Financial scene visual judgment system

Abstract

The invention relates to a financial scene visual judgment system, and the system comprises a visual monitoring mechanism which is installed in a two-dimensional code consumption scene, is used for carrying out the visual monitoring processing, and is used for obtaining and outputting a corresponding on-site monitoring image; the information conversion equipment is arranged at a far-end monitoring end of the scene and is used for intelligently identifying whether the field monitoring image is a RAW image or not by adopting an intelligent identification model based on multiple pieces of scene data of the two-dimensional code consumption scene and multiple pieces of information of the visual monitoring mechanism; the RAW image is image data which is sensed by an image sensor of the visual monitoring mechanism and is not subjected to image processing. According to the invention, whether the field monitoring image received by the remote end of the scene is the RAW image or not can be intelligently analyzed according to the multiple pieces of scene data of the two-dimensional code consumption scene and the multiple pieces of information of the visual monitoring mechanism by adopting the intelligent identification model which is designed in a targeted manner; therefore, valuable reference information is provided for remote image content judgment of the two-dimensional code consumption scene.

Description

Technical Field

[0001] The present invention relates to the field of visual monitoring, and in particular to a financial scene visual judgment system. Background Art

[0002] Two-dimensional barcode / QR code is a black and white graphic that records data symbol information distributed in a plane (two-dimensional direction) according to a certain pattern using certain geometric figures; in the compilation of the code, the concept of "0" and "1" bit streams that constitute the internal logic basis of the computer is cleverly used, and several geometric figures corresponding to binary are used to represent text numerical information, which is automatically read by image input equipment or photoelectric scanning equipment to realize automatic information processing: it has some common features of barcode technology: each code system has its specific character set; each character occupies a certain width; has certain verification functions, etc. At the same time, it also has the function of automatically identifying information in different rows and processing graphic rotation change points.

[0003] As a financial payment scenario, the QR code consumption scenario is a key scenario for video surveillance. In the prior art, in order to facilitate the beauty and transmission of the video screen, the video surveillance screen is generally optimized, so that the video surveillance screen received by the monitoring end is no longer a RAW image, that is, the image data sensed by the image sensor of the shooting mechanism without image processing, but a scene image after various processing and processing. Some details of the scene of the QR code consumption scenario may be eliminated or changed during the processing and processing. In the prior art, it is difficult to effectively identify whether the video surveillance screen received by the monitoring end of the QR code consumption scenario is a RAW image. Summary of the invention

[0004] In order to solve technical problems in related fields, the present invention provides a financial scene visual judgment system, which adopts a targeted intelligent identification model to intelligently analyze whether the on-site monitoring image received at the remote end of the scene is a RAW image based on multiple scene data of the QR code consumption scene and multiple information of the visualization monitoring mechanism. Specifically, the multiple scene data of the QR code consumption scene include the scene area of ​​the QR code consumption scene, the number of cashiers, the number of cashier counters and the standard pattern corresponding to the code scanning device. The multiple information of the visualization monitoring mechanism includes the shutter speed and ISO value of the visualization monitoring mechanism, thereby ensuring the reliability of the identification result of whether the image data is authentic, providing more realistic image data for the remote end of the QR code consumption scene, and helping the monitoring end to make more accurate image content judgments.

[0005] According to the present invention, a financial scene visual judgment system is provided, the system comprising: Visual monitoring mechanism, installed in the QR code consumption scene to perform visual monitoring processing, used to obtain and output corresponding on-site monitoring images; A customized learning mechanism, arranged at a remote monitoring end of the scene, for performing multiple learning actions on the deep residual network to obtain an intelligent identification model, wherein the intelligent identification model is a deep residual network that completes multiple learning actions; An information conversion device is arranged at a remote monitoring end of the scene and is respectively connected to the customized learning mechanism and the visual monitoring mechanism through a network, and is used to obtain multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and adopt an intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, wherein the RAW image is image data sensed by the image sensor of the visual monitoring mechanism without image processing; The on-site reporting device is connected to the information conversion device and is used to report the received intelligent identification results to the cloud service node via a wireless data connection; The customized learning mechanism is arranged at the remote monitoring end of the scene, and is used to perform multiple learning actions on the deep residual network to obtain an intelligent identification model, wherein the intelligent identification model is a deep residual network that completes multiple learning actions, including: in the deep residual network that completes multiple learning actions, the number of learning actions of the deep residual network is monotonically positively correlated with the horizontal resolution of the on-site monitoring image; Among them, obtaining multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and using the intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism includes: the multiple scene data of the QR code consumption scene include the scene area of ​​the QR code consumption scene, the number of cashiers, the number of cashier counters, and the standard pattern corresponding to the code scanning device; Among them, obtaining multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and using an intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism also includes: the multiple information of the visual monitoring mechanism includes the shutter speed and ISO value of the visual monitoring mechanism.

