A method for image fusion of resistor array infrared scene images

Abstract

The present invention relates to the field of image processing technology, and in particular to an image fusion method for resistor array infrared scene images, comprising: obtaining an infrared grayscale image sequence and a visible light grayscale image sequence, and matching them into image groups according to ordinal values. Calculating the grayscale value difference of corresponding pixel points in each group to determine the target area. Identifying non-target areas, further obtaining a number of surrounding connected domains, and evaluating the prominence of the target area according to the area of ​​the surrounding connected domains and the length of overlap with the edge of the target area. Based on this, the distribution weight of each pixel point in the target area of ​​the infrared grayscale image is calculated, and a fused image is generated by combining the grayscale values ​​of all pixel points. The present invention analyzes the grayscale values ​​of all pixel points in each frame of the infrared grayscale image and the visible light grayscale image, fuses the infrared grayscale image and the visible light grayscale image, and improves the clarity of the image.

Description

Technical Field

[0001] The invention relates to the technical field of image processing, and in particular to an image fusion method for resistor array infrared scene images. Background Art

[0002] Nowadays, people pay more and more attention to their own safety and personal property safety. Most people will choose to install a surveillance camera at home to protect their private property and prevent thieves from robbing their homes. Even if thieves are robbed, the thieves can be identified and arrested based on the images captured by the camera. This not only helps to protect citizens' private property, but also helps to maintain social order. Therefore, it is necessary to improve the clarity of the images captured by the surveillance camera.

[0003] Thieves often steal at night, and at night, the resolution of infrared grayscale images is low, and the details of the background environment around the target area will not be displayed clearly. When the target area is located in a high-temperature area such as indoors, the target area will be difficult to distinguish from the background area. Visible light grayscale images cannot display information in low-light areas very well. If the target area is hidden in a low-light area or behind some shelters, the target area will not be found in the visible light grayscale image. Therefore, it is necessary to fuse the visible light grayscale image and the infrared grayscale image. However, if the weights given to different images during the fusion process are inappropriate, the image fusion effect will be very poor. Summary of the invention

[0004] The invention provides an image fusion method of resistor array infrared scene images to solve the existing problems.

[0005] The image fusion method of a resistor array infrared scene image of the present invention adopts the following technical solution:

[0006] An embodiment of the present invention provides an image fusion method of a resistor array infrared scene image, the method comprising the following steps:

[0007] Acquire an infrared grayscale image sequence and a visible light grayscale image sequence;

[0008] In an infrared grayscale image sequence and a visible light grayscale image sequence, infrared grayscale images and visible light grayscale images with the same ordinal value are used to form an image group; according to the position of each pixel point in the infrared grayscale image and the visible light grayscale image of each image group, a plurality of pixel pairs of each image group are obtained; according to the grayscale values ​​of all pixels of each pixel pair of each image group, the difference of pixel grayscale values ​​of each pixel pair of each image group is obtained; according to the difference of pixel grayscale values ​​of each pixel pair of each image group, the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group is obtained;

[0009] In the infrared grayscale image and the visible light grayscale image of each image group, a number of non-target areas are obtained; according to the position of each non-target area, a number of surrounding connected domains of the infrared grayscale image and the visible light grayscale image of each image group are obtained; according to the area of ​​each surrounding connected domain of the infrared grayscale image and the visible light grayscale image of each image group and the length of overlap with the edge of the corresponding target area, the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group is obtained;

[0010] According to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group and the difference in the pixel grayscale values ​​of the pixel pairs corresponding to all the pixel points in the target area of ​​each frame of the infrared grayscale image, the allocation weight of each pixel point in the target area of ​​the infrared grayscale image in each image group is obtained;

[0011] According to the assigned weight of each pixel point of the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image, several frames of fused images are obtained.

[0012] Furthermore, the step of obtaining a plurality of pixel pairs of each image group according to the position of each pixel point in the infrared grayscale image and the visible light grayscale image of each image group includes the following specific steps:

[0013] In the In the image group, The corresponding pixel points in the same position in the infrared grayscale image and the visible light grayscale image of the image group are denoted as Pixel pairs of an image group.

