Method and apparatus for adjusting field of view of an exterior mirror

Abstract

The application provides a rearview mirror field of view adjusting method and device, the method comprises the following steps: step 1, determining a reference area of ORB feature detection; step 2, performing ORB feature detection according to the reference area to obtain a first feature point set; step 3, obtaining turning information of a target vehicle, and performing ORB feature detection in the reference area to obtain a second feature point set; step 4, matching the first feature point set and the second feature point set to determine a first quantity and a first direction of pixel movement; and step 5, adjusting a display field of view of the rearview mirror according to the first quantity and the first direction. The application realizes the follow-up field of view adjustment of a semi-trailer truck and improves driving safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electronic rearview mirror, and in particular to a method and device for adjusting the field of view of an outside rearview mirror. BACKGROUND

[0002] Due to the large blind area of the trailer of a semi-trailer truck, traffic accidents are prone to occur, and therefore how to reduce the blind area of the trailer has always been a hot research topic. The mainstream solution at present is to calculate the steering angle of the front wheels of the tractor through the steering wheel rotation angle and the steering ratio, collect the angle change relationship between the front wheel angle of the tractor and the vehicle body at different speeds, and then use an algorithm to cut and display the field of view of the trailer on the steering side, so that the display range of the field of view changes with the steering wheel angle and the vehicle speed. However, in practice, the front wheel angle of the tractor changes variously at different speeds, and it is difficult to obtain all the conditions. When the obtained data is not enough, the field of view display is very abrupt, and even sudden jumps occur, which cannot smoothly realize the follow-up field of view. Moreover, since the tractor and the trailer of the semi-trailer truck can be freely combined, the angle change relationship between the front wheel angle of the tractor and the vehicle body at different speeds collected in advance cannot be used when the tractor or the trailer is replaced, resulting in poor universality of this solution. SUMMARY

[0003] The present application provides a method and device for adjusting the field of view of an outside rearview mirror, aiming to solve the defects in the prior art, realize the follow-up field of view adjustment of a semi-trailer truck, and improve the driving safety.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present application is as follows:

[0005] In one aspect, the present application provides a method for adjusting the field of view of an outside rearview mirror, comprising:

[0006] Step 1: determining a reference area for ORB feature detection;

[0007] Step 2: performing ORB feature detection according to the reference area to obtain a first set of feature points;

[0008] Step 3: obtaining the turning information of a target vehicle, and performing ORB feature detection in the reference area to obtain a second set of feature points;

[0009] Step 4: matching the first set of feature points and the second set of feature points to determine a first number and a first direction of pixel movement;

[0010] Step 5: adjusting the display field of view of the outside rearview mirror according to the first number and the first direction.

[0011] Specifically, the step 1 comprises:

[0012] Step 101, obtaining a target rear image, wherein the target rear image contains a marker for locating a rear position;

[0013] Step 102, receiving a region selection instruction, and taking the marker and at least part of the rear image region as a reference region for ORB feature detection according to the region selection instruction.

[0014] Specifically, the step 2 comprises:

[0015] Step 201, taking a minimum circumscribed square of the reference region as a current feature extraction region;

[0016] Step 202, performing ORB feature extraction in the current feature extraction region, and judging whether the number of extracted features exceeds a preset number, and if yes, outputting the features as a first feature point set, and if not, entering the next step;

[0017] Step 203, judging whether the side length of the current feature extraction region has reached a maximum value, and if yes, judging that feature extraction fails, and if not, increasing the side length of the circumscribed square of the reference region by a preset step as a new current feature extraction region, and returning to step 202.

[0018] Specifically, the step 4 comprises determining a pixel movement number and a direction by a local BA algorithm.

[0019] Specifically, the step 5 comprises translating the image displayed by the outside rearview mirror by the first number in the same direction as the first direction.

[0020] Another aspect of the application provides an outside rearview mirror field of view adjusting device, comprising:

[0021] a feature detection module, and a reference region determination module, a vehicle control module and a feature matching module connected thereto, further comprising a movement determination module connected to the feature matching module, and an image adjusting module connected to the movement determination module;

[0022] The reference region determination module is configured to determine a reference region for ORB feature detection.

[0023] The feature detection module is configured to perform ORB feature detection, and obtain a first feature point set and a second feature point set.

