Construction method, device and equipment of building construction image and readable storage medium

By acquiring and processing construction background images and three-dimensional models, high-accurate construction images are generated, which solves the problem that batch construction site images cannot be generated quickly in the prior art, and accurately restores the construction site environment and identifying deep neural networks are realized. Improved effect.

CN118967991APending Publication Date: 2024-11-15GLODON CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410983693.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-22
Publication Date
2024-11-15

AI Technical Summary

Technical Problem

The prior art cannot quickly and accurately generate images that restore the construction site environment in batches, especially in disaster scenarios such as major fires.

Method used

By obtaining the three-dimensional model of the construction background image and construction equipment, determine the target adjustment parameters such as camera parameters, light source information, posture parameters and material texture parameters, and process and synthesis of images and models to generate the target construction background image and target construction equipment model Finally, the image synthesis is performed according to the synthetic area, the building construction image is generated and input into the deep neural network model for identification.

Benefits of technology

It realizes high-accuracy restoration of the construction site environment and automatic synthesis and labeling of images, can adjust the viewing angle and lighting environment factors, and improves the recognition effect of deep neural networks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118967991A_ABST
    Figure CN118967991A_ABST
Patent Text Reader

Abstract

The invention discloses a building construction image construction method, apparatus and device, and a readable storage medium. The method comprises the steps of obtaining a to-be-synthesized construction background image and a three-dimensional model corresponding to a construction device; determining a synthesis area of the construction equipment in the construction background image through the category of the construction equipment; respectively determining target adjustment parameters of the three-dimensional model corresponding to the construction background image and the construction equipment, respectively processing the construction background image and the three-dimensional model corresponding to the construction equipment through the target adjustment parameters, and generating a target construction background image and a target construction equipment model; synthesizing the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image; and inputting the batch building construction images into the initial deep neural network model as a training set, so as to realize the recognition of the building construction images. According to the method, the construction image training set with high reduction degree is generated by configuring the adjustment parameters, and the recognition effect of the deep learning model is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to a method, device, equipment and readable storage medium for constructing a building construction image. Background Art

[0002] With the gradual development of artificial intelligence technology, deep learning models have been widely used in various industries. In particular, the construction field also applies deep neural network models to actual construction scenarios. Since the training process of deep learning models relies on a large amount of labeled data, and the commonly used labeled data is mostly general field image data, and there is less image data in the construction field, the actual scene application has a great demand for data acquisition in the construction field.

[0003] There are generally two ways to obtain image data in the field of conventional construction: one is for relevant personnel to collect and annotate in the actual construction area. The execution process of this method is cumbersome and time-consuming, and the efficiency is low. At the same time, some demand scenarios cannot obtain real images, such as disaster scenes such as major fires, and it is difficult to obtain real image data through simulation in reality. The second method is to use synthetic image data technology, including two-dimensional data enhancement algorithms and deep learning models. Although the two-dimensional enhancement algorithm can achieve changes such as translation, rotation, and stretching of the image, this method is not easy to distinguish between the foreground and background, cannot observe the back of the object, and cannot adjust the material of the object. The deep learning model is a method of synthesizing new image or video data based on text or other images. This method cannot finely adjust the rotation angle, position, and posture information of the target. The annotation position is relatively fixed, and it is impossible to obtain accurate label information for deep learning tasks.

[0004] There is currently no effective solution to the technical problem that the existing technology cannot quickly and accurately generate batch images to restore the construction site environment. Summary of the invention

[0005] The purpose of the present invention is to provide a method, device, equipment and readable storage medium for constructing a building construction image, which can solve the technical problem that it is impossible to quickly and accurately generate batch images that restore the construction site environment.

[0006] One aspect of the present invention provides a method for constructing a building construction image, the method comprising: obtaining a construction background image to be synthesized and a three-dimensional model corresponding to construction equipment; determining a synthesis area of ​​the construction equipment in the construction background image according to the category of the construction equipment; respectively determining target adjustment parameters of the construction background image and the three-dimensional model corresponding to the construction equipment, respectively processing the construction background image and the three-dimensional model corresponding to the construction equipment according to the target adjustment parameters to generate a target construction background image and a target construction equipment model; synthesizing the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image, and inputting batches of building construction images as training sets into an initial deep neural network model to realize recognition of building construction images.

