Panoramic image processing method, device and equipment of lamp post, medium and program product

Abstract

The invention relates to a panoramic image processing method and device of a lamp post, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: acquiring images acquired by a plurality of cameras towards the periphery of the lamp pole; the plurality of cameras are arranged on the lamp post, and the plurality of cameras are arranged in various directions relative to the lamp post for acquiring the image; determining matching areas between the images in different directions; and splicing the images in all the directions by taking the matching areas as connection to obtain a panoramic image. The panoramic image obtained by the method can visually reflect the actual field scene around the lamp post, and information loss is reduced.

Description

Technical Field

[0001] The present application relates to the technical field of image processing, and particularly to a panoramic image processing method, apparatus, computer device, computer-readable storage medium, and computer program product for a lamp post. Background Art

[0002] With the development of intelligent lamp post technology, cameras on intelligent lamp posts can collect images in multiple directions to transmit information by integrating images from multiple angles.

[0003] In the traditional technology, the images collected by the cameras of intelligent lamp posts need to be adjusted into smaller rectangles, and then arranged in sequence along the direction of the screen into a larger rectangle, and this larger rectangle is used as the panoramic image. However, although this panoramic image can provide 360-degree information, it cannot reproduce the actual scene seen by the human eye at the position of the camera, resulting in excessive information loss. Summary of the Invention

[0004] Based on this, in view of the above technical problems, it is necessary to provide a panoramic image processing method, apparatus, computer device, computer-readable storage medium, and computer program product for a lamp post, which can more accurately restore the panoramic image of the real scene and reduce the information loss of the panoramic image.

[0005] In a first aspect, the present application provides a panoramic image processing method for a lamp post, including:

[0006] Obtaining images collected by a plurality of cameras around the lamp post; the plurality of cameras are arranged on the lamp post, and the directions of the plurality of cameras for collecting the images with respect to the lamp post are various;

[0007] Determining the matching regions between the images in different directions;

[0008] Using the matching regions as connections to splice the images in each of the directions to obtain a panoramic image.

[0009] In one embodiment, the panoramic image includes a spliced image region corresponding to a reference image; the determining the matching regions between the images in different directions includes:

[0010] Determining the matching regions between the reference image and at least two adjacent images; wherein, the reference image and the adjacent images are collected in adjacent directions, and each adjacent image and the reference image are collected in different directions;

[0011] The using the matching regions as connections to splice the images in each of the directions to obtain a panoramic image includes:

[0012] Taking the matching area as a connection, the reference image and the similar image are stitched to obtain the stitched image area corresponding to the reference image.

[0013] In one embodiment, the resolution of the reference image is less than the resolution of the similar image, the field of view angle of the reference image is greater than the field of view angle of the similar image, and the range corresponding to the reference image in the stitched image area is less than the range corresponding to each frame of the similar image in the stitched image area.

[0014] In one embodiment, before acquiring the images collected by multiple cameras around the lamp post, the method further includes:

[0015] Based on the first corrected images collected by multiple cameras around the lamp post, the corrected parameters of each camera are obtained, and based on the corrected parameters of each camera, the second corrected images in each direction are acquired;

[0016] According to the second corrected images in each direction, the image offset between each camera is corrected to obtain the corrected parameters between each camera;

[0017] The acquiring of the images collected by multiple cameras around the lamp post includes:

[0018] Based on the corrected parameters between each camera, the images collected by multiple cameras around the lamp post are obtained.

[0019] In one embodiment, the central area of the panoramic image is the position of the lamp post relative to each camera, and the central area of the panoramic image is close to the corresponding area of the first side of each image in the panoramic image, and the central area of the panoramic image is far from the corresponding area of the second side of each image in the panoramic image; the object closer to the lamp post is closer to the first side of the image; the object farther from the lamp post is closer to the second side of the image.

[0020] In one embodiment, the panoramic image includes control areas corresponding to each direction; the method further includes:

[0021] In response to a magnification instruction for the control area in the panoramic image, the target image in the direction matched by the control area is determined from the images in each direction;

[0022] Based on the target image for display.

[0023] In a second aspect, the present application further provides a panoramic image processing device for a lamp post, including:

[0024] An acquisition module, configured to acquire images collected by multiple cameras around the lamp post; the multiple cameras are arranged on the lamp post, and the directions of the multiple cameras for collecting the images relative to the lamp post are various;

[0025] A matching module, configured to determine a matching area between the images in different directions;

[0026] A processing module, configured to splice the images in each of the directions with the matching area as a connection to obtain a panoramic image.

[0027] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, the steps of panoramic image processing of the lamp post in any of the above embodiments are implemented.

[0028] In a fourth aspect, the present application further provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, and when the computer program is executed by a processor, the steps of panoramic image processing of the lamp post in any of the above embodiments are implemented.

[0029] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program, and when the computer program is executed by a processor, the steps of panoramic image processing of the lamp post in any of the above embodiments are implemented.

[0030] The panoramic image processing method, apparatus, computer device, computer-readable storage medium, and computer program product of the above-mentioned lamp post. Since multiple cameras are arranged on the lamp post and the directions of image acquisition by the multiple cameras relative to the lamp post are various; the multiple cameras can provide images of the outside of the lamp post in various directions, thereby accurately constructing the information around the lamp post, so that there are boundaries with overlapping information in the images in different directions. Furthermore, by determining the matching regions between the images in different directions, adaptive adjustment of the matching regions can be realized, so that the distortion degrees in different directions are reduced and tend to be consistent, thereby avoiding excessive differences between the splicing regions of the panoramic images and the regions outside the matching regions, and enabling information dissemination. Finally, with the matching regions as the splicing points, the images in each direction are spliced, so that the loss of the field of view angle in each direction during the splicing process is small, ensuring that the corresponding panoramic image matches the information around the lamp post during acquisition. Thus, through the above directions, the information around the lamp post during acquisition is more accurate; and determining the matching regions in different directions can represent the matching and calibration results of the overlapping information in different directions, avoiding unnatural regions in the panoramic image caused by the differences in the above directions during the splicing process, so as to avoid the matching process affecting the information accuracy; finally, directly retaining each direction for splicing can make the loss of the field of view angle in each direction during the splicing process small, thereby more accurately representing the information. Thus, the panoramic image is spliced according to the above directions, with a small gap from the actual scene and can intuitively reflect the actual picture. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the drawings required for use in the description of the embodiments of the present application or related technologies. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0032] Figure 1 It is an application environment diagram of the panoramic image processing method of the lamp post in an embodiment;

