Camera photo splicing method and device, computer equipment and storage medium

Abstract

The invention provides a camera photo splicing method and device, computer equipment and a storage medium, and belongs to the field of visual inspection, and the method specifically comprises the steps: receiving a plurality of shot images of the same scene shot by a plurality of cameras at the same time, the shot images carrying mechanical coordinate information; identifying the shot image by adopting a visual inspection system, and correspondingly storing an obtained identification result and the shot image; independently storing a shot image shot by each camera and a corresponding identification result in a time queue form; extracting shot images in a plurality of time queues at the same moment and corresponding identification results, and converting all the identification results into a virtual mechanical coordinate system through a calibration relationship between pixel coordinates and mechanical coordinates; and according to the mechanical coordinate system, determining a superposed view field of the plurality of shot images, and combining to obtain a spliced picture. Through the processing scheme provided by the invention, lossless splicing is carried out on photos of different camera fields of view.

Description

Technical Field

[0001] The present invention relates to the field of visual detection, and in particular to a camera photo stitching method, device, computer equipment and storage medium. Background Art

[0002] At present, many factories have realized the production of automated assembly lines. Generally, machine vision is used to track and identify the products on the automated assembly lines, and analyze and judge them. Machine vision is the science and technology of studying the computer simulation of biological macroscopic visual functions, that is, using machines such as cameras and computers to replace human eyes to measure, track, identify and judge targets. For general applications, users can realize functions such as product presence / absence judgment, surface / defect inspection, size measurement, barcode reading, etc. without programming. However, when items on the assembly line overlap, industrial smart cameras are prone to misjudgment, missed detection, inaccurate measurement and other problems. Therefore, it is necessary to improve the existing technology to overcome the defects in the existing technology. Although the assembly line can be inspected by using multiple cameras, the fields of view of multiple cameras may overlap, and the calculation results of each camera alone may cause data duplication. Forcibly cropping and splicing the image itself will cause the image quality to be damaged and distorted, which will affect the accuracy of visual inspection, so it is impossible to further expand the application. Summary of the invention

[0003] Therefore, in order to overcome the above-mentioned shortcomings of the prior art, the present invention provides a camera photo stitching method, device, computer equipment and storage medium for losslessly stitching photos of different camera fields of view.

[0004] In order to achieve the above-mentioned purpose, the present invention provides a camera photo stitching method, comprising: receiving multiple images of the same scene taken by multiple cameras at the same time, the images carrying mechanical coordinate information; using a visual detection system to identify the images, and storing the obtained recognition results in correspondence with the images; separately storing the images taken by each camera and the corresponding recognition results in the form of a time queue; extracting the images and the corresponding recognition results in multiple time queues at the same time, and converting all the recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates; determining the overlapping field of view of the multiple images according to the mechanical coordinate system, and merging them to obtain a stitched picture.

[0005] In one of the embodiments, converting the recognition result to the same coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates includes: determining the pixel coordinates corresponding to the mechanical coordinate information in the captured image and generating a calibration relationship; obtaining the result pixel coordinates corresponding to the recognition result in the captured image; and converting the recognition result in each of the captured images to the same mechanical coordinate system according to the calibration relationship and the result pixel coordinates.

[0006] In one of the embodiments, determining the overlapping field of view of the multiple captured images according to the mechanical coordinate system and merging them to obtain a stitched image includes: fitting the same object identified in the multiple captured images in the mechanical coordinate system to obtain a fitting coordinate range of the object; when it is determined that the fitting coordinate ranges of different objects at least partially overlap, taking the maximum value in the corresponding fitting coordinate ranges in the overlapping area to generate an overlapping stitching range; and merging to obtain a stitched image based on the overlapping stitching range.

[0007] In one embodiment, fitting the same object identified in multiple captured images in the mechanical coordinate system to obtain a fitting coordinate range of the object includes: adjusting the captured images according to the calibration relationship to obtain an adjusted image; extracting the coordinate values ​​of the same object identified on a specific plane according to the adjusted image, and generating a plane coordinate range of the same object on the specific plane, which is the fitting coordinate range of the object.