[0006] It can be seen that the present invention has at least the following three main inventive concepts: Inventive concept A: A targeted intelligent identification model is used to intelligently analyze whether the on-site monitoring image received by the remote end of the scene is a RAW image based on multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, thereby providing more realistic image data for the remote end of the QR code consumption scene; Inventive concept B: Specifically, the multiple scene data of the QR code consumption scene include the scene area occupied by the QR code consumption scene, the number of cashiers, the number of cashier counters, and the standard pattern corresponding to the code scanning device, and the multiple information of the visual monitoring mechanism includes the shutter speed and ISO value of the visual monitoring mechanism, thereby ensuring the reliability of the identification result of whether the image data is authentic; Inventive concept C: An intelligent identification model with a customized structure design is used to perform intelligent analysis of whether the on-site monitoring image received at the remote end of the scene is a RAW image. The intelligent identification model is a deep residual network that completes multiple learning actions, and in the deep residual network that completes multiple learning actions, the number of learning actions of the deep residual network is monotonically positively correlated with the horizontal resolution of the on-site monitoring image, thereby ensuring the stability and effectiveness of the intelligent analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein: Figure 1 It is a structural block diagram of a financial scene visual judgment system according to the primary embodiment of the present invention.

[0008] Figure 2 It is a structural block diagram of a financial scene visual judgment system according to a secondary embodiment of the present invention.

[0009] Figure 3 The figure is a structural block diagram of a financial scene visual judgment system according to another embodiment of the present invention. DETAILED DESCRIPTION

[0010] The following is a detailed description of an embodiment of the financial scene visual judgment system of the present invention with reference to the accompanying drawings.

[0011] Figure 1 FIG. 1 is a structural block diagram of a financial scene visual judgment system according to a primary embodiment of the present invention, wherein the system comprises: Visual monitoring mechanism, installed in the QR code consumption scene to perform visual monitoring processing, used to obtain and output corresponding on-site monitoring images; Specifically, the visual monitoring mechanism is installed in the QR code consumption scene to perform visual monitoring processing, and is used to obtain and output corresponding on-site monitoring images, including: the visual monitoring mechanism includes a parameter configuration interface, which is used to configure a variety of parameters of the working mode of the visual monitoring mechanism; A customized learning mechanism, arranged at a remote monitoring end of the scene, for performing multiple learning actions on the deep residual network to obtain an intelligent identification model, wherein the intelligent identification model is a deep residual network that completes multiple learning actions; An information conversion device is arranged at a remote monitoring end of the scene and is respectively connected to the customized learning mechanism and the visual monitoring mechanism through a network, and is used to obtain multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and adopt an intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, wherein the RAW image is image data sensed by the image sensor of the visual monitoring mechanism without image processing; The on-site reporting device is connected to the information conversion device and is used to report the received intelligent identification results to the cloud service node via a wireless data connection; The customized learning mechanism is arranged at the remote monitoring end of the scene, and is used to perform multiple learning actions on the deep residual network to obtain an intelligent identification model, wherein the intelligent identification model is a deep residual network that completes multiple learning actions, including: in the deep residual network that completes multiple learning actions, the number of learning actions of the deep residual network is monotonically positively correlated with the horizontal resolution of the on-site monitoring image; Among them, obtaining multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and using the intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism includes: the multiple scene data of the QR code consumption scene include the scene area of ​​the QR code consumption scene, the number of cashiers, the number of cashier counters, and the standard pattern corresponding to the code scanning device; Among them, obtaining multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and using the intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and the multiple information of the visual monitoring mechanism also includes: the multiple information of the visual monitoring mechanism includes the shutter speed and ISO value of the visual monitoring mechanism; Among them, the multiple scene data of the two-dimensional code consumption scene include the scene area occupied by the two-dimensional code consumption scene, the number of cashiers, the number of cashier counters, and the standard pattern corresponding to the code scanning device, including: the standard pattern corresponding to the code scanning device is numerically represented in BMP digital file format; And wherein, the multiple scene data of the QR code consumption scene include the scene area occupied by the QR code consumption scene, the number of cashiers, the number of cashier counters and the standard pattern corresponding to the code scanning device, and also include: different data acquisition components are used to respectively obtain the scene area occupied by the QR code consumption scene, the number of cashiers, the number of cashier counters and the standard pattern corresponding to the code scanning device.

