Fixed asset change state acquisition method and device

Abstract

The invention discloses a fixed asset change state acquisition method and device, which are applied to the field of financial science and technology, and the method comprises the steps: carrying out the motion amplification of a collected fixed asset video, which comprises a starting frame, to obtain a reconstruction frame; according to the starting frame and the reconstruction frame, an optical flow feature map is obtained, and the optical flow feature map comprises optical flow features representing dynamic changes of the fixed assets; and inputting the optical flow feature map into a pre-constructed fixed asset identification model to obtain a target fixed asset change state. According to the method, the optical flow feature map is obtained through the collected asset video, the optical flow feature map is input into the fixed asset identification model, and the change state of the fixed assets is accurately identified through the model, so that the change state of the fixed assets is accurately and efficiently obtained, and the safety of credit assets is improved.

Description

Technical Field

[0001] This invention relates to the field of financial technology, and in particular to a method and apparatus for obtaining the changing status of fixed assets. Background Technology

[0002] Credit business is the most important asset business for commercial banks and other related enterprises, and it is also the main source of profit for commercial banks. From loan application, investigation, approval, to disbursement, each step is closely linked. Before disbursement, investigations are conducted on the borrower's assets, especially fixed assets, to verify their repayment ability, which plays a crucial role in the security of credit assets.

[0003] Currently, the status of fixed assets, such as a company's factory buildings, is reported by loan applicants and verified on-site by auditors. This method is not only inefficient but also fails to accurately reflect changes in fixed assets. If current fixed assets meet the threshold for a loan, but the company's fixed assets have been significantly reduced in the past six months due to poor management, lending to such a company still poses a credit security risk. Therefore, accurately and efficiently obtaining information on changes in fixed assets plays a crucial role in ensuring credit security. Summary of the Invention

[0004] This invention provides a method for obtaining the status of fixed assets changes, so as to achieve accurate and efficient acquisition of the status of fixed assets changes.

[0005] According to a first aspect of the present invention, a method for obtaining the changing state of fixed assets is provided, comprising: performing motion magnification on a collected fixed asset video to obtain a reconstructed frame, wherein the fixed asset video includes a start frame;

[0006] An optical flow feature map is obtained based on the starting frame and the reconstructed frame, wherein the optical flow feature map contains optical flow features representing the dynamic changes of fixed assets;

[0007] The optical flow feature map is input into a pre-built fixed asset identification model to obtain the change status of the target fixed asset.

[0008] According to another aspect of the present invention, a device for acquiring the changing state of fixed assets is provided, comprising: a motion magnification module for performing motion magnification on a collected fixed asset video to acquire a reconstructed frame, wherein the fixed asset video includes a start frame;

[0009] An optical flow feature map acquisition module is used to acquire an optical flow feature map based on the starting frame and the reconstructed frame, wherein the optical flow feature map contains optical flow features representing the dynamic changes of fixed assets;

[0010] The fixed asset change status acquisition module is used to input the optical flow feature map into a pre-built fixed asset identification model to obtain the change status of the target fixed asset.

[0011] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: at least one processor; and

[0012] A memory communicatively connected to the at least one processor; wherein,

[0013] The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the method described in any embodiment of the present invention.

[0014] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the method described in any embodiment of the present invention.

[0015] According to another aspect of the present invention, a computer program product is provided, comprising a computer program that, when executed by a processor, implements the method described in any embodiment of the present invention.

[0016] The beneficial technical effect of this invention is that it obtains optical flow feature maps by collecting asset videos, and inputs the optical flow feature maps into a fixed asset identification model. The model accurately identifies the changing state of fixed assets, thereby accurately and efficiently obtaining the changing state of fixed assets and improving the security of credit assets.

[0017] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a flowchart of a method for obtaining the changing status of fixed assets according to Embodiment 1 of the present invention;

[0020] Figure 2 This is a schematic diagram illustrating an application scenario of the method for obtaining the status of fixed asset changes according to Embodiment 1 of the present invention.

