A mobile securities trading service system based on artificial intelligence

Abstract

The present invention provides a mobile securities trading service system based on artificial intelligence, comprising a data collection module, a classification management module and a trading module; wherein the data collection module is used to obtain securities data, and classify and store the securities data of each security, construct securities files, and construct a securities information database; the classification management module is used to obtain the user's screening feature parameters, and match the corresponding securities information from the securities information database according to the obtained screening feature parameters, and extract the corresponding securities files for display; the trading module is used to generate securities trading instructions according to the trading parameters filled in by the user, and transmit the securities trading instructions to the securities institution terminal, and receive the securities trading order information returned by the securities institution terminal. The present invention helps to promote users to complete securities trading, and at the same time improves the user's experience in the securities trading process.

Description

Technical Field

[0001] The present invention relates to the technical field of mobile securities trading services, and in particular to a mobile securities trading service system based on artificial intelligence. Background Art

[0002] As the functions of mobile terminals become more and more powerful, securities trading business is gradually shifting to mobile terminals. Users only need to operate mobile terminals to complete securities trading anytime and anywhere.

[0003] In the prior art, users only need to log in to the securities trading system through a mobile terminal to operate the purchase and sale of securities. However, in actual application scenarios, when users buy or sell related securities, they usually need to compare or consult a large amount of securities-related information to support the completion of securities trading. However, the current mobile securities trading system usually lacks a complete securities information display function, which makes it inconvenient for users to obtain specific securities information, thereby affecting the progress of securities trading. Summary of the invention

[0004] In view of the above problems, the present invention aims to provide a mobile securities trading service system based on artificial intelligence.

[0005] The purpose of the present invention is achieved by the following technical solutions:

[0006] The present invention proposes a mobile securities trading service system based on artificial intelligence, comprising a data collection module, a classification management module and a trading module; wherein:

[0007] The data collection module is used to obtain securities data, classify and store the securities data of each security, build securities archives, and build a securities information database; the securities data includes stock price data, expert forecast data, transaction data, and news information related to the securities;

[0008] The classification management module is used to obtain the user's screening feature parameters, match the corresponding securities information from the securities information database according to the obtained screening feature parameters, and extract the corresponding securities files for display;

[0009] The trading module is used to generate securities trading instructions according to the trading parameters filled in by the user, transmit the securities trading instructions to the securities institution terminal, and receive the securities trading order information returned by the securities institution terminal.

[0010] Preferably, the data acquisition module includes an interface unit, a mining unit and a database unit;

[0011] The interface unit is used to connect with the data interface of the securities trading system to obtain the stock price data and transaction data of each security from the securities trading system, wherein the stock price data includes historical stock price data and real-time stock price data, and the transaction data includes historical transaction data and real-time transaction data;

[0012] The mining unit is used to obtain news information and expert forecast data related to each security from the external information database;

[0013] The database unit is used to classify, store and manage the securities data of each security, build securities archives, and build a securities information database.

[0014] Preferably, the database unit includes a label unit;

[0015] The label unit is used to extract feature labels according to the securities data of each security in the securities information database, and to label the corresponding securities according to the extracted feature labels.

[0016] Preferably, the classification management module includes a keyword matching unit and a feature matching unit;

[0017] The keyword matching unit is used to match the corresponding securities files from the securities information database according to the keywords input by the user, and push the securities files to the user;

[0018] The feature matching unit is used to obtain the feature parameters input or selected by the user, and match and compare the feature parameters with the feature tags related to each security in the securities information database, obtain the securities files that meet the feature parameters, and push the securities files to the user.

[0019] Preferably, the transaction module includes an instruction unit and a confirmation unit;

[0020] The instruction unit is used to generate securities trading instructions according to the transaction parameters filled in by the user, and transmit the securities trading instructions to the securities institution terminal; wherein the securities trading instructions include the operation type, securities number, operation quantity, user confirmation information, etc.;

[0021] The confirmation unit is used to receive the securities trading order information returned by the securities institution terminal, wherein the trading order information is the securities trading order generated by the securities institution terminal according to the final securities operation information after completing the securities trading operation according to the acquired securities trading instruction.

[0022] Preferably, the system further comprises an auxiliary analysis module; wherein,

[0023] The auxiliary analysis module is used to perform artificial intelligence-based securities trend analysis based on the securities selected by the user, obtain the trend analysis results of the selected securities, and display them based on the trend analysis results.

[0024] Preferably, the auxiliary analysis module includes a selection unit, a feature extraction unit, a model analysis unit and a display unit; wherein,

[0025] The selection unit is used for the user to select securities that need to be analyzed and obtain the securities information to be analyzed;

[0026] The feature extraction unit is used to obtain feature data information of the security from the securities information database according to the security information to be analyzed, and to form an input vector set according to the obtained feature data information;

[0027] The model analysis unit is used to input the obtained input vector set into the trained securities trend analysis model to obtain the securities trend analysis results output by the securities trend analysis model;

[0028] The display unit is used to visualize the results of securities trend analysis.