[0014] Furthermore, the step of obtaining the pixel gray value difference of each pixel pair of each image group according to the gray values ​​of all pixels of each pixel pair of each image group includes the following specific steps:

[0015] The first The first The absolute value of the difference between the grayscale values ​​of the two pixels in the pixel pair is recorded as The first The difference in pixel grayscale values ​​between pairs of pixels.

[0016] Furthermore, the step of obtaining the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group according to the difference in the pixel grayscale values ​​of each pixel pair of each image group includes the following specific steps:

[0017] According to the difference in pixel grayscale values ​​of each pixel pair in each image group, a number of target pixel pairs are obtained;

[0018] In the connected domains formed by adjacent pixel points of all target pixel pairs of the infrared grayscale image and the visible light grayscale image of each image group, the connected domains with the largest areas in the infrared grayscale image and the visible light grayscale image are recorded as the target regions of the infrared grayscale image and the visible light grayscale image of each image group.

[0019] Furthermore, the step of obtaining a plurality of target pixel pairs according to the pixel gray value difference of each pixel pair of each image group includes the following specific steps:

[0020] Jordi The first The difference in the grayscale value of the pixel pair is greater than that of the The mean of the pixel gray value differences of all pixel pairs in the image group is The first pixel pairs, recorded as target pixel pairs.

[0021] Furthermore, the method of obtaining several surrounding connected domains of the infrared grayscale image and the visible light grayscale image of each image group according to the position of each non-target area includes the following specific steps:

[0022] In the infrared grayscale image and visible light grayscale image of each image group, all non-target areas adjacent to each target area are recorded as surrounding connected areas.

[0023] Furthermore, according to the area of ​​each surrounding connected domain of the infrared grayscale image and the visible light grayscale image of each image group and the length of overlap with the edge of the corresponding target area, the specific calculation formula corresponding to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group is obtained as follows:

[0024]

[0025] in, Indicates The highlighting degree of the target area of ​​the infrared grayscale image in the image group; Indicates The number of connected domains around the infrared grayscale images in the image group; Indicates The infrared grayscale image in the image group The area of ​​the surrounding connected domain; Indicates The infrared grayscale image in the image group The area of ​​the surrounding connected domain; Indicates The infrared grayscale image in the image group The length of the overlap between the surrounding connected domain and the edge of the target area; Indicates The length of the edge line of the target area of ​​the infrared grayscale image in the image group; Indicates The average grayscale value of all pixels in the target area of ​​the infrared grayscale image in the image group; Indicates The infrared grayscale image in the image group The average gray value of all pixels in the surrounding connected domain; is the absolute value function.

[0026] According to The method for obtaining the highlighting degree of the target area of ​​the infrared grayscale image of the image group is to obtain the The highlighting degree of the target area of ​​the visible light grayscale image in the image group.

[0027] Furthermore, the method of obtaining the distribution weight of each pixel point in the target area of ​​the infrared grayscale image in each image group according to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group and the pixel grayscale value difference of the pixel pairs corresponding to all the pixel points in the target area of ​​each frame of the infrared grayscale image includes the following specific steps:

[0028] In the In the image group, the difference between the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image is calculated, and the difference is added to the first The normalized value of the product of the grayscale value differences of the pixel pairs corresponding to the pixel points is recorded as The first image of the target area in the infrared grayscale image of the image group The weight assigned to each pixel.

[0029] Furthermore, the method of obtaining a plurality of fused images according to the assigned weight of each pixel of the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image includes the following specific steps:

[0030] The average of the grayscale values ​​of all pixels of each pixel pair outside the target area in each image group is recorded as the corrected pixel value;

[0031] Obtaining a corrected pixel value for each pixel pair in the target area of ​​each image group according to the assigned weight of each pixel point in the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image;

[0032] A fused image corresponding to each image group is constructed according to the corrected pixel values ​​of all pixel pairs in each image group.