[0021] Figure 3 This is a schematic diagram illustrating the principle of motion amplification according to Embodiment 1 of the present invention;

[0022] Figure 4 This is a schematic diagram of the training process of the fixed asset change status identification model provided in Embodiment 1 of the present invention;

[0023] Figure 5 This is a flowchart of a method for obtaining the changing status of fixed assets according to Embodiment 2 of the present invention;

[0024] Figure 6 This is a schematic diagram of a device for obtaining the changing status of fixed assets according to Embodiment 3 of the present invention;

[0025] Figure 7 This is a schematic diagram of the structure of the electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0026] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0027] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices. Additionally, the collected information is information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure, and application of related data all comply with the relevant laws, regulations, and standards of the relevant countries and regions, necessary confidentiality measures have been taken, public order and good morals have not been violated, and corresponding operation interfaces are provided for users to choose to authorize or refuse.

[0028] Example 1

[0029] Figure 1 This is a flowchart illustrating a method for obtaining the changing status of fixed assets according to Embodiment 1 of the present invention. This embodiment is applicable to situations where customer satisfaction is accurately assessed based on micro-expressions. This method can be executed by a device for obtaining the changing status of fixed assets, which can be implemented in hardware and / or software. Figure 1 As shown, the method includes:

[0030] Step S101: Motion magnification is performed on the collected fixed asset video to obtain reconstructed frames.

[0031] Optionally, the acquired fixed asset video is subjected to motion magnification to obtain reconstructed frames, including: analyzing the fixed asset video to extract the start frame and peak frame; obtaining the first shape representation information and texture representation information in the start frame, and the second shape representation information in the peak frame; performing shape representation motion magnification processing based on the first shape representation information and the second shape representation information by a modulator to obtain magnified shape representation information; and constructing reconstructed frames based on the texture representation information and the magnified shape representation information.

[0032] Specifically, the fixed asset videos collected in this embodiment can be remote sensing satellite videos or videos taken by drones. Regardless of the type, the videos need to have a relatively long time span because fixed assets, such as factory buildings, typically have long change cycles. Therefore, judgment needs to be made based on video information collected over a long period. Furthermore, to avoid losing information about subtle changes, this embodiment performs motion magnification processing on the fixed asset videos to better capture subtle changes and distinguish them from noise, mitigating problems such as distortion, excessive noise, and excessive blurring in the magnified results. Figure 2 The diagram shows an application scenario of a satisfaction evaluation method based on micro-expression recognition. It mainly involves motion magnification, optical flow feature acquisition, and backbone network. Of course, this embodiment is only an example and does not limit the relevant content involved in the process of obtaining the status of fixed assets changes.

[0033] Among them, such as Figure 3The diagram illustrates the principle of motion magnification. In this embodiment, after acquiring fixed asset video within a specified time range, the acquired fixed asset video is analyzed to extract the start frame and peak frame. Since each frame contains both shape representation information and texture representation information, shape representation information tends to represent geometric edges and boundaries, while texture representation information tends to represent color and attributes. However, during fixed asset movement, shape representation information is more closely related to the asset's movement, while texture representation information provides relatively less effective information, even less than the impact of noise. Therefore, this embodiment only amplifies the shape representation information, specifically through a modulator, while the decoder reconstructs the amplified frame. For example, when the first shape representation information g in the start frame is acquired... t-1 The texture representation information is f t-1 The second shape representation information in the peak frame is g. t Given that the amplification factor is α, the modulator specifically uses the following formula (1) to obtain the amplified shape representation information:

[0034] G(g t-1 ,g t ,α)=g t-1 +α (g t -g t-1 (1)

[0035] After obtaining the magnified shape representation information G, a decoder will be used to determine the magnified shape representation information G and the texture representation information f. t-1 The reconstruction frame is constructed. Of course, this embodiment is only an example and does not limit the specific method of obtaining the reconstruction frame. As long as the fixed asset information can be amplified, it is within the protection scope of this application.

