Object distance measurement method, system and device for panoramic image and storage medium

Abstract

The invention discloses an object distance measurement method, system and device for a panoramic image, and a storage medium, belongs to the technical field of distance measurement based on images, and solves the technical problem that the target distance in an intersection area of two adjacent panoramic cameras cannot be accurately measured. The method comprises the steps that no less than four surround view cameras are included; collecting target images in the crossed area through two adjacent all-round cameras respectively; performing clock synchronization on the target image; respectively extracting and matching image feature points of the target image; calculating the coordinates of the image feature points and the optical center of the look-around camera in a world coordinate system; constructing a target triangle according to the optical center and the world coordinate system coordinates of the image feature points; and according to the actual distance between the target triangle and the two optical centers, the distance between the automobile body and the image feature points is calculated by using a triangulation ranging method, so that the ranging precision of the panoramic image surround-view camera is greatly improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image-based distance measurement, and more specifically, to a method, system, device and storage medium for measuring the distance of objects in a panoramic image. Background Art

[0002] With the rapid development of road traffic, especially highways, the traffic accident rate is also on the rise, and traffic safety has become more and more the focus of people's attention. Therefore, the research on vehicle safety assisted driving technology provides vehicles with safety assisted driving functions, thereby providing intelligent technical services to reduce traffic accidents caused by subjective factors of drivers. Computer vision perception technology has gradually become the research focus of vehicle safety assisted driving technology due to its large amount of information and good stability. With the continuous development of computer vision technology, the application of computer vision perception technology to the on-board visual perception equipment of road vehicles or non-road vehicles can greatly improve the driving safety of automobiles.

[0003] Currently, the panoramic system is used to observe the environment around the vehicle body: the system uses the four fisheye cameras at the front, rear, left and right to perform data synchronization, distortion correction, graphics stitching and other software processing to form a 3D plane 360-degree panoramic surround effect, and sends the final image to the multimedia control unit for display on the multimedia display screen.

[0004] Since the original design idea of ​​panoramic cameras is to pursue a larger viewing angle, two adjacent surround-view cameras will form a certain area of ​​overlap, which is often a blind spot in the driver's field of vision, and objects in the overlapping area cannot be accurately measured. However, it is often difficult to achieve ideal accuracy by simply relying on a surround-view camera and a vehicle-mounted terminal to measure the distance of the obstacles ahead. Summary of the invention

[0005] The technical problem to be solved by the present invention is to address the above-mentioned deficiencies in the prior art. One purpose of the present invention is to provide a method for measuring the distance of objects in a panoramic image, which can solve the problem of low ranging accuracy of surround-view cameras in the prior art.

[0006] A second objective of the present invention is to provide a system for measuring the distance between objects in a panoramic image.

[0007] A third object of the present invention is to provide a computer device.

[0008] A fourth objective of the present invention is to provide a computer storage medium.

[0009] In order to achieve the above-mentioned first objective, the present invention provides a method for measuring the distance of an object in a panoramic image, comprising installing no less than four surround view cameras at different positions of a vehicle body to achieve panoramic image shooting, and further comprising the following steps:

[0010] Step 1. Use two adjacent surround view cameras to collect target images in the intersection area respectively;

[0011] Step 2. Perform clock synchronization on the target images captured by the two surround view cameras;

[0012] Step 3. Extract and match the image feature points of the target images of the two surround view cameras respectively;

[0013] Step 4. Calculate the coordinates of the image feature points in the world coordinate system and the coordinates of the optical centers of the two adjacent surround view cameras in the world coordinate system; construct a target triangle according to the world coordinate system coordinates of the two optical centers and the world coordinate system coordinates of the image feature points;

[0014] Step 5. According to the actual distance between the target triangle and the two optical centers, the distance between the vehicle body and the image feature point is calculated using triangulation.

[0015] As a further improvement, the surround view camera is a fisheye camera.

[0016] Furthermore, the surround view cameras are installed at different positions on the same horizontal plane.

[0017] Furthermore, the coordinates of the image feature points in the surround camera coordinate system are first obtained, and then the coordinates of the image feature points in the world coordinate system are calculated using coordinate transformation.

[0018] Furthermore, the actual distance between the optical centers of the two surround-view cameras is obtained by measuring the actual positions of the two surround-view cameras.

[0019] Furthermore, the image feature points refer to points in the image that have distinct characteristics and can effectively reflect the essential characteristics of the image and can identify the target objects in the image. The image feature points include: corner points, boundary points, and segmentation points; the image feature points can be extracted using a scale-invariant feature extraction method.

[0020] Further, let the target triangle be △ABD, where A and B are two optical centers respectively, and D is an image feature point;

[0021] Construct a right triangle △AOB with side AB as the common side, side AB as the hypotenuse, and O as the origin of the world coordinate system;

[0022] In △AOB:

[0023]

[0024] In △ABD:

[0025]

[0026] The straight-line distance between the image feature point and the coordinate origin O is:

[0027]

[0028] Right now:

[0029]

[0030] The coordinates of the image feature points are:

[0031] (X,Y)=(L2-L4*cos(∠ABO+∠ABD),L4*sin(∠ABO+∠ABD)).