[0029] Preferably, the system also includes a security control module; wherein,

[0030] The security control module is used to further verify the user's identity information when obtaining the transaction parameters filled in by the user. When the user's identity information is verified, it sends identity authentication feedback to the transaction module so that the transaction module can further generate securities trading instructions based on the transaction parameters.

[0031] Preferably, the security management and control model includes an identity management unit and an identity authentication unit; wherein,

[0032] The identity management unit is used to manage user identity information, wherein the user identity information includes standard user identity authentication information;

[0033] The identity authentication unit is used to obtain user identity authentication information, and compare and verify the obtained user identity authentication information or standard user identity authentication information. When the verification is passed, the identity authentication feedback is sent to the transaction module.

[0034] Preferably, the system also includes a secure transmission module; wherein,

[0035] The secure transmission module is used to securely encrypt the securities trading instructions when the trading module transmits the securities trading instructions to the securities institution terminal, and feed the securely encrypted securities trading instructions back to the trading module, so that the trading module can further transmit the securely encrypted securities trading instructions to the securities institution terminal.

[0036] The beneficial effects of the present invention are as follows: through the data collection module, the relevant information of each security is comprehensively collected and integrated into a security file, thereby improving the level of integrated management of security-related information. At the same time, based on the classification management module, the user can provide screening feature parameters according to his own needs, and the system will assist the user in screening out relevant security information and provide it to the user, assist the user in using the security-related information as a basis, and promote the user to complete the security transaction. Through the transaction module, the user can complete the transaction for the specified security and obtain the returned security transaction order information. Through the above-mentioned mobile transaction securities service system of the present invention, it is possible to assist the user in completing the integration and screening of security information through the mobile terminal, which is helpful to promote the user to complete the security transaction, and at the same time improve the user's experience in the security transaction process. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention is further described using the accompanying drawings, but the embodiments in the accompanying drawings do not constitute any limitation to the present invention. A person skilled in the art can obtain other drawings based on the following drawings without creative work.

[0038] Figure 1 This is a framework diagram of a mobile securities trading service system based on artificial intelligence as shown in an embodiment of the present invention;

[0039] Figure 2 This is a framework diagram of an artificial intelligence-based mobile securities trading service system according to another embodiment of the present invention. DETAILED DESCRIPTION

[0040] The present invention is further described in conjunction with the following application scenarios.

[0041] See also Figure 1 As shown in the embodiment, it shows a mobile securities trading service system based on artificial intelligence, including a data collection module, a classification management module and a trading module; wherein,

[0042] The data collection module is used to obtain securities data, classify and store the securities data of each security, build securities archives, and build a securities information database; the securities data includes stock price data, expert forecast data, transaction data, and news information related to the securities;

[0043] The classification management module is used to obtain the user's screening feature parameters, match the corresponding securities information from the securities information database according to the obtained screening feature parameters, and extract the corresponding securities files for display;

[0044] The trading module is used to generate securities trading instructions according to the trading parameters filled in by the user, transmit the securities trading instructions to the securities institution terminal, and receive the securities trading order information returned by the securities institution terminal.

[0045] The above-mentioned implementation mode of the present invention collects all relevant information of each security through the data collection module and integrates it into a security file, thereby improving the level of integrated management of security-related information. At the same time, based on the classification management module, the user can provide screening feature parameters according to his own needs, and the system will assist the user to screen out relevant security information and provide it to the user, assist the user to use the security-related information as a basis, and promote the user to complete the security transaction. Through the transaction module, the user can complete the transaction for the specified security and obtain the returned security transaction order information. Through the above-mentioned mobile transaction securities service system of the present invention, it can assist the user to complete the integration and screening of security information through the mobile terminal, which is helpful to promote the user to complete the security transaction and improve the user's experience in the security transaction process.

[0046] In one scenario, the above-mentioned mobile securities trading service system can be built based on a local server, a cloud server, a local data processing system, etc. to realize the corresponding data collection and data processing functions, which is not specifically limited in this application. Users can access the securities trading service system through mobile terminals to complete securities information acquisition or securities trading.

[0047] Preferably, the data acquisition module includes an interface unit, a mining unit and a database unit;

[0048] The interface unit is used to connect with the data interface of the securities trading system to obtain the stock price data and transaction data of each security from the securities trading system, wherein the stock price data includes historical stock price data and real-time stock price data, and the transaction data includes historical transaction data and real-time transaction data;

[0049] The mining unit is used to obtain news information and expert forecast data related to each security from the external information database;

[0050] The database unit is used to classify, store and manage the securities data of each security, build securities archives, and build a securities information database.