[0036] Step S102: Obtain optical flow feature maps based on the starting frame and the reconstructed frame.

[0037] Optionally, an optical flow feature map is obtained based on the start frame and the reconstructed frame, including: obtaining the horizontal and vertical optical flow velocities of the same pixel from the start frame to the reconstructed frame; using the horizontal and vertical optical flow velocities as optical flow features; and constructing an optical flow feature map based on the optical flow features.

[0038] Optionally, obtaining the horizontal and vertical optical flow velocities of the same pixel from the start frame to the reconstructed frame includes: obtaining the horizontal and vertical light intensity changes of the same pixel from the start frame to the reconstructed frame; obtaining the interval time between the start frame and the reconstructed frame; determining the horizontal optical flow velocity based on the interval time and the horizontal light intensity change, and determining the vertical optical flow velocity based on the interval time and the vertical light intensity change.

[0039] Specifically, since optical flow feature maps can describe pixel changes between two frames of images and capture subtle changes in fixed assets, this embodiment uses optical flow features to describe the spatiotemporal information of fixed assets. Optical flow features can capture the temporal information of fixed assets from a motion perspective, describing their dynamic changes. For example, the change in horizontal illumination intensity of the same pixel from the starting frame to the reconstructed frame is δa, and the change in vertical illumination intensity is δb. The interval between the starting frame and the reconstructed frame is t. Based on the interval and the change in horizontal illumination intensity, the horizontal optical flow velocity u = δa / t is determined, and the vertical optical flow velocity v = δb / t is determined. Therefore, the optical flow feature can be specifically described as [u, v]. Of course, this embodiment only uses one pixel as an example; the calculation method for the optical flow feature of other pixels is roughly the same, and will not be elaborated further in this embodiment.

[0040] It should be noted that, in this embodiment, after obtaining the optical flow features corresponding to each pixel, an optical flow feature map is constructed based on the obtained optical flow features. Therefore, the optical flow feature map obtained in this embodiment mainly contains optical flow features representing the dynamic changes of the fixed asset, that is, it contains the spatiotemporal information of the fixed asset captured from the perspective of motion. Of course, this embodiment is only an example and does not limit the specific method of obtaining the optical flow feature map. As long as it can represent the dynamic spatiotemporal information of the fixed asset, it is within the scope of protection of this application.

[0041] Step S103: Input the optical flow feature map into the pre-built specified asset identification model to obtain the change status of the target fixed asset.

[0042] Optionally, the fixed asset identification model includes a backbone network and a linear classifier; inputting the optical flow feature map into the pre-trained fixed asset identification model to obtain the target fixed asset change state includes: inputting the optical flow feature into the backbone network to obtain fixed asset features; inputting the fixed asset features into the linear classifier to obtain the probability value of each fixed asset change state, and taking the fixed asset change state corresponding to the maximum probability value as the target fixed asset change state.

[0043] In this embodiment, after obtaining the optical flow feature map, it is input into the backbone network of the fixed asset identification model, and the backbone network extracts the specified asset features from the optical flow feature map. Furthermore, to address the small sample size problem and reduce the risk of overfitting in the fixed asset change status identification task, the backbone network in this embodiment can adopt a convolutional neural network with a residual structure. The residual results can avoid the performance degradation problem of deep network structures. Moreover, the convolutional neural network uses 18 convolutional layers, making it a shallow network, which is more suitable for micro-expression recognition tasks in scenarios with sparse samples.

[0044] Furthermore, in this embodiment, after obtaining the fixed asset characteristics through the backbone network, the fixed asset characteristics are input into a linear classifier to obtain the probability values ​​of each fixed asset change state. For example, the probability value for a significant increase is 0.4, the probability value for a slight increase is 0.3, the probability value for no change is 0.1, and the probability value for a slight decrease is 0.2. Of course, this embodiment is only an example and does not limit the specific value of the probability value corresponding to each fixed asset change state. After obtaining the probability values ​​corresponding to each fixed asset change state, the fixed asset change state corresponding to the highest probability value, i.e., a significant increase, is taken as the target fixed asset change state, indicating that the company is currently carrying out large-scale factory expansion.