[0033] Figure 2 It is a flowchart of the panoramic image processing method of the lamp post in an embodiment;

[0034] Figure 3 It is a flowchart of image splicing in an embodiment;

[0035] Figure 4 It is a flowchart of image acquisition in an embodiment;

[0036] Figure 5 It is a regional diagram of the panoramic image in an embodiment;

[0037] Figure 6 Schematic diagram of the boundary of the panoramic image in an embodiment;

[0038] Figure 7 Schematic diagram of the relationship of the images collected by the camera in an embodiment;

[0039] Figure 8 Schematic diagram of the specific process of the panoramic image processing method of the lamp post in an embodiment;

[0040] Figure 9 Block diagram of the structure of the panoramic image processing device of the lamp post in an embodiment;

[0041] Figure 10 Internal structure diagram of a computer device in an embodiment. Specific implementation manners

[0042] In order to make the objectives, technical solutions and advantages of the present application clearer and more understandable, the present application will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0043] The panoramic image processing method of the lamp post provided by the embodiments of the present application can be applied to an application environment as Figure 1 shown. Among them, the lamp post device 102 communicates with the server 104 through a network, and the server 104 communicates with the terminal 106 through a network to display the panoramic image transmitted by the lamp post device 102 on the terminal 106. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed in the cloud or other network servers.

[0044] Among them, the lamp post device 102 can be, but is not limited to, various Internet of Things devices installed on street lamps. The lamp post device 102 can be a smart speaker, a smart monitor, a smart lamp, a smart street lamp, etc. The server 104 can be an independent physical server, a server cluster or a distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The lamp post device 102 communicates with the server 104 through a network. The data storage system can store the data that the server 104 needs to process. The data storage system can be integrated on the server 104, or placed on the cloud or other network servers. Among them, the terminal 106 can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, Internet of Things devices and portable wearable devices. The Internet of Things devices can be smart TVs, smart in-vehicle devices, projection devices, etc. The portable wearable devices can be smart watches, smart bracelets, head-mounted devices, etc. The head-mounted device can be a virtual reality (VR) device, an augmented reality (AR) device, smart glasses, etc.

[0045] In an exemplary embodiment, as Figure 2 shown, a panoramic image processing method for a lamp post is provided. Taking the method applied to the lamp post device 102 in Figure 1 as an example for illustration, this method can also be implemented by the server 104 or the terminal 106. This method includes the following steps 202 to 206. Among them:

[0046] Step 202, obtain images collected by multiple cameras around the lamp post; the multiple cameras are arranged on the lamp post, and the directions of the multiple cameras for collecting images relative to the lamp post are various.

[0047] The lamp post is a support structure for the lamp, and the lamp post is equipped with a lamp and multiple cameras. Each lamp post has a first part and a second part. The cameras are arranged close to the first part and far from the second part, and each camera collects images in the direction around the second part to collect the environment around the second part.

[0048] Exemplarily, when the first part is the top of the lamp post or the middle of the lamp post, the second part is the bottom of the lamp post, so as to collect pedestrians, vehicles and ground environment around the lamp post through the camera to form a top-down view; when the second part is the bottom of the lamp post or the middle of the lamp post, the second part is the top of the lamp post, so as to collect branches, birds and low-altitude environment or mid-air environment around the lamp post through the camera to form a bottom-up view.

[0049] The area around the lamp post is the outer side of the lamp post; among them, the outer side is the direction away from the lamp post. Correspondingly, the inner side is the direction towards the lamp post. The direction in which the camera captures images relative to the lamp post means the direction centered on the lamp post and with the camera facing the outer side of the lamp post. When there are multiple directions in which the cameras capture images relative to the lamp post, in this case, each direction corresponds to at least one camera, thereby forming a panoramic image of the outer side of the lamp post.

[0050] The camera is installed on the lamp post. Therefore, the images captured by the camera around the lamp post are captured towards the outer side of the lamp post. During this process, taking the position of the camera on the lamp post as the origin, image capture is performed with the field of view (FOV) of each camera, and an image positively correlated with the field of view is obtained; among them, the field of view represents the angular range of the area that the camera can cover. Since the camera is installed on the lamp post, the distance between the camera and the second part is also positively correlated with the size of the area that the camera can cover.

[0051] In some embodiments, obtaining the images captured by multiple cameras around the lamp post includes: when each camera captures images around the lamp post in the direction corresponding to each camera respectively, the lamp post device obtains the images captured by each camera respectively.

[0052] In some embodiments, obtaining the images captured by multiple cameras around the lamp post includes: obtaining the images captured by multiple cameras around the lamp post in the same area. Thus, through the geographical relationship, the control range of the lamp post device is set. Since the lamp post device belongs to an edge device, it can more efficiently control the image capture function of the camera.

[0053] On the premise that multiple cameras are installed on the lamp post, the lamp post integrates a video capture function and can capture images in specific directions. Since there are multiple directions in which the multiple cameras capture images relative to the lamp post, the multiple cameras can provide images of the outer side of the lamp post in multiple directions, thereby accurately constructing the information around the lamp post.

[0054] Step 204, determine the matching area between the above-mentioned images in different directions.

[0055] The different directions are different directions towards the outer side of the lamp post, used to provide information about each object in each direction, thereby gathering more extensive information. For example, due north centered on the lamp post and due south centered on the lamp post are different directions.

[0056] The matching region is the registration result of similar regions of images in different directions. When the camera captures images in different directions, due to the corresponding field of view angle and installation height of each camera, there is also a certain similarity between the images captured by the camera in different directions, and the matching region can be determined based on this. In different directions towards the outside of the lamp post, even the tightness of the camera screws will affect the image capture situation of the camera, causing distortion in the images captured by each camera. These distortions are inevitable. By matching the above-mentioned images in different directions, an adaptive adjustment of the matching region can be achieved, reducing the degree of inconsistency of distortions in different directions.