[0024] The vehicle control module is configured to control a target vehicle to turn.

[0025] The feature matching module is configured to match the first feature point set and the second feature point set.

[0026] The movement determination module is configured to determine a first number and a first direction of pixel movement.

[0027] The image adjusting module is configured to adjust the display field of view of the outside rearview mirror according to the number and direction of the pixel movement.

[0028] Specifically, the reference region determining module comprises an image unit and a man-machine interaction unit and a determining unit connected thereto.

[0029] The image unit is configured to acquire and display a target vehicle tail image.

[0030] The man-machine interaction unit is configured to receive a region selection instruction.

[0031] The determining unit is configured to take the marker and at least part of the vehicle tail image region as a reference region for ORB feature detection according to the region selection instruction.

[0032] Specifically, the feature detecting module comprises a feature judging unit, and a feature extracting unit, a feature output unit and an extraction region updating unit connected to the feature judging unit, and further comprises an extraction region determining unit connected to the feature extracting unit and the extraction region updating unit.

[0033] The extraction region determining unit is configured to take the minimum circumscribed square of the reference region as a current feature extraction region.

[0034] The feature extracting unit is configured to perform ORB feature extraction in the current feature extraction region.

[0035] The feature judging unit is configured to determine whether the number of extracted features exceeds a preset number.

[0036] The feature output unit is configured to output the detected features into a corresponding feature point set.

[0037] The extraction region updating unit is configured to increase the side length of the circumscribed square of the reference region by a preset step length.

[0038] Specifically, the feature matching module is a local BA algorithm module.

[0039] The present application has the beneficial effect that: the present application places a marker with high resolution at the tail of the trailer, selects the marker on the display screen, extracts ORB feature points near the marker in the image, determines the number of pixels moved between adjacent frames through matching inter-frame ORB characteristics, and finally translates the display image according to the number of pixels moved, thereby realizing the follow-up field of view adjustment of the semi-trailer truck and improving the driving safety. BRIEF DESCRIPTION OF DRAWINGS

[0040] Figure 1 is a flowchart of the outside rearview mirror field of view adjusting method of the present application.

[0041] Figure 2 is a structural schematic diagram of the rearview mirror field of view adjustment device of the present application;

[0042] Figure 3 is a structural schematic diagram of the reference area determination module of the present application;

[0043] Figure 4 is a structural schematic diagram of the feature detection module of the present application. DETAILED DESCRIPTION

[0044] The embodiments of the present application will be described in detail below with reference to the accompanying drawings, which are used only for reference and illustration, and do not constitute a limitation on the scope of patent protection of the present application.

[0045] In the flow described in the specification, claims or drawings of the present application, the serial numbers of the individual steps (such as steps 10, 20, etc.) are only used to distinguish the individual steps, and the serial numbers themselves do not represent any execution order. It should be noted that the "first", "second" and the like in the present text are only used to distinguish the description objects and the like, and do not represent the order of precedence.

[0046] Example 1

[0047] As shown in Figure 1 , the present embodiment provides a rearview mirror field of view adjustment method, comprising:

[0048] Step 1, determining the reference area of ORB feature detection.

[0049] In the present embodiment, the step 1 comprises:

[0050] Step 101, acquiring a target trailer image, the target trailer image containing a marker for positioning the trailer position.

[0051] In specific implementation, first, a semi-trailer is parked in a parking space, then a marker with a large color contrast with the trailer compartment is placed on the ground at the trailer tail, and the marker should be visible in the display image of the rearview mirror.

[0052] Step 102, receiving a region selection instruction, and according to the region selection instruction, taking the marker and at least part of the trailer image region as the reference area of ORB feature detection.

[0053] In a specific implementation, the processor of the exterior rearview mirror receives a preset shape (e.g., a circle) drawn by an operator (e.g., a driver) on the exterior rearview mirror display device through a human-computer interaction device (e.g., an adjusting knob, a touch screen, etc.) to enclose the marker and a part of the image of the vehicle tail, and the image in the part of the enclosed circular region is the reference region.

[0054] Step 2, performing ORB feature detection according to the reference region to obtain a first feature point set.

[0055] ORB (Oriented FAST and Rotated BRIEF) is a fast and efficient image feature detection algorithm that combines FAST corner detection and BRIEF descriptor and is improved to provide scale invariance, rotation invariance, and certain noise resistance, and thus is widely used in image processing.