[0045] Optionally, before inputting the optical flow feature map into the pre-built fixed asset identification model to obtain the change status of the target fixed asset, the method further includes: obtaining sample optical flow feature maps, and using data augmentation to perform sample augmentation on the sample optical flow feature maps to obtain training feature maps; using the training feature maps to train the backbone network, and determining that the backbone network training is complete when the contrast loss function value of the backbone network reaches the preset requirement.

[0046] Specifically, in this embodiment, before using the fixed asset identification model to identify optical flow features, the backbone network of the fixed asset identification model is first trained. During training, data-augmented samples are first trained through the backbone network to obtain feature representations. These representations are then further propagated through the projection network, and the supervised contrastive loss is calculated using its output projection. The use of the supervised contrastive loss function helps to appropriately reduce the impact of unnecessary information on model training when the number of training samples is small. This method can more effectively capture key information in fixed assets and achieve accurate classification of the changing states of fixed assets.

[0047] Among them, such as Figure 4 The diagram illustrates the training process of the fixed asset change status recognition model. The motion magnification and optical flow feature processing methods involved in processing the sample images are largely the same as those in the fixed asset change status recognition process described above, and will not be repeated here. However, after obtaining the sample optical flow feature maps during training, data augmentation is used to enhance the sample optical flow feature maps and obtain training feature maps. Specifically, data augmentation employs a series of random transformations to generate enhanced images, thereby constructing positive and negative sample pairs. Training feature maps are then obtained based on these constructed positive and negative sample pairs. This approach aims to force the model to ignore surface factors and focus on learning inherently consistent structural information, such as translation, color, and rotation invariance.

[0048] Furthermore, the training process involves a contrastive loss function. The core objective of this function is to learn an effective representation space where similar samples (samples of the same class) are closer together, while dissimilar samples (samples of different classes) are farther apart. During training, a batch of samples is used, each with a corresponding label. For each sample, data augmentation techniques are used to generate multiple augmented versions. Thus, N samples in a batch become 2N pairs of image samples. In each pair, one image is called an "anchor point," and the other image is considered a positive sample if it belongs to the same class as the anchor point, and a negative sample otherwise. Cosine similarity is used to calculate the distance between the anchor point and the augmented sample; if y is a positive sample, this distance is minimized; if y is a negative sample, the distance is maximized. This method uses a contrastive loss function, and through backpropagation and gradient descent, the network continuously adjusts the weights to minimize this loss function, thereby learning an effective representation space that brings similar samples closer together and widens the distance between dissimilar samples. This learning method helps improve the model's generalization ability and its ability to recognize new samples.

[0049] In this embodiment, optical flow feature maps are obtained by collecting asset videos, and these feature maps are input into a fixed asset identification model. The model accurately identifies the changing state of fixed assets, thereby accurately and efficiently obtaining the changing state of fixed assets and improving the security of credit assets.

[0050] Example 2

[0051] Figure 5 This is a flowchart of a satisfaction evaluation method based on micro-expression recognition provided in Embodiment 2 of the present invention. Based on the above embodiments, after inputting the optical flow feature map into a pre-constructed fixed asset recognition model to obtain the target fixed asset's change status, the method further includes: determining a loan risk coefficient based on the target fixed asset's change status, and determining a lending strategy based on the loan risk coefficient, such as... Figure 5 As shown, the method includes:

[0052] Step S201: Motion magnification is performed on the collected fixed asset video to obtain reconstructed frames.

[0053] Optionally, the acquired fixed asset video is subjected to motion magnification to obtain reconstructed frames, including: analyzing the fixed asset video to extract the start frame and peak frame; obtaining the first shape representation information and texture representation information in the start frame, and the second shape representation information in the peak frame; performing shape representation motion magnification processing based on the first shape representation information and the second shape representation information by a modulator to obtain magnified shape representation information; and constructing reconstructed frames based on the texture representation information and the magnified shape representation information.