[0008] In one of the embodiments, the method further includes: outputting the fitting coordinate range of the identified object to a clearing module so that the clearing module can clear the specific object.

[0009] A camera photo stitching device comprises: an image receiving module, used for receiving multiple images of the same scene taken by multiple cameras at the same time, wherein the images carry mechanical coordinate information; an identification module, used for identifying the images by using a visual detection system, and storing the obtained identification results in correspondence with the images; a storage module, used for separately storing the images taken by each camera and the corresponding identification results in the form of a time queue; an extraction module, used for extracting the images and the corresponding identification results in the multiple time queues at the same time, and converting all the identification results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates; and a stitching module, used for determining the overlapping fields of view of the multiple images according to the mechanical coordinate system, and merging them to obtain a stitched image.

[0010] In one embodiment, the extraction module includes: a calibration unit, used to determine the pixel coordinates corresponding to the mechanical coordinate information in the captured image and generate a calibration relationship; a coordinate acquisition unit, used to obtain the result pixel coordinates corresponding to the recognition result in the captured image; and a coordinate conversion unit, used to convert the recognition results in each of the captured images into the same coordinate system according to the calibration relationship and the result pixel coordinates.

[0011] In one embodiment, the stitching module includes: a fitting unit, used to fit the same object identified in multiple captured images in the mechanical coordinate system to obtain a fitting coordinate range of the object; a screening unit, used to take the maximum value of the corresponding fitting coordinate range in the overlapping area to generate an overlapping stitching range when it is determined that the fitting coordinate ranges of different objects at least partially overlap; and a stitching unit, used to obtain a stitched image based on the overlapping stitching range.

[0012] A computer device comprises a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0013] A computer-readable storage medium stores a computer program, which implements the steps of the above method when executed by a processor.

[0014] Compared with the prior art, the advantages of the present invention are: multiple images taken by multiple cameras at the same time are converted into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates, and then the recognition results are merged and spliced. The whole process only processes the detection results, and no cropping or splicing operations are performed on the images themselves, which will not cause damage or distortion to the image quality and affect the accuracy of visual detection; and because the images are lossless, they can also be convenient for subsequent re-inspection and verification. In addition, the results of the overlapping parts of the field of view under the splicing of multiple cameras are processed through the mechanical coordinate system, and the results will not be partially repeated due to the overlap of the field of view, so that the subsequent clearing process will not have unnecessary actions that affect the actual performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 is a schematic flow chart of a camera photo stitching method in one embodiment of the present invention;

[0017] Figure 2is a schematic diagram of fitting the same object identified in multiple captured images into a mechanical coordinate system in one embodiment of the present invention;

[0018] Figure 3 is a spliced ​​picture obtained by merging the overlapping fields of view in one embodiment of the present invention;

[0019] Figure 4 is a structural block diagram of a camera photo stitching device in one embodiment of the present invention;

[0020] Figure 5 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION

[0021] The embodiments of the present application are described in detail below with reference to the accompanying drawings.

[0022] The following describes the implementation methods of the present application through specific examples, and those skilled in the art can easily understand other advantages and effects of the present application from the contents disclosed in this specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all the embodiments. The present application can also be implemented or applied through other different specific implementation methods, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present application. It should be noted that, in the absence of conflict, the following embodiments and the features in the embodiments can be combined with each other. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without making creative work belong to the scope of protection of the present application.

[0023] It should be noted that various aspects of the embodiments within the scope of protection of the present invention are described below. It should be apparent that the aspects described herein can be embodied in a wide variety of forms, and any specific structure and / or function described herein is merely illustrative. Based on the present application, it should be understood by those skilled in the art that an aspect described herein can be implemented independently of any other aspect, and two or more of these aspects can be combined in various ways. For example, any number and aspect described herein can be used to implement the device and / or practice the method. In addition, other structures and / or functionalities other than one or more of the aspects described herein can be used to implement this device and / or practice this method.

[0024] It should also be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present application. The drawings only show components related to the present application rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0025] Additionally, in the following description, specific details are provided to facilitate a thorough understanding of the examples. However, it will be understood by those skilled in the art that the aspects described may be practiced without these specific details.