[0033] Furthermore, the method of obtaining the corrected pixel value of each pixel pair of the target area in each image group according to the assigned weight of each pixel point of the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image includes the following specific steps:

[0034] In the In the image group, calculate 1 minus the target area of ​​the infrared grayscale image The difference in the weights of the pixels is compared with the first pixel of the target area of ​​the visible light grayscale image. The product of the gray values ​​of pixels plus the gray value of the target area of ​​the infrared gray image The sum of the products of the weights and gray values ​​of pixels is recorded as The first image of the target area in the image group The corrected pixel value of the pixel pair.

[0035] The beneficial effects of the technical solution of the present invention are:

[0036] An infrared grayscale image sequence and a visible light grayscale image sequence are obtained; in the infrared grayscale image sequence and the visible light grayscale image sequence, the infrared grayscale images and the visible light grayscale images with the same ordinal value are formed into an image group, which ensures the consistency of time and provides an accurate basis for subsequent processing. According to the position of each pixel point in the infrared grayscale image and the visible light grayscale image of each image group, several pixel pairs of each image group are obtained, which ensures that the positions of all pixels in each pixel pair are the same. According to the grayscale values ​​of all pixels of each pixel pair of each image group, the difference in pixel grayscale values ​​of each pixel pair of each image group is obtained, which helps to accurately identify the target object in a complex background and improves the accuracy of target detection. According to the difference in pixel grayscale values ​​of each pixel pair of each image group, the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group is obtained, and several target areas are accurately obtained. In the infrared grayscale image and visible grayscale image of each image group, several non-target areas are obtained; according to the position of each non-target area, several surrounding connected domains of the infrared grayscale image and visible grayscale image of each image group are obtained, thereby improving the accuracy of the subsequent calculation of the prominence degree of the target area. According to the area of ​​each surrounding connected domain of the infrared grayscale image and visible grayscale image of each image group and the length of overlap with the edge of the corresponding target area, the prominence degree of the target area of ​​the infrared grayscale image and visible grayscale image in each image group is obtained, thereby improving the accuracy of the subsequent calculation of the weight distribution of each pixel point of the target area. According to the prominence degree of the target area of ​​the infrared grayscale image and visible grayscale image in each image group and the difference in pixel grayscale values ​​of pixel pairs corresponding to all pixel points of the target area of ​​each frame of infrared grayscale image, the weight distribution of each pixel point of the target area of ​​the infrared grayscale image in each image group is obtained, thereby ensuring the clarity and detail retention of the target area in the fused image, while balancing the information of different image sources and enhancing the overall quality of the image. According to the assigned weight of each pixel point of the target area of ​​the infrared grayscale image in each image group and the grayscale value of all pixels of each frame of infrared grayscale image and visible light grayscale image, several frames of fused images are obtained. The present invention analyzes the grayscale value of all pixels of each frame of infrared grayscale image and visible light grayscale image, fuses the infrared grayscale image and the visible light grayscale image, and improves the clarity of the image. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0038] Figure 1 The present invention is a flowchart of the steps of an image fusion method of a resistor array infrared scene image. DETAILED DESCRIPTION

[0039] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the image fusion method of a resistor array infrared scene image proposed by the present invention, its specific implementation method, structure, features and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0040] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0041] The specific scheme of the image fusion method of the resistor array infrared scene image provided by the present invention is described in detail below with reference to the accompanying drawings.

[0042] See also Figure 1 , which shows a flowchart of a method for image fusion of a resistor array infrared scene image provided by an embodiment of the present invention, the method comprising the following steps:

[0043] Step S001: Acquire an infrared grayscale image sequence and a visible light grayscale image sequence.

[0044] A dual-spectrum surveillance camera that integrates a visible light camera and an infrared camera based on a resistor array is installed at a suitable location to monitor and protect the target area, and real-time infrared video and visible light video taken by the infrared and visible light cameras at the same angle and at the same time are obtained, and the infrared video and the visible light video have the same size and resolution. Several frames of images are captured from the infrared video and the visible light video respectively, and the images are grayed, mean filtered, and interpolated to obtain infrared grayscale image sequences and visible light grayscale image sequences. Among them, mean filtering and interpolation algorithms are well-known technologies, and the specific methods are not introduced here.