[0057] In some embodiments, determining the matching region between the above-mentioned images in different directions includes: calculating the similarity between the above-mentioned images in different directions to obtain the image similarity between different directions; in the above-mentioned images in different directions, determining the matching region in different directions according to the image similarity; where different directions refer to two directions different from each image when the camera captures images towards the outside of the lamp post. Thus, when the camera captures images towards the outside of the lamp post, based on different directions on the outside, the similarity of the images is calculated, and under the corresponding similarity conditions, there are respective matching regions for the images captured by two different cameras in different directions, thereby forming a panoramic image.

[0058] In some embodiments, determining the matching region between the above-mentioned images in different directions includes: determining the matching region between the above-mentioned images in at least three directions according to the matching points. Thus, based on the images in each two directions, the respective matching regions are determined according to the time and the matching pixel points, thereby forming an image with a wider range.

[0059] In the case of the images captured by the camera around the lamp post, there are boundaries with overlapping information in these above-mentioned images in different directions. These boundaries can perform matching and distortion calibration on the above-mentioned images in different directions from different directions, achieve adaptive adjustment of the matching region, reduce the degree of distortion in different directions and make them tend to be consistent, thereby avoiding too large a difference between the connection region of the panoramic image and the region outside the matching region.

[0060] Step 206: Stitch the images in each direction with the matching region as the connection to obtain a panoramic image.

[0061] Connection means using the matching region as the boundary region of the images in different directions to stitch the images in each direction into a panoramic image. In the case of using the matching region as the connection, each frame of the images in each direction forms a similar connection relationship, thereby integrating the information of the multi-view cameras to form comprehensive information of multiple views.

[0062] A panoramic image is an image synthesized from images in multiple directions. When using the matching area as the connection, the consistency of each area of the panoramic image can be relatively high, so that the corresponding information can be retained more meticulously. And the relative position between the camera and the lamp post is used as a reference for each direction, so that the comprehensive viewing angle of the image acquisition by the camera is also referenced by the relative position between the camera and the lamp post. Furthermore, the field of view angle lost during the stitching process is relatively small, so that the information provided by the images in each direction can be better retained. The panoramic image and the images collected in each direction are two types of video streams, and these two types of video streams can be switched and coexist.

[0063] In some embodiments, each direction includes two directions, namely the first direction and the second direction. Both the first direction and the second direction perform image acquisition around the bottom of the lamp post outside the lamp post, and the first direction is different from the second direction. When there is a corresponding matching area between the image in the first direction and the image in the second direction, the images in the first direction and the second direction can be stitched into a corresponding panoramic image through the connection of this matching area.

[0064] For example, if the first direction is also located due south of the lamp post, then the second direction is towards a direction other than due south of the lamp post; the second direction can be a direction that is 115 degrees to 125 degrees off due south of the lamp post, or the second direction can be a direction that is 87 degrees to 93 degrees off due south of the lamp post. At this time, when there is a corresponding matching area between the image in the first direction and the image in the second direction, the images in the first direction and the second direction can be stitched into a corresponding panoramic image through the connection of this matching area.

[0065] In some embodiments, stitching the images in each direction with the matching area as the connection to obtain a panoramic image includes: stitching other matching areas of the images in each direction to their respective matching areas with the matching area as the connection to obtain a panoramic image. Thus, there are non-matching areas outside the matching areas in the images in each direction, and the matching areas are between the non-matching areas of each image. And the matching degree between the non-matching areas contained in the images in different directions is small and the similarity is weak, which helps to better retain the original information in each direction.

[0066] In some embodiments, stitching the images in each direction with the matching area as the connection to obtain a panoramic image includes: using the matching area as the connection area between the images in each direction and stitching the images in each direction with the lamp post as the center to obtain a panoramic image. Thus, with the lamp post as the center, the change in the direction of the images in each direction is less, so that the degree of information retention of the panoramic image is improved.

[0067] In the panoramic image processing method of the above-mentioned lamp post, on the premise that multiple cameras are arranged on the lamp post, the lamp post is integrated with a video acquisition function and can acquire images in a specific direction. Since the directions of image acquisition of multiple cameras relative to the lamp post are various, multiple cameras can provide images of the outside of the lamp post in multiple directions, thereby accurately constructing the information around the lamp post. In this case, there are boundaries with overlapping information in the images in different directions. These boundaries can match and correct the distortion of the above-mentioned images in different directions from different directions, realize the adaptive adjustment of the matching area, reduce the distortion degree in different directions and make it tend to be consistent, so as to avoid too large a difference between the splicing area of the panoramic image and the area outside the matching area. Furthermore, the consistency of each area of the panoramic image can be relatively high, and the field of view angle lost during the splicing process is also relatively small, so that the information provided by the images in each direction can be better retained. Thus, through the above directions, the information around the lamp post during acquisition is more accurate; and determining the matching areas in different directions can represent the matching and calibration results of the overlapping information in different directions, avoiding unnatural areas in the panoramic image caused by the above-mentioned differences during the splicing process, so as to avoid the influence of the matching process on the information accuracy; finally, directly retaining each direction for splicing can make the loss of the field of view angle in each direction during the splicing process smaller, so as to more accurately represent the information. Thus, the panoramic image is spliced according to the above directions, with a small gap from the actual scene and can intuitively reflect the actual picture.

[0068] In some embodiments, the panoramic image includes a spliced image area corresponding to the reference image; determining the matching area between the images in different directions includes step 302. Correspondingly, splicing the images in each direction with the matching area as the connection to obtain the panoramic image includes step 304; wherein:

[0069] Step 302, determining the matching area between the reference image and at least two adjacent images; wherein, the reference image and the adjacent images are acquired in the above-mentioned adjacent directions, and each adjacent image and the reference image are acquired in different above-mentioned directions.

[0070] Both the reference image and the adjacent images are images acquired by the camera itself towards the outside of the lamp post in the direction of the camera itself relative to the lamp post, and there are at least two adjacent above-mentioned directions for the reference image, thereby forming at least two adjacent images of the reference image. The adjacent above-mentioned directions and the above-mentioned direction of the reference image during acquisition meet the preset angle condition; meeting the preset angle condition includes, but is not limited to, the smallest angle difference between the adjacent above-mentioned directions and the above-mentioned direction of the reference image during acquisition, and the largest overlapping angle between the adjacent above-mentioned directions and the above-mentioned direction of the reference image during acquisition.

