Visible light image geometric radiation stitching processing method

Abstract

The invention discloses a visible light image geometric radiation stitching processing method, and belongs to the technical field of image processing. The method comprises the following steps: acquiring two images # imgabs0 # and # imgabs1 # which need to be spliced; taking images of three objects at the boundary of an image # imgabs3 # adjacent to the image # imgabs2 #, enabling the images to form a triangular image template, enabling a boundary image of an image # imgabs5 # adjacent to the image # imgabs4 # to be matched with the triangular image template, finding a triangular image matched with the triangular image template in the image # imgabs6 #, and enabling the triangular image to be matched with the triangular image template in the image # imgabs4 #; translating and rotating the image # imgabs9 # according to the position relationship between the triangular image template in the image # imgabs7 # and the triangular image matched with the triangular image template in the image # imgabs8 #, so that the images of the three objects at the boundary of the image # imgabs10 # and the image # imgabs11 # are overlapped to obtain a new image # imgabs12 #; and adaptively adjusting the pixel value of the overlapped boundary of the image # imgabs13 # and the image # imgabs14 # to obtain a spliced image. The method is small in splicing error, and the spliced image is good in transition effect.

Description

Technical Field

[0001] The invention relates to a visible light image geometric radiation splicing processing method, belonging to the technical field of image processing. Background Art

[0002] Digital image processing is a method and technology for removing noise, enhancing, restoring, segmenting, extracting features, etc. from images. The emergence and rapid development of digital image processing are mainly affected by three factors: one is the development of computers; the second is the development of mathematics (especially the creation and improvement of theories); the third is the growth of application needs in agriculture, animal husbandry, forestry, environment, military, industry and medicine. In the field of digital image processing, it is often necessary to process large scene images. Due to the limited wide-angle shooting of cameras, multiple cameras are usually used in a matrix distribution to shoot images at different positions, and the generated visible light images are geometrically spliced ​​to obtain a complete and high-definition image. In the field of virtual reality, three-dimensional panorama (also known as real-life virtual) is a real scene based on panoramic images. Panorama is a group or more groups of photos taken by a camera in a 360-degree circle stitched into a panoramic image, through the realization of a real scene restoration display method for all-round interactive viewing. In order to ensure the effect of three-dimensional panoramic images, it is also necessary to stitch multiple interconnected images. Therefore, image stitching processing methods are needed to stitch discrete images into a complete high-definition image.

[0003] In the image stitching method disclosed in CN 104331872 B, corresponding first and second homonymous points are respectively marked in a correction area, wherein the first homonymous point is located in a first image to be corrected, and the second homonymous point is located in a second image to be corrected, and then a point whose distance to the two homonymous points meets a preset requirement is confirmed as a reference point through the first and second homonymous points, and further, according to the positions of the reference points, the image where the first homonymous point is located is corrected in a manner of correcting the first homonymous point to the reference point, and the second image is corrected in a manner of correcting the second homonymous point to the reference point, and finally the two corrected images are stitched. In the method, firstly, the two images are of different sizes and are stitched only through one common point, so the stitching error is large, and the image quality of the effect image generated after stitching is poor.

[0004] The existing visible light image stitching method has the problem of large stitching errors in the stitching step and poor image effect after stitching. The stitched image needs to be processed twice, which is relatively troublesome. Summary of the invention

[0005] In order to overcome the shortcomings of the prior art, the purpose of the present invention is to provide a visible light image geometric radiation stitching processing method, which has a small stitching error and a good transition effect of the stitched image.

[0006] In order to overcome the shortcomings of the prior art, the present invention provides a method for geometric radiation stitching of visible light images, which comprises the following steps: S1-1: Divide the large scene to be photographed into M×N areas. Each area is equipped with a camera to capture images of the area. The images captured by the camera in each area are marked as ; S1-2: Obtain images taken by cameras in adjacent areas and , , ; In the same image Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate to get a new image , so that the image and images The images of the three objects at the boundary overlap, , ; Make the image With image The pixel values ​​of overlapping boundaries are adaptively adjusted; S1-3: Repeat step S1-2 to traverse the images taken by N cameras at the same time and stitch them into an image of a large scene.

[0007] To achieve the above-mentioned purpose, the present invention also provides a visible light image geometric radiation stitching processing method, which comprises the following steps: S2-1: Get the two images that need to be stitched and , , ; In the same image Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate to get a new image , so that the image and images The images of the three objects at the boundary overlap, , ; Make the image With image The pixel values ​​of the overlapping boundaries are adaptively adjusted to obtain the stitched large scene image.