[0054] Step S202: Obtain optical flow feature maps based on the starting frame and the reconstructed frame.

[0055] Optionally, an optical flow feature map is obtained based on the start frame and the reconstructed frame, including: obtaining the horizontal and vertical optical flow velocities of the same pixel from the start frame to the reconstructed frame; using the horizontal and vertical optical flow velocities as optical flow features; and constructing an optical flow feature map based on the optical flow features.

[0056] Optionally, obtaining the horizontal and vertical optical flow velocities of the same pixel from the start frame to the reconstructed frame includes: obtaining the horizontal and vertical light intensity changes of the same pixel from the start frame to the reconstructed frame; obtaining the interval time between the start frame and the reconstructed frame; determining the horizontal optical flow velocity based on the interval time and the horizontal light intensity change, and determining the vertical optical flow velocity based on the interval time and the vertical light intensity change.

[0057] Step S203: Input the optical flow feature map into the pre-built specified asset identification model to obtain the change status of the target fixed asset.

[0058] Optionally, the fixed asset identification model includes a backbone network and a linear classifier; inputting the optical flow feature map into the pre-trained fixed asset identification model to obtain the target fixed asset change state includes: inputting the optical flow feature into the backbone network to obtain fixed asset features; inputting the fixed asset features into the linear classifier to obtain the probability value of each fixed asset change state, and taking the fixed asset change state corresponding to the maximum probability value as the target fixed asset change state.

[0059] Optionally, before inputting the optical flow feature map into the pre-built fixed asset identification model to obtain the change status of the target fixed asset, the method further includes: obtaining sample optical flow feature maps, and using data augmentation to perform sample augmentation on the sample optical flow feature maps to obtain training feature maps; using the training feature maps to train the backbone network, and determining that the backbone network training is complete when the contrast loss function value of the backbone network reaches the preset requirement.

[0060] Step S204: Determine the loan risk coefficient based on the change status of the target fixed assets, and determine the lending strategy based on the loan risk coefficient.

[0061] Specifically, this implementation method sets different loan risk coefficients for different fixed asset change states, as shown in Table 1 below as an example of credit risk coefficients:

[0062] Table 1

[0063] Fixed asset change status Loan risk coefficient Significant expansion 0 Small amplification 0 No change 0.1 Slight reduction 0.3 Significantly reduced 0.5 … …

[0064] In this implementation, if the target fixed asset change status is determined to be a significant reduction, a loan risk coefficient of 0.5 can be determined by referring to a table. A loan risk coefficient of 0.5 indicates a high risk of credit asset security when lending to the current enterprise. Therefore, this implementation determines the lending strategy based on the loan risk coefficient. For example, for an enterprise with a credit risk coefficient of 0.5, a higher level of collateral and a longer period of credit history without issues are required when lending. This strengthens the loan approval process to ensure the enterprise can repay the loan on time and improve the security of the credit assets. Of course, this implementation is merely an example and does not limit the specific content of the lending strategy. Furthermore, the higher the credit risk coefficient, the stricter the corresponding lending strategy will be.

[0065] In this embodiment, optical flow feature maps are obtained by collecting asset videos, and these feature maps are input into a fixed asset identification model. The model accurately identifies the changing state of fixed assets, thereby accurately and efficiently obtaining the changing state of fixed assets and improving the security of credit assets.

[0066] Example 3

[0067] Figure 6 This is a schematic diagram of a satisfaction assessment device based on micro-expression recognition provided in Embodiment 3 of the present invention. Figure 6 As shown, the device includes: a motion amplification module 310, an optical flow feature map acquisition module 320, and a fixed asset change status acquisition module 330;

[0068] The motion magnification module 310 is used to perform motion magnification on the acquired fixed asset video to obtain reconstructed frames, wherein the fixed asset video includes a start frame.

