An object recognition system and method

Abstract

A method and system for object recognition are provided. The method can include obtaining three-dimensional scan data of a scanning chamber; identifying, using at least one item recognition model, at least one target item from the three-dimensional scan data; determining a model of the at least one target item; and generating a virtual reality model of the scanning chamber based on the three-dimensional scan data and the model of the at least one target item, the virtual reality model including the model of the at least one target item.

Description

Technical Field

[0001] This specification relates to the field of medical technology, and in particular to a recognition system and method for scanning indoor objects. Background Art

[0002] During the transport, placement, installation, commissioning, and subsequent maintenance of a nuclear magnetic resonance (NMR) scanner, it is necessary to determine the location, type, and characteristic information of objects within the scanning room, such as physical characteristics (e.g., electromagnetic characteristics and / or magnetic field characteristics). Currently, this information is determined manually, and a model of the scanning room is drawn. This is very slow and inconvenient to modify later. Therefore, the present application seeks to provide a system and method for efficient object recognition within a scanning room. Summary of the Invention

[0003] One embodiment of this specification provides an object recognition method. The method includes acquiring three-dimensional scan data of a scanning room; identifying at least one target object from the three-dimensional scan data using at least one object recognition model; determining the model of the at least one target object; and generating a virtual reality model of the scanning room based on the three-dimensional scan data and the model of the at least one target object, the virtual reality model including the model of the at least one target object.

[0004] In some embodiments, the object recognition model may identify at least one target object in the three-dimensional scan data based on the three-dimensional scan data. In some embodiments, the object recognition model may further output three-dimensional scan data corresponding to each of the at least one target object.

[0005] In some embodiments, the object recognition model is obtained through the following training process: obtaining an initial object recognition model; obtaining sample three-dimensional scanning data, wherein the sample three-dimensional scanning data includes at least one target object; obtaining an object type label corresponding to at least one target object in the sample three-dimensional scanning data; training the initial object recognition model based on the sample three-dimensional scanning data and the object type label of the at least one target object to obtain an object recognition model.

[0006] In some embodiments, the model recognition model can extract features of the target 3D scan data of the target object and determine the model of the target object based on the features. In some embodiments, the features of the target 3D scan data can be extracted from projection data at one or more standard angles.

[0007] In some embodiments, a model recognition model can determine the model of a target object based on target 3D scan data of the target object and at least two reference models. The at least two reference models can correspond to at least two candidate models of the target object. The reference models can include reference 3D scan data of the candidate model objects or 3D models constructed based thereon. The reference 3D scan data can be obtained through actual measurement or simulation.

[0008] In some embodiments, the virtual reality model of the scanning room can include generating a three-dimensional model based on three-dimensional scanning data of the scanning room. In some embodiments, the virtual reality model of the scanning room can also include reference information. This reference information can include one or more of the target object's model, characteristic information, and physical field distribution information of the scanning room. The characteristic information of the target object can include physical characteristics, material information, whether it is movable, etc. The physical characteristic information can include electromagnetic characteristics and / or magnetic field characteristics. In some embodiments, the physical field distribution within the scanning room during operation of the MRI scanner can be determined based on the three-dimensional scanning data, and the virtual reality model can be further generated based on the physical field distribution. In some embodiments, the physical field emission model of the MRI scanner and the physical field characteristics of each region in the scanning room can be obtained, and the physical field distribution information of the scanning room during operation of the MRI scanner can be determined using a physical field simulation model. In some embodiments, the processing device can determine the physical field emission model based on performance parameters of one or more components of the MRI scanner that emit magnetic and / or electromagnetic fields. The physical field characteristics of a region can include at least one of physical field absorption characteristics and physical field reflection characteristics of the region.

[0009] One embodiment of this specification provides an object recognition system, comprising: at least one storage device for storing computer instructions; and at least one processor for executing the computer instructions to implement the object recognition method.

[0010] One embodiment of this specification provides an object recognition system. The object recognition system includes an acquisition module for acquiring three-dimensional scan data of a scanning room; an identification module for identifying at least one target object from the three-dimensional scan data using at least one object recognition model; a determination module for determining the model of the at least one target object; and a generation module for generating a virtual reality model of the scanning room based on the three-dimensional scan data and the model of the at least one target object, the virtual reality model including the model of the at least one target object.

[0011] Some additional features of the present application may be explained in the following description. Some additional features of the present application will be apparent to those skilled in the art from a study of the following description and accompanying drawings, or from the production or operation of the embodiments. The features of the present application may be realized and obtained by practicing or using various aspects of the methods, means, and combinations described in the following detailed examples. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] This specification will be further described in the form of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not limiting, and in these embodiments, the same numbers represent the same structures, wherein:

[0013] Figure 1 is a schematic diagram of an exemplary scanning room according to some embodiments of this specification;

[0014] Figure 2 is a schematic diagram of modules of an exemplary object recognition system according to some embodiments of this specification;

[0015] Figure 3 is a flowchart of an exemplary object recognition method according to some embodiments of this specification;

[0016] Figure 4 is a schematic diagram of an exemplary virtual reality model according to some embodiments of this specification; and

[0017] Figure 5 It is a flowchart of identifying the model of an object according to the exemplary model recognition model shown in some embodiments of this specification. DETAILED DESCRIPTION

[0018] To more clearly illustrate the technical solutions of the embodiments of this specification, the following briefly describes the drawings required for describing the embodiments. Obviously, the drawings described below are merely examples or embodiments of this specification. Those skilled in the art can apply this specification to other similar scenarios based on these drawings without inventive effort. Unless otherwise apparent from the context or otherwise noted, the same reference numerals in the figures represent the same structure or operation.

[0019] It should be understood that the terms "system," "device," "unit," and / or "module" used herein are a method for distinguishing different components, elements, parts, portions, or assemblies at different levels. However, if other terms can achieve the same purpose, the terms may be replaced by other expressions.

[0020] As used in this specification and claims, unless the context clearly indicates otherwise, the words "a," "an," "an," and / or "the" do not refer to the singular but also include the plural. Generally speaking, the terms "comprises" and "include" only indicate the inclusion of the steps and elements specifically identified, and these steps and elements do not constitute an exclusive list. A method or apparatus may also include other steps or elements.

[0021] Flowcharts are used throughout this specification to illustrate the operations performed by systems according to embodiments of this specification. It should be understood that preceding or following operations do not necessarily need to be performed in exact order. Instead, the steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more operations may be removed from these processes.