[0026] like Figure 1 As shown, the embodiment of the present application provides a camera photo stitching method, which can be applied to a server or a terminal. The terminal can be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable smart devices. The server can be implemented as an independent server or a server cluster composed of multiple servers. Taking the application of the method to a server as an example, the method includes the following steps:

[0027] Step 101 : receiving a plurality of images of the same scene taken by a plurality of cameras simultaneously, wherein the images carry mechanical coordinate information.

[0028] The server can receive multiple images of the same scene taken by multiple cameras at the same time in real time, or passively generate image retrieval instructions according to the callback function and obtain the captured images according to the image retrieval instructions. The captured images carry mechanical coordinate information. A reference object with a clear scale is set in the scene. The reference object can be a caliper, a ruler or a reference object of a specific length set at a specific position. When the camera shoots the scene, the reference object will also be captured in the captured image. Therefore, according to the length of the reference object and the specific position determined in advance, the coordinate information of other objects in the captured image can be clearly determined, so the captured image carries mechanical coordinate information. The reference object is set on one side of the automated assembly line. Multiple cameras simultaneously capture the reference object and the object moving on the automated assembly line on the same side. The object can be a product to be processed in the subsequent process or debris that needs to be removed.

[0029] Step 102: Use a visual inspection system to identify the captured image, and store the obtained recognition result in correspondence with the captured image.

[0030] The server uses a visual inspection system to identify the captured image and stores the obtained recognition results in correspondence with the captured image. The visual inspection system can be an AI visual inspection system or a visual inspection system obtained by machine learning training based on items on the production line. The server uses a visual inspection system to identify the captured image and stores the obtained recognition results in correspondence with the captured image. Each captured image can identify at least one recognition result, and sometimes multiple recognition results.

[0031] Step 103 , storing the images captured by each camera and the corresponding recognition results separately in the form of a time queue.

[0032] The server stores the captured images and corresponding recognition results of each camera in the form of a time queue. For example, in the time queue of each camera, the captured images, the shooting time, and at least one recognition result are stored.

[0033] Step 104 , extracting the captured images and corresponding recognition results in multiple time queues at the same time, and converting all recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates.

[0034] The server extracts the captured images and corresponding recognition results from multiple time queues at the same time, and converts all recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates. The server can extract different captured images of the same scene captured by different cameras through the time information in the time queue. The server determines the different captured images captured by different cameras at the same time information as the same group of images to be spliced, and then splices the group of images to be spliced ​​to obtain a spliced ​​picture.

[0035] The server can determine the calibration relationship between pixel coordinates and mechanical coordinates based on the reference object or mechanical coordinate information in the recognition result, and then convert all recognition results into a virtual mechanical coordinate system based on the calibration relationship. The mechanical coordinate system is fitted by the server, but the position and size of each object in the recognition result in the mechanical coordinate system are completely consistent with the actual coordinates of each object in the automated assembly line.

[0036] Step 105 , determining the overlapping fields of view of the multiple captured images according to the mechanical coordinate system, and merging them to obtain a stitched image.

[0037] The server determines the overlapping fields of view of multiple captured images according to the mechanical coordinate system and merges them to obtain a stitched image. The server can determine the size data of other objects in the recognition results according to the reference object or mechanical coordinate information in each image to be stitched, and then stitch the images to be stitched based on the reference object or mechanical coordinate information. When there are overlapping fields of view in the captured images, the server will optimize the overlapping fields of view and finally obtain a stitched image.

[0038] The above method converts all recognition results of multiple images taken by multiple cameras at the same time into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates, and then merges and splices the recognition results. The entire process only processes the detection results, and no cropping or splicing operations are performed on the images themselves, which will not cause damage or distortion to the image quality and affect the accuracy of visual detection; and because the images are lossless, they can also be easily re-inspected and verified later. In addition, the results of the overlapping parts of the field of view under multi-camera splicing are processed through the mechanical coordinate system, so that there will be no duplication of the results due to the overlap of the field of view, and the subsequent clearing process will not cause unnecessary actions that affect the actual performance.

