CN114390270B - Real-time intelligent site panorama survey method, device and electronic equipment - Google Patents

Abstract

The embodiment of the invention provides a real-time intelligent site panoramic investigation method, a device and electronic equipment, wherein the method comprises the following steps: receiving a survey task, wherein the survey task comprises machine room internal survey modeling, pole surrounding environment modeling and pole large-scale environment modeling; acquiring images of the survey targets corresponding to the survey tasks, wherein the acquisition equipment and the acquisition mode for acquiring the images of the survey targets corresponding to the survey tasks are determined based on the natural space conditions of the predetermined survey targets; and carrying out three-dimensional model recovery based on the image, and sending the three-dimensional model as a task result to a survey platform for release. The method, the device and the electronic equipment provided by the embodiment of the invention realize the high efficiency, low labor cost and large information quantity of the site panoramic survey result, and are easy to display and convenient to share.

Description

Real-time intelligent site panorama survey method, device and electronic equipment

technical field

The invention relates to the technical field of site survey, in particular to a real-time intelligent site panorama survey method, device and electronic equipment.

Background technique

The communication site includes the computer room and the pole. To conduct a panoramic survey of the site, it is necessary to collect the information of the indoor computer room and the surrounding environment of the outdoor pole. The surrounding environment of the pole holding is divided into the surrounding environment of the pole holding (that is, the small area, close to the holding pole) and the large-scale environment of the holding pole (that is, a large area, far away from the holding pole). The small area and the large area are based on the site Scale setting. The existing site survey method needs to measure the size of all facilities and equipment, record all equipment numbers and draw drawings, all of which are done manually, with heavy workload, low efficiency and high difficulty, and cannot visually display the situation of the site, and Drawings only include two-dimensional information, which is not easy to spread and share.

Therefore, how to avoid the trouble of low efficiency and high labor cost caused by manual method for site survey, and the survey results in the form of drawings with little information and inconvenient sharing are still problems to be solved urgently by those skilled in the art.

Contents of the invention

Embodiments of the present invention provide a real-time intelligent site panorama survey method, device and electronic equipment to solve the problems of low efficiency and high labor costs caused by the existing manual method for surveying sites and the lack of information in the survey results in the form of drawings. Problems that are not easy to share.

In the first aspect, the embodiment of the present invention provides a real-time intelligent site panorama survey method, including:

Receive survey tasks, the survey tasks include internal survey modeling of the computer room, modeling of the surrounding environment of the pole, and modeling of the large-scale environment of the pole;

Collecting the image of the survey target corresponding to the survey task, wherein the collection equipment and collection method for collecting the image of the survey task corresponding to the survey target are determined based on the predetermined natural space conditions of the survey target;

The three-dimensional model is restored based on the image, and the three-dimensional model is sent to the survey platform as a task result for release.

Preferably, in this method, also include:

If the survey target is inside the machine room, the natural space conditions are the intensity of natural light and the richness of equipment textures;

If the survey target is the surrounding environment of the pole, the natural space condition is the rain and fog around the pole;

If the survey target is a large-scale environment with a pole, the natural space condition is the volume of an object in the large-scale environment with a pole.

Preferably, in this method, the acquisition equipment and acquisition method for acquiring the image of the survey target corresponding to the survey task are determined based on the predetermined natural space conditions of the survey target, specifically including:

When the survey task is the internal survey modeling of the computer room,

If the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room;

If the natural light intensity is appropriate and the equipment texture is rich, use the SLAM robot to carry the binocular camera to move and shoot in the computer room;

If any of the above conditions are not met, use the SLAM robot to carry the camera and lidar to move and shoot in the computer room;

When the survey task is modeling the surrounding environment of the pole,

If there is no rain and fog, use the drone to carry the camera and lidar to fly around the pole and take pictures;

When the survey task is modeling a large-scale environment with poles,

If the object is large in the large-scale environment of the pole, use the drone to carry the camera to fly around and shoot in the large-scale environment of the pole;

If the object is small in the large-scale environment of the pole, use a drone equipped with a camera and lidar to fly around and shoot in the large-scale environment of the pole.

Preferably, in the method, the restoration of the three-dimensional model based on the image specifically includes:

Carrying out feature matching of the image based on the SIFT operator to obtain a pair of matching feature points;

The matching feature point pair adopts the RANSAC eight-point method to calculate the basic matrix, and the matching feature point pair that does not satisfy the basic matrix is eliminated;

Then use the CMVS and PMVS algorithms to reconstruct the dense point cloud of the matched feature point pairs after elimination to obtain a 3D model.