[0022] Figure 1 FIG. 1 is a schematic diagram of an exemplary scanning room according to some embodiments of this specification. Figure 1 As shown, the scanning room 100 may be provided with an imaging device 110 , a cabinet 120 , a computer 130 , a printer 140 , an air conditioner 150 , a spare parts cabinet 160 , a table 170 , a chair 180 and a scanning device 190 .

[0023] The imaging device 110 can be a non-invasive scanning imaging device for disease diagnosis or research purposes. In some embodiments, the imaging device 110 can scan a target object within a detection area or a scanning area to obtain scan data of the target object. In some embodiments, the imaging device 110 can include a magnetic resonance imaging (MRI) device, an X-ray imaging-magnetic resonance imaging (X-ray-MRI) device, a single photon emission computed tomography-magnetic resonance imaging (SPECT-MRI) device, a digital subtraction angiography-magnetic resonance imaging (DSA-MRI) device, etc. The following description takes an MRI device as an example. The imaging device 110 includes a scanning bed, and the target object can be moved into the scanning area by moving the scanning bed.

[0024] The cabinet 120 can control various aspects of the MRI scan, such as the emission of radio frequency pulses, the application of gradient magnetic fields, and the acquisition of magnetic resonance signals. In some embodiments, the cabinet 120 can also perform tasks such as data processing and image reconstruction. In some embodiments, the cabinet 120 can also acquire three-dimensional scan data of the scanning room and identify at least one target object from the three-dimensional scan data using an object recognition model. In some embodiments, the cabinet 120 can also determine the model of the at least one target object. In some embodiments, the cabinet 120 can also generate a virtual reality model of the scanning room 100 based on the three-dimensional scan data and the model of the at least one target object. The cabinet 120 can be connected to the imaging device 110 via a network to control the operation of various components of the imaging device 110. The cabinet 120 can also be connected to the computer 130 via a network to exchange information and / or data. In some embodiments, the cabinet 120 can execute control instructions stored in a storage device (e.g., a storage device within the cabinet) to perform its functions.

[0025] Computer 130 can be used by medical imaging technicians to send instructions to the console, view, process and store medical images. Processing medical images may include marking medical images, selecting regions of interest, etc. Computer 130 may include a tablet computer, a laptop computer, a desktop computer, a mobile device, etc. In some embodiments, computer 130 can receive data from cabinet 120 and perform data processing and image reconstruction. In some embodiments, computer 130 can be connected to imaging device 110 via a network and can send information and / or data to each other. For example, computer 130 can be connected to the scanning bed control part of imaging device 110 to control the movement of the scanning bed. In some embodiments, computer 130 can communicate with other devices outside the scanning room via a network.

[0026] The printer 140 can be used to print medical reports and medical images. The printer 140 can include color printing, black and white printing, 3D printing, etc.

[0027] The air conditioner 150 can be used to provide the temperature and humidity required for the operation of the MRI equipment in the scanning room 100. The air conditioner 150 can include a floor-standing air conditioner, a wall-mounted air conditioner, a central air conditioner, and the like.

[0028] The spare parts cabinet 160 can be used to store spare parts and consumables for the MRI equipment. It can also be used to store radio frequency coils. It can also be used to store paper documents such as product specifications, usage records, and maintenance records for the MRI equipment.

[0029] The table 170 and the chair 180 are used by medical imaging technicians during their work.

[0030] Scanning device 190 may include a camera and / or a radar, and is configured to scan scanning room 100 to obtain three-dimensional scan data of scanning room 100. The three-dimensional scan data may include point cloud data, depth data, a three-dimensional image, and / or a three-dimensional model. For example, scanning device 190 may be a depth camera for collecting depth data of scanning room 100. For another example, scanning device 190 may be a radar for collecting point cloud data of scanning room 100. Scanning device 190 may transmit the three-dimensional scan data of scanning room 100 to cabinet 120, computer 130, or another computer.

[0031] It should be noted that the scanning room 100 is provided for illustrative purposes only and is not intended to limit the scope of this specification. For those skilled in the art, various modifications or variations can be made based on the description of this specification. For example, the scanning room 100 can also include an equipotential box. For another example, the scanning room 100 can be divided into multiple sub-areas, such as an imaging device scanning sub-area, a computer control sub-area, and other sub-areas. The multiple sub-areas can be separated by walls, etc. However, these changes and modifications do not deviate from the scope of this specification. For example, one or more devices in the scanning room 100 can be omitted, and / or can further include one or more other devices. For another example, multiple devices in the scanning room 100 (for example, a computer 130 and a cabinet 120) can be integrated into a single device.

[0032] Figure 2 is a schematic diagram of an exemplary object recognition system according to some embodiments of this specification.

[0033] like Figure 2 As shown, in some embodiments, object recognition system 200 may include an acquisition module 210, an identification module 220, a determination module 230, and a generation module 240. In some embodiments, object recognition system 200 may be implemented by a processing device. For example, acquisition module 210, identification module 220, determination module 230, and generation module 240 may be modules within the processing device. The processing device may acquire 3D scan data within the scanning room and perform subsequent processing. For example, the processing device may include computer 130, cabinet 120, or other computers. In some embodiments, the processing device may be part of a scanning device (e.g., radar, camera, etc.) and / or an imaging device.

[0034] The acquisition module 210 can be used to acquire three-dimensional scanning data of the scanning room. The three-dimensional scanning data of the scanning room can be related to the internal spatial structure of the scanning room and one or more target objects located inside the scanning room. In some embodiments, the three-dimensional scanning data can be obtained by a scanning device (e.g., Figure 1More details about obtaining the three-dimensional scanning data of the scanning room can be found in step 310, which will not be repeated here.

[0035] Identification module 220 can use at least one object recognition model to identify at least one target object from the three-dimensional scan data. In some embodiments, identifying the target object can include determining the type of at least one target object in the scanning room. For example, identification module 220 can use the object recognition model to identify one or more target objects in the scanning room, including imaging device 110, cabinet 120, computer 130, printer 140, air conditioner 150, spare parts cabinet 160, table 170, and chair 180. In some embodiments, identifying the target object can include determining target three-dimensional scan data corresponding to the target object based on the three-dimensional scan data of the scanning room. For example, identification module 220 can use the object recognition model to segment a portion corresponding to the target object from the three-dimensional scan data of the scanning room to provide the target three-dimensional scan data for the target object. For more information on identifying at least one target object from the three-dimensional scan data using at least one object recognition model, see step 320 and will not be repeated here.

