Infrared image non-overlapping splicing method and device

Abstract

The present application belongs to the technical field of image processing, and particularly relates to an infrared image non-overlapping splicing method and device. The traditional image splicing technology is a technology of splicing a plurality of images with overlapping parts, including images obtained at different times, from different angles or by different sensors, into a seamless panoramic image or a high-resolution image. The present application provides a non-overlapping image splicing method and device. The infrared image non-overlapping splicing method of the present application splices a plurality of continuous infrared images into a panoramic image, and the scene is continuous, unbroken and non-repeated, and the brightness is smooth and has no splicing seam. The brightness curve does not need to be manually adjusted, the method is simple to calculate, and is highly adaptable.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of image processing, and particularly relates to an infrared image non-overlapping splicing method and device. BACKGROUND

[0002] The panoramic all-around scanning monitoring device is different from the traditional turntable system in that the rotation speed is faster, up to ≥180 rps. In order to avoid imaging trailing, a fast mirror is provided to compensate for the imaging process of each frame, and then the panoramic image is spliced. The traditional splicing method generally splices two or more images with overlapping regions together by feature matching, that is, splicing the images with the same feature points (SIFT feature points) together, transforming the images taken from multiple different angles to the same angle, and splicing into a wide-view image. The basic process of image splicing includes (1) taking multiple / sequence images for a certain scene; (2) calculating the transformation relationship between the second image and the first image; (3) transforming the second image into the coordinate system of the first image; (4) transforming the fused / synthesized image; (5) repeating the above process in a multi-image scene. The whole process is complex and time-consuming, which is not conducive to embedding and transplantation. If the scene is distorted, ghosting is easy to occur in the fusion area.

[0003] The invention patent CN202310158914.X discloses a non-overlapping panoramic image splicing method, system and terminal, which belongs to the field of infrared imaging. Image information is obtained to form an image sequence; the image information in the image sequence is arranged in a predetermined order; the image is converted into a gray image; the brightness difference between the right edge of the previous image and the left edge of the next image is counted as the initial threshold for adjusting the brightness; the image is preprocessed, and the brightness adjustment curve of the image is defined; according to the brightness of the image, the brightness changes from dark to light from top to bottom, and the initial threshold for adjusting the brightness is combined to splice the images in the image sequence to form a panoramic image. The present application can make the brightness of the image connection similar after image splicing by taking the image center as a reference point and adjusting the image brightness of the image around the boundary in combination with the edge brightness of the adjacent two images, which can meet the actual use requirements.

[0004] The application method has the following problems in the application process of splicing the adjacent two images of the scene:

[0005] (1) the brightness of the joint is not uniform, and the boundary of the object in the scene, such as buildings, power poles and other objects with large temperature difference with the background, cannot achieve uniform brightness at the joint; (2) the method of defining the image brightness adjustment curve with the center of the image as the reference point and the Gaussian circle diameter as the distance ignores the real brightness difference at the joint of adjacent frames, and cannot guarantee the uniform brightness of the joint after splicing; (3) the method of defining the image brightness curve and adjusting the brightness curve needs manual intervention, and the control effect is limited. Therefore, in the actual use of the device such as the circular scanning light, it is difficult to meet the requirement of completely uniform brightness of the joint of adjacent frames. SUMMARY

[0006] Therefore, the technical problem to be solved by the present application is to provide an image splicing method and device without overlap, which does not produce brightness difference at the object boundary of the joint, does not need manual adjustment of the brightness curve, and obtains a panoramic image with completely uniform brightness at the object boundary of the joint.

[0007] The device of the present application is composed of an image acquisition module, an image processing module and an image display module. The image acquisition module includes a horizontal scanning module, a fast feedback compensation module, an optical lens module and a camera; the image processing module is composed of an image cropping module and an image splicing module; and the image display module is composed of a panoramic image display module.

[0008] The working process of the device of the present application includes the following steps:

[0009] (1) image acquisition: acquiring the sequence of circular scanning photoelectric images after fast feedback compensation;

[0010] (2) image cropping: determining how many columns of pixels are cropped to meet the requirements of panoramic splicing through initial image splicing;

[0011] (3) image splicing: performing brightness statistical analysis on the joint of adjacent frames without overlap, adjusting, and then splicing;

[0012] (4) image display: displaying the completed panoramic image.