[0007] Optionally, target adjustment parameters of the three-dimensional model corresponding to the construction background image and the construction equipment are determined respectively, including: obtaining a preset configuration file, wherein the preset configuration file includes adjustment items and adjustment parameters corresponding to the adjustment items; matching the construction background image and the three-dimensional model corresponding to the construction equipment with the adjustment items of the preset configuration file respectively, and obtaining target adjustment parameters of the construction background image and target adjustment parameters of the three-dimensional model corresponding to the construction equipment respectively, wherein the target adjustment parameters of the three-dimensional model corresponding to the construction equipment are camera parameters, light source information, posture parameters, material texture parameters and size information.

[0008] Optionally, if the target adjustment parameter is a camera parameter, the three-dimensional model corresponding to the construction equipment is processed through the target adjustment parameter, including: obtaining real camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the real camera parameters include the placement position and orientation angle of the real camera; determining reference camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the reference camera parameters are observation angles of monitoring equipment simulating actual construction scenes, including the placement position and orientation angle of the reference camera; establishing a mapping relationship between the placement position and orientation angle of the real camera and the pixel points of the three-dimensional model corresponding to the construction equipment; determining the pixel point distribution position under the reference camera through the mapping relationship and the placement position and orientation angle of the reference camera, and forming a new three-dimensional model corresponding to the construction equipment based on the determined pixel point distribution position.

[0009] Optionally, if the target adjustment parameter is light source information, the three-dimensional model corresponding to the construction equipment is processed through the target adjustment parameter, including: obtaining the first lighting position and the first lighting intensity of the fixed light source, processing the three-dimensional model corresponding to the construction equipment through the first lighting position and the first lighting intensity, and generating the three-dimensional model corresponding to the construction equipment under the fixed light source; obtaining the second lighting position and the second lighting intensity of the mobile light source, processing the three-dimensional model corresponding to the construction equipment under the fixed light source through the second lighting position and the second lighting intensity, and generating the three-dimensional model corresponding to the construction equipment under the superimposed light source.

[0010] Optionally, the three-dimensional model corresponding to the construction equipment is processed through target adjustment parameters, and the further step includes: obtaining preset posture parameters to adjust the angle of the three-dimensional model corresponding to the construction equipment, and generating three-dimensional models corresponding to construction equipment at different angles; obtaining material texture information associated with the construction equipment, and adjusting the three-dimensional models corresponding to construction equipment at different angles through the material texture information, and generating three-dimensional models of construction equipment of different material categories; determining a target synthesis size of the three-dimensional model of the construction equipment, and adjusting the three-dimensional models of construction equipment of different materials through the target synthesis size, and generating multiple target construction equipment models.

[0011] Optionally, the target construction background image and the target construction equipment model are synthesized according to the synthesis area to generate a building construction image, including: converting the target construction equipment model into a two-dimensional construction equipment image; determining a target synthesis area corresponding to the synthesis area in the target construction background image; comparing the size of the target synthesis area with the two-dimensional construction equipment image; if the size of the target synthesis area is greater than or equal to the size of the two-dimensional construction equipment image, replacing all / part of the pixel points of the target synthesis area with the pixel points of the two-dimensional construction equipment image; if the size of the target synthesis area is smaller than the size of the two-dimensional construction equipment image, determining the final synthesis area by the center point of the target synthesis area and the size of the two-dimensional construction equipment image, and replacing the pixel points in the final synthesis area with the pixel points of the two-dimensional construction equipment image.

[0012] Another aspect of the present invention provides a device for constructing a building construction image, which includes: an acquisition module for acquiring a construction background image to be synthesized and a three-dimensional model corresponding to construction equipment; a determination module for determining a synthesis area of ​​the construction equipment in the construction background image according to the category of the construction equipment; an adjustment module for respectively determining target adjustment parameters of the construction background image and the three-dimensional model corresponding to the construction equipment, and processing the construction background image and the three-dimensional model corresponding to the construction equipment according to the target adjustment parameters to generate a target construction background image and a target construction equipment model; a synthesis module for synthesizing the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image; and a training module for inputting a batch of building construction images as a training set into an initial deep neural network model to realize recognition of building construction images.

[0013] Optionally, the synthesis module is also used to: convert the target construction equipment model into a two-dimensional construction equipment image; determine the target synthesis area corresponding to the synthesis area in the target construction background image; compare the size of the target synthesis area with the two-dimensional construction equipment image; if the size of the target synthesis area is greater than or equal to the size of the two-dimensional construction equipment image, replace all / part of the pixel points of the target synthesis area with the pixel points of the two-dimensional construction equipment image; if the size of the target synthesis area is smaller than the size of the two-dimensional construction equipment image, determine the final synthesis area through the center point of the target synthesis area and the size of the two-dimensional construction equipment image, and replace the pixel points in the final synthesis area with the pixel points of the two-dimensional construction equipment image.