[0036] The determination module 230 can be used to determine the model of at least one target item. In some embodiments, the processing device can obtain planar image data of at least one target item. For example, a planar image taken by a two-dimensional camera. Further, the determination module 230 can use optical character recognition (OCR) technology to identify the characters of the relevant model in the planar image, thereby determining the model of the target item. In some embodiments, for each type of target item, the determination module 230 can use the model recognition model corresponding to the target item to determine the model of the target item. For example, the determination module 230 can input the target three-dimensional scanning data of the target item into the model recognition model, and the model recognition model can output the model of the target item. In some embodiments, the determination module 230 can also obtain at least two reference models. The at least two reference models can correspond to at least two alternative models of the target item. The determination module 230 can determine the model of the target item from the at least two alternative models based on the target three-dimensional scanning data of the target item and the above-mentioned at least two reference models using the model recognition model. For more descriptions on determining the model of at least one target item, please refer to steps 330 and Figure 5 , I will not go into details here.

[0037] Generation module 240 can be used to generate a virtual reality model of the scanning room based on the three-dimensional scan data and the model of at least one target object, where the virtual reality model includes the model of the at least one target object. The virtual reality model of the scanning room can be a virtual architectural space model that simulates the interior of the scanning room. It can be presented using a virtual reality device, allowing the wearer of the virtual reality device to have a realistic understanding of the interior of the scanning room. In some embodiments, generation module 240 can first generate a three-dimensional model based on the three-dimensional scan data of the scanning room. The three-dimensional model of the scanning room can include a three-dimensional model representing the scanning room itself and a three-dimensional model representing the target object inside the scanning room. More details on generating the virtual reality model of the scanning room can be found in step 340 and will not be repeated here.

[0038] It should be understood that Figure 2 The system and its modules shown can be implemented in various ways. For example, in some embodiments, the system and its modules can be implemented by hardware, software, or a combination of software and hardware.

[0039] It should be noted that the above description of the system and its modules is for convenience of description only and is not intended to limit this specification to the scope of the embodiments. It is understood that those skilled in the art, after understanding the principles of the system, may arbitrarily combine the modules or form subsystems connected to other modules without departing from the principles. For example, in some embodiments, Figure 2 The above modules disclosed in the specification may be different modules in one system, or one module may realize the functions of two or more modules mentioned above. For example, each module may share a storage module, or each module may have its own storage module. Such variations are within the scope of protection of this specification. For another example, in some embodiments, one or more modules in the object recognition system 200 may be implemented by other systems. That is, the above one or more modules may not be included in the object recognition system 200. For example, the acquisition module 210 may be a module in other systems (for example, a radar or a camera).

[0040] Figure 3 is a flowchart of an exemplary object recognition method according to some embodiments of this specification. In some embodiments, one or more steps of process 300 may be performed in Figure 1 The scanning room 100 shown is implemented or composed of Figure 2 For example, the process 300 may be performed by a module within the object recognition system 200. Figure 3 As shown, the process 300 may include the following steps.

[0041] In step 310 , three-dimensional scanning data of the scanning room may be acquired. In some embodiments, step 310 may be performed by the processing device or acquisition module 210 .

[0042] The three-dimensional scanning data of the scanning room may be related to the internal spatial structure of the scanning room and one or more target objects located inside the scanning room. The internal spatial structure of the scanning room may include walls and floors. The one or more target objects located in the scanning room may include one or more of the imaging device 110, cabinet 120, computer 130, printer 140, air conditioner 150, spare parts cabinet 160, table 170, and chair 180. For example, the three-dimensional scanning data may show the position, shape, size, etc. of the objects in the scanning room. In some embodiments, the three-dimensional scanning data may be generated by a scanning device (e.g., Figure 1 The scanning device 190 described above is used to collect data.

[0043] In some embodiments, the three-dimensional scanning data may include point cloud data, depth data, a three-dimensional image, and / or a three-dimensional model. Each data point in the point cloud data may correspond to a physical point or an area of ​​the interior scene of the scanning chamber. The data points in the point cloud data may include information related to the corresponding physical point (or physical area), such as the location of the physical point, the object to which the physical point belongs, etc. In some embodiments, the point cloud data may be acquired by a sensor (e.g., LiDAR). For example, the sensor may emit laser pulses to scan the interior space of the scanning chamber. The laser pulses may be reflected by physical points in the interior space of the scanning chamber and return to the sensor. The sensor may generate point cloud data representing the scanning chamber based on one or more characteristics of the returned laser pulses. In some embodiments, during the point cloud data collection process, the sensor may rotate within a scanning angle range (e.g., 360 degrees, 180 degrees, 120 degrees) and scan the interior space of the scanning chamber at a specific scanning frequency (e.g., 10 Hz, 15 Hz, 20 Hz).

[0044] The depth data may include the distance from a point in the scanning room or a point on an object to the camera. Since the position of the camera is determined, three-dimensional data of the point or object in the scanning room can be obtained based on the depth data. The depth data can be acquired using a depth camera. In some embodiments, the three-dimensional scanning data of the scanning room can be generated based on multiple two-dimensional images. In some embodiments, the multiple two-dimensional images can be images taken in real time or in advance. The processing device can reconstruct the three-dimensional scanning data by three-dimensional reconstruction technology based on the multiple two-dimensional images taken in advance. Exemplary three-dimensional reconstruction technologies may include texture shape (SFT) method, light and shade reconstruction three-dimensional shape method, multi-view stereo (MVS) method, motion recovery structure (SFM) method, time of flight (ToF) method, structured light method, moiré method, etc., or any combination thereof.

[0045] In step 320 , at least one object recognition model may be used to identify at least one target object from the three-dimensional scan data. In some embodiments, step 320 may be performed by the processing device or the recognition module 220 .

[0046] In some embodiments, identifying the target object may include determining the type of at least one target object in the scanning room. For example, the processing device may use the object recognition model to identify one or more target objects in the scanning room, including imaging device 110, cabinet 120, computer 130, printer 140, air conditioner 150, spare parts cabinet 160, table 170, and chair 180. In some embodiments, identifying the target object may include determining target 3D scan data corresponding to the target object based on the 3D scan data of the scanning room. For example, the processing device may use the object recognition model to segment a portion corresponding to the target object from the 3D scan data of the scanning room to provide the target 3D scan data for the target object.