[0045] Step S002: In the infrared grayscale image sequence and the visible light grayscale image sequence, the infrared grayscale images and the visible light grayscale images with the same ordinal value are formed into an image group; according to the position of each pixel point in the infrared grayscale image and the visible light grayscale image of each image group, a plurality of pixel pairs of each image group are obtained; according to the grayscale values ​​of all pixels of each pixel pair of each image group, the difference in pixel grayscale values ​​of each pixel pair of each image group is obtained; according to the difference in pixel grayscale values ​​of each pixel pair of each image group, the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group is obtained.

[0046] It should be noted that the advantage of infrared grayscale images is that they are not affected by light or darkness and have a penetrating effect. Even at night or in a foggy environment, they can clearly show the location and outline of the target area. However, their disadvantages are also obvious, that is, the resolution is low, the details of the surrounding background environment of the target area are not clearly displayed, and if the target area is located in a high-temperature area such as indoors, the target area will be difficult to distinguish from the background area.

[0047] It is necessary to further explain that: the advantage of visible light is that it can provide detailed and high-definition texture information. When the target area is close to the doorway and other areas with high brightness and no cover, it will be displayed very clearly; but the disadvantage is that it cannot capture information in low-light areas. When the target area is hidden in a low-light area or behind some cover, the target area will not be found in the visible light grayscale image. The advantages and disadvantages of the two are complementary, so the fusion of the two images can obtain a high-quality image with a clear target area and a clear background texture area.

[0048] What needs to be further explained is that the target area is first distinguished from the background area. From the above analysis, it can be seen that no matter the target area (thief) is in an area with hidden light or behind an obstruction, or in an area with bright light but high temperature at the door, the difference between the target area in the infrared grayscale image and the visible light grayscale image will be very obvious. Therefore, the target area can be located first according to the difference in grayscale values ​​between the infrared and visible light images.

[0049] In the infrared grayscale image sequence and the visible light grayscale image sequence, all infrared grayscale images and visible light grayscale images with the same ordinal value form a plurality of image groups. The same ordinal value means the same time frame.

[0050] First Take the image group as an example, The corresponding pixel points in the same position in the infrared grayscale image and the visible light grayscale image of the image group are denoted as Pixel pairs of an image group.

[0051] Through the above process, several pixel pairs are obtained for each image group.

[0052] First The first For example, the first The first The specific calculation formula corresponding to the difference in pixel grayscale values ​​of pixel pairs is:

[0053]

[0054] in, Indicates The first The difference in pixel grayscale values ​​of pixel pairs; Indicates The first The gray value of the pixel point in the infrared gray image of a pixel pair; Indicates The first The grayscale value of the pixel point of the visible light grayscale image in the pixel pair; is the absolute value function.

[0055] What needs to be explained is: and Proportional, The larger the value of The first The greater the difference in pixel grayscale values ​​between pixel pairs.

[0056] Through the above process, we obtain The first The difference in pixel grayscale values ​​between pairs of pixels.

[0057] Each pixel pair of each image group is operated according to the above process to obtain the pixel gray value difference of each pixel pair of each image group.

[0058] Jordi The first The difference in pixel grayscale values ​​of the pixel pairs is greater than that of the The mean of the pixel gray value differences of all pixel pairs in the image group is The first pixel pairs, recorded as target pixel pairs.

[0059] Through the above process, several target pixel pairs are obtained for each image group.

[0060] In the connected domains formed by adjacent pixel points of all target pixel pairs of the infrared grayscale image and the visible light grayscale image of each image group, the connected domains with the largest areas in the infrared grayscale image and the visible light grayscale image are recorded as the target regions of the infrared grayscale image and the visible light grayscale image of each image group.