[0071] Optionally, at least two of the above-mentioned adjacent directions may include the above-mentioned directions on the left and right sides of the reference image, so as to form a panoramic image from a top-down or bottom-up perspective. The above-mentioned directions on the left and right sides of the reference image may respectively have an angular difference from the direction of the reference image during acquisition; at least two of the above-mentioned adjacent directions may also include the above-mentioned directions on the upper and lower sides of the reference image. The above-mentioned directions on the upper and lower sides of the reference image may respectively have an angular difference from the direction of the reference image during acquisition.

[0072] Optionally, the above-mentioned adjacent directions refer to the images collected by adjacent cameras with a camera orientation difference less than a preset range, based on the direction in which the camera itself collects images towards the outside of the lamp post.

[0073] In one example, the reference image is an image in which the camera itself faces the outside of the lamp post and is within 60 degrees of deviation from the due north direction. On this basis, one adjacent image of the reference image is an image in which the camera itself faces the outside of the lamp post and is within the range of -60 degrees to -180 degrees of deviation from the due north direction, and another adjacent image of the reference image is an image in which the camera itself faces the outside of the lamp post and is within the range of +60 degrees to +180 degrees of deviation from the due north direction. Thus, a panoramic image from a top-down or bottom-up perspective is formed through the images in the three above-mentioned directions.

[0074] In another example, the reference image is an image in which the camera itself faces the outside of the lamp post and is within 135 degrees of deviation from the due north direction, that is, from a deviation of -135 degrees to a deviation of +135 degrees; on this basis, one adjacent image of the reference image is an image in which the camera itself faces the outside of the lamp post and is within the range of -60 degrees to -180 degrees of deviation from the due north direction, that is, from a deviation of -60 degrees to a deviation of -180 degrees; another adjacent image of the reference image is an image in which the camera itself faces the outside of the lamp post and is within the range of +60 degrees to +180 degrees of deviation from the due north direction, that is, from a deviation of +60 degrees to a deviation of +180 degrees. Thus, a panoramic image from a top-down or bottom-up perspective is formed through the images in the three above-mentioned directions, and there is a large overlapping area, so that the matching area is large, thereby ensuring that the pixels in the image stitching area are smoother.

[0075] In some embodiments, determining a matching region between a reference image and at least two adjacent images includes: performing feature point matching on the reference image and each adjacent image of the reference image to obtain the feature points corresponding to each adjacent image; the feature points corresponding to each adjacent image include the feature points of the reference image and the corresponding points of the feature points of the reference image in each adjacent image; determining the matching region between the reference image and at least two adjacent images according to the feature points corresponding to each adjacent image. Thus, based on the feature point matching between the reference image and its adjacent images, the matching region is related to the feature points corresponding to each adjacent image, so that the matching region is not necessarily a whole block region, but can also be multiple pixel points, thereby ensuring a better registration effect of the matching region.

[0076] In some embodiments, determining a matching region between a reference image and at least two adjacent images includes: calculating the similarity of each region between the reference image and each adjacent image of the reference image to obtain a reference image region and an adjacent image region whose similarity meets a preset similarity condition; registering the reference image region and the adjacent image region whose similarity meets the preset similarity condition to obtain the matching region between the reference image and each adjacent image of the reference image; wherein, the similarity meeting the preset similarity condition may mean that the similarity of a certain region is the largest, or it may mean that the similarity of a certain region is greater than a preset value. Thus, corresponding regions are screened by similarity, and the pixel difference between the reference image and its adjacent images is reduced through registration, so that the difference between the connection region and the reference image is relatively small, and the difference between the connection region and the above adjacent images is relatively small, thereby reducing the situation that the panoramic image has a sudden change due to the difference in the above direction, and thus expressing information more accurately.

[0077] Step 304: Using the matching region as a connection, splicing the reference image and the adjacent image to obtain a spliced image region corresponding to the reference image.

[0078] The spliced image region is at least part of the panoramic image. When the matching region is used as a connection, the splicing process of the reference image and the adjacent image is relatively natural. Therefore, the spliced image region can retain relatively high information of the reference image and the adjacent image, thereby ensuring accurate information transmission.

[0079] In some embodiments, taking the matching region as the connection, the reference image and the similar images are spliced to obtain the spliced image region corresponding to the reference image, including: taking the first matching region as the connection, splicing the reference image and the first similar image to obtain the first spliced image region corresponding to the reference image; taking the second matching region as the connection, splicing the first spliced image region and the second similar image to obtain the second spliced image region corresponding to the reference image; at this time, the second spliced image region is the spliced image region corresponding to the reference image. Thus, two splicings are performed successively to form the spliced image region corresponding to the reference image, and thus, one splicing process is omitted.

[0080] In some embodiments, taking the matching region as the connection, the reference image and the similar images are spliced to obtain the spliced image region corresponding to the reference image, including: taking the first matching region as the connection, splicing the reference image and the first similar image to obtain the first spliced image region corresponding to the reference image; taking the second matching region as the connection, splicing the reference image and the second similar image to obtain the second spliced image region corresponding to the reference image; both the first spliced image region and the second spliced image region are the spliced image regions corresponding to the reference image. Thus, the reference image is spliced twice to form the spliced image region corresponding to the reference image. At this time, there may be a further splicing or matching process between the first spliced image region and the second spliced image region.

[0081] Among them, the first matching region is the matching region between the reference image and the first similar image, and the second matching region is the matching region between the reference image and the second similar image; the first similar image and the second similar image are different similar images of the reference image, and the first similar image and the second similar image are acquired at different directions of the camera relative to the lamp post.

[0082] In this embodiment, both sides of each frame of the reference image are spliced with different similar images respectively. Since both the reference image and the similar images are acquired at the direction relative to the lamp post, the center of the corresponding spliced image region is the lamp post, and the reference image and each frame of the similar images can form an approximately annular panoramic image, so as to simulate the shooting angle of a person at the corresponding position for the field of view angle at the corresponding camera position and more finely characterize the corresponding information.