[0047] The object recognition model can identify at least one target object in the 3D scan data based on the 3D scan data. The input of the object recognition model can be the 3D scan data, and the output can be the type of at least one target object and / or the target 3D scan data. In some embodiments, different types of target objects can correspond to different object recognition models. The processing device can use multiple object recognition models to identify different target objects. For example, the processing device can use multiple object recognition models, such as an imaging device recognition model, a computer recognition model, and a cabinet recognition model, to respectively identify an imaging device, a computer, and a cabinet. In some embodiments, different types of target objects can correspond to a single object recognition model. The processing device can use the object recognition model to identify different types of target objects. For example, the processing device can use the object recognition model to identify imaging device 110, computer 130, and cabinet 120.

[0048] In some embodiments, the item recognition model can be a machine learning model. In some embodiments, the item recognition model can be a Faster R-CNN model.

[0049] In some embodiments, the processing device may input the 3D scan data of the scanning room into at least one object recognition model. The at least one object recognition model may output a category of at least one target object in the scene within the scanning room. In some embodiments, the at least one object recognition model may also output 3D scan data corresponding to each of the at least one target object. In some embodiments, the processing device may extract corresponding planar image data from the depth data based on the 3D scan data corresponding to the target object.

[0050] In some embodiments, the object recognition model can be obtained by training an initial model using the first training sample. In some embodiments, the object recognition model can be pre-trained by the manufacturer. In some embodiments, the object recognition model can be updated in real time. For example, if the object recognition model identifies an error, the processing device can input the correct category label and update the object recognition model based on the 3D scan data corresponding to the category label.

[0051] In some embodiments, an object recognition model can be obtained by a model training device through the following training process. The model training device can be a computer executing process 300 or other computer. The model training device can obtain an initial object recognition model. The model training device can obtain at least two first training samples. The first training sample can include sample three-dimensional scanning data and an object type label of a target object in the sample three-dimensional scanning data. The sample three-dimensional scanning data includes at least one target object. The object type label can be manually added or confirmed. The model training device trains the initial object recognition model based on the sample three-dimensional scanning data and the object type label of at least one target object to obtain an object recognition model.

[0052] In some embodiments, when an object recognition model is used to identify different types of target objects, the sample three-dimensional scanning data used to train its initial object recognition model may include three-dimensional scanning data of target objects of different types. In some embodiments, when an object recognition model is used to identify only one type of target object, the sample three-dimensional scanning data used to train its initial object recognition model may include only three-dimensional scanning data of target objects of one type. In some embodiments, when an object recognition model is used to identify only one type of target object, the sample three-dimensional scanning data used to train its initial object recognition model may include three-dimensional scanning data of target objects of multiple types. When the target object in the sample three-dimensional scanning data is of the type of object to be identified by the object recognition model, its label may be a positive label or an object type label. Conversely, when the target object in the sample three-dimensional scanning data is not of the type of object to be identified by the object recognition model, its label may be a negative label.

[0053] Step 330 , determining the model of at least one target item. In some embodiments, step 330 may be performed by the processing device or the determination module 230 .

[0054] In some embodiments, the processing device may receive a manually input model number of at least one target item.

[0055] In some embodiments, the processing device may obtain planar image data of at least one target item, such as a planar image captured by a two-dimensional camera. Furthermore, the processing device may use optical character recognition (OCR) technology to identify characters related to the model number in the planar image, thereby determining the model number of the target item.

[0056] In some embodiments, for each type of target item, the processing device may use the model recognition model corresponding to the target item to determine the model of the target item. For example, the processing device may input the target three-dimensional scanning data of the target item into the model recognition model, and the model recognition model may output the model of the target item. In some embodiments, the processing device may also obtain at least two reference models. The at least two reference models may correspond to at least two alternative models of the target item. The processing device may use the model recognition model to determine the model of the target item from the at least two alternative models based on the target three-dimensional scanning data of the target item and the above-mentioned at least two reference models. For more descriptions of the model recognition model and its use, please refer to Figure 5 .

[0057] Step 340 : Generate a virtual reality model of the scanning room based on the three-dimensional scanning data and the model of at least one target object. In some embodiments, step 340 may be performed by the processing device or the generation module 240 .

[0058] The virtual reality model of the scanning room may be a virtual architectural space model that simulates the interior scene of the scanning room, which may be presented using a virtual reality device so that a wearer of the virtual reality device can have a realistic understanding of the interior scene of the scanning room.

[0059] In some embodiments, the processing device may first generate a three-dimensional model based on the three-dimensional scan data of the scanning room. The three-dimensional model of the scanning room may include a three-dimensional model representing the scanning room itself and a three-dimensional model representing the target objects within the scanning room. For example, the processing device may process the three-dimensional scan data using three-dimensional modeling techniques (such as box modeling, polygonal modeling, surface modeling, photogrammetry, Boolean modeling, procedural modeling, modular modeling, etc.) to construct a three-dimensional model representing the scanning room. In some embodiments, three-dimensional models of target objects of different models may be pre-generated. Based on the model of the target object determined in step 330, the processing device may obtain a three-dimensional model corresponding to that model to generate the three-dimensional model of the scanning room. Furthermore, the processing device may utilize virtual reality technology to expand and process the three-dimensional model of the scanning room, assigning material and texture features to the interior of the scanning room and one or more target objects located within the scanning room, thereby generating a three-dimensional virtual view of the scanning room. The processing device may perform three-dimensional rendering on the three-dimensional virtual view of the scanning room to present a vivid three-dimensional virtual rendering, i.e., a virtual reality model. As just an example, the processing device may render corresponding areas in the three-dimensional virtual space using the same color as that of various areas in the scanning room, so that the constructed three-dimensional virtual space is as close to the scanning room as possible.

[0060] After the virtual reality model of the scanning room is generated, the processing device can determine the reference information contained therein. For example only, Figure 4 illustrative reference information included in a virtual reality model according to some embodiments of this specification is shown. Figure 4 The virtual reality model may include one or more of the model and feature information of the target object and the physical field distribution information of the scanning room.

[0061] In some embodiments, the model of the target item can be labeled with characters. For example, the virtual reality model can be labeled with the model information of one or more target items in the scanning room (e.g., imaging device 110, cabinet 120, computer 130, printer 140, air conditioner 150), making it easier for the wearer to understand the target items in the scanning room. In some embodiments, the model of the target item can be displayed directly or displayed when the user clicks or looks at the target item.