[0056] In this embodiment, the step 2 includes:

[0057] Step 201, taking the minimum circumscribed square of the reference region as a current feature extraction region.

[0058] Due to the requirements of regulations, the color resolution of the trailer compartment is generally high, and a reflective strip or the like is generally pasted at the vehicle tail to provide a clear marker, so the color at this position often has a certain texture change from the surrounding environment, and a sufficient number of effective ORB features can be easily extracted near the marker, and thus taking the minimum circumscribed square of the reference region as the current initial feature extraction region can greatly reduce the computation amount of the image.

[0059] Step 202, performing ORB feature extraction in the current feature extraction region, and determining whether the number of extracted features exceeds a preset number, if yes, outputting the features as a first feature point set, and if no, proceeding to the next step.

[0060] Step 203, determining whether the side length of the current feature extraction region has reached a maximum value, if yes, determining that the feature extraction fails, and if no, increasing the side length of the circumscribed square of the reference region by a preset step as a new current feature extraction region, and returning to step 202.

[0061] In this embodiment, the size of the feature extraction region is determined by the number of extracted ORB features, and the lower limit of the number of extracted features is set to n, when the number of extracted features is less than n, the square side length of the extraction region is continuously increased by δ until the number of extracted features meets the number requirement.

[0062] Step 3, obtaining turning information of the target vehicle, and performing ORB feature detection in the reference region to obtain a second feature point set.

[0063] The method for obtaining the second feature point set is the same as that in step 2, and will not be repeated here.

[0064] Step 4, matching the first feature point set and the second feature point set to determine a first quantity and a first direction of pixel movement.

[0065] In this embodiment, the quantity and direction of pixel movement are determined by a local BA (Bundle Adjustment) algorithm.

[0066] Step 5, adjusting the display field of view of the outside rearview mirror according to the first quantity and the first direction.

[0067] In this embodiment, the step 5 includes: translating the image displayed by the outside rearview mirror by the first quantity in the same direction as the first direction.

[0068] The outside rearview mirror field of view adjustment method of this embodiment can be executed by a processor built in the outside rearview mirror itself, or by a processor outside the outside rearview mirror, and the execution result is returned to the outside rearview mirror, which is not limited by the present application.

[0069] Example 2

[0070] As shown in Figure 2 this embodiment provides an outside rearview mirror field of view adjustment device, which includes:

[0071] a feature detection module, a reference area determination module, a vehicle control module, a feature matching module connected to the feature detection module, a movement determination module connected to the feature matching module, and an image adjustment module connected to the movement determination module;

[0072] The reference area determination module is configured to determine a reference area for ORB feature detection.

[0073] The feature detection module is configured to perform ORB feature detection to obtain a first feature point set and a second feature point set.

[0074] The vehicle control module is configured to control the target vehicle to turn.

[0075] The feature matching module is configured to match the first feature point set and the second feature point set.

[0076] The movement determination module is configured to determine a first quantity and a first direction of pixel movement.

[0077] The image adjustment module is configured to adjust the display field of view of the outside rearview mirror according to the quantity and direction of pixel movement.

[0078] The working process of the rearview mirror field of view adjustment device described in this embodiment is as shown in the rearview mirror field of view adjustment method in Embodiment 1, and will not be repeated here.

[0079] like Figure 3 As shown, in another embodiment of the present invention, the reference area determination module includes an image unit and a human-computer interaction unit and a determination unit connected thereto;

[0080] The image unit is used to acquire and display the image of the target vehicle's rear end;

[0081] The human-computer interaction unit is used to receive region selection instructions;

[0082] The determining unit is used to use the marker and at least part of the rear image area as the reference area for ORB feature detection according to the area selection instruction.

[0083] like Figure 4 As shown, in another embodiment of the present invention, the feature detection module includes a feature judgment unit, and a feature extraction unit, a feature output unit, and an extraction region update unit connected thereto, and also includes an extraction region determination unit connected to the feature extraction unit and the extraction region update unit.

[0084] The extraction region determination unit is used to take the smallest circumscribed square of the reference region as the current feature extraction region.

[0085] The feature extraction unit is used to extract ORB features in the current feature extraction region;

[0086] The feature judgment unit is used to determine whether the number of extracted features exceeds a preset number.