[0061] Step S003: Acquire several non-target areas in the infrared grayscale image and visible light grayscale image of each image group; obtain several surrounding connected domains of the infrared grayscale image and visible light grayscale image of each image group according to the position of each non-target area; obtain the prominence degree of the target area of ​​the infrared grayscale image and visible light grayscale image of each image group according to the area of ​​each surrounding connected domain of the infrared grayscale image and visible light grayscale image of each image group and the length of overlap with the edge of the corresponding target area.

[0062] It should be noted that: through the above process, the target areas of the infrared grayscale image and the visible grayscale image of each image group are obtained. The next step is to analyze whether the target area is more prominent in the infrared grayscale image or in the visible grayscale image. When the target area (thief) is in a shelter or dark environment, the visible grayscale image is unclear because it cannot penetrate objects and the light is dim, while the infrared grayscale image is not affected by light and can penetrate objects. The temperature of the thief's body is greatly different from the ambient temperature, so the infrared grayscale image is more prominent at this time, and the grayscale value difference between the target area and its surrounding area in the infrared grayscale image is large; when the target area (thief) needs to enter the room to steal, because the temperature in the room is high, the difference between the target area and its surrounding area in the infrared grayscale image will become smaller, and as the indoor light increases, the difference between the target area and its surrounding area in the visible grayscale image will become larger, so the visible grayscale image is more prominent at this time. Therefore, according to the difference between the target area and its surrounding connected domain in the infrared grayscale image and the visible grayscale image, it can be determined in which image the target area is more prominent at this time.

[0063] The canny operator is used to perform edge detection on the area outside the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group to obtain several edge pixel points of the infrared grayscale image and the visible light grayscale image of each image group. The canny operator is a well-known technology and the specific method is not introduced here.

[0064] According to several edge pixel points of the infrared grayscale image and visible light grayscale image of each image group, the contour extraction algorithm is used to extract the contour of the infrared grayscale image and visible light grayscale image of each image group, and several closed contour areas (edge ​​pixel points can form a closed area with the boundary of the target area) of the infrared grayscale image and visible light grayscale image of each image group are recorded as non-target areas. Among them, the contour extraction algorithm is a well-known technology, and the specific method is not introduced.

[0065] In the infrared grayscale image and visible light grayscale image of each image group, all non-target areas adjacent to each target area are recorded as surrounding connected areas.

[0066] It should be noted that: the larger the area of ​​the surrounding connected domain, the higher the degree of its influence on the target area. However, if there is a surrounding connected domain with a large area, but the borderline adjacent to the target area is short, its effective influence area on the target area will not be large, and only the part of the area adjacent to the target area will have an impact on the target area. Therefore, it is necessary to consider not only the size of the surrounding connected domain, but also the length of the borderline adjacent to the target area, and combine the two to consider the degree of its influence on the target area.

[0067] First For example, the target area of ​​the infrared grayscale image in the image group The specific calculation corresponding to the highlighting degree of the target area of ​​the infrared grayscale image in the image group is:

[0068]

[0069] in, Indicates The highlighting degree of the target area of ​​the infrared grayscale image in the image group; Indicates The number of connected domains around the infrared grayscale images in the image group; Indicates The infrared grayscale image in the image group The area of ​​the surrounding connected domain; Indicates The infrared grayscale image in the image group The area of ​​the surrounding connected domain; Indicates The infrared grayscale image in the image group The length of the overlap between the surrounding connected domain and the edge of the target area; Indicates The length of the edge line of the target area of ​​the infrared grayscale image in the image group; Indicates The average grayscale value of all pixels in the target area of ​​the infrared grayscale image in the image group; Indicates The infrared grayscale image in the image group The average gray value of all pixels in the surrounding connected domain; is the absolute value function.

[0070] What needs to be explained is: and Proportional, The larger the value of The more the target area of ​​the infrared grayscale image in the image group is highlighted, the more and Proportional, The larger the value of The more the target area of ​​the infrared grayscale image in the image group is highlighted, the more and Proportional, The larger the value of The more the target area of ​​the infrared grayscale image in the image group is highlighted, the more

[0071] Through the above process, we obtain The highlighting degree of the target area of ​​the infrared grayscale image in the image group.