[0062] The characteristic information of the target object may include physical characteristic information, material information, whether it is movable, etc. The physical characteristic information may include electromagnetic characteristic information and / or magnetic field characteristic information. In some embodiments, the electromagnetic characteristic information and / or magnetic field characteristic information may include the maximum magnetic field strength that the target object can withstand, the magnetic field absorptivity, the magnetic field reflectivity, etc. The electromagnetic characteristic information and / or magnetic field characteristic information may also include the electromagnetic radiation intensity and magnetic field intensity generated by the target object. The material information may include the material of the target object, such as wood, plastic, ferromagnetic material, etc. Whether it is movable may include whether the target object is movable and / or whether it is easy to move. For example, the computer 130 can be moved, but the equipotential box cannot be moved.

[0063] In some embodiments, the processing device may obtain characteristic information of at least one target item based on the model of the at least one target item. For example, characteristic information corresponding to different models of the target item may be stored in a lookup table or database. The processing device may access the lookup table or database and retrieve the characteristic information of the target item based on the model of the target item. In some embodiments, the processing device may mark the characteristic information of the target item in text and / or digital form on or near the target item. In some embodiments, the characteristic information of the target item may be displayed directly or displayed when the user clicks or gazes at the target item. In some embodiments, the processing device may use color information on the target item to represent or mark material information or physical characteristic information.

[0064] In some embodiments, the processing device can obtain physical characteristic information of the target item based on the model of the target item. The processing device can also obtain the usage record of the target item. The usage record of the target item may include usage time, number of uses, usage frequency, maintenance records, etc. The maintenance record may further include the cause of damage, accessory replacement records, number of repairs, etc. Furthermore, the processing device can adjust the physical characteristic information of the target item based on the usage record and use the adjusted physical characteristic information as the characteristic information of the target item. For example, for a target item coated with an anti-magnetic layer, increased usage time and number of uses will cause the anti-magnetic coating to age, and its magnetic field absorption rate will increase. For another example, if the target item replaces accessories, its magnetic field absorption rate and reflectivity will change due to the accessory adjustment. In some embodiments, the physical characteristic information of the target item can be marked on or near the target object using characters. The physical characteristic information of the target item can be displayed directly or displayed when the user clicks or looks at the target item.

[0065] In some embodiments, the scanning room includes a nuclear magnetic resonance apparatus. The processing device can determine the physical field distribution in the scanning room when the nuclear magnetic resonance apparatus is in operation based on the three-dimensional scanning data, and the virtual reality model is further generated based on the physical field distribution. When the nuclear magnetic resonance apparatus is in operation, it will emit a magnetic field (for example, a main magnetic field generated by a main magnet) and an electromagnetic field (for example, a gradient field generated by a gradient system) into the scanning room where it is located. The physical field in this application includes at least one of a magnetic field and an electromagnetic field. For example, the physical field may include a magnetic field. For another example, the physical field may include the sum of a magnetic field and an electromagnetic field (for example, a maximum electromagnetic field). The physical field distribution may refer to the distribution of physical field intensities at different locations in the scanning room when the nuclear magnetic resonance apparatus is in operation. When the nuclear magnetic resonance apparatus is in operation, it will radiate physical fields to the surroundings, and these physical fields may be reflected or absorbed by other devices in the scanning room to form the final physical field distribution. In some embodiments, the processing device can determine the above-mentioned physical field distribution based on the physical field radiation generated by the operation of the nuclear magnetic resonance apparatus and the physical characteristic information of at least one target object.

[0066] For example, finite element analysis (FEA) can be used to perform magnetic resonance physics simulation to determine the physical field distribution in the scanning room where the MRI scanner is located. Physical field simulation based on FEA primarily decomposes a 2D or 3D environment into a series of nodes or points. In each calculation, the values ​​of adjacent nodes or points are calculated, and a series of different algorithms are used to iterate and determine the physical field distribution.

[0067] For another example, a physical field emission model of a nuclear magnetic resonance apparatus and the physical field characteristics of various regions in a scanning room can be obtained, and the physical field simulation model can be used to determine the physical field distribution information in the scanning room when the nuclear magnetic resonance apparatus is in operation. The physical field emission model can represent the physical field emission characteristics of the nuclear magnetic resonance apparatus during operation. For example, the physical field emission model can include the intensity of the physical field emitted by the apparatus at different distances during operation. In some embodiments, because the intensity of the physical field emitted by the nuclear magnetic resonance apparatus during operation varies, the processing device can determine the physical field emission model based on the maximum intensity of the physical field emitted by the nuclear magnetic resonance apparatus during operation. In some embodiments, the processing device can determine the physical field emission model based on the performance parameters of one or more components of the nuclear magnetic resonance apparatus that emit magnetic fields and / or electromagnetic fields. For example, the processing device can obtain the magnetic field intensity emitted by the main magnet at different locations and / or the maximum electromagnetic field intensity (i.e., the maximum gradient field intensity) that the gradient system can emit at different locations during operation. For each location in the scanning room, the processing device can use the sum of the magnetic field intensity and / or the maximum electromagnetic field intensity at that location as the physical field intensity at that location, thereby determining the physical field emission model. The physical field characteristics of a region may include at least one of the physical field absorption characteristics and physical field reflection characteristics of the region, for example, magnetic field absorptivity and magnetic field reflectivity. The physical field simulation model may be a model for determining the physical field distribution, and the physical field simulation model may include a convolutional neural network (CNN), a residual network (ResNet), etc. In some embodiments, the physical field emission model of the nuclear magnetic resonance apparatus and the physical field characteristics of each region in the scanning room may be input into the physical field simulation model, and the physical field simulation model may output information related to the physical field distribution. For example, the physical field simulation model may output the physical field intensity corresponding to each region. The physical field distribution may be determined based on the physical field intensity corresponding to each region.

[0068] In some embodiments, the processing device can display the above-mentioned information related to the physical field distribution in the virtual reality model. In some embodiments, a physical field intensity contour map can be used to represent the physical field distribution in the scanning room when the nuclear magnetic resonance apparatus is running. In the physical field intensity contour map, position points with equal or similar physical field intensities in the scanning room will be connected to form a closed curve, and the corresponding physical field intensities will be marked on different closed-loop curves. In some embodiments, the physical field intensity contour map can be a three-dimensional contour map or a two-dimensional contour map. In some embodiments, different colors can be set for different areas in the virtual reality model based on the physical field distribution to display the physical field distribution.