[0087] The feature output unit is used to output the detected features into the corresponding feature point set;

[0088] The extraction region update unit is used to increase the side length of the circumscribed square of the reference region by a preset step size.

[0089] In another embodiment of the present invention, the feature matching module is a local BA algorithm module.

[0090] The rearview mirror field of view adjustment device described in this embodiment can be integrated into the processor built into the rearview mirror itself, or into a processor outside the rearview mirror, or partly integrated into the processor built into the rearview mirror itself and partly integrated into a processor outside the rearview mirror. This invention does not limit this.

[0091] The above merely provides the preferred embodiment of the present application, and cannot allude to the protection scope of the present application; therefore, any equivalent change made according to the patent application scope of the present application shall still fall within the scope of the present application.

Claims

1. A method for adjusting the field of view of an exterior rearview mirror, characterized in that, include: Step 1: Determine the baseline region for ORB feature detection; Step 2: Perform ORB feature detection based on the reference region to obtain the first feature point set; Step 3: Obtain the turning information of the target vehicle and perform ORB feature detection within the reference area to obtain the second feature point set; Step 4: Match the first feature point set and the second feature point set to determine the first number and first direction of pixel movement; Step 5: Adjust the field of view of the exterior rearview mirror according to the first quantity and the first direction; Step 1 includes: Step 101: Obtain the target vehicle rear image, wherein the target vehicle rear image includes a marker for locating the vehicle rear position; Step 102: Receive region selection instruction, and use the sign and at least part of the rear image region as the reference region for ORB feature detection according to the region selection instruction; Step 2 includes: Step 201: Take the smallest bounding square of the reference region as the current feature extraction region; Step 202: Perform ORB feature extraction in the current feature extraction region and determine whether the number of extracted features exceeds the preset number. If yes, output the features as the first feature point set; otherwise, proceed to the next step. Step 203: Determine whether the side length of the current feature extraction region has reached the maximum value. If yes, the feature extraction is determined to have failed. Otherwise, the side length of the circumscribed square of the reference region is increased by a preset step size as the new current feature extraction region, and the process returns to step 202.

2. The method for adjusting the field of view of the exterior rearview mirror according to claim 1, characterized in that, Step 4 includes: determining the number and direction of pixel movement using a local BA algorithm.

3. The method for adjusting the field of view of the exterior rearview mirror according to claim 2, characterized in that, Step 5 includes: shifting the image displayed by the exterior rearview mirror by the first amount in the same direction as the first direction.

4. A rearview mirror field of view adjustment device, characterized in that, include: The feature detection module, and the reference region determination module, vehicle control module, and feature matching module connected thereto, also include a motion determination module connected to the feature matching module and an image adjustment module connected to the motion determination module; The reference region determination module is used to determine the reference region for ORB feature detection. The feature detection module is used to perform ORB feature detection and obtain a first feature point set and a second feature point set; The vehicle control module is used to control the target vehicle to turn; The feature matching module is used to match the first feature point set and the second feature point set; The movement determination module is used to determine a first number and a first direction of pixel movement; The image adjustment module is used to adjust the field of view of the external rearview mirror according to the first number and first direction of the pixel movement; The reference region determination module includes an image unit and a human-computer interaction unit and a determination unit connected thereto. The image unit is used to acquire and display a target vehicle rear image, the target vehicle rear image including a marker for locating the vehicle rear position; The human-computer interaction unit is used to receive region selection instructions; The determining unit is used to use the marker and at least part of the rear image area as the reference area for ORB feature detection according to the area selection instruction. The feature detection module includes a feature judgment unit, and a feature extraction unit, a feature output unit, and an extraction region update unit connected thereto, and also includes an extraction region determination unit connected to the feature extraction unit and the extraction region update unit. The extraction region determination unit is used to take the smallest circumscribed square of the reference region as the current feature extraction region. The feature extraction unit is used to extract ORB features in the current feature extraction region; The feature judgment unit is used to determine whether the number of extracted features exceeds a preset number. The feature output unit is used to output the detected features into the corresponding feature point set; The extraction region update unit is used to increase the side length of the circumscribed square of the reference region by a preset step size.

5. The rearview mirror field of view adjustment device according to claim 4, characterized in that, The feature matching module is a local BA algorithm module.

Images