[0012] Figure 2It is a structural block diagram of a financial scene visual judgment system according to a secondary embodiment of the present invention.

[0013] Compared to Figure 1 , Figure 2 The financial scene visual judgment system may also include: A current sensor device is connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, respectively, and is used to measure the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively; Wherein, the current sensor device is respectively connected with the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to respectively measure the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, including: the current sensor device includes a plurality of current measurement units, which are respectively connected with the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, so as to complete the measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism; Wherein, the current sensor device includes a plurality of current measuring units, which are used to be respectively connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism to complete the respective measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, including: the plurality of current measuring units are a plurality of current sensing circuits, which are used to be respectively connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism to complete the respective measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism; Wherein, the multiple current measurement units are multiple current sensing circuits, which are used to be connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, so as to complete the measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, including: the structures of the multiple current sensing circuits are the same; And wherein, the multiple current measurement units are multiple current sensing circuits, which are used to respectively connect with the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism to complete the separate measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and also include: the multiple current sensing circuits have the same current measurement upper limit value and current measurement lower limit value.

[0014] Figure 3 The figure is a structural block diagram of a financial scene visual judgment system according to another embodiment of the present invention.

[0015] Compared to Figure 1 , Figure 3 The financial scene visual judgment system may also include: A real-time display component, connected to the multiple current sensing circuits of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, respectively, for displaying the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism in real time; Wherein, the real-time display component is respectively connected to the multiple current sensing circuits of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to display the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism in real time, including: the real-time display component is an LED display array composed of multiple LED display units; And wherein, the real-time display component is respectively connected to multiple current sensing circuits of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to display the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism in real time, including: the real-time display component is an LCD display array composed of multiple LCD display units.

[0016] In addition, in the financial scene visual judgment system, multiple scene data of the QR code consumption scene and multiple information of the visual monitoring agency are obtained, and an intelligent identification model is used to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring agency, wherein the RAW image is image data sensed by the image sensor of the visual monitoring agency without image processing, including: inputting the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring agency into the intelligent identification model to execute the intelligent identification model, so as to obtain the identification mark output by the intelligent identification model indicating whether the on-site monitoring image is a RAW image.

[0017] The financial scene visual judgment system of the present invention is used to address the technical problem in the prior art that it is difficult to effectively identify whether the on-site monitoring images corresponding to the QR code consumption scene fully reflect various on-site details. By adopting a targeted intelligent identification model, it intelligently analyzes whether the on-site monitoring images received at the remote end of the scene are RAW images based on multiple scene data of the QR code consumption scene and multiple information of the visualization monitoring organization, thereby providing valuable reference information for the image content judgment at the remote end of the QR code consumption scene.

[0018] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure, rather than to limit them. Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present disclosure, and they should all be included in the scope of the claims and specification of the present disclosure.