[0051] In the above implementation mode, by setting up the interface unit, the service system can directly access the real-time trading data and stock price data related to securities from the securities trading system; and by setting up the mining unit and utilizing the network data capture technology, the relevant information of each security can be captured from the external network or external information library in a targeted manner as an information supplement for building and improving the securities archives; and finally, by integrating the relevant data and information obtained for securities into securities archives, it can help to improve the level of collection and management of securities-related data and information, and lay the foundation for subsequently providing users with relevant reference information on securities.

[0052] Preferably, the database unit includes a label unit;

[0053] The label unit is used to extract feature labels according to the securities data of each security in the securities information database, and to label the corresponding securities according to the extracted feature labels.

[0054] Among them, according to the acquired securities data or related information, feature tags can be extracted and marked adaptively to improve the management level of securities feature information.

[0055] In one scenario, for example, if the acquired securities trading volume data continues to grow over a period of time, where the growth level exceeds the preset label standard, the securities trading volume data is marked with a relevant label of "attention" to facilitate subsequent matching or analysis.

[0056] Preferably, the classification management module includes a keyword matching unit and a feature matching unit;

[0057] The keyword matching unit is used to match the corresponding securities files from the securities information database according to the keywords input by the user, and push the securities files to the user;

[0058] The feature matching unit is used to obtain the feature parameters input or selected by the user, and match and compare the feature parameters with the feature tags related to each security in the securities information database, obtain the securities files that meet the feature parameters, and push the securities files to the user.

[0059] When a user needs to obtain relevant information about a specific security, or wants to obtain securities information of a certain type of characteristics, he or she can search through keyword matching or feature matching. The system will match the relevant information from the securities information database and display it visually to the user for further reference.

[0060] Preferably, the transaction module includes an instruction unit and a confirmation unit;

[0061] The instruction unit is used to generate securities trading instructions according to the transaction parameters filled in by the user, and transmit the securities trading instructions to the securities institution terminal; wherein the securities trading instructions include the operation type, securities number, operation quantity, user confirmation information, etc.;

[0062] The confirmation unit is used to receive the securities trading order information returned by the securities institution terminal, wherein the trading order information is the securities trading order generated by the securities institution terminal according to the final securities operation information after completing the securities trading operation according to the acquired securities trading instruction.

[0063] Preferably, the confirmation unit is also used to transmit the acquired securities trading order information back to the user terminal.

[0064] Users can select designated securities through the trading module and conduct trading operations on the securities. After completing the trading operations, the system generates securities trading instructions based on the user's operation information and transmits the securities trading instructions to the securities institution terminal, which then conducts real-time securities trading operations based on the contents of the securities trading instructions. Based on the feedback information obtained after the securities trading operations are finally completed, a securities trading order is generated and sent back to the user, so that the user can update his or her securities holdings and retain trading vouchers based on the securities trading order.

[0065] Preferably, the system further comprises an auxiliary analysis module; wherein,

[0066] The auxiliary analysis module is used to perform artificial intelligence-based securities trend analysis based on the securities selected by the user, obtain the trend analysis results of the selected securities, and display them based on the trend analysis results.

[0067] Among them, in order to further improve the level of intelligence in providing securities information to users, the service system of the present invention is also equipped with an auxiliary analysis module, which uses artificial intelligence technology to perform securities trend analysis based on relevant information in the securities information database, and can realize intelligent prediction of the stock price change trend of securities, which can be used as a reference for users in securities trading.

[0068] Preferably, the auxiliary analysis module includes a selection unit, a feature extraction unit, a model analysis unit and a display unit; wherein,

[0069] The selection unit is used for the user to select securities that need to be analyzed and obtain the securities information to be analyzed;

[0070] The feature extraction unit is used to obtain feature data information of the security from the securities information database according to the security information to be analyzed, and to form an input vector set according to the obtained feature data information;

[0071] The model analysis unit is used to input the obtained input vector set into the trained securities trend analysis model to obtain the securities trend analysis results output by the securities trend analysis model;

[0072] The display unit is used to visualize the results of securities trend analysis.

[0073] Among them, when conducting securities trend analysis based on artificial intelligence, firstly, the securities information to be analyzed is obtained according to the selected unit, and the data and information related to the securities to be analyzed are obtained from the securities information database based on the feature extraction unit, and the input vector set is adaptively composed, and the input vector set is used as the basis for input into the pre-trained securities trend analysis model, and the securities trend analysis model performs analysis based on the input vector set to obtain relevant trend analysis results. The securities trend analysis is completed through the above analysis, which can improve the accuracy and intelligence level of trend analysis based on the massive data in the securities information database and artificial intelligence technology.