[0008] Compared with the prior art, the present invention has the following beneficial effects: The present invention is achieved by Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate the image and images The images of the three objects at the boundary overlap to obtain a new image , , ; Make the image With image The pixel values ​​of the overlapping boundaries are adaptively adjusted, and the above steps are repeated to traverse the images taken by N cameras at the same time, and stitch them into an image of a large scene, so that the stitching speed is fast, the stitching error is small, and the transition effect of the stitched image is good. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 It is a flow chart of a visible light image geometric radiation stitching processing method provided by the first embodiment of the present invention.

[0010] Figure 2 It is a flow chart of a visible light image geometric radiation stitching processing method provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0011] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0012] In the description of the present invention, it should be understood that the terms "upper", "middle", "outer", "inner", "lower", "all around" and the like indicating directions or positional relationships are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the components or elements referred to must have a specific direction, be constructed and operated in a specific direction, and therefore should not be understood as limiting the present invention.

[0013] First embodiment

[0014] Figure 1 : is a flow chart of the visible light image geometric radiation stitching processing method provided by the present invention. Figure 1 As shown, a visible light image geometric radiation stitching processing method comprises the following steps: S1-1: Divide the large scene to be photographed into M×N areas. Each area is equipped with a camera to capture images of the area. The images captured by the camera in each area are marked as ; S1-2: Obtain images taken by cameras in adjacent areas and , , ; In the same image Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate to get a new image , so that the image and images The images of the three objects at the boundary overlap, , ; Make the image With image The pixel values ​​of overlapping boundaries are adaptively adjusted; S1-3: Repeat step S1-2 to traverse the images taken by N cameras at the same time and stitch them into an image of a large scene.

[0015] In the first embodiment of the present invention, the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S1-4: By dividing the image into a pixel window including K×K pixels With image Overlapping boundaries are split into Regions, , , where K is the number of pixels in a row and column of the pixel window, and L is the image With image The number of pixels in a row of overlapping boundaries, J is the image With image The number of pixels in a column of overlapping borders, Represents the floor function.

[0016] In the first embodiment of the present invention, the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S1-5: Calculate the image according to the following formula The value of the pixel in the pixel window at row x and column y in the overlap boundary: , In the formula, , For images The pixel in the pixel window at row x and column y of the overlap boundary The value of For images The pixel in the pixel window at row x and column y of the overlap boundary The value of; , .

[0017] In the first embodiment of the present invention, the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S1-6: The image is calculated according to the following formula The value of the pixel in the pixel window at row x and column y in the overlap boundary: , In the formula, , For images The pixel in the pixel window at row x and column y of the overlap boundary The value of For images The pixel in the pixel window at row x and column y of the overlap boundary The value of .

[0018] In the first embodiment of the present invention, the pixel window is determined according to the required stitching boundary accuracy. The higher the accuracy requirement, the smaller the pixel window.

[0019] The first embodiment of the present invention uses the above technical solution to stitch images of multiple areas acquired by multiple cameras into an image of a large scene, with a fast stitching speed, a small stitching error, and a good transition effect of the stitched image.

[0020] Second embodiment

[0021] The second embodiment of the present invention only scans the contents that are different from those of the first embodiment, and the same contents are not repeated.

[0022] Figure 2 FIG. 2 is a flow chart of a method for geometric radiation stitching of visible light images provided by the second embodiment of the present invention. Figure 2 As shown, the second embodiment of the present invention provides a visible light image geometric radiation stitching processing method comprising the following steps: S2-1: Get the two images that need to be stitched and , , ; In the same image Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate to get a new image , so that the image and images The images of the three objects at the boundary overlap, , ; Make the image With image The pixel values ​​of the overlapping boundaries are adaptively adjusted to obtain the stitched large scene image.

[0023] The second embodiment of the present invention can stitch images of different areas acquired by the same camera or different cameras into an image of a large scene, with a fast stitching speed, a small stitching error, and a good transition effect of the stitched image.

[0024] Third embodiment

[0025] A third embodiment of the present invention provides a system, which includes a storage medium and one or more processors. The storage medium stores a computer program, and the computer program is called by the one or more processors to implement the methods in the first and second embodiments.

[0026] The beneficial effects of the third embodiment of the present invention are the same as those of the first embodiment or the second embodiment, and will not be described again here.