[0039] In one embodiment, the recognition result is converted into the same coordinate system by calibrating the relationship between the pixel coordinates and the mechanical coordinates, including the following steps:

[0040] Determine the pixel coordinates corresponding to the mechanical coordinate information in the captured image and generate a calibration relationship;

[0041] Obtain the pixel coordinates of the recognition result in the captured image;

[0042] The recognition results in each captured image are converted into the same mechanical coordinate system according to the calibration relationship and the resulting pixel coordinates.

[0043] The server determines the pixel coordinates corresponding to the mechanical coordinate information in the captured image and generates a calibration relationship. For example, the server may first determine the pixel coordinates of the reference object corresponding to the mechanical coordinate information in the captured image. Since the mechanical coordinate information is known, the calibration relationship between the pixel coordinates and the mechanical coordinates may be determined.

[0044] The server obtains the pixel coordinates of the recognition result in the captured image. For example, the plane where the captured image is located is the xy plane, and the direction perpendicular to the xy plane is the z axis. Figure 2 Two objects 1 and 2 are shown in the figure. Object 1 and object 2 are offset in the z-axis direction, so they overlap in the xy plane. The pixel coordinates of object 1 in the xy plane are (x0b, y0b) and (x1b, y1b), and the pixel coordinates of object 2 in the xy plane are (x0a, y0a) and (x1a, y1a).

[0045] The server converts the recognition results in each captured image at the same time into the same mechanical coordinate system according to the calibration relationship and the result pixel coordinates.

[0046] The above method only processes the recognition results, and does not perform any cropping or splicing operations on the image itself, which will not cause damage or distortion to the image quality, nor will it affect the accuracy of visual detection.

[0047] In one embodiment, the overlapping fields of view of multiple captured images are determined according to a mechanical coordinate system, and then merged to obtain a stitched image, including the following steps:

[0048] Fitting the same object identified in multiple captured images into the mechanical coordinate system to obtain a fitting coordinate range of the object;

[0049] When it is determined that the fitting coordinate ranges of different objects at least partially overlap, taking the maximum value of the corresponding fitting coordinate ranges in the overlapping area to generate the overlapping splicing range;

[0050] Based on the overlapping stitching range, the stitching images are merged.

[0051] The server fits the same object identified in multiple captured images into the mechanical coordinate system to obtain the fitting coordinate range of the object. For example, when Figure 2 After fitting the pattern of the same object identified in multiple captured images, the pixel coordinates (x0b, y0b) and (x1b, y1b) of object 1 in the xy plane are the fitting coordinate range of object 1; the pixel coordinates (x0a, y0a) and (x1a, y1a) of object 2 in the xy plane are the fitting coordinate range of object 2.

[0052] At this time, the fitting coordinate ranges of object 1 and object 2 partially overlap in (x0b, y0a) and (x1b, y1a). The server takes the maximum value of the corresponding fitting coordinate range in the overlapping area to generate the overlapping stitching range. Therefore, the overlapping stitching ranges (x0a, y0a) and (x1a, y1a) are obtained. In this embodiment, since there are only two objects, this range is exactly the same as the fitting coordinate range of object 2; when there are more objects, there may be differences between the overlapping stitching range and the fitting coordinate range of the object.

[0053] The server merges the overlapping stitching ranges to obtain a stitched image. The server also retains the fitting coordinate range that is not included in the overlapping stitching range in the recognition result, and merges it with the overlapping stitching range to form a rectangular combination stacked in the y-axis direction. For example, the server also Figure 3 The gray area of ​​object 1 that is not included in the overlapping stitching range is retained and merged with the overlapping stitching range (white area) to form a rectangular combination stacked in the y-axis direction.

[0054] In one embodiment, fitting the same object identified in multiple captured images into a mechanical coordinate system to obtain a fitting coordinate range of the object includes:

[0055] Adjusting the captured image according to the calibration relationship to obtain an adjusted image;

[0056] The coordinate values ​​of the same object on a specific plane are extracted and identified according to the adjustment image, and a plane coordinate range of the same object on the specific plane is generated, and the plane coordinate range is a fitting coordinate range of the object.