[0074] In one scenario, the feature extraction unit forms an input vector set according to the feature data of the selected securities, wherein the formed input vector set includes: S i (t) = {X i (t); hp i (t); Tp i (t); mst i (t)}; where S i (t) represents the feature vector set of the i-th security corresponding to time t; X i (t) represents the stock price sequence of the ith security before time t; hp i (t) represents the stock price change sequence of the i-th security before time t; Tp i (t) represents the trading volume data of the i-th security at time t; mst i (t) expert evaluation data of the i-th security corresponding to time t; further inputting the obtained input vector set into the trained securities trend analysis model to obtain the securities trend analysis results output by the securities trend analysis model.

[0075] Among them, the securities trend analysis model is built based on the LSTM network. By training the LSTM network with a training set constructed using real data from the securities information database, the securities trend analysis model can accurately predict the trend of securities stock price changes, which can serve as a reference for users' securities trading operations and improve user experience.

[0076] In one scenario, the securities trend analysis model can also be built using an LSTM network that has been built in the prior art and completed deep learning training, and this application does not make specific limitations here.

[0077] Preferably, the system also includes a security control module; wherein,

[0078] The security control module is used to further verify the user's identity information when obtaining the transaction parameters filled in by the user. When the user's identity information is verified, it sends identity authentication feedback to the transaction module so that the transaction module can further generate securities trading instructions based on the transaction parameters.

[0079] Among them, considering that during the mobile securities trading process, the user's securities property safety is affected by the security of the mobile terminal, once the mobile terminal is stolen, it may cause serious property losses to the user; therefore, when the user needs to perform securities trading operations, the service system first uses the security management module to verify the user's identity information in real time to ensure that the user's securities trading is performed by himself, thereby improving the security of the user's securities trading.

[0080] When a user intends to initiate a securities transaction, the user's identity information is first verified to ensure that the operating user's identity is correct and to improve the security of securities transactions.

[0081] Preferably, the security management and control model includes an identity management unit and an identity authentication unit; wherein,

[0082] The identity management unit is used to manage user identity information, wherein the user identity information includes standard user identity authentication information;

[0083] The identity authentication unit is used to obtain user identity authentication information, and compare and verify the obtained user identity authentication information or standard user identity authentication information. When the verification is passed, the identity authentication feedback is sent to the transaction module.

[0084] When registering, users complete real-name authentication and enter registration information into the system. The system stores and manages the verified user identity information as the user's standard identity verification information so that it can be called during subsequent identity verification.

[0085] Preferably, the identity verification unit includes a face recognition unit;

[0086] The face recognition unit is used to obtain the user's face image in real time and perform face recognition processing based on the user's face image to complete user identity authentication.

[0087] The user face image is collected by the mobile terminal in real time, and the collected user face image is transmitted to the mobile securities trading service system in real time.

[0088] User identity authentication through facial recognition can effectively avoid the risk of information leakage caused by traditional identity authentication methods such as account numbers and passwords, and improve the reliability of user identity authentication.

[0089] Preferably, in the face recognition unit, face recognition processing is performed based on the user face image, specifically including:

[0090] According to the obtained user ID, the standard face image reserved by the user during registration is extracted from the registration information;

[0091] Perform facial feature extraction processing based on the user's facial image acquired in real time to obtain the user's facial feature information;

[0092] The acquired user facial feature information is compared with the standard user facial feature information corresponding to the standard facial image for verification. When the similarity between the two meets the preset standard value, the identity authentication result is passed.

[0093] Among them, the user face features include face texture features extracted based on the user face image, including LBP features and LBPH features.

[0094] When performing user identity authentication, the user's facial image is also collected based on the mobile terminal. The system performs identity authentication based on the user's facial image obtained by the mobile terminal to ensure that the identity of the user initiating the securities transaction is correct, so as to improve the reliability of identity authentication and further protect the security of transaction data.

[0095] Among them, in actual operation, since the mobile terminal obtains user facial images in a variety of scenarios and is affected by the scene environment, the quality of user facial images is easily uneven, which makes the system susceptible to interference when performing identity authentication based on user facial images (for example, when the user collects facial images through a mobile terminal, the light emitted by the mobile terminal screen will be reflected on the user's face, causing abnormal reflective areas to appear on the face, etc.), reducing the accuracy and reliability of identity recognition based on user facial images and affecting the user experience.

[0096] Preferably, the face recognition unit, before performing face feature extraction processing based on the user face image acquired in real time, further includes:

[0097] The received user face image is adaptively adjusted, specifically including:

[0098] Sa1 extracts the foreground part of the acquired user face image, and acquires the foreground part of the user face image as the face area A1; and acquires the background part of the user face image as the background area A2;

[0099] Sa2 obtains edge feature information in the face region image according to the grayscale information of the pixel points in the face region image A1, extracts edge pixel points in the face region image, and records the edge pixel points into the set JH b middle;

[0100] Sa3 converts the user's face image into the HSV color space and extracts the H component value, S component value, and V component value of each pixel point respectively;

[0101] Sa4 classifies the pixels in the face area image A1 based on the obtained V component value and S component, wherein the pixel classification function used is:

[0102] source3:V(x,y)-0.1S(x,y)≤0.3

[0103] source2:V(x,y)-0.1S(x,y)>0.3andV(x,y)<0.9

[0104] sourse1: V(x,y)≥0.9

[0105] Among them, source1, source2, and source3 represent the classification labels of the pixels. When the pixel (x, y) meets the conditional function corresponding to the classification label, the pixel (x, y) is marked as the pixel of the corresponding classification label; V(x, y) and S(x, y) represent the V component value and S component value of the pixel (x, y) respectively;

[0106] Sa5 performs a first adjustment process on the V component value of the pixel point of source1, wherein the first adjustment function used is:

[0107] V *1 (x,y)=V s(x,y)

[0108] Among them, V *1 (x, y) represents the V component value of the pixel point (x, y) after the first adjustment process, (x, y) ∈ source1; V s(x,y) Represents the average V component value of each pixel in the surrounding area of ​​the pixel point (x, y);

[0109] Sa6 forms a temporary set JH for the pixels of source1 and source2 after the first adjustment process c , for the temporary set JH c The pixels of are sorted from large to small according to the V component value to obtain a pixel sequence Hlc, wherein in Hlc, the pixel with a larger V component value is ranked higher;

[0110] Sa7 extracts the V component value V of the top 10% of the pixel points according to the pixel point sequence Hlc Hlc-10％ and temporary collection JH c The average V component value of each pixel in like Where VTH Indicates the preset standard V component value, V TH ∈[0.1,0.3]; then the second adjustment process is performed on the V component values ​​of the first 10% of the pixels in the sorting, otherwise Then complete the temporary set JH c Adjustment of the middle pixel;

[0111] The second adjustment function used is:

[0112] V *2 (x,y)=V c(x,y)

[0113] Among them, V *2 (x, y) represents the V component value of the pixel point (x, y) after the second adjustment process, c(x, y) represents the adaptive neighborhood range centered on the pixel point (x, y), where the adaptive neighborhood range is a rectangular range centered on the pixel point (x, y) with a side length of d, where the side length d satisfies: d = argmin(num ＝Hlc-10％ (c(x,y)) <num ≠Hlc-10％ (c(x,y))), d means that the function num ＝Hlc-10％ (c(x,y)) <num ≠Hlc-10％ The minimum value for which (c(x,y)) holds, num ＝Hlc-10％ (c(x,y)) represents the number of pixels in the top 10% of the sequence Hlc within the adaptive neighborhood centered on the pixel point (x,y). ≠Hlc-10％ (c(x,y)) represents the number of pixels in the adaptive neighborhood centered on the pixel point (x,y) that are not in the top 10% of the sequence Hlc;

[0114] After the second adjustment process is completed, the V component value of the pixel point (x, y) after the second adjustment process updates the pixel point sequence Hlc to obtain a new pixel point sequence and repeats step Sa7 based on the new pixel point sequence until the temporary set JH is completed. c Adjustment of the middle pixel;

[0115] Sa9 performs a third adjustment process on the V component value of the pixel point of source3, wherein the third adjustment function used is:

[0116] V *3 (x,y)=V dt +V(x,y)×V(x,y)

[0117] Among them, V *3(x, y) represents the V component value of the pixel point (x, y) after the third adjustment process, (x, y) ∈ source3; V(x, y) represents the V component value of the pixel point (x, y) before the third adjustment process, V dt Indicates the preset darkness V component value, where V dt ∈[0.01,0.1];

[0118] Sa10 adjusts the overall brightness of the face area image based on the current face area image, and the overall brightness adjustment function used is:

[0119]

[0120] Among them, V *Z (x, y) represents the V component value of the pixel point (x, y) after the overall brightness adjustment, V(x, y) represents the V component value of the current pixel point (x, y), V bt Indicates the preset standard V component value, V bt ∈[0.5,0.7],d a2b (x, y) represents the pixel distance between the pixel point (x, y) and its nearest edge pixel point (a, b), where (a, b) ∈ JH b ;d at Indicates the preset distance standard value, where d at ∈[3,5];

[0121] Reconverting the user face image after overall brightness adjustment to the RGB color space to obtain the user face image after adaptive adjustment;

[0122] The facial feature extraction process is further performed based on the user facial image after the adaptive adjustment process to obtain the user facial feature information.

[0123] Preferably, the foreground part extraction includes using a foreground part extraction algorithm based on edge detection, or a foreground part extraction algorithm based on depth information, etc. to complete the foreground part extraction, which is not specifically limited in the present invention.

[0124] Preferably, acquiring edge feature information based on the face region image includes detecting edge pixels using an LBP operator, a Sobel operator, a Canny operator, etc., which is not specifically limited in the present invention.

[0125] Preferably, s(x, y) represents the surrounding area of ​​the pixel point (x, y), wherein the surrounding area includes a rectangular area with a specific side length of 3×3, 5×5, 7×7, etc. centered on the pixel point, or the entire face area.