In the second aspect, an embodiment of the present invention provides a real-time intelligent site panorama survey device, including:

The receiving unit is used to receive survey tasks, and the survey tasks include survey modeling inside the computer room, modeling of the surrounding environment of the pole, and modeling of the large-scale environment of the pole;

The acquisition unit is configured to acquire the images of the survey targets corresponding to the survey tasks, wherein the acquisition equipment and acquisition method for acquiring the images of the survey tasks corresponding to the survey targets are determined based on the predetermined natural space conditions of the survey targets;

The computing unit is configured to restore the 3D model based on the image, and send the 3D model as a task result to the survey platform for release.

Preferably, the device also includes:

If the survey target is inside the machine room, the natural space conditions are the intensity of natural light and the richness of equipment textures;

If the survey target is the surrounding environment of the pole, the natural space condition is the rain and fog around the pole;

If the survey target is a large-scale environment with a pole, the natural space condition is the volume of an object in the large-scale environment with a pole.

Preferably, in the device, the acquisition equipment and acquisition method for acquiring images corresponding to the survey targets of the survey tasks are determined based on the predetermined natural space conditions of the survey targets, specifically including:

When the survey task is the internal survey modeling of the computer room,

If the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room;

If the natural light intensity is appropriate and the equipment texture is rich, use the SLAM robot to carry the binocular camera to move and shoot in the computer room;

If any of the above conditions are not met, use the SLAM robot to carry the camera and lidar to move and shoot in the computer room;

When the survey task is modeling the surrounding environment of the pole,

If there is no rain and fog, use the drone to carry the camera and lidar to fly around the pole and take pictures;

When the survey task is modeling a large-scale environment with poles,

If the object is large in the large-scale environment of the pole, use the drone to carry the camera to fly around and shoot in the large-scale environment of the pole;

If the object is small in the large-scale environment of the pole, use a drone equipped with a camera and lidar to fly around and shoot in the large-scale environment of the pole.

Preferably, in the device, the restoration of the three-dimensional model based on the image specifically includes:

Carrying out feature matching of the image based on the SIFT operator to obtain a pair of matching feature points;

The matching feature point pair adopts the RANSAC eight-point method to calculate the basic matrix, and the matching feature point pair that does not satisfy the basic matrix is eliminated;

Then use the CMVS and PMVS algorithms to reconstruct the dense point cloud of the matched feature point pairs after elimination to obtain a 3D model.

In a third aspect, an embodiment of the present invention provides an electronic device, including a memory, a processor, and a computer program stored in the memory and operable on the processor. When the processor executes the program, the computer program described in the first aspect is implemented. The steps of the real-time intelligent site panorama survey method provided.

In a fourth aspect, an embodiment of the present invention provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the real-time intelligent site panorama survey method as provided in the first aspect is implemented. step.

The method, device, and electronic equipment provided by the embodiments of the present invention receive survey tasks, and the survey tasks include the internal survey modeling of the computer room, the surrounding environment modeling of the pole holding, and the large-scale environment modeling of the pole holding; the survey tasks corresponding to the survey tasks are collected The image of the target, wherein the acquisition equipment and acquisition method for acquiring the image of the survey target corresponding to the survey task are determined based on the predetermined natural space conditions of the survey target; the three-dimensional model is restored based on the image, and the The 3D model is sent to the survey platform as a task result for publication. In this way, according to the different types of survey tasks and the different natural space conditions of the survey targets corresponding to the survey tasks, the appropriate image acquisition equipment and methods are selected, and finally based on the collected images for identification or recognition results obtained by 3D modeling or 3D The modeling is returned as the result of the survey task, which realizes the automation of the survey process and the automatic return and sharing of the results, and the 3D model as the result of the task completion can also provide more information for easy display. Therefore, the real-time intelligent site panorama survey method, device, and electronic equipment provided by the embodiments of the present invention realize high efficiency, low labor cost, large amount of survey results, and easy display and sharing of site panorama survey.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description These are some embodiments of the present invention. Those skilled in the art can also obtain other drawings based on these drawings without creative work.

FIG. 1 is a schematic flow diagram of a real-time intelligent site panorama survey method provided by an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a real-time intelligent site panoramic survey device provided by an embodiment of the present invention;

FIG. 3 is a schematic structural diagram of an electronic device provided by an embodiment of the present invention.