[0072] According to The method for obtaining the highlighting degree of the target area of ​​the infrared grayscale image of the image group is to obtain the The highlighting degree of the target area of ​​the visible light grayscale image in the image group.

[0073] The infrared grayscale image and the visible light grayscale image in each image group are operated according to the above process to obtain the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group.

[0074] Step S004: Obtain the allocation weight of each pixel point in the target area of ​​the infrared grayscale image in each image group according to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group and the difference in pixel grayscale values ​​of pixel pairs corresponding to all pixel points in the target area of ​​each frame of the infrared grayscale image.

[0075] It should be noted that: when the target area is more prominent in the infrared grayscale image, more features in the infrared grayscale image should be retained during the fusion process of the target area of ​​the image. Similarly, if the target area is more prominent in the visible light grayscale image, more features of the visible light grayscale image should be retained during the fusion process.

[0076] First For example, the target area of ​​the infrared grayscale image in the image group The first image of the target area in the infrared grayscale image of the image group The specific calculation formula corresponding to the allocation weight of each pixel is:

[0077]

[0078] in, Indicates The first image of the target area in the infrared grayscale image of the image group The weight of each pixel; Indicates The first image of the target area in the infrared grayscale image of the image group The difference in pixel grayscale values ​​of the pixel pairs corresponding to the pixel points; Indicates The highlighting degree of the target area of ​​the infrared grayscale image in the image group; Indicates The highlighting degree of the target area of ​​the visible light grayscale image in the image group; Represents the normalization function.

[0079] What needs to be explained is: and Proportional, The larger the value of The pixel allocation weight of the target area of ​​the infrared grayscale image in the image group is greater; and Proportional, The larger the value of The first image of the target area in the infrared grayscale image of the image group The greater the weight assigned to each pixel, the greater the weight assigned to each pixel.

[0080] Through the above process, we obtain The first image of the target area in the infrared grayscale image of the image group The weight assigned to each pixel.

[0081] Each pixel point of the target area of ​​the infrared grayscale image in each image group is calculated according to the above process to obtain the allocation weight of each pixel point of the target area of ​​the infrared grayscale image in each image group.

[0082] Step S005: obtaining several frames of fused images according to the assigned weight of each pixel point of the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image.

[0083] It should be noted that: according to the distribution weight of each pixel point in the target area of ​​the infrared grayscale image in each image group obtained by the above process, the grayscale value of the pixel point in the target area is corrected.

[0084] First In the image group, the The specific calculation formula corresponding to the corrected pixel value of a pixel pair is:

[0085]

[0086] in, Indicates The first image of the target area in the image group Corrected pixel values ​​of pixel pairs; Indicates The first image of the target area in the infrared grayscale image of the image group The weight of each pixel; Indicates The first image of the target area in the infrared grayscale image of the image group The gray value of each pixel; Indicates The first image of the target area in the visible light grayscale image of the image group The gray value of a pixel.

[0087] What needs to be explained is: The larger the value of The first image of the target area in the infrared grayscale image of the image group Pixels are more prominent; The larger the value of The first image of the target area in the visible light grayscale image of the image group Pixels are more prominent.

[0088] Through the above process, we obtain The first image of the target area in the image group The corrected pixel value of the pixel pair.

[0089] Each pixel pair in the target area of ​​each image group is operated according to the above process to obtain a corrected pixel value of each pixel pair in the target area of ​​each image group.

[0090] The average of the grayscale values ​​of all pixels of each pixel pair outside the target area in each image group is recorded as the corrected pixel value.

[0091] Through the above process, the corrected pixel value of each pixel pair in each image group is obtained.

[0092] A fused image corresponding to each image group is constructed according to the corrected pixel values ​​of all pixel pairs in each image group.