[0032] In order to achieve the above-mentioned second objective, the present invention provides a panoramic image object distance measurement system, comprising no less than 4 surround view cameras installed at different positions of a vehicle body to achieve panoramic image shooting, and also comprising:

[0033] A collection module, used to collect target images in the intersection area through two adjacent surround view cameras;

[0034] A synchronization module, used for performing clock synchronization on the target images captured by the two surround view cameras;

[0035] An extraction module, used to respectively extract and match image feature points of the target images of the two surround view cameras;

[0036] A triangle construction module is used to calculate the coordinates of the image feature points in the world coordinate system and the coordinates of the optical centers of the two adjacent surround view cameras in the world coordinate system; a target triangle is constructed according to the world coordinate system coordinates of the two optical centers and the world coordinate system coordinates of the image feature points;

[0037] The calculation module is used to calculate the distance between the vehicle body and the image feature point by using a triangulation method according to the actual distance between the target triangle and the two optical centers.

[0038] In order to achieve the third objective above, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the above-mentioned method for measuring the distance of objects in a panoramic image when executing the computer program.

[0039] In order to achieve the fourth objective above, the present invention provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the computer program implements the above-mentioned method for measuring the distance of objects in a panoramic image.

[0040] Beneficial Effects

[0041] Compared with the prior art, the present invention has the following advantages:

[0042] The present invention calculates the distance between the vehicle and the target in front through a triangle formed by two adjacent surround-view cameras and the target in front of the vehicle. The calculation process is a precise calculation, which greatly improves the ranging accuracy of the front-view camera. In addition, the present invention is reasonably designed, which improves the detection performance while reducing the manufacturing and use costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the present invention;

[0044] Figure 2 It is a schematic diagram of constructing a target triangle in the present invention.

[0045] Among them: 1-surround view camera, 2-intersection area, 3-car body. DETAILED DESCRIPTION

[0046] The present invention will be further described below with reference to the specific embodiments in the accompanying drawings.

[0047] See also Figure 1-2 A method for measuring the distance of an object in a panoramic image includes installing no less than four surround view cameras 1 at different positions of a vehicle body 3 to achieve panoramic image shooting. The method also includes the following steps 1 to 5:

[0048] Step 1. Use two adjacent surround view cameras 1 to respectively capture target images in the intersection area 2, such as pedestrians, vehicles, and animals in front of the car.

[0049] Step 2. Since different surround view cameras are located at different positions on the car, different surround view cameras have different positions relative to the front target when shooting the front target, and different surround view cameras also shoot the front target at different angles. Therefore, the images of the front target shot by different surround view cameras are different.

[0050] The target images captured by the two surround view cameras 1 need to be clock synchronized.

[0051] If the two cameras capture the same target at different times, the position of the target in the image will be different. In order to ensure the accuracy of distance measurement, the images captured by the surround view camera must be synchronized. Only when the images are captured at the same time can the distance between the target in front and the vehicle body be accurately measured and calculated.

[0052] Step 3. Extract and match the image feature points of the target images of the two surround view cameras 1 respectively.

[0053] Image feature points refer to points in an image that have distinct characteristics and can effectively reflect the essential features of the image and identify the target object in the image. The image feature points include: corner points, boundary points, and segmentation points. Image feature points can be extracted using scale-invariant feature extraction methods, such as ORB feature extraction method, SIFT feature extraction method, and SURF feature extraction method.

[0054] After extracting the image feature points captured by the surround view camera, they are matched to find the matching image feature points, that is, the same points of the front target in the images captured by two adjacent surround view cameras at the same time.

[0055] Step 4. Calculate the coordinates of the image feature points in the world coordinate system and the coordinates of the optical centers of two adjacent surround view cameras 1 in the world coordinate system; construct the target triangle based on the world coordinate system coordinates of the two optical centers and the world coordinate system coordinates of the image feature points.

[0056] Step 5. According to the actual distance between the target triangle and the two optical centers, the distance between the vehicle body 3 and the image feature point is calculated using the triangulation method.

[0057] In this embodiment, the surround view camera 1 is a fisheye camera, and the surround view camera 1 is installed at different positions on the same horizontal plane, so as to reduce the influence of different height differences.

[0058] In step 4, the coordinates of the image feature points in the coordinate system of the surround view camera 1 are first obtained, and then the coordinates of the image feature points in the world coordinate system are calculated using coordinate transformation.

[0059] The actual distance between the optical centers of the two surround view cameras 1 is obtained by measuring the actual positions of the two surround view cameras 1 .

[0060] Triangulation distance calculation Figure 2 As shown, let the target triangle be △ABD, where A and B are the two optical centers respectively, and D is the image feature point.

[0061] Construct a right triangle △AOB with side AB as the common side, side AB as the hypotenuse, and O as the origin of the world coordinate system.