[0069] The optical flow feature map acquisition module 320 is used to acquire optical flow feature maps based on the start frame and the reconstructed frame, wherein the optical flow feature maps contain optical flow features representing the dynamic changes of fixed assets;

[0070] The fixed asset change status acquisition module 330 is used to input the optical flow feature map into the pre-built fixed asset identification model to obtain the change status of the target fixed asset.

[0071] Optional, a motion magnification module is used to analyze fixed asset videos and extract the start frame and peak frame;

[0072] Obtain the first shape representation information and texture representation information in the starting frame, and the second shape representation information in the peak frame;

[0073] Amplified shape representation information is obtained by performing shape representation motion amplification processing based on the first shape representation information and the second shape representation information using a modulator.

[0074] The reconstructed frame is built based on texture representation information and magnified shape representation information.

[0075] Optionally, an optical flow feature map acquisition module is used to acquire the horizontal and vertical optical flow velocities of the same pixel from the starting frame to the reconstructed frame.

[0076] Horizontal and vertical optical flow velocities are used as optical flow characteristics;

[0077] Construct an optical flow feature map based on optical flow characteristics.

[0078] Optionally, the optical flow feature map acquisition module is also used to acquire the horizontal and vertical light intensity changes of the same pixel from the starting frame to the reconstructed frame.

[0079] Obtain the time interval between the start frame and the reconstructed frame;

[0080] The horizontal optical flow velocity is determined based on the interval time and the change in horizontal illumination intensity, and the vertical optical flow velocity is determined based on the interval time and the change in vertical illumination intensity.

[0081] Optionally, the fixed asset identification model includes a backbone network and a linear classifier; a satisfaction assessment module is used to input optical flow features into the backbone network to obtain fixed asset features;

[0082] The fixed asset features are input into a linear classifier to obtain the probability value of each fixed asset change state, and the fixed asset change state corresponding to the highest probability value is taken as the target fixed asset change state.

[0083] Optionally, the device also includes a lending strategy determination module for determining the loan risk coefficient based on the change status of the target fixed assets;

[0084] The lending strategy is determined based on the loan risk coefficient.

[0085] Optionally, the device also includes a model training module for acquiring sample optical flow feature maps and using data augmentation to perform sample augmentation on the sample optical flow feature maps to obtain training feature maps.

[0086] The backbone network is trained using training feature maps. The training is considered complete when the contrastive loss function value of the backbone network reaches the preset requirement.

[0087] The satisfaction assessment device based on micro-expression recognition provided in the embodiments of the present invention can execute the satisfaction assessment method based on micro-expression recognition provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0088] Example 4

[0089] Figure 7 A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processors, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0090] like Figure 7 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12 or a random access memory (RAM) 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the ROM 12 or loaded from storage unit 18 into the RAM 13. The RAM 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, ROM 12, and RAM 13 are interconnected via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0091] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0092] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as methods for obtaining the changing status of fixed assets.

[0093] In some embodiments, the method for obtaining the status of changes in fixed assets may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via ROM 12 and / or communication unit 19. When the computer program is loaded into RAM 13 and executed by processor 11, one or more steps of the method for obtaining the status of changes in fixed assets described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the method for obtaining the status of changes in fixed assets by any other suitable means (e.g., by means of firmware).

[0094] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), payload-programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0095] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0096] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fibers, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0097] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0098] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0099] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a hosting product within the cloud computing service system to address the shortcomings of traditional physical hosts and VPS services, such as high management difficulty and weak business scalability.

[0100] Example 5

[0101] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the method for obtaining the status of changes in fixed assets as provided in any embodiment of this application.

[0102] In implementing the computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as C or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0103] It should be noted that in the embodiments of this application, certain software, components, models and other existing solutions in the industry may be mentioned. These should be regarded as exemplary and are only intended to illustrate the feasibility of implementing the technical solution of this application. However, it does not mean that the applicant has used or necessarily used the solution.