[0014] Another aspect of the present invention provides a computer device, which includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method for constructing a construction image of any of the above-mentioned embodiments when executing the computer program.

[0015] Another aspect of the present invention provides a computer storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the method for constructing a building construction image of any of the above embodiments is implemented. Further, the computer-readable storage medium may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function, etc.; the data storage area may store data created according to the use of a blockchain node, etc.

[0016] The present invention can automatically synthesize and annotate images of any magnitude by configuring adjustment parameters; it can adjust the image's viewing angle, lighting and other environmental factors to obtain more visual information. At the same time, the three-dimensional model can realize arbitrary transformation of the construction equipment image, improving the accuracy of restoring the real construction scene. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0018] Figure 1 An optional flow chart of a method for constructing a building construction image provided by the first embodiment of the present invention is shown;

[0019] Figure 2 A structural block diagram showing a device for constructing a building construction image provided by a second embodiment of the present invention; and

[0020] Figure 3 A block diagram of a computer device suitable for implementing a method for constructing a building construction image provided by a third embodiment of the present invention is shown. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical scheme and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0022] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or device including the element.

[0023] Embodiment 1

[0024] This embodiment provides a method for constructing a building construction image. Figure 1 A flow chart showing the construction method of the building construction image is shown in FIG. Figure 1 As shown, the method for constructing the building construction image may include steps S101 to S105, wherein:

[0025] Step S101, obtaining a construction background image to be synthesized and a three-dimensional model corresponding to the construction equipment;

[0026] In order to simulate the real construction environment, the designers collected a batch of construction background images and construction equipment images in advance. Different construction background images have different attribute information, namely, shooting / collection time, construction scene category, and shooting angle; construction equipment images contain attribute information such as equipment name, placement, and display angle.

[0027] When implementing this step, the system first receives the image acquisition instruction input by the designer at the front end. The image instruction includes the background information of the image data to be synthesized (a certain attribute information combination of the construction background image) and the construction equipment information (a certain attribute information combination of the construction equipment image). Then, the database is traversed to match the information in the image acquisition instruction with the information stored in the database. When the match is consistent, the corresponding construction background image and construction equipment image are obtained.

[0028] In order to improve the accuracy of the construction image in restoring the actual environment, the obtained construction background image and construction equipment image are not directly synthesized, but the construction equipment image is further processed through the image to 3D model tool to generate a three-dimensional model corresponding to the construction equipment image. The three-dimensional model can be directly edited, and its mapping and texture are adjusted after the model is obtained to make the model appearance closer to the real world, which greatly improves the accuracy of the construction image synthesis. Among them, the image to 3D model tool can be a conversion tool such as Embossify, Blender, Tinkercad, etc., or other tools that convert 2D to 3D, which are not limited here. The three-dimensional image of the construction equipment can also be directly obtained by scanning the object with a mobile phone or scanner.

[0029] Step S102, determining a synthesis area of ​​the construction equipment in the construction background image according to the category of the construction equipment;

[0030] Since the functions and application scenarios of construction equipment are different, their placement in the real construction environment will also be different. Here, we take the fire extinguisher box in the construction site background image as an example. The fire extinguisher box is usually placed in a location that is easy to use directly, has a flat ground, is ventilated and cool, and can avoid direct sunlight. Based on this placement condition, by using multimodal deep learning technology, the area suitable for placing the fire extinguisher box is marked in the construction background image and recorded to prepare for subsequent synthesis.

[0031] The implementation principle of this step can be: the preset construction equipment placement table records the mapping relationship between the category of construction equipment and the placement position. First, the placement position of the construction equipment is determined in the preset construction equipment placement table according to the category of the construction equipment to be synthesized, and then the placement position and the construction background image are input into the trained multimodal deep learning model for annotation, so as to accurately obtain the synthesis area in the construction background image of the construction equipment.