Detailed ways

In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

The existing method of surveying sites by manual surveying and drawing generally has the problems of low efficiency, high labor costs, less information in the survey results in the form of drawings, and inconvenient sharing. In this regard, the embodiment of the present invention provides a method for ensuring the consistency of a telecommunications service contract based on blockchain, and the subject of execution of the method is an authoritative service device. Fig. 1 is a schematic flow chart of a real-time intelligent site panorama survey method provided by an embodiment of the present invention. As shown in Fig. 1, the method includes:

In step 110, a survey task is received, and the survey task includes survey and modeling inside the computer room, modeling of the surrounding environment of the pole, and modeling of the large-scale environment of the pole.

Specifically, the survey task is received, and then the survey task category and the corresponding survey target are determined. Among them, the survey task category is divided into three categories: the internal survey modeling of the computer room, the surrounding environment modeling of the pole holding, and the large-scale environment modeling of the pole holding. The survey targets corresponding to the three types of survey tasks are: the interior of the computer room, the surrounding environment of the pole, and the large-scale environment of the pole. All the above tasks need to return a 3D model, that is, perform 3D modeling on the survey target, and then return the 3D model as the survey result. After receiving the survey task, the task type and the corresponding survey target can be obtained by analyzing the task.

Step 120, collecting the images of the survey targets corresponding to the survey tasks, wherein the acquisition equipment and acquisition method for acquiring the images of the survey tasks corresponding to the survey targets are determined based on the predetermined natural space conditions of the survey targets.

Specifically, since the site panorama survey method provided by the embodiment of the present invention is fully automatic, there is no need to manually measure and draw survey targets, and the collected images are uniformly used to process the collected graphics to obtain corresponding 3D models. Since the survey tasks include the internal survey modeling of the computer room, the surrounding environment modeling of the poles, and the large-scale environment modeling of the poles, different survey tasks and different environmental conditions of the survey targets need to choose suitable image acquisition equipment and methods . Especially for site environment modeling, because it is necessary to restore the 3D model based on the image, and the 3D model requires rich image information, and has high requirements for the collected images. Therefore, the natural light intensity inside the computer room and the richness of equipment texture will affect the internal survey of the computer room. The equipment and acquisition method for collecting images inside the computer room during modeling, and the spatial environment conditions around the pole and the large-scale environment of the pole (such as weather conditions and the size of objects in the environment) will also affect the surrounding environment of the pole and the size of the pole. Modeling of scoped environments. For the interior of the computer room with different light conditions, different texture conditions, and outdoor environments with different weather conditions, you can choose the corresponding image acquisition technology. For example, the structured light depth camera is more suitable for close-range shooting where the light is normal and the object equipment lacks texture. The binocular camera It is more suitable for close-range shooting with rich textures of objects. Camera plus lidar is suitable for any situation, but the cost is higher. Outdoor image collection is usually carried by drones with cameras to fly around and then shoot. Therefore, the use of different equipment and image acquisition methods is more suitable for the survey target, and avoids the trouble of poor image shooting effect caused by the use of unified acquisition equipment and the need for powerful image processing algorithms to process images of different quality.

Step 130, restore the 3D model based on the image, and send the 3D model as a task result to the survey platform for release.

Specifically, for the above-mentioned survey tasks, after determining the image acquisition equipment and image acquisition method suitable for the survey target, the image acquisition is carried out, and the subsequent process of reconstructing the 3D model of the acquired image is usually to obtain a point cloud based on the depth data. Data, then point cloud registration and fusion, and finally the process of generating the surface, and finally send the generated 3D model as the task result to the viewing platform for release. Specifically, set up a server on the intranet, upload the 3D model to the server, and then those who need to view the 3D model can log in to the server website to view without downloading, for example: use Register360 and TrueView Enterprise for automatic stitching, distance measurement and publishing results. When the person who needs to view the 3D model logs in to the server website and selects the model to be viewed, he can use the mouse wheel to zoom in and out, move the mouse to adjust the viewing angle, click the collection point to jump, select the distance measurement function from the menu, and then click the remove button from the model. Any two points outside the sky can measure the distance and the X-axis, Y-axis and Z-axis distance.

