Business handling operation evaluation method and device, storage medium and electronic equipment

Abstract

The application discloses a kind of service operation evaluation method, device, storage medium and electronic equipment.It relates to the field of financial technology, and the method comprises: obtaining the gesture video when the first object is for the second object to handle service, and obtaining the service information of the target service actually handled by the second object;According to the first standard gesture information matched according to the service information and the gesture information in the gesture video, it is judged whether the gesture of the first object conforms to the initial specification;In the case where it is determined that the gesture of the first object conforms to the initial specification, it is judged whether the object type of the second object is the target type;In the case where the object type is the target type, according to the second standard gesture information matched according to the target type and the gesture information in the gesture video, target information is obtained.The application solves the technical problem of low evaluation accuracy in the prior art when evaluating the gesture operation of the staff during the service handling process.

Description

Technical Field

[0001] This invention relates to the field of financial technology, and more specifically, to an evaluation method, apparatus, storage medium, and electronic device for business processing operations. Background Technology

[0002] With the development of modern technology, people's demand for financial services is increasing. However, in this digital age, the deaf and mute community faces a restrictive predicament. Because they cannot hear or speak, it is difficult for them to communicate effectively with staff in financial institutions.

[0003] To address this issue, financial institutions have begun using sign language services to meet the needs of deaf and hard-of-hearing users. By providing specially trained staff to communicate with users using sign language, financial institutions can better provide financial services to the deaf and hard-of-hearing. However, current technologies often fail to adequately assess the accuracy of staff sign language during use within financial institutions, which can negatively impact the user experience.

[0004] There is currently no effective solution to the above problems. Summary of the Invention

[0005] This invention provides a method, apparatus, storage medium, and electronic device for evaluating business processing operations, in order to at least solve the technical problem of low evaluation accuracy in the prior art when evaluating the gesture operations of staff during business processing.

[0006] According to one aspect of the present invention, an evaluation method for business processing operations is provided, comprising: acquiring a gesture video of a first object processing a business for a second object, and acquiring business information of the target business actually processed by the second object; determining whether the gesture of the first object conforms to an initial specification based on first standard gesture information matched with the business information and gesture information in the gesture video; if the gesture of the first object conforms to the initial specification, determining whether the object type of the second object is a target type; if the object type is a target type, obtaining target information based on second standard gesture information matched with the target type and gesture information in the gesture video, wherein the target information is used to characterize whether the gesture of the first object conforms to the target specification.

[0007] Furthermore, the evaluation method for business processing operations also includes: identifying the coordinates of bone points in multiple frames of images in a gesture video using a hand skeleton prediction model to obtain gesture information; inputting the gesture information into a gesture classification model to obtain a gesture classification result, wherein the gesture classification result is used to characterize the gesture categories corresponding to the multiple frames of images in sequence; determining the gesture category of each gesture in the gesture process of matching business information to obtain first standardized gesture information; and determining initial information based on the gesture categories in the gesture classification result and the gesture categories in the first standardized gesture information, wherein the initial information is used to characterize whether the gesture of the first object conforms to the initial standard.

[0008] Furthermore, the evaluation method for business processing operations also includes: determining the first gesture execution order based on the gesture category in the gesture classification result; determining the second gesture execution order based on the gesture category in the first standard gesture information; if the first gesture execution order includes the second gesture execution order, determining the initial information as information indicating that the gesture of the first object conforms to the initial standard; if the first gesture execution order does not include the second gesture execution order, determining the initial information as information indicating that the gesture of the first object does not conform to the initial standard.

[0009] Furthermore, the gesture classification result is a gesture category sequence. In the case of the target type being the elderly, the evaluation method for business processing operations also includes: extracting a target gesture category sequence that conforms to the second gesture execution order from the gesture category sequence; determining the sequence length of the target gesture category sequence, wherein the sequence length is related to the number of images corresponding to the target gesture category sequence, and the number of images is related to the time factor; determining the length of the first target sequence matching the target business from the second standardized gesture information; and determining the target information based on the relationship between the first target sequence length and the sequence length.

[0010] Furthermore, the evaluation method for business processing operations also includes: when the sequence length is greater than or equal to the first target sequence length, determining the target information as information indicating that the gesture of the first object conforms to the target specification; when the sequence length is less than the first target sequence length, determining the target information as information indicating that the gesture of the first object does not conform to the target specification.