[0032] Step S103, respectively determining target adjustment parameters of the construction background image and the three-dimensional model corresponding to the construction equipment, and processing the construction background image and the three-dimensional model corresponding to the construction equipment respectively by using the target adjustment parameters to generate a target construction background image and a target construction equipment model;

[0033] Specifically, the processing method of the construction background image is a two-dimensional enhancement algorithm, including adjustment parameters such as brightness, saturation, contrast, and size; the adjustment parameters of the three-dimensional model corresponding to the construction equipment include camera parameters, light source information, posture parameters, material texture parameters, and size information. Since the construction background image is a two-dimensional image and the three-dimensional model corresponding to the construction equipment is a three-dimensional image, the adjustment methods of the two are also different. By setting different adjustment parameters for the construction background image and the three-dimensional model corresponding to the construction equipment, targeted processing of the construction background and construction equipment is achieved, which is conducive to restoring the real construction environment.

[0034] In particular, although traditional image processing technology can achieve changes such as translation, rotation, and stretching of images, it is not easy to distinguish the foreground and background, cannot observe the back of objects, and cannot adjust the material of objects, so it cannot effectively restore the real construction scene. This embodiment provides a separate processing for the three-dimensional model corresponding to the construction equipment, which can be edited from different dimensions of the construction equipment and restore the real state of the construction equipment in the construction environment.

[0035] Step S104, synthesizing the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image.

[0036] From the above, it can be seen that the synthesis area represents the placement position of the construction equipment in the real construction environment. Synthesizing the target construction background image and the target construction equipment model based on the synthesis area can greatly improve the accuracy of restoring the real construction environment.

[0037] Step S105: input the batch of building construction images as training sets into the initial deep neural network model to achieve recognition of the building construction images.

[0038] By configuring and adjusting parameters, a construction image training set with a high degree of restoration can be generated. The generated multiple construction images are input into the initial deep neural network model for training, which can improve the recognition effect of the deep neural network.

[0039] Preferably, step S103 may include step S1031 to step S1032, wherein:

[0040] Step S1031, obtaining a preset configuration file, wherein the preset configuration file includes adjustment items and adjustment parameters corresponding to the adjustment items;

[0041] Adjustment item categories include construction background and construction equipment;

[0042] Step S1032, respectively match the construction background image and the three-dimensional model corresponding to the construction equipment with the adjustment items of the preset configuration file, and respectively obtain the target adjustment parameters of the construction background image and the target adjustment parameters of the three-dimensional model corresponding to the construction equipment, wherein the target adjustment parameters of the three-dimensional model corresponding to the construction equipment are camera parameters, light source information, posture parameters, material texture parameters and size information.

[0043] The preset configuration file provides a variety of adjustment methods for image synthesis. By writing scripts to execute the adjustment parameters in the above preset configuration files, you can control whether each adjustment parameter changes randomly, as well as the range and frequency of random changes. Through unified management of scripts, the appearance of the model can be closer to the real world. Taking the fire extinguisher object as an example, you can set the object's rotation angle, size, different lighting angles and intensities, different materials, contamination conditions, etc., as well as the number of images synthesized in subsequent steps, and the number, size, position and other information of objects that need to be synthesized in each background image. This can make the synthesized images more diverse and help train robust models.

[0044] Preferably, if the target adjustment parameter is a camera parameter, the three-dimensional model corresponding to the construction equipment is processed by the target adjustment parameter, including steps A1 to A4, wherein:

[0045] Step A1, obtaining real camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the real camera parameters include the placement position and orientation angle of the real camera;

[0046] The placement position can be the ceiling, floor, or corner of the construction space. The orientation angles include upward, horizontal, and downward.

[0047] Step A2, determining reference camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the reference camera parameters are observation angles of the monitoring equipment simulating the actual construction scene, including the placement position and orientation angle of the reference camera;

[0048] Step A3, establishing a mapping relationship between the placement position and orientation angle of the real camera and the pixel points of the three-dimensional model corresponding to the construction equipment;

[0049] Step A4, determining the pixel distribution position under the reference camera through the mapping relationship and the placement position and orientation angle of the reference camera, and forming a three-dimensional model corresponding to the new construction equipment based on the determined pixel distribution position.

[0050] By using fixed and random camera angles within a certain range to simulate the angles at which monitoring equipment observes construction equipment in different scenarios in the real world, the accuracy of restoring the real construction scene is improved.