The method provided by the embodiment of the present invention receives a survey task, and the survey task includes survey modeling inside the computer room, modeling of the surrounding environment of the pole, and modeling of the large-scale environment of the pole; collecting images of the survey target corresponding to the survey task, wherein The acquisition equipment and acquisition method for acquiring images of the survey targets corresponding to the survey tasks are determined based on the predetermined natural space conditions of the survey targets; the 3D model is restored based on the images, and the 3D model is used as the task result Send to survey platform for publication. In this way, according to the different types of survey tasks and the different natural space conditions of the survey targets corresponding to the survey tasks, the appropriate image acquisition equipment and methods are selected, and finally based on the collected images for identification or recognition results obtained by 3D modeling or 3D The modeling is returned as the result of the survey task, which realizes the automation of the survey process and the automatic return and sharing of the results, and the 3D model as the result of the task completion can also provide more information for easy display. Therefore, the real-time intelligent site panorama survey method provided by the embodiment of the present invention realizes high efficiency of site panorama survey, low labor cost, large amount of survey results and easy display and sharing.

Based on the foregoing embodiments, the method also includes:

If the survey target is inside the machine room, the natural space conditions are the intensity of natural light and the richness of equipment textures;

If the survey target is the surrounding environment of the pole, the natural space condition is the rain and fog around the pole;

If the survey target is a large-scale environment with a pole, the natural space condition is the volume of an object in the large-scale environment with a pole.

Specifically, for different survey targets, the natural space conditions that will affect the image acquisition equipment and acquisition methods are also different. When the survey target is inside the computer room, the impact of natural light intensity and equipment texture richness on image acquisition is mainly considered. When the survey target is the surrounding environment of the pole, the influence of the weather conditions around the pole, mainly rain and fog, is mainly considered. When surveying When the target is a large-scale environment with poles, the influence of the volume of objects in the large-scale environment with poles is mainly considered. The above-mentioned natural space conditions are pre-determined and stored state data that can be directly obtained.

Based on any of the above-mentioned embodiments, in this method, the acquisition equipment and acquisition method for acquiring the image of the survey target corresponding to the survey task are determined based on the predetermined natural space conditions of the survey target, specifically including:

Specifically include:

When the survey task is the internal survey modeling of the computer room,

If the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room;

If the natural light intensity is appropriate and the equipment texture is rich, use the SLAM robot to carry the binocular camera to move and shoot in the computer room;

If any of the above conditions are not met, use the SLAM robot to carry the camera and lidar to move and shoot in the computer room;

When the survey task is modeling the surrounding environment of the pole,

If there is no rain and fog, use the drone to carry the camera and lidar to fly around the pole and take pictures;

When the survey task is modeling a large-scale environment with poles,

If the object is large in the large-scale environment of the pole, use the drone to carry the camera to fly around and shoot in the large-scale environment of the pole;

If the object is small in the large-scale environment of the pole, use a drone equipped with a camera and lidar to fly around and shoot in the large-scale environment of the pole.