[0093] According to the above-obtained several frames of fused images, firstly, whether there is a thief in the fused image, it can be judged whether there is a thief. If there is, it is recorded and a warning is sent to the user to ask whether to call the police.

[0094] So far, the present invention is completed.

[0095] In summary, in an embodiment of the present invention, an infrared grayscale image sequence and a visible light grayscale image sequence are obtained and matched into image groups according to ordinal values. The grayscale value difference of corresponding pixels in each group is calculated to determine the target area. The non-target area is identified, and a number of surrounding connected domains are further obtained. The prominence of the target area is evaluated according to the area of ​​the surrounding connected domains and the length of overlap with the edge of the target area. Based on this, the allocation weight of each pixel in the target area of ​​the infrared grayscale image is calculated, and the grayscale values ​​of all pixels are combined to generate a fused image. The present invention analyzes the grayscale values ​​of all pixels in each frame of infrared grayscale image and visible light grayscale image, fuses the infrared grayscale image and the visible light grayscale image, and improves the clarity of the image.

[0096] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for image fusion of resistor array infrared scene images, characterized in that: The method comprises the following steps: Acquire an infrared grayscale image sequence and a visible light grayscale image sequence; In an infrared grayscale image sequence and a visible light grayscale image sequence, infrared grayscale images and visible light grayscale images with the same ordinal value are used to form an image group; according to the position of each pixel point in the infrared grayscale image and the visible light grayscale image of each image group, a plurality of pixel pairs of each image group are obtained; according to the grayscale values ​​of all pixels of each pixel pair of each image group, the difference of pixel grayscale values ​​of each pixel pair of each image group is obtained; according to the difference of pixel grayscale values ​​of each pixel pair of each image group, the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group is obtained; In the infrared grayscale image and the visible light grayscale image of each image group, a number of non-target areas are obtained; according to the position of each non-target area, a number of surrounding connected domains of the infrared grayscale image and the visible light grayscale image of each image group are obtained; according to the area of ​​each surrounding connected domain of the infrared grayscale image and the visible light grayscale image of each image group and the length of overlap with the edge of the corresponding target area, the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group is obtained; According to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group and the difference in the pixel grayscale values ​​of the pixel pairs corresponding to all the pixel points in the target area of ​​each frame of the infrared grayscale image, the allocation weight of each pixel point in the target area of ​​the infrared grayscale image in each image group is obtained; According to the assigned weight of each pixel point of the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image, a plurality of frames of fused images are obtained; The specific steps of obtaining the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group according to the pixel grayscale value difference of each pixel pair of each image group are as follows: According to the difference in pixel grayscale values ​​of each pixel pair in each image group, a number of target pixel pairs are obtained; In the connected domains formed by adjacent pixel points of all target pixel pairs in the infrared grayscale image and the visible light grayscale image of each image group, the connected domains with the largest areas in the infrared grayscale image and the visible light grayscale image are recorded as the target areas of the infrared grayscale image and the visible light grayscale image of each image group; The canny operator is used to perform edge detection on the area outside the target area of ​​the infrared grayscale image and the visible light grayscale image of each image group to obtain a number of edge pixel points of the infrared grayscale image and the visible light grayscale image of each image group; According to several edge pixel points of the infrared grayscale image and the visible light grayscale image of each image group, the contour extraction algorithm is used to extract the contour of the infrared grayscale image and the visible light grayscale image of each image group, and several closed contour areas of the infrared grayscale image and the visible light grayscale image of each image group are recorded as non-target areas; According to the area of ​​each surrounding connected domain of the infrared grayscale image and the visible light grayscale image of each image group and the length of overlap with the edge of the corresponding target area, the specific calculation formula corresponding to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group is obtained as follows: in, Indicates Infrared grayscale images in the image group The prominence of the target area 1; Indicates Infrared grayscale images in the image group The number of connected domains around ; Indicates Infrared grayscale images in the image group No. The area of ​​the surrounding connected domain; Indicates The infrared grayscale image in the image group The area of ​​the surrounding connected domain; Indicates Infrared grayscale images in the image group No. The length of the overlap between the surrounding connected domain and the edge of the target area; Indicates Infrared grayscale images in the image group The length of the edge line of the target area 1; Indicates Infrared grayscale images in the image group The average gray value of all pixels in the target area 1; Indicates Infrared grayscale images in the image group No. The average gray value of all pixels in the surrounding connected domain; is the absolute value function; According to The method for obtaining the highlighting degree of the target area of ​​the infrared grayscale image of the image group is to obtain the The highlighting degree of the target area of ​​the visible light grayscale image in the image group; The specific steps of obtaining the distribution weight of each pixel point in the target area of ​​the infrared grayscale image in each image group according to the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image in each image group and the pixel grayscale value difference of the pixel pairs corresponding to all the pixel points in the target area of ​​each frame of the infrared grayscale image are as follows: In the In the image group, the difference between the highlighting degree of the target area of ​​the infrared grayscale image and the visible light grayscale image is calculated, and the difference is added to the first The normalized value of the product of the grayscale value differences of the pixel pairs corresponding to the pixel points is recorded as The first image of the target area in the infrared grayscale image of the image group The weight of each pixel; The specific steps of obtaining a plurality of fused images according to the distribution weight of each pixel point of the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image are as follows: The average of the grayscale values ​​of all pixels of each pixel pair outside the target area in each image group is recorded as the corrected pixel value; Obtaining a corrected pixel value for each pixel pair in the target area of ​​each image group according to the assigned weight of each pixel point in the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image; A fused image corresponding to each image group is constructed according to the corrected pixel values ​​of all pixel pairs in each image group.