[0051] Preferably, if the target adjustment parameter is light source information, the three-dimensional model corresponding to the construction equipment is processed by the target adjustment parameter, including steps B1 to B2, wherein:

[0052] Step B1, obtaining a first illumination position and a first illumination intensity of a fixed light source, processing a three-dimensional model corresponding to the construction equipment according to the first illumination position and the first illumination intensity, and generating a three-dimensional model corresponding to the construction equipment under the fixed light source;

[0053] The number of fixed light sources may be 1, 2 or more. When processing the three-dimensional model corresponding to the construction equipment through the fixed light sources, the three-dimensional model corresponding to the construction equipment may be projected based on the first illumination positions and first illumination intensities of multiple fixed light sources turned on at the same time; or the three-dimensional model corresponding to the construction equipment may be projected by shielding some of the fixed light sources and retaining some of the fixed light sources.

[0054] Step B2, obtaining the second lighting position and the second lighting intensity of the mobile light source, processing the three-dimensional model corresponding to the construction equipment under the fixed light source through the second lighting position and the second lighting intensity, and generating the three-dimensional model corresponding to the construction equipment under the superimposed light source.

[0055] The mobile light source may be sunlight. The second illumination position of the mobile light source may be a window area in the three-dimensional model corresponding to the construction equipment.

[0056] By using fixed and random light source information within a certain range, the visual information of different scenes under the influence of different lighting environments is simulated. Usually, two or more point light sources are required to simulate the complex lighting environment in the real world.

[0057] It should be noted that the three-dimensional model corresponding to the construction equipment is cut out using the mask function in the image processing tool during the adjustment process, which can ensure the integrity of the outline of the three-dimensional model.

[0058] Preferably, the three-dimensional model corresponding to the construction equipment is processed by the target adjustment parameter, including steps C1 to C3, wherein:

[0059] Step C1, obtaining preset posture parameters to adjust the angle of the three-dimensional model corresponding to the construction equipment, and generating three-dimensional models corresponding to the construction equipment at different angles;

[0060] Configure fixed and random posture parameters of the construction equipment to be tested, such as rotation angle, etc. Through fixed and random posture parameters within a certain range, simulate the different postures of different construction equipment in different scenes.

[0061] Step C2, obtaining material texture information associated with the construction equipment, adjusting the three-dimensional models corresponding to the construction equipment at different angles according to the material texture information, and generating three-dimensional models of construction equipment of different material categories;

[0062] Configure the material and texture information of the construction equipment, including rough textures with frosted textures, smooth textures with metallic textures, etc. By adjusting the different materials of the construction equipment, the texture and appearance of the model are closer to the real world, and can simulate the changes in appearance such as aging and wear caused by the construction site environment in reality.

[0063] Step C3, determining a target synthesis size of the three-dimensional model of the construction equipment, adjusting the three-dimensional models of the construction equipment of different materials respectively according to the target synthesis size, and generating a plurality of target construction equipment models.

[0064] According to the standards in the field of deep learning target detection, the synthetic size of the three-dimensional model of construction equipment is divided into three categories: large targets, medium targets and small targets. The specific size variation ranges and occurrence frequencies of these three types of targets are set to be close to the sizes and occurrence frequencies in realistic construction scenes, making the data distribution of the synthetic data set closer to reality.

[0065] Preferably, step S104 may include steps S1041 to S1045, wherein:

[0066] Step S1041, converting the target construction equipment model into a two-dimensional construction equipment image;

[0067] Step S1042, determining a target synthesis area corresponding to the synthesis area in the target construction background image;

[0068] Step S1043, comparing the size of the target synthesis area with the two-dimensional construction equipment image;

[0069] Step S1044, if the size of the target synthesis area is greater than or equal to the size of the two-dimensional construction equipment image, all / part of the pixel points of the target synthesis area are replaced by the pixel points of the two-dimensional construction equipment image;

[0070] Step S1045, if the size of the target synthesis area is smaller than the size of the two-dimensional construction equipment image, the final synthesis area is determined by the center point of the target synthesis area and the size of the two-dimensional construction equipment image, and the pixel points in the final synthesis area are replaced by the pixel points of the two-dimensional construction equipment image.

[0071] In addition, the number of construction equipment in the construction background image can also be set to be close to the actual construction scene.

[0072] This embodiment automatically synthesizes and annotates images of any magnitude by configuring adjustment parameters; it can adjust the image's viewing angle, lighting and other environmental factors to obtain more visual information. At the same time, the three-dimensional model can realize arbitrary transformation of the construction equipment image, improving the accuracy of restoring the real construction scene. By configuring the adjustment parameters, a construction image training set with a high degree of restoration can be generated and input into the initial deep neural network model for training, which can improve the recognition effect of the deep neural network.