Specifically, when the survey task is modeling the internal survey of the computer room, if the natural light intensity is low and the equipment texture is lacking, the SLAM robot is used to carry the structured light depth camera to move and shoot in the computer room. Among them, low natural light intensity and lack of equipment texture are pre-determined natural space conditions that can be directly obtained, that is, the natural light intensity and equipment in the computer room are determined in advance through the detection and determination of the intensity of natural light and the richness of equipment texture. The richness of the texture, the above detection and determination methods are various and common, and will not be described in detail here. The SLAM robot carrying a structured light depth camera to shoot in the computer room refers to a robot that uses the SLAM (Simultaneous Localization And Mapping, simultaneous positioning and mapping) algorithm to carry a structured light depth camera to shoot in the computer room. The SLAM algorithm was first applied to robots. field, its goal is to construct a map of the surrounding environment based on sensor data without any prior knowledge, and at the same time infer its own position based on this map, assuming that the robot carries a sensor (camera) to move in an unknown environment, and the movement time t = 1, 2, ..., k, corresponding to the position x1, x2, ..., xk of the robot. Assuming that the map consists of n landmarks y1, y2, ..., yn, the camera will measure a part of the landmark data at each moment, and solve the positioning problem (estimating x) and mapping problem (estimating y) through motion measurement and sensor readings . When the robot performs motion shooting in the computer room, it will use the preset route and the length of stay at the collection point for motion shooting. Usually, the collection point is usually on the ground between the cabinets, which needs to be close to important equipment. Every two The distance between the collection points cannot exceed the measurement distance of the camera. The measurement range of the cameras at all collection points can cover the entire computer room. The more collection points, the more accurate the 3D model will be. According to the measurement distance of the camera and the computer room Depending on the complexity of the environment, each point generally stays within one minute, and the entire collection process generally takes within ten minutes. When the survey task is the internal survey and modeling of the computer room, if the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room, and use the preset route and the length of stay at the collection point to For sports shooting, usually, the selection of collection points is generally on the ground between the cabinets, which need to be close to important equipment. The distance between every two collection points cannot exceed the measurement distance of the camera, and the measurement range of the camera at all collection points is added up. It can cover the entire computer room, and the more the number of collection points, the more accurate the 3D model will be. According to the measurement distance of the camera and the complexity of the computer room environment, each point usually stays for several minutes, and the entire collection process generally takes half an hour to one hour. Hour. When the survey task is the internal survey modeling of the computer room, it is not the natural space conditions of low natural light intensity and lack of equipment texture, nor the natural space conditions of appropriate natural light intensity and rich equipment texture, so the SLAM robot is used to carry the camera and laser Radar shot in motion in the engine room. LiDAR is more suitable as a depth information acquisition device in the case of a bad natural environment. It will use the preset route and the length of stay at the collection point for motion shooting. Usually, the selection of the collection point is generally on the ground between the cabinets. , needs to be close to important equipment, and the distance between every two collection points cannot exceed the measurement distance of the camera and lidar. The measurement range of the camera and lidar at all collection points can cover the entire computer room. The more collection points, the generated The more accurate the 3D model will be, according to the measurement distance of the camera and lidar and the complexity of the computer room environment, generally each point stays within one minute, and the entire acquisition process generally takes within ten minutes. When the survey task is to model the surrounding environment of the pole, if there is no rain and fog, use a drone equipped with a camera and a laser radar to fly around the pole and take pictures; if there is rain or fog, the weather is bad, and image collection cannot be performed , so the survey task cannot be carried out, and we can only wait until the weather returns to normal before continuing to collect. When the survey task is to model the large-scale environment of the pole, if the object in the large-scale environment of the pole is large, use the drone equipped with a camera to fly around and shoot in the large-scale environment of the pole. Objects can be recognized even when shooting with an ordinary camera, especially under the condition that the modeling accuracy is not high. The drone equipped with the camera flies around and shoots in the pole environment, and the range of collection is tens of meters in the air according to the needs Fly until all parts of the environment are covered, and it is necessary to fly around multiple times; if the object is small in a large-scale environment with a pole, use a drone equipped with a camera and a lidar to fly around and shoot in a large-scale environment with a pole. The small size of the object requires a camera and a laser radar. Especially under the condition of high modeling accuracy, a drone equipped with a camera and a laser radar flies around and shoots in a pole-holding environment. The range of collection is tens of Meters of airflight until all parts of the environment are covered, and multiple rounds are required.

Based on any of the above embodiments, in the method, the restoration of the 3D model based on the image specifically includes:

Carrying out feature matching of the image based on the SIFT operator to obtain a pair of matching feature points;

The matching feature point pair adopts the RANSAC eight-point method to calculate the basic matrix, and the matching feature point pair that does not satisfy the basic matrix is eliminated;

Then use the CMVS and PMVS algorithms to reconstruct the dense point cloud of the matched feature point pairs after elimination to obtain a 3D model.

Specifically, image feature extraction and matching are performed first: the SIFT (Scale Invariant Feature Transform) operator is mainly used for extraction. The SIFT feature is based on some local appearance points of interest on the object and has nothing to do with the size and rotation of the image. It has scale and rotation invariance, high tolerance to light, noise, and micro viewing angle changes, and strong robustness. It is suitable for extracting various image feature point information of scale transformation and rotation angle, and its accuracy is strong; secondly, Perform sparse point cloud reconstruction: In order to perform reconstruction, the wrong matches in the matched points must first be removed. A feature point in the image may match multiple feature points, and there will be a many-to-one situation. In fact, there should be one-to-one correspondence between feature points. In order to solve this many-to-one situation, sampling consistency can be used for matching points Algorithm The RANSAC eight-point method calculates the fundamental matrix, and eliminates the matching pairs that do not satisfy the fundamental matrix. After a pair of correct matching points is obtained, the depth information of the corresponding three-dimensional point can be recovered by triangulation, and more images can be added, and The previous images are matched, and the three-dimensional point information is calculated to form a sparse point cloud; finally, the dense point cloud reconstruction is performed: after the sparse point cloud reconstruction, CMVS (clustering multi-view stereo) and PMVS (patch-based multi-view stereo) are generally used. -view stereo) to perform dense point cloud reconstruction and restore more 3D information to better build a 3D model. Among them, CMVS filters out images with a high degree of overlap through image clustering algorithms, reducing the need for subsequent PMVS algorithms to process The number of images, PMVS obtains the final dense point cloud through multi-view matching of small pieces of images.