[0083] In some embodiments, the resolution of the reference image is less than the resolution of the similar images, the field of view angle of the reference image is greater than the field of view angle of the similar images, and the range of the reference image corresponding to the spliced image region is less than the range of each frame of the similar images corresponding to the spliced image region.

[0084] The resolution represents the total number of pixels in an image, which can be expressed as the product of the width and the height, or as a specific value. The higher the resolution of an image, the larger the number of pixels, and the higher the degree of information refinement, resulting in a smaller information range provided by each pixel. Therefore, the higher the resolution of an image, the lower the probability of successful matching with other images; the lower the resolution of an image, the smaller the number of pixels, and the lower the degree of information refinement, resulting in a larger information range provided by each pixel. Therefore, the lower the resolution of an image, the higher the probability of successful matching with other images.

[0085] Since the reference image needs to be stitched with at least two adjacent images respectively, the resolution of the reference image is relatively small. On the one hand, this makes the matching area between the reference image and each adjacent image relatively large. On the other hand, it enables the corresponding range of the adjacent images to provide more information, thereby ensuring the accuracy of the information through the corresponding range of the adjacent images. Therefore, the resolution of the reference image is smaller than that of the adjacent images, which can make the matching accuracy of the matching area better and the accuracy of the information conveyed by the stitched image area better.

[0086] The field of view angle represents the viewing angle range of the image captured by the camera. The field of view angle can be expressed in degrees or represented by a larger range of the image. The larger the field of view angle of an image, the larger the physical area that each pixel needs to represent, the larger the information range of a single pixel, and the lower the degree of information refinement. Therefore, the larger the field of view angle of an image, the higher the probability of successful matching with other images; the smaller the field of view angle of an image, the smaller the physical area that each pixel needs to represent, the smaller the information range of a single pixel, and the higher the degree of information refinement. Therefore, the smaller the field of view angle of an image, the higher the probability of successful matching with other images.

[0087] Since the reference image needs to be stitched with at least two adjacent images respectively, the field of view angle of the reference image is relatively large. On the one hand, this makes the matching area between the reference image and each adjacent image relatively large. On the other hand, it enables the corresponding range of the adjacent images to provide more information, thereby ensuring the accuracy of the information through the corresponding range of the adjacent images. Therefore, the field of view angle of the reference image is larger than that of the adjacent images, which can make the matching accuracy of the matching area better and the accuracy of the information conveyed by the stitched image area better.

[0088] The corresponding range of the reference image in the stitched image area means that the feature difference between the stitched image area and the reference image within this range is smaller than the feature difference between the stitched image area and the adjacent image within this range. The feature difference includes, but is not limited to, the difference in pixel values and the difference in channel values. For example, the difference between the pixel value of coordinate A in the stitched image area and the pixel value of coordinate A in the reference image is 30, and the difference between the pixel value of coordinate A in the stitched image area and the pixel value of coordinate A in the adjacent image is 40.

[0089] The range corresponding to the adjacent image in the stitching image area means that the feature difference between the stitching image area and the reference image within this range is greater than the feature difference between the stitching image area and the adjacent image within this range. This feature difference includes, but is not limited to, the difference in pixel values and the difference in channel values. For example, the gap between the pixel value of coordinate A in the stitching image area and the pixel value of coordinate A in the reference image is 40, which is greater than the gap between the pixel value of coordinate A in the stitching image area and the pixel value of coordinate A in the adjacent image, which is 30.

[0090] In one example, the reference image is obtained by collecting with a 5 - megapixel fisheye camera with a 270 - degree field of view angle, and the adjacent image is obtained by collecting with an 8 - megapixel camera with a 180 - degree field of view angle. Thus, the range of the area guaranteed by the fisheye lens is relatively large, and the corresponding image is obtained.

[0091] In another example, the reference image is obtained by collecting with a 4 - megapixel fisheye camera with a 180 - degree field of view angle, and the adjacent image is obtained by collecting with an 8 - megapixel camera with a 120 - degree field of view angle. Thus, by combining the fisheye lens with a wide - angle lens, the corresponding image is obtained.

[0092] In this embodiment, compared with the adjacent image, the reference image has a relatively smaller resolution and a relatively larger field of view angle, making the granularity of the information represented by each pixel in the reference image coarser. This can make the overlapping range between the reference image and its adjacent image larger, making it easier to match, thereby ensuring that there are fewer unnatural situations in the panoramic image caused by the stitching process, avoiding the introduction of abnormal information, and thus ensuring that the information represented by the panoramic image is more accurate. At the same time, the range corresponding to each adjacent image in the stitching image area is larger, making the stitching image area closer to the information with finer granularity, thereby ensuring more accurate information.

[0093] In some embodiments, before obtaining the images collected by multiple cameras around the lamp post, the method further includes steps 402 - 404. Correspondingly, obtaining the images collected by multiple cameras around the lamp post includes step 406, where:

[0094] Step 402: Based on the first corrected images collected by multiple cameras around the lamp post, obtain the corrected parameters of each camera, and based on the corrected parameters of each camera, obtain the second corrected images in each direction.

[0095] The first corrected image is the image used for correcting each camera. The first corrected image is collected separately by each camera and is used to correct the image acquisition parameters of each camera itself to achieve the internal parameter correction of the camera.

[0096] The calibrated parameters of each camera are the image acquisition parameters obtained by each camera independently based on its own calibration. Exemplarily, the calibrated parameters of each camera include, but are not limited to, the distortion parameter calibration of each camera, the principal point calibration of the camera, etc.

[0097] The second calibrated image is an image used to calibrate the relative relationship between different cameras. The second calibrated image is independently acquired by each camera and is used to calibrate the relative relationship of the image acquisition parameters between different cameras, so that the matching degree between the images acquired by different cameras is relatively high.

[0098] In some embodiments, the step of obtaining the first calibrated image includes: each camera acquires images in each direction based on its own parameters to be calibrated, and transmits the images acquired by the camera based on its own parameters to be calibrated to the lamp post device, so that the lamp post device obtains the first calibrated image.

[0099] In some embodiments, the step of obtaining the first calibrated image includes: in response to an acquisition instruction for a panoramic image, the lamp post device obtains the first calibrated image. Thus, when a panoramic image needs to be displayed, the calibration step is executed so that there are more matching regions in the stitching process of the panoramic image.