2. The image fusion method of a resistor array infrared scene image according to claim 1, characterized in that: The method of obtaining a plurality of pixel pairs of each image group according to the position of each pixel point in the infrared grayscale image and the visible light grayscale image of each image group includes the following specific steps: In the In the image group, The corresponding pixel points in the same position in the infrared grayscale image and the visible light grayscale image of the image group are denoted as Pixel pairs of an image group.

3. The image fusion method of a resistor array infrared scene image according to claim 1, characterized in that: The step of obtaining the pixel gray value difference of each pixel pair of each image group according to the gray values ​​of all pixels of each pixel pair of each image group includes the following specific steps: The first The first The absolute value of the difference between the grayscale values ​​of the two pixels in the pixel pair is recorded as The first The difference in pixel grayscale values ​​between pairs of pixels.

4. The image fusion method of a resistor array infrared scene image according to claim 1, characterized in that: The specific steps of obtaining a plurality of target pixel pairs according to the pixel gray value difference of each pixel pair of each image group are as follows: Jordi The first The difference in the grayscale value of the pixel pair is greater than that of the The mean of the pixel gray value differences of all pixel pairs in the image group is The first pixel pairs, recorded as target pixel pairs.

5. The image fusion method of a resistor array infrared scene image according to claim 1, characterized in that: The specific steps of obtaining a plurality of surrounding connected domains of the infrared grayscale image and the visible light grayscale image of each image group according to the position of each non-target area are as follows: In the infrared grayscale image and visible light grayscale image of each image group, all non-target areas adjacent to each target area are recorded as surrounding connected areas.

6. The image fusion method of a resistor array infrared scene image according to claim 1, characterized in that: The method of obtaining the corrected pixel value of each pixel pair in the target area of ​​each image group according to the assigned weight of each pixel point in the target area of ​​the infrared grayscale image in each image group and the grayscale values ​​of all pixels of each frame of the infrared grayscale image and the visible light grayscale image includes the following specific steps: In the In the image group, calculate 1 minus the target area of ​​the infrared grayscale image The difference in the weights of the pixels is compared with the first pixel of the target area of ​​the visible light grayscale image. The product of the gray values ​​of pixels plus the gray value of the target area of ​​the infrared gray image The sum of the products of the weights and gray values ​​of pixels is recorded as The first image of the target area in the image group The corrected pixel value of the pixel pair.

Images