Based on any of the above-mentioned embodiments, an embodiment of the present invention provides a real-time intelligent site panoramic survey device, and FIG. 2 is a schematic structural diagram of the real-time intelligent site panoramic survey device provided by the embodiment of the present invention. As shown in FIG. 2, the device includes a receiving unit 210, an acquisition unit 220, and a computing unit 230, wherein,

The receiving unit 210 is configured to receive survey tasks, and the survey tasks include survey modeling inside the computer room, modeling of the surrounding environment of the pole, and modeling of the large-scale environment of the pole;

The collection unit 220 is configured to collect the images of the survey targets corresponding to the survey tasks, wherein the collection equipment and collection method for collecting the images of the survey tasks corresponding to the survey targets are based on the predetermined natural space conditions of the survey targets definite;

The calculation unit 230 is configured to restore the 3D model based on the image, and send the 3D model as a task result to the survey platform for release.

The device provided by the embodiment of the present invention receives a survey task, and the survey task includes the internal survey modeling of the computer room, the surrounding environment modeling of the pole holding, and the large-scale environment modeling of the pole holding; the image of the survey target corresponding to the survey task is collected, wherein ,

The acquisition equipment and acquisition method for acquiring images corresponding to the survey targets of the survey task are determined based on the predetermined natural space conditions of the survey targets; the 3D model is restored based on the images, and the 3D model is sent as a task result to the survey platform for publication. In this way, according to the different types of survey tasks and the different natural space conditions of the survey targets corresponding to the survey tasks, the appropriate image acquisition equipment and methods are selected, and finally based on the collected images for identification or recognition results obtained by 3D modeling or 3D The modeling is returned as the result of the survey task, which realizes the automation of the survey process and the automatic return and sharing of the results, and the 3D model as the result of the task completion can also provide more information for easy display. Therefore, the real-time intelligent site panorama survey device provided by the embodiment of the present invention realizes high efficiency of site panorama survey, low labor cost, large amount of survey results and easy display and sharing.

Based on any of the above embodiments, the device further includes:

If the survey target is inside the machine room, the natural space conditions are the intensity of natural light and the richness of equipment textures;

If the survey target is the surrounding environment of the pole, the natural space condition is the rain and fog around the pole;

If the survey target is a large-scale environment with a pole, the natural space condition is the volume of an object in the large-scale environment with a pole.

Based on any of the above-mentioned embodiments, in this device, the acquisition equipment and acquisition method for acquiring the image of the site environment modeling corresponding to the survey target are determined based on the predetermined natural space conditions of the survey target, specifically including:

When the survey task is the internal survey modeling of the computer room,

If the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room;

If the natural light intensity is appropriate and the equipment texture is rich, use the SLAM robot to carry the binocular camera to move and shoot in the computer room;

If any of the above conditions are not met, use the SLAM robot to carry the camera and lidar to move and shoot in the computer room;

When the survey task is modeling the surrounding environment of the pole,

If there is no rain and fog, use the drone to carry the camera and lidar to fly around the pole and take pictures;

When the survey task is modeling a large-scale environment with poles,

If the object is large in the large-scale environment of the pole, use the drone to carry the camera to fly around and shoot in the large-scale environment of the pole;

If the object is small in the large-scale environment of the pole, use a drone equipped with a camera and lidar to fly around and shoot in the large-scale environment of the pole.

Based on any of the above-mentioned embodiments, in this device, the acquisition equipment and acquisition method for acquiring the image of the survey target corresponding to the survey task are determined based on the predetermined natural space conditions of the survey target, specifically including:

Carrying out feature matching of the image based on the SIFT operator to obtain a pair of matching feature points;

The matching feature point pair adopts the RANSAC eight-point method to calculate the basic matrix, and the matching feature point pair that does not satisfy the basic matrix is eliminated;

Then use the CMVS and PMVS algorithms to reconstruct the dense point cloud of the matched feature point pairs after elimination to obtain a 3D model.