[0069] In some embodiments, the scanning room may include other imaging devices besides an MRI device, such as a computed tomography (CT) device, an X-ray scanner, or other imaging devices that emit radiation that could be harmful to the human body or electronic devices. The processing device may determine the radiation distribution in the scanning room when the other imaging devices are operating, and generate a virtual reality model based on the radiation distribution.

[0070] In some embodiments, the processing device or display module can display the virtual reality model using a virtual reality device. Exemplary virtual reality devices may include AR devices, VR devices, etc. In some embodiments, a user can view the virtual reality model through the virtual reality device. For example, a user can wear VR glasses to view the model and physical characteristics of one or more target items in the virtual reality model. For another example, a user can use the virtual reality device to view the distribution of the physical field in the scanning room and the physical field strength at the location of the target item.

[0071] When arranging the nuclear magnetic resonance imaging equipment and other equipment in the scanning room, the staff can use the virtual reality device to view the physical field distribution generated when the nuclear magnetic resonance imaging equipment is placed in different positions to see whether it will affect the target equipment. In some embodiments, when the position of the nuclear magnetic resonance imaging equipment is fixed, the staff can use the virtual reality device to view the physical field distribution of the nuclear magnetic resonance imaging equipment and adjust the positions of other equipment based on the physical field distribution. In some embodiments, the physical field distribution can be determined by simulating the operating conditions of the nuclear magnetic resonance imaging equipment. In some embodiments, the physical field distribution can be generated by measurement during the actual operation of the nuclear magnetic resonance imaging equipment. When the physical field strength at the location of the target item is greater than or close to the maximum physical field strength, for example, when the physical field strength exceeds the maximum physical field strength that the target item can withstand or 80% or 90% of the maximum physical field strength, the virtual reality device can issue a voice warning and / or display a warning message in the virtual reality model.

[0072] In some embodiments, a user can enter a scanning room while an MRI machine is operating and, using wearable devices (e.g., smart clothing), experience the strength of the physical field at different locations within the scanning room. In some embodiments, a user can enter a scanning room while an MRI machine is operating and, using a virtual reality device, can issue alerts based on the user's real-time location and the distribution of the physical field. For example, if a user approaches an area with a strong physical field, the virtual reality device can issue a warning. In some embodiments, the virtual reality device can directly sound an alarm to alert the user wearing the device. In some embodiments, the virtual reality device can send a prompt to a monitoring person or monitoring device, prompting the appropriate device or person to take timely action. In some embodiments, a user can use a wearable device and a virtual reality device to simulate the process of entering a scanning room. For example, the virtual reality device can present a virtual reality model of the scanning room to the user, and the user can use controls to control an avatar within the model. As the avatar moves to different locations within the scanning room, the wearable device can provide specific feedback (e.g., applying a certain force) to the user, allowing the user to experience the strength of the physical field at different locations.

[0073] In some embodiments, the virtual reality device uses different colors or numbers to indicate the physical field strength in different areas and the maximum physical field strength that an object can withstand. This allows users to easily check in the virtual reality model whether the physical field strength at the object's location exceeds the maximum physical field strength that the object can withstand. Each color or number can represent a physical field strength value or range.

[0074] In some embodiments, the processing device can identify magnetic objects made of materials such as metal that are not suitable for placement near an MRI scanner. The virtual reality device can determine a space suitable for placing magnetic objects in the virtual reality model. For example, a space where the physical field strength is below a threshold. In some embodiments, the space can be represented by a cube or a plane frame. The virtual reality device can instruct the user through voice or visual instructions to move magnetic objects such as a wheelchair to the above-mentioned space. In some embodiments, if the processing device determines that the above-mentioned space does not exist in the scanning room 100, the virtual reality device can instruct the user to move the magnetic objects out of the scanning room.

[0075] It should be noted that the above description of process 300 is for illustration and purpose only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and alterations to process 300 under the guidance of this specification. However, such modifications and alterations are still within the scope of this specification.

[0076] Figure 5FIG. 5 is a flow chart illustrating an exemplary method of identifying an object model using a model recognition model according to some embodiments of this specification. In some embodiments, process 500 may be executed by a processing device or a determination module 240. Figure 5 As shown, process 500 may include the following steps.

[0077] In step 510 , the target 3D scan data of the target object may be input into the model recognition model. In some embodiments, step 510 may be performed by the processing device or the determination module 230 .

[0078] In some embodiments, the target 3D scan data of the target object can be obtained from the 3D scan data of the scanning room. Figure 3 For example, the processing device can obtain target 3D scanning data of the target object based on the object recognition model. For another example, the processing device can segment the target 3D scanning data corresponding to the target object from the 3D scanning data of the scanning room based on the target object segmentation model.

[0079] In some embodiments, the processing device may pre-process the target three-dimensional scan data of the target object. In some embodiments, the pre-processing may include interpolation, normalization, projection, etc. of the target three-dimensional scan data. Projection refers to projecting the target three-dimensional scan data of the target object to a standard angle. For example, the target three-dimensional scan data may be collected from the side of the target object, and the processing device may project the target three-dimensional scan data of the target object to obtain the target three-dimensional scan data corresponding to the front of the object. In some embodiments, the processing device may obtain multiple sets of projection data of the target object at multiple standard angles. The processing device may use the projection data as processed target three-dimensional scan data and input it into the model recognition model. Since the point cloud data or depth map data of the target object is usually collected from a specific angle, there may be cases where it cannot be matched with an object of a known model. After projecting the above-mentioned scan data to obtain its projection data at its standard angle, the matching problem in this case can be solved.

[0080] In some embodiments, each type of item may correspond to a model identification model. For example, computer 130 may correspond to a computer model identification model, and air conditioner 150 may correspond to an air conditioner model identification model. In some embodiments, the model identification model may be a machine learning model. In some embodiments, the model identification model may be a neural network model. For example, the model identification model may be a Faster R-CNN model. In some embodiments, the model identification model and the item identification model may be different parts of the same model. For example, the model identification model may be connected to the item identification model via a fully connected layer.

[0081] In some embodiments, the model recognition model can extract features of the target 3D scan data of the target object and determine the model of the target object based on the features. In some embodiments, the features of the target 3D scan data can be extracted from projection data at one or more standard angles.

