate 9, as it is shown in FIG. 1 to FIG. 4 if the computer program product is executed on a processor of the electronic computing device 3.