FIG. 3 is a schematic diagram of the physical structure of the electronic device provided by the embodiment of the present invention. As shown in FIG. The bus 304 , wherein the processor 301 , the communication interface 302 , and the memory 303 communicate with each other through the communication bus 304 . The processor 301 can invoke a computer program stored in the memory 303 and runnable on the processor 301 to execute the real-time intelligent site panorama survey method provided by the above-mentioned embodiments, for example, including: receiving a survey task, the survey task including the computer room Internal survey modeling, surrounding environment modeling of the pole holding and large-scale environment modeling of the pole holding; collecting the image of the survey target corresponding to the survey task, wherein, the collection equipment and collection method for collecting the image of the survey task corresponding to the survey target are Determined based on the predetermined natural space conditions of the survey target; restore the 3D model based on the image, and send the 3D model as a task result to the survey platform for release.

In addition, the above-mentioned logic instructions in the memory 303 may be implemented in the form of software functional units and may be stored in a computer-readable storage medium when sold or used as an independent product. Based on this understanding, the technical solution of the embodiment of the present invention is essentially or the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program codes. .

An embodiment of the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to perform the real-time intelligent site panorama survey method provided by the above-mentioned embodiments, for example, including : receiving a survey task, the survey task includes the internal survey modeling of the computer room, the surrounding environment modeling of the pole holding and the large-scale environment modeling of the pole holding; collecting the image of the survey target corresponding to the survey task, wherein the collection of the survey task corresponding to The acquisition equipment and acquisition method of the image of the survey target are determined based on the predetermined natural space conditions of the survey target; the 3D model is restored based on the image, and the 3D model is sent to the survey platform as a task result for release .

The device embodiments described above are only illustrative, and the units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in One place, or it can be distributed to multiple network elements. Part or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. It can be understood and implemented by those skilled in the art without any creative efforts.

Through the above description of the implementations, those skilled in the art can clearly understand that each implementation can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the prior art can be embodied in the form of software products, and the computer software products can be stored in computer-readable storage media, such as ROM/RAM, magnetic discs, optical discs, etc., including several instructions to make a computer device (which may be a personal computer, server, or network device, etc.) execute the methods described in various embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it can still be Modifications are made to the technical solutions described in the foregoing embodiments, or equivalent replacements are made to some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims (8)