[0057] The server adjusts the captured image according to the calibration relationship to obtain an adjusted image. Different cameras have different calibration relationships due to their different angles with the automated assembly line, and need to be adjusted one by one. However, since different cameras use the same mechanical coordinate information, the adjusted images of the same object in different captured images are consistent. The server extracts the coordinate values ​​of the same object identified on a specific plane based on the adjustment image, and generates a plane coordinate range of the same object on a specific plane, which is the fitting coordinate range of the object.

[0058] In one embodiment, the method further includes: outputting the fitting coordinate range of the identified object to the cleaning module so that the cleaning module can clean the specific object. The cleaning module can be a high-pressure valve group. For example, the automated assembly line is a conveyor belt moving in the y direction, and a high-pressure valve group arranged in the x direction is arranged at the end for blowing. The high-pressure valve group receives the fitting coordinate range of the identified object. When it is necessary to blow certain objects in the spliced ​​image, the high-pressure valve group can blow these objects according to the overlapping splicing range and / or fitting coordinate range on the spliced ​​image.

[0059] In one embodiment, Figure 4 As shown, a camera photo stitching device is provided, including a picture receiving module 401, a recognition module 402, a storage module 403, an extraction module 404 and a stitching module 405.

[0060] The image receiving module 401 is used to receive multiple images of the same scene taken by multiple cameras at the same time, and the captured images carry mechanical coordinate information.

[0061] The recognition module 402 is used to recognize the captured image using a visual detection system, and store the obtained recognition result in correspondence with the captured image.

[0062] The storage module 403 is used to store the images taken by each camera and the corresponding recognition results separately in the form of a time queue.

[0063] The extraction module 404 is used to extract the captured images and corresponding recognition results in multiple time queues at the same time, and transform all the recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates.

[0064] The stitching module 405 is used to determine the overlapping fields of view of the multiple captured images according to the mechanical coordinate system, and merge them to obtain a stitched image.

[0065] In one embodiment, the extraction module includes:

[0066] The calibration unit is used to determine the pixel coordinates corresponding to the mechanical coordinate information in the captured image and generate a calibration relationship.

[0067] The coordinate acquisition unit is used to acquire the result pixel coordinates corresponding to the recognition result in the captured image.

[0068] The coordinate conversion unit is used to convert the recognition results in each captured image into the same mechanical coordinate system according to the calibration relationship and the result pixel coordinates.

[0069] In one embodiment, the splicing module includes:

[0070] The fitting unit is used to fit the same object identified in multiple captured images in the mechanical coordinate system to obtain a fitting coordinate range of the object.

[0071] The screening unit is used to take the maximum value of the corresponding fitting coordinate ranges in the overlapping area to generate the overlapping splicing range when it is determined that the fitting coordinate ranges of different objects at least partially overlap.

[0072] The stitching unit is used to obtain a stitched image based on the overlapping stitching range.

[0073] In one embodiment, the splicing module includes:

[0074] The adjustment unit is used to adjust the captured image according to the calibration relationship to obtain an adjusted image.

[0075] The fitting unit is used to extract the coordinate values ​​of the same object on a specific plane according to the adjusted image, and generate a plane coordinate range of the same object on the specific plane, which is the fitting coordinate range of the object.

[0076] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 5 As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. 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 database of the computer device is used to store time queue data, etc. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a camera photo stitching method is implemented.

[0077] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor implements the following steps when executing the computer program: receiving multiple images of the same scene taken by multiple cameras at the same time, the images carrying mechanical coordinate information; using a visual detection system to recognize the images, and storing the obtained recognition results in correspondence with the images; separately storing the images taken by each camera and the corresponding recognition results in the form of a time queue; extracting the images and the corresponding recognition results in multiple time queues at the same time, and converting all the recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates; determining overlapping fields of view for the multiple images according to the mechanical coordinate system, and merging them to obtain a spliced ​​image.

[0078] 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 following steps are implemented: receiving multiple images of the same scene taken by multiple cameras at the same time, the images carrying mechanical coordinate information; using a visual detection system to recognize the images, and storing the obtained recognition results in correspondence with the images; storing the images taken by each camera and the corresponding recognition results separately in the form of a time queue; extracting the images and the corresponding recognition results in multiple time queues at the same time, and converting all the recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates; determining the overlapping fields of view of the multiple images according to the mechanical coordinate system, and merging them to obtain a stitched image.