[0013] The infrared image non-overlapping splicing method of the present application includes the following steps:

[0014] (1) receiving the sequence of continuous images collected;

[0015] (2) sorting the frames to confirm the left and right images for adjacent frame splicing;

[0016] (3) calculating the brightness average value nGrayLeftRM(i) of nNum pixels in each row at the right boundary of the left image; calculating the brightness average value nGrayRightLM(i) of nNum pixels in each row at the left boundary of the right image; wherein i is the row number;

[0017] (4) Calculate the adjusted brightness target of each row nGrayGoal(i) = (nGrayLeftRM(i) + nGrayRightLM(i)) / 2;

[0018] (5) Calculate the left border of the right image of each row of pixel brightness correction value nGrayCorrectLeft(i) = nGryaGoal(i) - nGrayLeftRM(i); Calculate the right border of the left image of each row of pixel brightness correction value nGrayCorrectRight(i) = nGryaGoal(i) - nGrayRightLM(i);

[0019] (6) Smooth the left border of the right image of each row of pixel brightness correction value nGrayCorrectLeft(i) using the adjacent row of the left image, to obtain nGrayCorrectLeftNew(i); Smooth the left border of the right image of each row of pixel brightness correction value nGrayCorrectRight(i) using the adjacent row of the right image, to obtain nGrayCorrectRightNew(i);

[0020] (7) The left border of the right image is nonlinearly smoothed in brightness according to a certain width window nWinWidthLeft; the left border of the right image is nonlinearly smoothed in brightness according to a certain width window nWinWidthRight;

[0021] (8) The left and right images are pixel spliced, and the adjacent two frame images are spliced; the sequence image splicing is sequentially performed to generate a panoramic image.

[0022] The traditional image splicing technology is a technology of splicing a plurality of frames of images with overlapping parts, including images obtained at different times, different angles or different sensors, into a seamless panoramic image or a high-resolution image. The present application provides a non-overlapping image splicing method and device. The non-overlapping infrared image splicing method of the present application splices a plurality of continuous infrared images into a panoramic image, and the scene is continuous, unbroken and non-repetitive, and the brightness is smooth and has no splicing seam. The brightness curve does not need to be manually adjusted, the method is simple to calculate and has strong adaptability. BRIEF DESCRIPTION OF DRAWINGS

[0023] Figure 1 The device module diagram of the present application.

[0024] Figure 2 The device workflow diagram of the present application.

[0025] Figure 3 The non-overlapping image panoramic splicing algorithm flowchart of the present application.

[0026] Figure 4 The direct splicing effect of non-overlapping images.

[0027] Figure 5 The method of the present application is spliced. DETAILED DESCRIPTION

[0028] The device of the embodiment of the present application is composed of an image acquisition module, an image processing module and an image display module, and can acquire 360-degree panoramic images. The image acquisition module is composed of a horizontal circumferential scanning module, a fast feedback compensation module, an optical lens module and a camera; the image processing module is composed of an image cropping module and an image splicing module; and the image display module is composed of a panoramic image display module. The working process of the device comprises the following steps:

[0029] (1) Image acquisition. The circumferential scanning photoelectric image sequence after fast feedback compensation is acquired, 50 frames of images are acquired per second, and the resolution of each frame of image is 640 pixels wide and 512 pixels high;

[0030] (2) Image cropping. According to the determination of how many columns of pixels meet the splicing panoramic requirements, 16 columns of pixels are cropped on the left and right of each frame of image as needed, and the resolution of each frame of image is 608 pixels wide and 512 pixels high;

[0031] (3) Image splicing. The adjacent frame image splicing seams without overlap are subjected to brightness statistical analysis, adjusted, and then spliced;

[0032] (4) Image display. The completed panoramic image is displayed, and the width is 30400 pixels and the height is 512 pixels.

[0033] The present application splices a plurality of continuous infrared images into a panoramic image, and the scene is continuous, unbroken and non-repetitive, and the brightness is smooth without seams. The splicing method comprises the following steps:

[0034] (1) receiving the acquired continuous image sequence at 50 frames per second;

[0035] (2) performing frame sorting to confirm the left and right images for adjacent frame splicing;

[0036] (3) calculating the brightness average value nGrayLeftRM(i) of nNum=5 pixels in each row of the right boundary of the left image; calculating the brightness average value nGrayRightLM(i) of nNum=5 pixels in each row of the left boundary of the right image; wherein i is the row number;

[0037] (4) calculating the adjustment brightness target nGrayGoal(i) of each row = (nGrayLeftRM(i)+nGrayRightLM(i)) / 2;