[0082] In some embodiments, a model recognition model can determine the model of a target item based on the target 3D scan data of the target item and at least two reference models. The input to the model recognition model may include the target 3D scan data of the target item and at least two reference models. The output of the model recognition model may include the model of the target item. The at least two reference models may correspond to at least two alternative models of the target item. The reference models may include reference 3D scan data of the alternative model items or 3D models constructed accordingly. The reference 3D scan data may be obtained through actual measurement or simulation. In some embodiments, the reference 3D scan data may include 3D scan data of the entire surface and / or interior of the alternative model items. In actual model recognition, due to the sparse 3D scan data of some target items, it may not be possible to extract sufficient effective features, making it difficult to accurately identify the model of the target item. When the input to the model recognition model includes reference models of the alternative models of the target item, the 3D scan data of the target item can be compared with the reference models of the alternative models, making it easier to determine the model of the target item from the alternative models.

[0083] In some embodiments, the model recognition model may include a feature extraction layer and a judgment layer (also referred to as an output layer). The feature extraction layer may extract extracted features of the target object and reference features of the at least two reference models based on the target 3D scan data of the target object and at least two reference models. In some embodiments, the feature extraction layer may include one or more convolutional layers. In some embodiments, the feature extraction layer may be a convolutional neural network (CNN). The input to the feature extraction layer may include the target 3D scan data of the target object and at least two reference models. The output of the feature extraction layer may include the extracted features of the target object and the reference features of the at least two reference models. In some embodiments, the extracted features of the target 3D scan data and the reference features of the reference models may be extracted from projection data at one or more common standard angles. The judgment layer may determine the model of the target object based on the extracted features of the target object and the reference features of the at least two reference models. The input to the judgment layer may include the extracted features of the target object and the reference features of the at least two reference models. The output of the judgment layer may include the model of the target object. In some embodiments, the output of the judgment layer may include one or more preliminary models of the target object and their corresponding confidence levels. The judgment layer may select the preliminary model with the highest confidence level as the model of the target object.

[0084] Taking computers as an example, the input of the computer model recognition model can also include Figure 5 Computer reference models 131, 132, and 133 are shown in FIG. Computer reference models 131, 132, and 133 correspond to three candidate computer models, respectively. The computer reference models can be generated based on reference 3D scan data of each candidate computer model. For example, computer reference model 131 may include point cloud data of a laptop computer. This point cloud data can be obtained through actual measurement or simulation.

[0085] In some embodiments, the model recognition model can be obtained by training an initial model using the second training sample. In some embodiments, the model recognition model can be pre-trained by the manufacturer. In some embodiments, the model recognition model can be updated in real time. For example, if the model recognition model identifies an error, the processing device can input the correct model label and update the model based on the 3D scan data corresponding to the model label.

[0086] In some embodiments, the model recognition model can be obtained by a model training device through the following training process. The model training device can be a computer that executes process 300 or other computer. The model training device can obtain an initial model recognition model. The model training device can obtain at least two second training samples.

[0087] In some embodiments, each of the at least two training samples may include sample target 3D scanning data of a sample target item, at least two sample reference models, and an item model label, wherein the at least two sample reference models correspond to the at least two candidate models of the target item. The item model labels may be manually labeled. The model training device may train the initial model recognition model based on the sample target 3D scanning data, the at least two sample reference models, and the item model labels.

[0088] The training of the initial model recognition model may include one or more iterations, and each iteration may include updating the model parameters of the initial model recognition model based on the training samples. In some embodiments, the optimization goal of the initial model recognition model training may include adjusting the model parameters so that the value of the loss function becomes smaller (for example, minimizing the value of the loss function). The initial model recognition model can output a predicted model. The loss function can be used to characterize the accuracy of the model predicted by the initial model recognition model. Exemplarily, the loss function may include a focal loss function, a logarithmic loss function, a cross entropy loss, etc.

[0089] In some embodiments, if the initial model recognition model meets the termination condition in a certain iteration, the training can be stopped. Exemplarily, the termination condition may include any one of the following or a combination thereof: the value of the loss function obtained in a certain iteration is less than a threshold, a certain number of iterations have been performed, the loss function converges (for example, the difference between the value of the loss function obtained in the previous iteration and the value of the loss function obtained in this iteration is within a preset threshold), etc. In some embodiments, when the iteration does not meet the termination condition, the model training device may further update the initial model recognition model for the next iteration according to a preset algorithm (for example, a back propagation algorithm). If the termination condition is met in the current iteration, the model training device can complete the training of the initial model recognition model.

[0090] In step 520 , the model recognition model outputs the predicted model of the target item. In some embodiments, step 520 may be performed by the processing device or the determination module 230 .

[0091] It should be noted that the above description of process 500 is for illustration and purpose only and does not limit the scope of application of this specification. Those skilled in the art may make various modifications and variations to process 500 under the guidance of this specification. However, such modifications and variations are still within the scope of this specification.

[0092] In some embodiments of the present specification, a virtual reality model of a scanning room is generated by an object recognition model and a model recognition type. The beneficial effects that may be brought about by the embodiments of the present specification include but are not limited to: (1) after determining the type of the object, the model recognition model corresponding to the object type is used to identify the model of the object, which can improve the accuracy of model recognition; (2) in the process of identifying the model of the object using the model recognition model, when determining the model of the object based on at least two reference models corresponding to at least two alternative models, even when the three-dimensional scanning data of the actual scanned object is relatively sparse, such as a depth map, the model of the object can be more accurately identified; (3) after obtaining the model of the object, the physical feature information of the object can be automatically obtained based on the model of the object, so that the physical feature information of the object can be easily obtained without manual search; (4) a virtual reality model can be generated based on the three-dimensional spatial model of the scanning room, the physical feature information of the object, and the physical field radiation generated by the operation of the nuclear magnetic resonance apparatus, and presented on the virtual reality device, which can facilitate the staff to view the physical field distribution of the scanning room.

[0093] While the basic concepts have been described above, it will be apparent to those skilled in the art that the detailed disclosure is merely illustrative and does not limit this specification. Although not explicitly stated herein, various modifications, improvements, and revisions to this specification may be made by those skilled in the art. Such modifications, improvements, and revisions are suggested in this specification and remain within the spirit and scope of the exemplary embodiments of this specification.

[0094] This specification also uses specific terms to describe the embodiments of this specification. For example, "one embodiment," "an embodiment," and / or "some embodiments" refer to a feature, structure, or characteristic associated with at least one embodiment of this specification. Therefore, it should be emphasized and noted that references to "one embodiment," "an embodiment," or "an alternative embodiment" two or more times in different locations in this specification do not necessarily refer to the same embodiment. Furthermore, certain features, structures, or characteristics of one or more embodiments of this specification may be appropriately combined.