1. A real-time intelligent site panorama survey method, characterized in that, comprising: Receive survey tasks, the survey tasks include computer room internal identification and site environment modeling, wherein the computer room internal identification includes computer room cabinet vacancy identification and computer room equipment identification, and the site environment modeling includes computer room internal survey modeling, pole holding Surrounding environment modeling and pole-holding large-scale environment modeling; Collecting the image of the survey target corresponding to the survey task, wherein the collection equipment and collection method for collecting the image of the site environment modeling corresponding to the survey target are determined based on the predetermined natural space conditions of the survey target; If the survey task is internal identification in the computer room, then perform text recognition and object recognition on the image, and return the recognition result as the task result; If the survey task is site environment modeling, the three-dimensional model is restored based on the image, and the three-dimensional model is sent to the viewing platform as a task result for release; Restoring the three-dimensional model based on the image specifically includes: Carrying out feature matching of the image based on the SIFT operator to obtain a pair of matching feature points; The matching feature point pair adopts the RANSAC eight-point method to calculate the basic matrix, and the matching feature point pair that does not satisfy the basic matrix is eliminated; Use the triangulation method to restore the depth information of the corresponding three-dimensional point of the matched feature point after elimination, add more images, match with the previous image, calculate the three-dimensional point information, and form a sparse point cloud; Then use the CMVS and PMVS algorithms to reconstruct the dense point cloud of the matched feature points after elimination to obtain a 3D model; Among them, the CMVS algorithm filters out images with a high degree of overlap through the image clustering algorithm, reducing the number of images that need to be processed when using the PMVS algorithm in the future. The PMVS algorithm obtains the final dense point cloud through multi-view matching of small pictures. . 2. The real-time intelligent site panorama survey method according to claim 1, further comprising: If the survey target is inside the machine room, the natural space conditions are the intensity of natural light and the richness of equipment textures; If the survey target is the surrounding environment of the pole, the natural space condition is the rain and fog around the pole; If the survey target is a large-scale environment with a pole, the natural space condition is the volume of an object in the large-scale environment with a pole. 3. The real-time intelligent site panorama survey method according to claim 2, characterized in that, the acquisition equipment and acquisition method of the image corresponding to the survey target of the site environment modeling are based on the predetermined survey target The natural space conditions are determined, specifically including: When the survey task is the internal survey modeling of the computer room, If the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room; If the natural light intensity is appropriate and the equipment texture is rich, use the SLAM robot to carry the binocular camera to move and shoot in the computer room; If any of the above conditions are not met, use the SLAM robot to carry the camera and lidar to move and shoot in the computer room; When the survey task is modeling the surrounding environment of the pole, If there is no rain and fog, use the drone to carry the camera and lidar to fly around the pole and take pictures; When the survey task is modeling a large-scale environment with poles, If the object is large in the large-scale environment of the pole, use the drone to carry the camera to fly around and shoot in the large-scale environment of the pole; If the object is small in the large-scale environment of the pole, use a drone equipped with a camera and lidar to fly around and shoot in the large-scale environment of the pole. 4. A real-time intelligent site panorama survey device, characterized in that it comprises: The receiving unit is configured to receive a survey task, the survey task includes computer room internal identification and site environment modeling, wherein the computer room internal identification includes computer room cabinet vacancy identification and computer room equipment identification, and the site environment modeling includes computer room internal survey Modeling, modeling of the surrounding environment of the pole and the modeling of the large-scale environment of the pole; The acquisition unit is configured to acquire the images of the survey targets corresponding to the survey task, wherein the acquisition equipment and acquisition method for acquiring the images of the site environment modeling corresponding to the survey targets are determined based on the predetermined natural space conditions of the survey targets of; A computing unit, configured to perform character recognition and object recognition on the image if the survey task is internal recognition in the computer room, and return the recognition result as a task result; if the survey task is site environment modeling, then based on the The image is restored to a three-dimensional model, and the three-dimensional model is sent to the viewing platform as a task result for publication; Restoring the three-dimensional model based on the image specifically includes: Carrying out feature matching of the image based on the SIFT operator to obtain a pair of matching feature points; The matching feature point pair adopts the RANSAC eight-point method to calculate the basic matrix, and the matching feature point pair that does not satisfy the basic matrix is eliminated; Use the triangulation method to restore the depth information of the corresponding three-dimensional point of the matched feature point after elimination, add more images, match with the previous image, calculate the three-dimensional point information, and form a sparse point cloud; Then use the CMVS and PMVS algorithms to reconstruct the dense point cloud of the matched feature points after elimination to obtain a 3D model; Among them, the CMVS algorithm filters out images with a high degree of overlap through the image clustering algorithm, reducing the number of images that need to be processed when using the PMVS algorithm in the future. The PMVS algorithm obtains the final dense point cloud through multi-view matching of small pictures. . 5. real-time smart site panorama survey device according to claim 4, is characterized in that, also comprises: If the survey target is inside the machine room, the natural space conditions are the intensity of natural light and the richness of equipment textures; If the survey target is the surrounding environment of the pole, the natural space condition is the rain and fog around the pole; If the survey target is a large-scale environment with a pole, the natural space condition is the volume of an object in the large-scale environment with a pole. 6. The real-time intelligent site panorama survey device according to claim 5, characterized in that, the acquisition equipment and acquisition method of the images corresponding to the site environment modeling corresponding to the survey target are based on the predetermined survey target The natural space conditions are determined, specifically including: When the survey task is the internal survey modeling of the computer room, If the natural light intensity is low and the equipment texture is lacking, use the SLAM robot to carry the structured light depth camera to move and shoot in the computer room; If the natural light intensity is appropriate and the equipment texture is rich, use the SLAM robot to carry the binocular camera to move and shoot in the computer room; If any of the above conditions are not met, use the SLAM robot to carry the camera and lidar to move and shoot in the computer room; When the survey task is modeling the surrounding environment of the pole, If there is no rain and fog, use the drone to carry the camera and lidar to fly around the pole and take pictures; When the survey task is modeling a large-scale environment with poles, If the object is large in the large-scale environment of the pole, use the drone to carry the camera to fly around and shoot in the large-scale environment of the pole; If the object is small in the large-scale environment of the pole, use a drone equipped with a camera and lidar to fly around and shoot in the large-scale environment of the pole. 7. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and operable on the processor, characterized in that, when the processor executes the program, it realizes any of claims 1 to 3. A step of the real-time intelligent site panorama survey method. 8. A non-transitory computer-readable storage medium, on which a computer program is stored, characterized in that, when the computer program is executed by a processor, the real-time intelligent site panorama according to any one of claims 1 to 3 is realized The steps of the survey method.