[0062] In △AOB:

[0063]

[0064] In △ABD:

[0065]

[0066] The straight-line distance between the image feature point and the coordinate origin O is:

[0067]

[0068] Right now:

[0069]

[0070] The coordinates of the image feature points are:

[0071] (X,Y)=(L2-L4*cos(∠ABO+∠ABD),L4*sin(∠ABO+∠ABD))

[0072] Through the above calculation, accurate distance measurement of objects in the intersection area 2 of two adjacent surround view cameras 1 can be obtained.

[0073] A panoramic image object distance measurement system includes at least four surround view cameras 1 installed at different positions of a vehicle body 3 to achieve panoramic image shooting, and also includes:

[0074] A collection module, used to collect target images in the intersection area 2 through two adjacent surround view cameras 1 respectively;

[0075] A synchronization module, used for clock synchronization of target images captured by two surround view cameras 1;

[0076] An extraction module, used to extract and match image feature points of target images of two surround view cameras 1 respectively;

[0077] A triangle construction module is used to calculate the coordinates of the image feature points in the world coordinate system and the coordinates of the optical centers of two adjacent surround view cameras 1 in the world coordinate system; a target triangle is constructed according to the world coordinate system coordinates of the two optical centers and the world coordinate system coordinates of the image feature points;

[0078] The calculation module is used to calculate the distance between the vehicle body 3 and the image feature point by using the triangulation method according to the actual distance between the target triangle and the two optical centers.

[0079] A computer device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, the above-mentioned method for measuring the distance between objects in a panoramic image is implemented.

[0080] A computer-readable storage medium stores a computer program, which implements the above-mentioned method for measuring the distance of objects in a panoramic image when executed by a processor.

[0081] The above are only preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the structure of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent.

Claims

1. A method for measuring the distance of an object in a panoramic image, comprising installing no less than four surround view cameras (1) at different positions of a vehicle body (3) to achieve panoramic image shooting, characterized in that: The following steps are also included: Step 1. Using two adjacent surround view cameras (1), respectively collect target images within the intersection area (2); Step 2. Performing clock synchronization on the target images captured by the two surround view cameras (1); Step 3. respectively extracting and matching image feature points of the target images of the two surround view cameras (1); Step 4. Calculate the coordinates of the image feature points in the world coordinate system and the coordinates of the optical centers of the two adjacent surround view cameras (1) in the world coordinate system; construct a target triangle according to the world coordinate system coordinates of the two optical centers and the world coordinate system coordinates of the image feature points; Step 5. According to the actual distance between the target triangle and the two optical centers, the distance between the vehicle body (3) and the image feature point is calculated using a triangulation method.

2. The method for measuring the object distance of a panoramic image according to claim 1, characterized in that: The surround view camera (1) is a fisheye camera.

3. The method for measuring object distance in a panoramic image according to claim 1, characterized in that: The surround view cameras (1) are installed at different positions on the same horizontal plane.

4. The method for measuring the object distance of a panoramic image according to claim 1, characterized in that: First, the coordinates of the image feature points in the coordinate system of the surround view camera (1) are obtained, and then the coordinates of the image feature points in the world coordinate system are calculated using coordinate transformation.

5. The method for measuring the object distance of a panoramic image according to claim 1, characterized in that: The actual distance between the optical centers of the two surround-view cameras (1) is obtained by measuring the actual positions of the two surround-view cameras (1).

6. The method for measuring object distance in a panoramic image according to claim 1, characterized in that: The image feature points refer to points in the image that have distinct characteristics and can effectively reflect the essential characteristics of the image and can identify the target objects in the image. The image feature points include: corner points, boundary points, and segmentation points; the image feature points can be extracted using a feature extraction method based on scale invariance.

7. The method for measuring object distance in a panoramic image according to claim 1, characterized in that: Let the target triangle be △ABD, where A and B are two optical centers respectively, and D is an image feature point; Construct a right triangle △AOB with side AB as the common side, side AB as the hypotenuse, and O as the origin of the world coordinate system; In △AOB: In △ABD: The straight-line distance between the image feature point and the coordinate origin O is: Right now: The coordinates of the image feature points are: (X,Y)=(L2-L4*cos(∠ABO+∠ABD),L4*sin(∠ABO+∠ABD)).

8. A panoramic image object distance measurement system, comprising at least four surround view cameras (1) installed at different positions on a vehicle body (3) to achieve panoramic image shooting, characterized in that: Also includes: A collection module, used for collecting target images in the intersection area (2) respectively through two adjacent surround-view cameras (1); A synchronization module, used for performing clock synchronization on the target images captured by the two surround view cameras (1); An extraction module, used for respectively extracting and matching image feature points of target images of the two surround view cameras (1); A triangle construction module is used to calculate the coordinates of the image feature points in the world coordinate system and the coordinates of the optical centers of the two adjacent surround view cameras (1) in the world coordinate system; and to construct a target triangle according to the world coordinate system coordinates of the two optical centers and the world coordinate system coordinates of the image feature points; A calculation module is used to calculate the distance between the automobile body (3) and the image feature point using a triangulation method according to the actual distance between the target triangle and the two optical centers.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the method for measuring the distance of objects in a panoramic image as described in any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for measuring the distance of objects in a panoramic image according to any one of claims 1 to 7 is implemented.

Images