[0104] Note that the above description is merely a preferred embodiment of the present invention and the technical principles employed. Those skilled in the art will understand that the present invention is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in detail through the above embodiments, the present invention is not limited to the above embodiments, and may include many other equivalent embodiments without departing from the concept of the present invention, the scope of which is determined by the scope of the appended claims.

Claims

1. A method for obtaining the changing status of fixed assets, characterized in that the method... include: The collected fixed asset video is motion-zoomed to obtain reconstructed frames, wherein the fixed asset video includes a start frame; An optical flow feature map is obtained based on the starting frame and the reconstructed frame, wherein the optical flow feature map contains optical flow features representing the dynamic changes of fixed assets; The optical flow feature map is input into a pre-built fixed asset identification model to obtain the change status of the target fixed asset.

2. The method according to claim 1, characterized in that, The process of performing motion magnification on the acquired fixed asset video to obtain reconstructed frames includes: The starting frame and peak frame are extracted from the fixed asset video. Obtain the first shape representation information and texture representation information in the starting frame, and the second shape representation information in the peak frame; Amplified shape representation information is obtained by performing shape representation motion amplification processing based on the first shape representation information and the second shape representation information using a modulator. The reconstructed frame is constructed based on the texture representation information and the magnified shape representation information.

3. The method according to claim 1, characterized in that, The step of obtaining the optical flow feature map based on the starting frame and the reconstructed frame includes: Obtain the horizontal and vertical optical flow velocities of the same pixel from the starting frame to the reconstructed frame; The horizontal optical flow velocity and the vertical optical flow velocity are used as the optical flow characteristics; The optical flow feature map is constructed based on the optical flow features.

4. The method according to claim 3, characterized in that, The step of obtaining the horizontal and vertical optical flow velocities of the same pixel from the starting frame to the reconstructed frame includes: Obtain the horizontal and vertical light intensity changes of the same pixel from the starting frame to the reconstructed frame; Obtain the time interval between the starting frame and the reconstructed frame; The horizontal optical flow velocity is determined based on the interval time and the change in horizontal illumination intensity, and the vertical optical flow velocity is determined based on the interval time and the change in vertical illumination intensity.

5. The method according to claim 1, characterized in that, The fixed asset identification model includes a backbone network and a linear classifier; The step of inputting the optical flow feature map into a pre-trained fixed asset identification model to obtain the change status of the target fixed asset includes: The optical flow characteristics are input into the backbone network to obtain fixed asset characteristics; The fixed asset features are input into the linear classifier to obtain the probability value of each fixed asset change state, and the fixed asset change state corresponding to the highest probability value is taken as the target fixed asset change state.

6. The method according to claim 1, characterized in that, After inputting the optical flow feature map into the pre-built fixed asset identification model to obtain the change status of the target fixed asset, the method further includes: The loan risk coefficient is determined based on the change status of the target fixed assets. The lending strategy is determined based on the aforementioned loan risk coefficient.

7. The method according to claim 5, characterized in that, Before inputting the optical flow feature map into the pre-built fixed asset identification model to obtain the change status of the target fixed asset, the method further includes: Obtain sample optical flow feature maps, and use data augmentation to perform sample augmentation on the sample optical flow feature maps to obtain training feature maps; The backbone network is trained using the training feature maps. When the contrastive loss function value of the backbone network reaches a preset requirement, the backbone network training is considered complete.

8. A device for obtaining the changing status of fixed assets, characterized in that, The device includes: The motion magnification module is used to perform motion magnification on the acquired fixed asset video to obtain reconstructed frames, wherein the fixed asset video includes a start frame; An optical flow feature map acquisition module is used to acquire an optical flow feature map based on the starting frame and the reconstructed frame, wherein the optical flow feature map contains optical flow features representing the dynamic changes of fixed assets; The fixed asset change status acquisition module is used to input the optical flow feature map into a pre-built fixed asset identification model to obtain the change status of the target fixed asset.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the method of any one of claims 1-7.

11. A computer program product, characterized in that, Includes a computer program that, when executed by a processor, implements the method of any one of claims 1-7.

Images