[0079] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by any technician familiar with the technical field within the technical scope disclosed in the present application should be included in the protection scope of the present application.

Claims

1. A camera photo stitching method, characterized in that: include: Receiving multiple images of the same scene taken by multiple cameras simultaneously, wherein the images carry mechanical coordinate information; Using a visual detection system to identify the captured image, and storing the obtained recognition result in correspondence with the captured image; The images captured by each camera and the corresponding recognition results are stored separately in the form of a time queue; Extracting the captured images and corresponding recognition results in the multiple time queues at the same time, and converting all the recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates; The overlapping fields of view of the plurality of captured images are determined according to the mechanical coordinate system, and the images are merged to obtain a spliced ​​image.

2. The camera photo stitching method according to claim 1, characterized in that: The converting the recognition result into the same coordinate system through the calibration relationship between the pixel coordinates and the mechanical coordinates includes: Determine pixel coordinates corresponding to the mechanical coordinate information in the captured image, and generate a calibration relationship; Obtaining the result pixel coordinates corresponding to the recognition result in the captured image; The recognition results in each of the captured images are converted into the same mechanical coordinate system according to the calibration relationship and the result pixel coordinates.

3. The camera photo stitching method according to claim 1, characterized in that: Determining the overlapping fields of view of the plurality of captured images according to the mechanical coordinate system and merging them to obtain a spliced ​​image includes: Fitting the same object identified in the plurality of captured images into the mechanical coordinate system to obtain a fitting coordinate range of the object; When it is determined that the fitting coordinate ranges of different objects at least partially overlap, taking the maximum value of the fitting coordinate ranges corresponding to the overlapping area to generate the overlapping splicing range; Based on the overlapping stitching ranges, a stitching picture is obtained by merging.

4. The camera photo stitching method according to claim 3, characterized in that: The step of fitting the same object identified in the plurality of captured images into the mechanical coordinate system to obtain a fitting coordinate range of the object includes: Adjusting the captured image according to the calibration relationship to obtain an adjusted image; The coordinate values ​​of the same object on a specific plane are extracted and identified according to the adjustment image, and a plane coordinate range of the same object on the specific plane is generated, and the plane coordinate range is a fitting coordinate range of the object.

5. The camera photo stitching method according to claim 4, characterized in that: Also includes: The identified fitting coordinate range of the object is output to a clearing module so that the clearing module can clear the specific object.

6. A camera photo stitching device, characterized in that: include: An image receiving module, used to receive multiple images of the same scene taken by multiple cameras at the same time, wherein the images carry mechanical coordinate information; A recognition module, used for recognizing the captured image by using a visual detection system, and storing the obtained recognition result in correspondence with the captured image; A storage module, used to store the images taken by each camera and the corresponding recognition results separately in the form of a time queue; An extraction module, used to extract the captured images and corresponding recognition results in the multiple time queues at the same time, and transform all the recognition results into a virtual mechanical coordinate system through the calibration relationship between pixel coordinates and mechanical coordinates; The stitching module is used to determine the overlapping fields of view of the multiple captured images according to the mechanical coordinate system, and merge them to obtain a stitched picture.

7. The camera photo stitching device according to claim 6, characterized in that: The extraction module includes: A calibration unit, used to determine the pixel coordinates corresponding to the mechanical coordinate information in the captured image, and generate a calibration relationship; A coordinate acquisition unit, used to acquire the result pixel coordinates corresponding to the recognition result in the captured image; A coordinate conversion unit is used to convert the recognition results in each of the captured images into the same mechanical coordinate system according to the calibration relationship and the result pixel coordinates.

8. The camera photo stitching device according to claim 6, characterized in that: The splicing module includes: A fitting unit, used for fitting the same object identified in the plurality of captured images into the mechanical coordinate system to obtain a fitting coordinate range of the object; A screening unit, configured to, when it is determined that the fitting coordinate ranges of different objects at least partially overlap, take the maximum value of the fitting coordinate ranges corresponding to the overlapping area to generate an overlapping splicing range; The stitching unit is used to obtain a stitched picture based on the overlapping stitching range.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 5 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 5 are implemented.

Images