[0126] Preferably, in Sa7, the neighborhood range of a pixel point is a rectangular area with a side length of 5×5 centered on the pixel point.

[0127] The above-mentioned embodiment of the present invention proposes a technical solution for adaptively adjusting and processing user facial images. It can perform targeted processing on the situation where the facial image of the user is reflected due to the influence of terminal screen lighting during the process of collecting user facial images through a mobile terminal, which affects the accuracy of subsequent feature extraction and facial recognition based on the user facial image, thereby effectively improving the reliability and robustness of user identity authentication.

[0128] Among them, when performing adaptive adjustment processing, the face area in the user's face image is first processed as the key part, and the texture feature information of the face area is obtained based on the grayscale features of the face area, which serves as the basis for subsequent overall adjustment; further, based on the obtained user face image, it is converted into HSV color space, and the pixels are classified based on the saturation component and brightness component characteristics of the pixels. The proposed pixel classification function can classify the pixels into sourse1 classification with obvious reflection, sourse2 classification that needs to be focused on, and sourse3 classification of heavy shadow parts. The pixels of sourse1 classification with obvious reflection are first adjusted, and the brightness of the obvious reflection points is suppressed through the idea of ​​brightness balance to eliminate the influence of facial skin reflection or eyeball reflection. For the adjusted sourse1 pixels and sourse2 pixels, , the sign is further based on the idea of ​​sorting the brightness values ​​of pixels to detect the reflection phenomenon of the pixels in the face area with the largest proportion. Based on the characteristics of the reflection phenomenon (according to the reflection of the screen, a local range of brightness abnormality will appear in the face part), the part with prominent brightness abnormality is traversed and adjusted to reduce the abnormality of the pixels (reducing the probability of being misjudged as facial texture features due to the reflection situation); at the same time, based on the iterative adjustment method, it can adaptively complete the processing of abnormal reflection areas and avoid the excessive flatness of brightness caused by over-adjustment (excessive elimination of facial texture features that should have been recognized), which helps to retain the facial details to the greatest extent when removing the abnormal reflection area; finally, for the middle shadow area (such as the texture area of ​​the facial features, such as the corners of the eyes, nostrils, lips, etc.), the reflection of the texture features is highlighted by brightness stretching to improve the performance level of the facial features. At the same time, combined with the overall brightness adjustment processing, based on the texture feature information in the face area, the brightness level of the non-texture area is adaptively increased, which helps to improve the clarity of the entire face area and improve the representation level of the facial texture features. The server finally performs further feature extraction processing based on the user face image after adaptive adjustment, which can effectively eliminate the influence of abnormal texture features caused by light reflection caused by the mobile terminal during the user face image acquisition process, improve the reliability of face feature extraction, and also improve the accuracy of user identity authentication obtained by subsequent further comparison and verification based on face features, effectively improving the reliability of identity authentication in the securities trading process.

[0129] Preferably, the system also includes a secure transmission module; wherein,

[0130] The secure transmission module is used to securely encrypt the securities trading instructions when the trading module transmits the securities trading instructions to the securities institution terminal, and feed the securely encrypted securities trading instructions back to the trading module, so that the trading module can further transmit the securely encrypted securities trading instructions to the securities institution terminal.

[0131] The system also has a secure transmission module that can encrypt transaction data transmitted between the system and the user's mobile terminal, thereby improving the security of the data transmission process.

[0132] Preferably, the secure transmission module includes an encryption unit;

[0133] The encryption unit is used to perform secure encryption processing on securities trading instructions to obtain securely encrypted securities trading instructions.

[0134] It should be noted that each functional unit / module in each embodiment of the present invention may be integrated into one processing unit / module, or each unit / module may exist physically separately, or two or more units / modules may be integrated into one unit / module. The above-mentioned integrated unit / module may be implemented in the form of hardware or in the form of software functional unit / module.

[0135] Through the description of the above implementation mode, it can be clearly understood by those skilled in the art that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code or any appropriate combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, other electronic units designed to implement the functions described herein or their combination. For software implementation, part or all of the process of the embodiment can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable medium or transmitted as one or more instructions or codes on a computer-readable medium. Computer-readable media include computer storage media and communication media, wherein the communication medium includes any medium that facilitates the transmission of a computer program from one place to another. The storage medium can be any available medium that can be accessed by a computer. Computer-readable media can include, but are not limited to, RAM, ROM, EEPROM, CD-ROM or other optical disk storage, disk storage media or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0136] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention, rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should analyze that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