[0038] (5) Calculate the brightness correction value of each row of pixels on the right border of the left image nGrayCorrectLeft(i)=nGryaGoal(i)-nGrayLeftRM(i), and the brightness correction value of each row of pixels on the left border of the right image nGrayCorrectRight(i)=nGryaGoal(i)-nGrayRightLM(i);

[0039] (6) Smooth the brightness correction value of each row of pixels on the right border of the left image nGrayCorrectLeft(i) with adjacent rows to obtain nGrayCorrectLeftNew(i), and smooth the brightness correction value of each row of pixels on the left border of the right image nGrayCorrectRight(i) with adjacent rows to obtain nGrayCorrectRightNew(i);

[0040] (7) Nonlinearly smooth the brightness of each row on the right border of the left image according to a certain width window nWinWidthLeft=100 pixels, and nonlinearly smooth the brightness of each row on the left border of the right image according to a certain width window nWinWidthRight=100 pixels, wherein a quadratic curve is used for nonlinear brightness smoothing;

[0041] (8) Pixel splicing is performed on the left and right images to complete the splicing of adjacent two frames of images; sequence image splicing is sequentially performed to generate a panoramic image.

[0042] The key points of the present application are:

[0043] (1) The brightness adjustment targets of the left and right seams of adjacent frames are consistent, and there is no brightness difference;

[0044] (2) In the case that there is an imaging target edge at the seam, adjacent brightness targets are smoothed;

[0045] (3) The transition method of the image brightness adjustment range adopts nonlinear brightness transition, no brightness adjustment trace is added in the image, and the brightness transition is more natural.

Claims

1. An infrared image non-overlapping stitching method, characterized in that, The steps include: (1) receiving a continuous image sequence collected; (2) performing frame sequencing to confirm left and right images for adjacent frame splicing; (3) calculating the brightness average nGrayLeftRM(i) of nNum pixels in each row of the right border of the left image; calculating the brightness average nGrayRightLM(i) of nNum pixels in each row of the left border of the right image; Where i is the row number; (4) calculating the adjusted brightness target nGrayGoal(i) of each row = (nGrayLeftRM(i)+nGrayRightLM(i)) / 2; (5) calculating the pixel brightness correction value nGrayCorrectLeft(i) of each row of the right border of the left image = nGryaGoal(i)-nGrayLeftRM(i); calculating the pixel brightness correction value nGrayCorrectRight(i) of each row of the left border of the right image = nGryaGoal(i)-nGrayRightLM(i); (6) smoothing the pixel brightness correction value nGrayCorrectLeft(i) of each row of the right border of the left image using adjacent rows of the left image to obtain nGrayCorrectLeftNew(i); smoothing the pixel brightness correction value nGrayCorrectRight(i) of each row of the left border of the right image using adjacent rows of the right image to obtain nGrayCorrectRightNew(i); (7) performing nonlinear brightness smoothing on each row of the right border of the left image according to a certain width window nWinWidthLeft; performing nonlinear brightness smoothing on each row of the left border of the right image according to a certain width window nWinWidthRight; (8) performing pixel splicing on the left and right images to complete the splicing of adjacent two frames of images; sequentially performing sequence image splicing to generate a panoramic image.

2. The method of claim 1, wherein, The value of nNum in step (3) is 5.

3. The method of claim 1, wherein, In step (6), the smoothing is performed by using the neighborhood average method.

4. The method of claim 1, wherein, In step (7), nWinWidthLeft=100pixel, nWinWidthRight=100pixel, and a quadratic curve is used for nonlinear brightness smoothing.

5. An infrared image non-overlapping splicing device, characterized in that, The image acquisition module, the image processing module, and the image display module are included, wherein the image acquisition module includes a horizontal circumferential scanning module, a fast feedback compensation module, an optical lens module, and a camera; the image processing module is composed of an image cropping module and an image splicing module; the image display module is composed of a panoramic image display module; the working process of the splicing device includes the following steps: (1) image acquisition: collecting a circumferential scanning photoelectric image sequence after fast feedback compensation; (2) image cropping: determining how many columns of pixels are cropped to meet the panoramic splicing requirements through initial image splicing; (3) image splicing: performing brightness statistical analysis, adjustment, and splicing on the non-overlapping adjacent frame image splicing seam according to the method of claim 1; (4) image display: displaying the completed panoramic image.

Images