[0073] Embodiment 2

[0074] Embodiment 2 of the present invention further provides a device for constructing a building construction image, which corresponds to the method for constructing a building construction image provided in the above embodiment 1. The corresponding technical features and technical effects are not described in detail in this embodiment, and the relevant parts can refer to the above embodiment 1. Specifically, Figure 2 FIG. 4 shows a structural block diagram of the device for constructing the building construction image. Figure 2 As shown, the construction image construction device 200 includes an acquisition module 201, a determination module 202, an adjustment module 203, a synthesis module 204 and a training module 205, wherein:

[0075] An acquisition module 201 is used to acquire a construction background image to be synthesized and a three-dimensional model corresponding to the construction equipment;

[0076] The determination module 202 is connected to the acquisition module 201 and is used to determine the synthesis area of ​​the construction equipment in the construction background image according to the category of the construction equipment;

[0077] The adjustment module 203 is connected to the determination module 202 and is used to determine target adjustment parameters of the three-dimensional model corresponding to the construction background image and the construction equipment respectively, and process the three-dimensional model corresponding to the construction background image and the construction equipment respectively according to the target adjustment parameters to generate a target construction background image and a target construction equipment model;

[0078] The synthesis module 204 is connected to the adjustment module 203 and is used to synthesize the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image.

[0079] The training module 205 is connected to the synthesis module 204 and is used to input the batch of building construction images as training sets into the initial deep neural network model to realize the recognition of building construction images.

[0080] Optionally, the adjustment module is also used to: obtain a preset configuration file, wherein the preset configuration file includes adjustment items and adjustment parameters corresponding to the adjustment items; match the construction background image and the three-dimensional model corresponding to the construction equipment with the adjustment items of the preset configuration file, respectively, to obtain target adjustment parameters of the construction background image and target adjustment parameters of the three-dimensional model corresponding to the construction equipment, respectively, wherein the target adjustment parameters of the three-dimensional model corresponding to the construction equipment are camera parameters, light source information, posture parameters, material texture parameters and size information.

[0081] Optionally, the adjustment module is also used to: obtain real camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the real camera parameters include the placement position and orientation angle of the real camera; determine the reference camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the reference camera parameters are observation angles of monitoring equipment that simulate actual construction scenes, including the placement position and orientation angle of the reference camera; establish a mapping relationship between the placement position and orientation angle of the real camera and the pixel points of the three-dimensional model corresponding to the construction equipment; determine the pixel point distribution position under the reference camera through the mapping relationship and the placement position and orientation angle of the reference camera, and form a new three-dimensional model corresponding to the construction equipment based on the determined pixel point distribution position.

[0082] Optionally, the adjustment module is also used to: obtain the first lighting position and the first lighting intensity of the fixed light source, process the three-dimensional model corresponding to the construction equipment through the first lighting position and the first lighting intensity, and generate the three-dimensional model corresponding to the construction equipment under the fixed light source; obtain the second lighting position and the second lighting intensity of the mobile light source, process the three-dimensional model corresponding to the construction equipment under the fixed light source through the second lighting position and the second lighting intensity, and generate the three-dimensional model corresponding to the construction equipment under the superimposed light source.

[0083] Optionally, the adjustment module is also used to: obtain preset posture parameters to adjust the angle of the three-dimensional model corresponding to the construction equipment, and generate three-dimensional models corresponding to construction equipment at different angles; obtain material texture information associated with the construction equipment, and adjust the three-dimensional models corresponding to construction equipment at different angles according to the material texture information, and generate three-dimensional models of construction equipment of different material categories; determine the target synthesis size of the three-dimensional model of the construction equipment, and adjust the three-dimensional models of construction equipment of different materials according to the target synthesis size to generate multiple target construction equipment models.

[0084] Optionally, the synthesis module is also used to: convert the target construction equipment model into a two-dimensional construction equipment image; determine the target synthesis area corresponding to the synthesis area in the target construction background image; compare the size of the target synthesis area with the two-dimensional construction equipment image; if the size of the target synthesis area is greater than or equal to the size of the two-dimensional construction equipment image, replace all / part of the pixel points of the target synthesis area with the pixel points of the two-dimensional construction equipment image; if the size of the target synthesis area is smaller than the size of the two-dimensional construction equipment image, determine the final synthesis area through the center point of the target synthesis area and the size of the two-dimensional construction equipment image, and replace the pixel points in the final synthesis area with the pixel points of the two-dimensional construction equipment image.