Claims

1. A mobile securities trading service system based on artificial intelligence, characterized in that: It includes data collection module, classification management module, transaction module and security control module; among them, The data collection module is used to obtain securities data, and to classify, store and manage the securities data of each security, build securities archives, and build a securities information database; the securities data includes stock price data, expert forecast data, transaction data and news information related to the securities; The classification management module is used to obtain the user's screening feature parameters, match the corresponding securities information from the securities information database according to the obtained screening feature parameters, and extract the corresponding securities files for display; The trading module is used to generate securities trading instructions according to the trading parameters filled in by the user, transmit the securities trading instructions to the securities institution terminal, and receive the securities trading order information returned by the securities institution terminal; The security control module is used to further verify the user's identity information when obtaining the transaction parameters filled in by the user. When the user's identity information is verified, it sends identity authentication feedback to the transaction module so that the transaction module can further generate securities trading instructions based on the transaction parameters; The security control module includes an identity management unit and an identity authentication unit; The identity management unit is used to manage user identity information, wherein the user identity information includes standard user identity authentication information; The identity verification unit is used to obtain user identity verification information, and compare and verify the obtained user identity verification information or standard user identity verification information. When the verification is passed, the identity verification feedback is sent to the transaction module; Wherein, the identity verification unit includes a face recognition unit; The face recognition unit is used to obtain the user's face image in real time, and perform face recognition processing based on the user's face image to complete the user's identity verification; The user face image is collected by the mobile terminal in real time, and the collected user face image is transmitted to the mobile securities trading service system in real time; In the face recognition unit, face recognition processing is performed based on the user's face image, specifically including: According to the obtained user ID, the standard face image reserved by the user during registration is extracted from the registration information; Perform facial feature extraction processing based on the user's facial image acquired in real time to obtain the user's facial feature information; The acquired user facial feature information is compared with the standard user facial feature information corresponding to the standard facial image for verification. When the similarity between the two meets the preset standard value, the identity verification result is obtained as passed; Wherein, in the face recognition unit, before performing face feature extraction processing according to the user face image acquired in real time, it also includes: The received user face image is adaptively adjusted, specifically including: Sa1 extracts the foreground part of the acquired user face image, and acquires the foreground part of the user face image as the face area A1; and acquires the background part of the user face image as the background area A2; Sa2 obtains edge feature information in the face region image according to the grayscale information of the pixel points in the face region image A1, extracts edge pixel points in the face region image, and records the edge pixel points into the set JH b middle; Sa3 converts the user's face image into the HSV color space and extracts the H component value, S component value, and V component value of each pixel point respectively; Sa4 classifies the pixels in the face area image A1 based on the obtained V component value and S component value, wherein the pixel classification function used is: source3:V(x,y)-0.1S(x,y)≤0.3 source2:V(x,y)-0.1S(x,y)>0.3 and V(x,y)<0.9 sourse1: V(x,y)≥0.9 Among them, source1, source2, and source3 represent the classification labels of the pixels. When the pixel (x, y) meets the conditional function corresponding to the classification label, the pixel (x, y) is marked as the pixel of the corresponding classification label; V(x, y) and S(x, y) represent the V component value and S component value of the pixel (x, y) respectively; Sa5 performs a first adjustment process on the V component value of the pixel point of source1, wherein the first adjustment function used is: V *1 (x,y)=V s(x,y) Among them, V *1 (x, y) represents the V component value of the pixel point (x, y) after the first adjustment process, (x, y) ∈ source1; V s(x,y) Represents the average V component value of each pixel in the surrounding area of ​​the pixel point (x, y); Sa6 forms a temporary set JH for the pixels of source1 and source2 after the first adjustment process c , for the temporary set JH c The pixels of are sorted from large to small according to the V component value to obtain a pixel sequence Hlc, wherein in Hlc, the pixel with a larger V component value is ranked higher; Sa7 extracts the V component value V of the top 10% of the pixel points according to the pixel point sequence Hlc Hlc-10％ and temporary collection JH c The average V component value of each pixel in like Where V TH Indicates the preset standard V component value, V TH ∈[0.1,0.3]; then the second adjustment process is performed on the V component values ​​of the first 10% of the pixels in the sorting, otherwise Then complete the temporary set JH c Adjustment of the middle pixel; The second adjustment function used is: V *2 (x,y)=V c(x,y) Among them, V *2 (x, y) represents the V component value of the pixel point (x, y) after the second adjustment process, c(x, y) represents the adaptive neighborhood range centered on the pixel point (x, y), where the adaptive neighborhood range is a rectangular range centered on the pixel point (x, y) with a side length of d, where the side length d satisfies: d = argmin(num ＝Hlc-10％ (c(x,y)) <num ≠Hlc-10％ (c(x,y))), d means that the function num ＝Hlc-10％ (c(x,y)) <num ≠Hlc-10％ The minimum value for which (c(x,y)) holds, num ＝Hlc-10％ (c(x,y)) represents the number of pixels in the top 10% of the sequence Hlc within the adaptive neighborhood centered on the pixel point (x,y). ≠Hlc-10％ (c(x,y)) represents the number of pixels in the adaptive neighborhood centered on the pixel point (x,y) that are not in the top 10% of the sequence Hlc; After the second adjustment process is completed, the V component value of the pixel point (x, y) after the second adjustment process updates the pixel point sequence Hlc to obtain a new pixel point sequence and repeats step Sa7 based on the new pixel point sequence until the temporary set JH is completed. c Adjustment of the middle pixel; Sa9 performs a third adjustment process on the V component value of the pixel point of source3, wherein the third adjustment function used is: V *3 (x,y)=V dt +V(x,y)×V(x,y) Among them, V *3 (x, y) represents the V component value of the pixel point (x, y) after the third adjustment process, (x, y) ∈ source3; V(x, y) represents the V component value of the pixel point (x, y) before the third adjustment process, V dt Indicates the preset darkness V component value, where V dt ∈[0.01,0.1]; Sa10 adjusts the overall brightness of the face area image based on the current face area image, and the overall brightness adjustment function used is: Among them, V *Z (x, y) represents the V component value of the pixel point (x, y) after the overall brightness adjustment, V(x, y) represents the V component value of the current pixel point (x, y), V bt Indicates the preset standard V component value, V bt ∈[0.5,0.7],d a2b (x, y) represents the pixel distance between the pixel point (x, y) and its nearest edge pixel point (a, b), where (a, b) ∈ JH b ;d at Indicates the preset distance standard value, where d at ∈[3,5]; Reconverting the user face image after overall brightness adjustment to the RGB color space to obtain the user face image after adaptive adjustment; The facial feature extraction process is further performed based on the user facial image after the adaptive adjustment process to obtain the user facial feature information.