[0100] In some embodiments, obtaining the calibrated parameters of each camera based on the first calibrated images acquired by multiple cameras around the lamp post includes: based on the first calibrated image in each direction, calibrating the distortion parameters included in the parameters of each camera to which the first calibrated image belongs, to obtain the calibrated parameters of each camera. Among them, the distortion parameters are used to characterize and correct the distortion generated during the imaging process of the camera, including but not limited to radial distortion parameters and tangential distortion parameters.

[0101] In some embodiments, obtaining the second calibrated images in each direction based on the calibrated parameters of each camera includes: after each camera calibrates the image to be calibrated acquired by itself based on its own calibrated parameters, the lamp post device uses the image to be calibrated as the second calibrated images in each direction.

[0102] Step 404, calibrate the image offset between each camera according to the second calibrated images in each direction, to obtain the calibrated parameters between each camera.

[0103] Image offset refers to the information deviation situation between the images acquired by different cameras. The image offset includes at least the offset of pixels or pixel values, and may also include the offset of distortion parameters, as well as the pixel relationship at adjacent positions.

[0104] The calibrated parameters between each camera are the image acquisition parameters obtained by each camera based on the joint calibration with other cameras. The calibrated parameters between each camera may involve further fine-tuning of the calibrated parameters of each camera, so that the images collected by different cameras can be more matched. The calibrated parameters of each camera include, but are not limited to, the offset calibration parameters of pixels or pixel values, the offset calibration parameters of distortion parameters, etc.

[0105] In some embodiments, the image offset between each camera is corrected according to the second calibrated images in each direction to obtain the calibrated parameters between each camera, including: when the first camera acquires the first calibrated image in the first direction and the second camera acquires the second calibrated image in the second direction, the image offset between the first camera and the second camera is corrected according to the pixel offset value of the overlapping area between the second calibrated image acquired in the first direction and the second calibrated image acquired in the second direction, so as to obtain the calibrated parameters between the first camera and the second camera. Thus, the first direction and the second direction are the directions for different cameras to acquire images, so the first direction and the second direction are at least two different directions. Then, through the corresponding pixel offset value of the overlapping area, the pixel offset rule between the first camera and the second camera is determined, and the calibrated parameters between at least two cameras are obtained, so that the image acquisition parameters of different cameras are matched with each other.

[0106] Step 406: Based on the calibrated parameters between each camera, obtain the images collected by multiple cameras around the lamp post.

[0107] In some embodiments, to obtain the second calibrated images in each direction based on the calibrated parameters of each camera, including: after each camera corrects the to-be-calibrated image collected by itself based on the calibrated parameters between each camera, the lamp post device uses the to-be-calibrated image as the to-be-stitched image in each direction; the to-be-stitched image is used to determine the matching area for stitching the panoramic image.

[0108] In this embodiment, calibration is first performed based on the first corrected image of each camera to achieve the internal parameter calibration of the individual camera, avoiding the influence of component problems such as the tightness of the screws inside the camera on image registration. Furthermore, according to the second corrected images in each direction, the image offsets between the cameras are corrected to obtain the corrected parameters between the cameras, which can make problems such as the relative degree of screw tightness of the cameras affect registration, so that the images collected in different above directions can be accurately joined together. Furthermore, based on the corrected parameters between the cameras, images collected by multiple cameras around the lamp post are obtained. Thus, the image will neither be abnormal due to the installation conditions of the components inside the camera nor be abnormal due to the relative installation conditions of the components between the cameras, ensuring accuracy.

[0109] In some embodiments, the central region of the panoramic image is the position of the lamp post relative to each camera, and the central region of the panoramic image is close to the corresponding region of the first side of each image in the panoramic image, and the central region of the panoramic image is far from the corresponding region of the second side of each image in the panoramic image; the closer the object is to the lamp post, the closer it is to the first side of the image; the farther the object is from the lamp post, the closer it is to the second side of the image.

[0110] The central region is the region at the central position of the panoramic image. When the camera is set on the lamp post, this central region is used to represent the position of the lamp post in the panoramic image, and the images for stitching are stitched with this central region as the center.

[0111] The first side and the second side are the image boundaries with the lamp post as a reference, and this image boundary is the boundary of the image collected in the direction of the camera relative to the lamp post. The distance between the object and the first side is used to represent that the object is close to the lamp post; the distance between the object and the second side is used to represent that the object is far from the lamp post. Optionally, the first side and the second side are two relatively set sides.

[0112] The corresponding region of the first side in the panoramic image represents the region of the first side in the panoramic image after the images in each direction are matched and stitched. The corresponding region of the second side in the panoramic image represents the region of the second side in the panoramic image after the images in each direction are matched and stitched.

[0113] In the Figure 5 shown panoramic image example, three regions are included around the central region. Region A is the corresponding range of the reference image in the panoramic image, and regions B and C are the corresponding ranges of two adjacent images of the reference image in the panoramic image. Among them, in the Figure 6 shown panoramic image example, Figure 5In the A area, B area, and C area in [the text], there are respective first sides and second sides; the corresponding area 601 of the first side in the panoramic image and the corresponding area 602 of the second side in the panoramic image are formed.

[0114] Since the camera is located on the lamp post and the camera captures images based on its own direction relative to the lamp post, when a person views the surrounding area of the lamp post at the position of the camera, the person is located in the central area of the panoramic image. At this time, the closer an object is to the lamp post, the closer it is to the first side of the image, and the closer the person can see the object to the corresponding area of the first side in the panoramic image; the farther an object is from the lamp post, the closer it is to the second side of the image, and the closer the person can see the object to the corresponding area of the second side in the panoramic image. Thus, it more realistically simulates the perspective of a person at the camera position to reduce information loss.

[0115] In this embodiment, the central area of the panoramic image is used to determine the central area when each image is stitched, so that an approximately circular panoramic image can be directly formed, thus more realistically simulating the panorama seen by a person at the camera position, thereby ensuring the normal transmission of information; on this basis, through the second side where the first side and the second side of the image are relatively arranged, a regional correspondence relationship between the image to be stitched and the panoramic image is formed, so as to control the expression of the real information of the object based on the distance between the object and the lamp post, thus more realistically simulating the perspective of a person at the camera position, thereby ensuring the normal transmission of information.