[0085] Embodiment 3

[0086] Figure 3 A block diagram of a computer device suitable for implementing a construction image construction method provided by the third embodiment of the present invention is shown. In this embodiment, the computer device 300 may be a smart phone, tablet computer, laptop computer, desktop computer, rack server, blade server, tower server or cabinet server (including an independent server or a server cluster composed of multiple servers) that executes a program. Figure 3 As shown, the computer device 300 of this embodiment includes at least but is not limited to: a memory 301, a processor 302, and a network interface 303 that can be interconnected through a system bus. It should be noted that Figure 3 Only computer device 300 having components 301 - 303 is shown, but it should be understood that implementing all of the components shown is not a requirement, and more or fewer components may alternatively be implemented.

[0087] In this embodiment, the memory 303 includes at least one type of computer-readable storage medium, and the readable storage medium includes flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, etc. In some embodiments, the memory 301 can be an internal storage unit of the computer device 300, such as a hard disk or memory of the computer device 300. In other embodiments, the memory 301 can also be an external storage device of the computer device 300, such as a plug-in hard disk equipped on the computer device 300, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Of course, the memory 301 can also include both the internal storage unit of the computer device 300 and its external storage device. In this embodiment, the memory 301 is generally used to store the operating system and various application software installed on the computer device 300, such as the program code of the construction method of the building construction image, etc.

[0088] In some embodiments, the processor 302 may be a central processing unit (CPU), a controller, a microcontroller, a microprocessor, or other data processing chips. The processor 302 is generally used to control the overall operation of the computer device 300. For example, it performs control and processing related to data interaction or communication with the computer device 300. In this embodiment, the processor 302 is used to run the program code of the steps of the construction method of the building construction image stored in the memory 301.

[0089] In this embodiment, the construction method of the building construction image stored in the memory 301 can also be divided into one or more program modules and executed by one or more processors (processor 302 in this embodiment) to complete the present invention.

[0090] The network interface 303 may include a wireless network interface or a wired network interface, and the network interface 303 is generally used to establish a communication link between the computer device 300 and other computer devices. For example, the network interface 303 is used to connect the computer device 300 to an external terminal through a network, and to establish a data transmission channel and a communication link between the computer device 300 and the external terminal. The network can be a wireless or wired network such as an intranet, the Internet, the Global System of Mobile communication (GSM), Wideband Code Division Multiple Access (WCDMA), 4G network, 5G network, Bluetooth, Wi-Fi, etc.

[0091] Embodiment 4

[0092] This embodiment also provides a computer-readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (for example, SD or DX memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, optical disk, server, App application store, etc., on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for constructing a building construction image are implemented.

[0093] Obviously, those skilled in the art should understand that the modules or steps of the above-mentioned embodiments of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, and optionally, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. In this way, the embodiments of the present invention are not limited to any specific combination of hardware and software.

[0094] It should be noted that the serial numbers of the embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0095] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method.

[0096] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A method for constructing a building construction image, It is characterized in that The method comprises: Obtaining a construction background image to be synthesized and a three-dimensional model corresponding to the construction equipment; Determining a synthesis area of ​​the construction equipment in the construction background image according to the category of the construction equipment; Determining target adjustment parameters of the construction background image and the three-dimensional model corresponding to the construction equipment respectively, and processing the construction background image and the three-dimensional model corresponding to the construction equipment respectively according to the target adjustment parameters to generate a target construction background image and a target construction equipment model; Synthesize the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image; Batches of construction images are input as training sets into the initial deep neural network model to achieve recognition of construction images.

2. The method according to claim 1, It is characterized in that The step of respectively determining target adjustment parameters of the three-dimensional model corresponding to the construction background image and the construction equipment includes: Obtaining a preset configuration file, wherein the preset configuration file includes adjustment items and adjustment parameters corresponding to the adjustment items; The construction background image and the three-dimensional model corresponding to the construction equipment are respectively matched with the adjustment items of the preset configuration file to obtain target adjustment parameters of the construction background image and the target adjustment parameters of the three-dimensional model corresponding to the construction equipment, wherein the target adjustment parameters of the three-dimensional model corresponding to the construction equipment are camera parameters, light source information, posture parameters, material texture parameters and size information.