[0095] In addition, unless expressly stated in the claims, the order of the processing elements and sequences, the use of alphanumeric characters, or the use of other names described in this specification are not intended to limit the order of the processes and methods of this specification. Although the above disclosure discusses some of the invention embodiments currently considered useful through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments. On the contrary, the claims are intended to cover all modifications and equivalent combinations that are consistent with the spirit and scope of the embodiments of this specification. For example, although the system components described above can be implemented by hardware devices, they can also be implemented only by software solutions, such as installing the described system on an existing server or mobile device.

[0096] Similarly, it should be noted that, in order to simplify the presentation of this specification and thus facilitate understanding of one or more embodiments of the invention, the foregoing descriptions of the embodiments of this specification sometimes combine multiple features into a single embodiment, figure, or description thereof. However, this disclosure method does not imply that the subject matter of this specification requires more features than those recited in the claims. In fact, an embodiment may have fewer features than all of the features of a single disclosed embodiment.

[0097] In some embodiments, numbers are used to describe the quantity of components and attributes. It should be understood that such numbers used in the description of the embodiments are modified by the modifiers "about", "approximately" or "substantially" in some examples. Unless otherwise stated, "about", "approximately" or "substantially" indicate that the numbers are allowed to vary by ±20%. Accordingly, in some embodiments, the numerical parameters used in the description and claims are approximate values, which may change according to the required characteristics of individual embodiments. In some embodiments, the numerical parameters should take into account the specified significant digits and adopt the general method of retaining digits. Although the numerical domains and parameters used to confirm the breadth of their range in some embodiments of this specification are approximate values, in specific embodiments, the settings of such numerical values ​​are as accurate as possible within the feasible range.

[0098] Each patent, patent application, patent application publication, and other materials, such as articles, books, specifications, publications, and documents, cited in this specification is hereby incorporated by reference in its entirety. This includes application history documents that are inconsistent with or conflict with the content of this specification, as well as documents (currently or subsequently attached to this specification) that limit the broadest scope of the claims of this specification. It should be noted that if the descriptions, definitions, and / or terminology used in the accompanying materials are inconsistent or conflicting with the content of this specification, the descriptions, definitions, and / or terminology used in this specification will control.

[0099] Finally, it should be understood that the embodiments described in this specification are intended only to illustrate the principles of the embodiments of this specification. Other variations may also fall within the scope of this specification. Therefore, by way of example and not limitation, alternative configurations of the embodiments of this specification may be considered consistent with the teachings of this specification. Accordingly, the embodiments of this specification are not limited to the embodiments explicitly described and illustrated in this specification.

Claims

1. A method for identifying an object, executed by at least one processor, characterized in that: The scanning room includes a nuclear magnetic resonance apparatus, and the method includes: Acquire three-dimensional scanning data of the scanning room; Identifying at least one target object from the three-dimensional scan data using at least one object recognition model; Determining the model of the at least one target item, and based on the model of the at least one target item, acquiring physical characteristic information of the at least one target item, the physical characteristic information including electromagnetic characteristic information and / or magnetic field characteristic information; determining, based on the three-dimensional scanning data, a physical field distribution within the scanning chamber when the nuclear magnetic resonance apparatus is in operation; and Based on the three-dimensional scanning data, the model of the at least one target item, the physical characteristic information of the at least one target item, and the physical field distribution, a virtual reality model of the scanning room is generated, wherein the virtual reality model displays the model of the at least one target item, the physical characteristic information of the at least one target item, and the physical field distribution.

2. The method according to claim 1, wherein The acquiring of the physical feature information of the at least one target object based on the model of the at least one target object includes: Based on the model of the at least one target item, obtaining physical feature information of the at least one target item; and The physical feature information is adjusted based on the usage record of the at least one target item, and the adjusted physical feature information is used as the physical feature information of the at least one target item.

3. The method according to claim 1, wherein Determining the model of the at least one target item includes: Acquiring target three-dimensional scanning data of the at least one target object; and For each of the at least one target object, the model of the target object is determined based on the target three-dimensional scanning data of the target object and using a model recognition model corresponding to the target object.

4. The method according to claim 3, wherein The acquiring target three-dimensional scanning data of the at least one target object comprises: Segmenting initial three-dimensional scan data of the at least one target object from the three-dimensional scan data; and The initial three-dimensional scanning data is pre-processed to generate the target three-dimensional scanning data.

5. The method according to claim 3, wherein The determining the model of the target object based on the target three-dimensional scanning data of the target object and using the model recognition model corresponding to the target object includes: Acquire at least two reference models, where the at least two reference models correspond to at least two candidate models of the target object; Based on the target three-dimensional scanning data of the target object and at least two reference models, the model of the target object is determined from the at least two candidate models using the model recognition model.

6. The method according to claim 5, wherein The model recognition model corresponding to the target item is obtained through the following training process: Obtain an initial model recognition model; Obtain at least two training samples, each training sample including sample 3D scanning data of a sample target object, at least two sample reference models, and an item model label, wherein the at least two sample reference models correspond to the at least two candidate models of the target object; The initial model recognition model is trained using the at least two training samples to obtain the model recognition model.

7. An object identification system, characterized in that: The system includes at least one processor and at least one memory; The at least one memory is for storing computer instructions; The at least one processor is configured to execute at least part of the computer instructions to implement the method according to any one of claims 1 to 6.

8. An object identification system, characterized in that: The scanning room includes a nuclear magnetic resonance apparatus, and the system includes: An acquisition module, used for acquiring three-dimensional scanning data of the scanning room; an identification module, configured to identify at least one target object from the three-dimensional scan data using at least one object identification model; a determination module, configured to determine the model of the at least one target object, and based on the model of the at least one target object, obtain physical characteristic information of the at least one target object, the physical characteristic information including electromagnetic characteristic information and / or magnetic field characteristic information; and determine, based on the three-dimensional scanning data, a physical field distribution within the scanning chamber when the nuclear magnetic resonance apparatus is in operation; and a generation module for generating a virtual reality model of the scanning room based on the three-dimensional scanning data, the model of the at least one target item, the physical characteristic information of the at least one target item, and the physical field distribution, wherein the virtual reality model displays the model of the at least one target item, the physical characteristic information of the at least one target item, and the physical field distribution.

Images