[0116] In some embodiments, the panoramic image includes control areas corresponding to each direction; the method further includes: in response to a magnification instruction for the control area in the panoramic image, determining a target image in the direction that the control area matches from the images in each direction; and performing display based on the target image.

[0117] The magnification instruction is an instruction for indicating the refinement of information in a certain area. The control area is an area on the panoramic image for receiving the magnification instruction, and each control area corresponds to one or more images in a direction. Optionally, the control area includes the corresponding area of the first side of each image in the panoramic image and the area between the corresponding areas of the second side of each image in the panoramic image. The control area can also be a button or other identifier on the panoramic image. The target image is at least one of the images in the above-mentioned directions, and the target image matches the control area.

[0118] In one example, as Figure 7 shown, cameras in three directions respectively capture image A with 5 million pixels, image B with 8 million pixels, and image C with 8 million pixels. One of these three images can be used as the target image, and these three images respectively have corresponding ranges in the panoramic image.

[0119] In some embodiments, in response to a zoom-in instruction for a panoramic image in a control area, it includes: when it is detected that there is a position indicated by an input device in the control area of the panoramic image, determining the zoom-in instruction for the panoramic image in the control area. Exemplarily, when it is detected that the control area of the panoramic image contains a mouse pointer, determining the zoom-in instruction for the panoramic image in the control area.

[0120] In some embodiments, determining a target image in the direction matched by the control area from images in all directions includes: determining a target image of a camera matched by the control area from images in all directions. Thus, through the correspondence between the control area and the camera, the correspondence between the control area and the matched direction is reflected, so as to accurately determine the target image.

[0121] In some embodiments, determining a target image in the direction matched by the control area from images in all directions includes: taking the image corresponding to the control area as the target image from images in all directions. Thus, according to the images used to stitch the control area, the target image is determined to reduce visual elements such as buttons, so as to ensure the accuracy of information transmission.

[0122] In some embodiments, performing a display based on the target image includes: when the camera sends the target image to the lamp post device, the lamp post device displays the target image.

[0123] In some embodiments, performing a display based on the target image includes: when the camera sends the target image to the lamp post device, the lamp post device displays the target image and the panoramic image.

[0124] It should be understood that in the actual stitching process, there will be overlapping areas between the images collected by cameras in all directions. Coupled with the problem of the camera's field of view angle, some pixels and information of the picture content are still lost in the stitched panoramic image. Therefore, when it is necessary to view certain areas in further detail, clicking on the picture can push up the original picture of this camera. The original picture has a higher resolution and field of view angle, can view a larger range of pictures, and provides detailed picture details.

[0125] In this embodiment, the panoramic image has control areas matched in all directions. The control area can control the data stream to be displayed, so that the target image in a specific matched direction is displayed. When the target image exists in the form of a video stream, the clarity of the target image is relatively high, so as to avoid information loss caused by the traditional digital zoom process, thereby ensuring the accuracy of information transmission. Thus, the panoramic image has a zoom-in function, and the clarity of the zoomed-in image remains unchanged or has less loss.

[0126] In an example, as Figure 8 shown, it includes:

[0127] Step 802: The lamp post device collects images from three cameras in real time, stitches the images from the three cameras into a panoramic image from a top-down perspective, and sends the panoramic image to the display platform running on the terminal. Among them, the steps of image stitching can be implemented through steps 302 - 306, and the display platform can be a client.

[0128] Step 804: On the display platform, the terminal displays the corresponding panoramic image for the user to view.

[0129] Step 806: When it is detected that there is a click in the control area of the panoramic image, the terminal obtains the high-resolution image corresponding to the control area where the click acts as the target image for display from the multi-channel high-resolution images of the three cameras.

[0130] In this embodiment, it intuitively reflects the actual picture, and the later stitched panoramic picture has a zoom function. In this way, three cameras are used to achieve 360-degree picture coverage, and the 360-degree top-down perspective stitched picture can be used to intuitively reflect the actual on-site scene. Moreover, the panoramic picture has a click-to-zoom function, providing a clearer and wider-angle detailed picture.

[0131] It should be understood that although the steps in the flowcharts involved in the above-described embodiments are shown in sequence according to the arrows, these steps do not necessarily have to be executed in the order indicated by the arrows. Unless there is a clear description in this article, the execution of these steps has no strict order limit, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-described embodiments may include multiple steps or multiple stages. These steps or stages do not necessarily have to be executed at the same time, but can be executed at different times. The execution order of these steps or stages does not necessarily have to be sequential, but can be executed alternately or in turn with at least a part of other steps or steps or stages in other steps.

[0132] Based on the same inventive concept, the embodiment of the present application also provides a panoramic image processing device for a lamp post for implementing the panoramic image processing method for the lamp post involved above. The implementation solutions provided by this device to solve problems are similar to the implementation solutions described in the above method. Therefore, the specific limitations in one or more embodiments of the following panoramic image processing device for a lamp post can refer to the limitations on the panoramic image processing method for a lamp post in the above text, and will not be repeated here.

[0133] In an exemplary embodiment, as Figure 9 shown, a panoramic image processing device for a lamp post is provided, including:

[0134] An acquisition module 902, configured to acquire images collected by multiple cameras around the lamp post; the multiple cameras are arranged on the lamp post, and the directions of the multiple cameras for collecting the images relative to the lamp post are various;

[0135] A matching module 904, configured to determine a matching area between the images in different directions;

[0136] A processing module 906, configured to splice the images in each direction with the matching area as a connection to obtain a panoramic image.

[0137] In one embodiment, the panoramic image includes a spliced image area corresponding to a reference image; the matching module 904 is configured to:

[0138] Determine a matching area between the reference image and at least two adjacent images; wherein, the reference image and the adjacent images are collected in adjacent directions, and each adjacent image and the reference image are collected in different directions;

[0139] The processing module 906 is configured to:

[0140] Splice the reference image and the adjacent images with the matching area as a connection to obtain a spliced image area corresponding to the reference image.