Claims

1. A financial scene visual judgment system, characterized in that: The system comprises: Visual monitoring mechanism, installed in the QR code consumption scene to perform visual monitoring processing, used to obtain and output corresponding on-site monitoring images; A customized learning mechanism, arranged at a remote monitoring end of the scene, for performing multiple learning actions on the deep residual network to obtain an intelligent identification model, wherein the intelligent identification model is a deep residual network that completes multiple learning actions; An information conversion device is arranged at a remote monitoring end of the scene and is respectively connected to the customized learning mechanism and the visual monitoring mechanism through a network, and is used to obtain multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and adopt an intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, wherein the RAW image is image data sensed by the image sensor of the visual monitoring mechanism without image processing; The on-site reporting device is connected to the information conversion device and is used to report the received intelligent identification results to the cloud service node via a wireless data connection; The customized learning mechanism is arranged at the remote monitoring end of the scene, and is used to perform multiple learning actions on the deep residual network to obtain an intelligent identification model, wherein the intelligent identification model is a deep residual network that completes multiple learning actions, including: in the deep residual network that completes multiple learning actions, the number of learning actions of the deep residual network is monotonically positively correlated with the horizontal resolution of the on-site monitoring image; Among them, obtaining multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and using the intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism includes: the multiple scene data of the QR code consumption scene include the scene area of ​​the QR code consumption scene, the number of cashiers, the number of cashier counters, and the standard pattern corresponding to the code scanning device; Among them, obtaining multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism, and using an intelligent identification model to intelligently identify whether the on-site monitoring image is a RAW image based on the multiple scene data of the QR code consumption scene and multiple information of the visual monitoring mechanism also includes: the multiple information of the visual monitoring mechanism includes the shutter speed and ISO value of the visual monitoring mechanism.

2. The financial scene visual judgment system according to claim 1, characterized in that: The multiple scene data of the two-dimensional code consumption scene include the scene area occupied by the two-dimensional code consumption scene, the number of cashiers, the number of cashier counters, and the standard pattern corresponding to the code scanning device, including: the standard pattern corresponding to the code scanning device is numerically represented in BMP digital file format; Among them, the multiple scene data of the QR code consumption scene include the scene area occupied by the QR code consumption scene, the number of cashiers, the number of cashier counters and the standard pattern corresponding to the code scanning device, and also include: different data acquisition components are used to respectively obtain the scene area occupied by the QR code consumption scene, the number of cashiers, the number of cashier counters and the standard pattern corresponding to the code scanning device.

3. The financial scene visual judgment system according to claim 2, characterized in that: The system further comprises: A current sensor device is connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, respectively, and is used to measure the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively; Among them, the current sensing device is respectively connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to respectively measure the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism. The current sensing device includes multiple current measurement units, which are used to be respectively connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism to complete the respective measurements of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism.

4. The financial scene visual judgment system according to claim 3, characterized in that: The current sensor device includes multiple current measuring units, which are used to be connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, so as to complete the separate measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, including: the multiple current measuring units are multiple current sensing circuits, which are used to be connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, so as to complete the separate measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively.

5. The financial scene visual judgment system according to claim 4, characterized in that: The multiple current measurement units are multiple current sensing circuits, which are used to be connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, so as to complete the respective measurements of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, including: the structures of the multiple current sensing circuits are the same.

6. The financial scene visual judgment system according to claim 5, characterized in that: The multiple current measurement units are multiple current sensing circuits, which are used to be connected to the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, so as to complete the separate measurement of the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism respectively, and also include: the multiple current sensing circuits have the same current measurement upper limit value and current measurement lower limit value.

7. The financial scene visual judgment system according to claim 2, characterized in that: The system further comprises: The real-time display component is respectively connected to multiple current sensing circuits of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to display the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism in real time.

8. The financial scene visual judgment system according to claim 7, characterized in that: A real-time display component is respectively connected to multiple current sensing circuits of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to display the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism in real time. The real-time display component is an LED display array composed of multiple LED display units.

9. The financial scene visual judgment system according to claim 7, characterized in that: A real-time display component is respectively connected to multiple current sensing circuits of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism, and is used to display the current real-time current values ​​of the on-site reporting device, the information conversion device, the customized learning mechanism and the visual monitoring mechanism in real time. The real-time display component includes an LCD display array composed of multiple LCD display units.

Images