2. The mobile securities trading service system based on artificial intelligence according to claim 1, characterized in that: The data acquisition module includes an interface unit, a mining unit and a database unit; The interface unit is used to connect with the data interface of the securities trading system to obtain the stock price data and transaction data of each security from the securities trading system, wherein the stock price data includes historical stock price data and real-time stock price data, and the transaction data includes historical transaction data and real-time transaction data; The mining unit is used to obtain news information and expert forecast data related to each security from the external information database; The database unit is used to classify, store and manage the securities data of each security, build securities archives, and build a securities information database.

3. The mobile securities trading service system based on artificial intelligence according to claim 2, characterized in that: The database unit includes a label unit; The label unit is used to extract feature labels according to the securities data of each security in the securities information database, and to label the corresponding securities according to the extracted feature labels.

4. The mobile securities trading service system based on artificial intelligence according to claim 1, characterized in that: The classification management module includes a keyword matching unit and a feature matching unit; The keyword matching unit is used to match the corresponding securities files from the securities information database according to the keywords input by the user, and push the securities files to the user; The feature matching unit is used to obtain the feature parameters input or selected by the user, and match and compare the feature parameters with the feature tags related to each security in the securities information database, obtain the securities files that meet the feature parameters, and push the securities files to the user.

5. The mobile securities trading service system based on artificial intelligence according to claim 1, characterized in that: The transaction module includes an instruction unit and a confirmation unit; The instruction unit is used to generate securities trading instructions according to the transaction parameters filled in by the user, and transmit the securities trading instructions to the securities institution terminal; wherein the securities trading instructions include the operation type, securities number, operation quantity, and user confirmation information; The confirmation unit is used to receive the securities trading order information returned by the securities institution terminal, wherein the trading order information is the securities trading order generated by the securities institution terminal according to the final securities operation information after completing the securities trading operation according to the acquired securities trading instruction.

6. The mobile securities trading service system based on artificial intelligence according to claim 1, characterized in that: It also includes an auxiliary analysis module; wherein, The auxiliary analysis module is used to perform artificial intelligence-based securities trend analysis based on the securities selected by the user, obtain the trend analysis results of the selected securities, and display them based on the trend analysis results.

7. The mobile securities trading service system based on artificial intelligence according to claim 6, characterized in that: The auxiliary analysis module includes a selection unit, a feature extraction unit, a model analysis unit and a display unit; wherein, The selection unit is used for the user to select securities that need to be analyzed and obtain the securities information to be analyzed; The feature extraction unit is used to obtain feature data information of the securities from the securities information database according to the securities information to be analyzed, and to form an input vector set according to the obtained feature data information; The model analysis unit is used to input the obtained input vector set into the trained securities trend analysis model to obtain the securities trend analysis results output by the securities trend analysis model; The display unit is used to visualize the results of securities trend analysis.

8. The mobile securities trading service system based on artificial intelligence according to claim 1, characterized in that: It also includes a secure transmission module; wherein, The secure transmission module is used to securely encrypt the securities trading instructions when the trading module transmits the securities trading instructions to the securities institution terminal, and feed the securely encrypted securities trading instructions back to the trading module, so that the trading module can further transmit the securely encrypted securities trading instructions to the securities institution terminal.

Images