[0141] In one embodiment, the resolution of the reference image is less than the resolution of the adjacent images, the field of view angle of the reference image is greater than the field of view angle of the adjacent images, and the range of the reference image corresponding to the spliced image area is less than the range of each adjacent image corresponding to the spliced image area.

[0142] In one embodiment, the acquisition module 902 is configured to:

[0143] Based on the first corrected images collected by multiple cameras around the lamp post, obtain the corrected parameters of each camera, and based on the corrected parameters of each camera, obtain the second corrected images in each direction;

[0144] According to the second corrected images in each direction, correct the image offset between the cameras to obtain the corrected parameters between the cameras;

[0145] Based on the corrected parameters between the cameras, acquire the images collected by multiple cameras around the lamp post.

[0146] In one embodiment, the central region of the panoramic image is the position of the light pole relative to each of the cameras, and the central region of the panoramic image is close to the first side of each image in the corresponding region of the panoramic image, and the central region of the panoramic image is far from the second side of each image in the corresponding region of the panoramic image; the object closer to the light pole is closer to the first side of the image; the object farther from the light pole is closer to the second side of the image.

[0147] In one embodiment, the panoramic image includes control regions corresponding to each of the directions; the processing module 906 is configured to:

[0148] In response to a magnification instruction for the control region in the panoramic image, determine, from the images in each of the directions, a target image in the direction matched by the control region;

[0149] Perform display based on the target image.

[0150] Each module in the above panoramic image processing device of the light pole can be implemented in whole or in part by software, hardware, and their combination. Each of the above modules can be embedded in the processor in the computer device in hardware form or be independent of the processor, or can be stored in the memory in the computer device in software form, so that the processor can call and execute the operations corresponding to each of the above modules.

[0151] In an exemplary embodiment, a computer device is provided. The computer device can be a light pole device, and its internal structure diagram can be as Figure 10 shown. The computer device includes a processor, a memory, an input / output interface (Input / Output, abbreviated as I / O), and a communication interface. Among them, the processor, the memory, and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used for exchanging information between the processor and external devices. The communication interface of the computer device is used for communicating with an external terminal through a network connection. When the computer program is executed by the processor, it implements a panoramic image processing method of a light pole.

[0152] Those skilled in the art can understand, Figure 10The structure shown is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0153] In one embodiment, a computer device is further provided, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps in the above method embodiments are implemented.

[0154] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0155] In one embodiment, a computer program product is provided, including a computer program, and when the computer program is executed by a processor, the steps in the above method embodiments are implemented.

[0156] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use, and processing of relevant data need to comply with relevant regulations.

[0157] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods. Among them, any reference to a memory, database, or other medium used in the embodiments provided in the present application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include Read-Only Memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, Resistive Random Access Memory (ReRAM), Magnetoresistive Random Access Memory (MRAM), Ferroelectric Random Access Memory (FRAM), Phase Change Memory (PCM), graphene memory, etc. Volatile memory can include Random Access Memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can be in various forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM), etc. The databases involved in the embodiments provided in the present application can include at least one of relational databases and non-relational databases. Non-relational databases can include distributed databases based on blockchain, etc., without limitation. The processors involved in the embodiments provided in the present application can be general-purpose processors, central processors, graphics processors, digital signal processors, programmable logic devices, data processing logics based on quantum computing, Artificial Intelligence (AI) processors, etc., without limitation.

[0158] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in the present application.

[0159] The above-described embodiments merely represent several implementation manners of the present application. The description thereof is relatively specific and detailed, but it should not be construed as a limitation on the patent scope of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all fall within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the appended claims.

Claims

1. A panoramic image processing method for a lamp post, characterized in that, The method includes: Obtaining images collected by multiple cameras around the lamp post; the multiple cameras are arranged on the lamp post, and the directions of collecting the images by the multiple cameras relative to the lamp post are various; Determining matching regions between the images in different directions; Taking the matching regions as joints to splice the images in each direction to obtain a panoramic image.

2. The method according to claim 1, wherein The panoramic image includes a spliced image region corresponding to a reference image; The determining of the matching regions between the images in different directions includes: Determining matching regions between a reference image and at least two adjacent images; wherein, the reference image and the adjacent images are collected in adjacent directions, and each adjacent image and the reference image are collected in different directions; The taking the matching regions as joints to splice the images in each direction to obtain a panoramic image includes: Taking the matching regions as joints to splice the reference image and the adjacent images to obtain the spliced image region corresponding to the reference image.

3. The method according to claim 2, wherein The resolution of the reference image is less than that of the adjacent images, the field of view angle of the reference image is greater than that of the adjacent images, and the range corresponding to the reference image in the spliced image region is less than the range corresponding to each adjacent image in the spliced image region.

4. The method according to claim 1, wherein Before obtaining the images collected by multiple cameras around the lamp post, the method further includes: Based on the first corrected images collected by multiple cameras around the lamp post, obtaining the corrected parameters of each camera, and based on the corrected parameters of each camera, obtaining the second corrected images in each direction; According to the second corrected images in each direction, correcting the image offsets between the cameras to obtain the corrected parameters between the cameras; The obtaining of the images collected by multiple cameras around the lamp post includes: Based on the corrected parameters between the cameras, obtaining the images collected by multiple cameras around the lamp post.

5. The method according to claim 1, wherein The central region of the panoramic image is the position of the lamp post relative to each camera, and the central region of the panoramic image is close to the corresponding region of the first side of each image in the panoramic image, and the central region of the panoramic image is far from the corresponding region of the second side of each image in the panoramic image; the object closer to the lamp post is closer to the first side of the image; The object farther from the lamp post is closer to the second side of the image.

6. The method according to claim 1, wherein The panoramic image includes control regions corresponding to each direction; the method further includes: In response to a zoom-in instruction for the panoramic image in the control region, determining a target image in the direction matching the control region from the images in each direction; Performing display based on the target image.

7. A panoramic image processing device for a lamp post, characterized in that, The device includes: An obtaining module, configured to obtain images collected by multiple cameras around the lamp post; the multiple cameras are arranged on the lamp post, and the directions of collecting the images by the multiple cameras relative to the lamp post are various; A matching module for determining a matching area between the images in different directions; A processing module for stitching the images in each of the directions with the matching area as a connection to obtain a panoramic image.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, the steps of the method according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented.

Images