3. The method according to claim 2, It is characterized in that If the target adjustment parameter is a camera parameter, the three-dimensional model corresponding to the construction equipment is processed by the target adjustment parameter, including: Acquire real camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the real camera parameters include the placement position and orientation angle of the real camera; Determine reference camera parameters of the three-dimensional model corresponding to the construction equipment, wherein the reference camera parameters are observation angles of the monitoring equipment simulating the actual construction scene, including the placement position and orientation angle of the reference camera; Establishing a mapping relationship between the placement position and orientation angle of the real camera and the pixel points of the three-dimensional model corresponding to the construction equipment; The pixel distribution position under the reference camera is determined by the mapping relationship and the placement position and orientation angle of the reference camera, and a three-dimensional model corresponding to the new construction equipment is formed based on the determined pixel distribution position.

4. The method according to claim 2 or 3, It is characterized in that If the target adjustment parameter is light source information, processing the three-dimensional model corresponding to the construction equipment according to the target adjustment parameter includes: Acquire a first illumination position and a first illumination intensity of a fixed light source, and process the three-dimensional model corresponding to the construction equipment according to the first illumination position and the first illumination intensity to generate a three-dimensional model corresponding to the construction equipment under the fixed light source; A second illumination position and a second illumination intensity of the mobile light source are obtained, and the three-dimensional model corresponding to the construction equipment under the fixed light source is processed according to the second illumination position and the second illumination intensity to generate a three-dimensional model corresponding to the construction equipment under the superimposed light source.

5. The method according to claim 2, It is characterized in that Processing the three-dimensional model corresponding to the construction equipment by using the target adjustment parameter also includes: Obtaining preset posture parameters to adjust the angle of the three-dimensional model corresponding to the construction equipment, and generating three-dimensional models corresponding to the construction equipment at different angles; Acquire material texture information associated with the construction equipment, and adjust the three-dimensional models corresponding to the construction equipment at different angles according to the material texture information to generate three-dimensional models of construction equipment of different material categories; The target synthesis size of the three-dimensional model of the construction equipment is determined, and the three-dimensional models of construction equipment of different materials are adjusted respectively according to the target synthesis size to generate a plurality of target construction equipment models.

6. The method according to claim 1, It is characterized in that The step of synthesizing the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image includes: converting the target construction equipment model into a two-dimensional construction equipment image; Determining a target synthesis area corresponding to the synthesis area in the target construction background image; Comparing the sizes of the target synthesis area and the two-dimensional construction equipment image; If the size of the target synthesis area is greater than or equal to the size of the two-dimensional construction equipment image, all / part of the pixel points of the target synthesis area are replaced by the pixel points of the two-dimensional construction equipment image; If the size of the target synthesis area is smaller than the size of the two-dimensional construction equipment image, the final synthesis area is determined by the center point of the target synthesis area and the size of the two-dimensional construction equipment image, and the pixel points in the final synthesis area are replaced by the pixel points of the two-dimensional construction equipment image.

7. A device for constructing a building construction image, It is characterized in that The device comprises: An acquisition module, used to acquire the construction background image to be synthesized and the three-dimensional model corresponding to the construction equipment; A determination module, used for determining a synthesis area of ​​the construction equipment in the construction background image according to the category of the construction equipment; An adjustment module, used to respectively determine target adjustment parameters of the construction background image and the three-dimensional model corresponding to the construction equipment, and process the construction background image and the three-dimensional model corresponding to the construction equipment respectively according to the target adjustment parameters to generate a target construction background image and a target construction equipment model; A synthesis module, used for synthesizing the target construction background image and the target construction equipment model according to the synthesis area to generate a building construction image; The training module is used to input batches of construction images as training sets into the initial deep neural network model to achieve recognition of construction images.

8. The device according to claim 7, It is characterized in that The synthesis module is also used for: converting the target construction equipment model into a two-dimensional construction equipment image; Determining a target synthesis area corresponding to the synthesis area in the target construction background image; Comparing the sizes of the target synthesis area and the two-dimensional construction equipment image; If the size of the target synthesis area is greater than or equal to the size of the two-dimensional construction equipment image, all / part of the pixel points of the target synthesis area are replaced by the pixel points of the two-dimensional construction equipment image; If the size of the target synthesis area is smaller than the size of the two-dimensional construction equipment image, the final synthesis area is determined by the center point of the target synthesis area and the size of the two-dimensional construction equipment image, and the pixel points in the final synthesis area are replaced by the pixel points of the two-dimensional construction equipment image.

9. A computer device, wherein the computer device include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method according to any one of claims 1 to 6 when executing the computer program.

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