[0027] Fourth embodiment

[0028] A fourth embodiment of the present invention provides a computer program product, which uses a computer language to compile the methods of the first embodiment and the second embodiment into a computer program that is called and executed by one or more processors.

[0029] The beneficial effects of the fourth embodiment of the present invention are the same as those of the first embodiment or the second embodiment, and will not be described again here.

[0030] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the contents of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that technicians in the relevant technical field can well understand and use the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A visible light image geometric radiation stitching processing method, characterized in that: The following steps are involved: S1-1: Divide the large scene to be photographed into M×N areas. Each area is equipped with a camera to capture images of the area. The images captured by the camera in each area are marked as ; S1-2: Obtain images taken by cameras in adjacent areas and , , ; In the same image Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate to get a new image , so that the image and images The images of the three objects at the boundary overlap, , ; Make the image With image The pixel values ​​of overlapping boundaries are adaptively adjusted; S1-3: Repeat step S1-2 to traverse the images taken by N cameras at the same time and stitch them into an image of a large scene.

2. The visible light image geometric radiation stitching processing method according to claim 1, characterized in that: Make the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S1-4: By dividing the image into a pixel window including K×K pixels With image Overlapping boundaries are split into Regions, , , where K is the number of pixels in a row and column of the pixel window, and L is the image With image The number of pixels in a row of overlapping boundaries, J is the image With image The number of pixels in a column of overlapping borders, Represents the floor function.

3. The visible light image geometric radiation stitching processing method according to claim 2, characterized in that: Make the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S1-5: Calculate the image according to the following formula The value of the pixel in the pixel window at row x and column y in the overlap boundary: , In the formula, , For images The pixel in the pixel window at row x and column y of the overlap boundary The value of For images The pixel in the pixel window at row x and column y of the overlap boundary The value of; , .

4. The visible light image geometric radiation stitching processing method according to claim 3 is characterized in that: Make the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S1-6: The image is calculated according to the following formula The value of the pixel in the pixel window at row x and column y in the overlap boundary: , In the formula, , For images The pixel in the pixel window at row x and column y of the overlap boundary The value of For images The pixel in the pixel window at row x and column y of the overlap boundary The value of .

5. The visible light image geometric radiation stitching processing method according to claim 4, characterized in that: The pixel window is determined according to the required stitching boundary accuracy. The higher the accuracy requirement, the smaller the pixel window.

6. A visible light image geometric radiation stitching processing method, characterized in that: The following steps are involved: S2-1: Get the two images that need to be stitched and , , ; In the same image Adjacent images Take three images of the object at the boundary of the image to form a triangular image template, so that Adjacent images The boundary image is matched with the triangle image template. Find the triangle image that matches the triangle image template, according to the image Medium triangle image template and image The positional relationship between the triangle image and the triangle image template matching, for the image Translate and rotate to get a new image , so that the image and images The images of the three objects at the boundary overlap, , ; Make the image With image The pixel values ​​of the overlapping boundaries are adaptively adjusted to obtain the stitched large scene image.

7. The visible light image geometric radiation stitching processing method according to claim 6, characterized in that: Make the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S2-2: The image is transformed into a pixel window including K×K pixels. With image Overlapping boundaries are split into Regions, , , where K is the number of pixels in a row and column of the pixel window, and L is the image With image The number of pixels in a row of overlapping boundaries, J is the image With image The number of pixels in a column of overlapping borders, Represents the floor function.

8. The visible light image geometric radiation stitching processing method according to claim 7, characterized in that: Make the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S2-3: Calculate the image according to the following formula The value of the pixel in the pixel window at row x and column y in the overlap boundary: , In the formula, , For images The pixel in the pixel window at row x and column y of the overlap boundary The value of For images The pixel in the pixel window at row x and column y of the overlap boundary The value of; , .

9. The visible light image geometric radiation stitching processing method according to claim 8, characterized in that: Make the image With image The adaptive adjustment of pixel values ​​of overlapping boundaries specifically includes the following steps: S2-4: The image is calculated according to the following formula The value of the pixel in the pixel window at row x and column y in the overlap boundary: , In the formula, , For images Overlap border pixel window at row x, column y The value of For images The pixel in the pixel window at row x and column y of the overlap boundary The value of .

10. The visible light image geometric radiation stitching processing method according to claim 9, characterized in that: The pixel window is determined according to the required stitching boundary accuracy. The higher the accuracy requirement, the smaller the pixel window.

Images