[0011] Furthermore, the evaluation method for business processing operations also includes: determining the key gesture category and the second target sequence length matching the key gesture category from the second standard gesture information; determining the maximum sequence length of the sequence in which the key gesture category appears consecutively in the target gesture category sequence; and determining the target information as information representing that the gesture of the first object conforms to the target standard when the maximum sequence length is greater than or equal to the second target sequence length.

[0012] Furthermore, the evaluation method for business processing operations also includes: after determining whether the object type of the second object is the target type, if the object type is not the target type, determining that the gesture of the first object conforms to the target specification.

[0013] According to another aspect of the present invention, an evaluation device for business processing operations is also provided, comprising: an acquisition module, configured to acquire a gesture video of a first object processing a business for a second object, and acquire business information of the target business actually processed by the second object; a first judgment module, configured to determine whether the gesture of the first object conforms to an initial specification based on the first specification gesture information matched with the business information and the gesture information in the gesture video; a second judgment module, configured to determine whether the object type of the second object is a target type if the gesture of the first object conforms to the initial specification; and a processing module, configured to obtain target information based on the second specification gesture information matched with the target type and the gesture information in the gesture video if the object type is a target type, wherein the target information is used to characterize whether the gesture of the first object conforms to the target specification.

[0014] According to another aspect of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, and the computer program is configured to perform the above-described evaluation method for business processing operations when it is run.

[0015] According to another aspect of the present invention, an electronic device is also provided, the electronic device including one or more processors; a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the one or more processors are configured to run the programs, wherein the programs are configured to perform the above-described evaluation method for business processing operations during runtime.

[0016] In this embodiment of the invention, a multi-dimensional evaluation method for business processing operations is adopted. By acquiring a gesture video of a first object processing a business for a second object, and acquiring the business information of the target business actually processed by the second object, the first standard gesture information matched with the business information and the gesture information in the gesture video are used to determine whether the gesture of the first object conforms to the initial standard. Then, if the gesture of the first object conforms to the initial standard, it is determined whether the object type of the second object is the target type. If the object type is the target type, the target information is obtained based on the second standard gesture information matched with the target type and the gesture information in the gesture video. The target information is used to characterize whether the gesture of the first object conforms to the target standard.

[0017] In the above process, a preliminary assessment of the first object's gestures in handling business is achieved by matching the first standardized gesture information with business information, thus effectively evaluating the accuracy of the first object's gestures in the business dimension. Further evaluation of the first object's gestures is achieved by matching the second standardized gesture information with the target type, thus effectively evaluating the accuracy of the first object's gestures in the specific service demand dimension. Therefore, accurate evaluation of the first object's business handling operations is achieved, avoiding the problem that when evaluating gesture accuracy only for business purposes, neglecting the object displaying the gesture leads to a lack of judgment on specific features in the gesture, thereby affecting the accuracy of the business handling operation evaluation.

[0018] Therefore, the solution provided in this application achieves the goal of evaluating business processing operations from multiple dimensions, thereby improving the technical effect of evaluation accuracy and solving the technical problem of low evaluation accuracy in the prior art when evaluating the gesture operations of staff during business processing. Attached Figure Description

[0019] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, illustrate exemplary embodiments of the invention and, together with their description, serve to explain the invention and do not constitute an undue limitation thereof. In the drawings:

[0020] Figure 1 This is a schematic diagram of an optional business processing operation evaluation method according to an embodiment of the present invention;

[0021] Figure 2 This is a schematic diagram illustrating the training of an optional gesture classification model according to an embodiment of the present invention;

[0022] Figure 3 This is a schematic diagram of an optional business processing operation evaluation device according to an embodiment of the present invention;

[0023] Figure 4 This is a schematic diagram of an optional electronic device according to an embodiment of the present invention. Detailed Implementation

[0024] 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.

[0025] 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 a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0026] It should be noted that the evaluation methods, devices, storage media and electronic equipment for business processing operations disclosed herein can be used in the financial technology field, or in any field other than the financial technology field. The application fields of the evaluation methods, devices, storage media and electronic equipment for business processing operations disclosed herein are not limited.

[0027] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation portals are provided for users to choose to authorize or refuse.

[0028] Example 1

[0029] According to an embodiment of the present invention, an embodiment of an evaluation method for business processing operations is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] Figure 1 This is a schematic diagram of an optional business processing operation evaluation method according to an embodiment of the present invention, such as... Figure 1 As shown, the method includes the following steps:

[0031] Step S101: Obtain the gesture video of the first object handling business for the second object, and obtain the business information of the target business actually handled by the second object.

[0032] Optionally, servers, application systems, and electronic devices can be used as the executing entities for the aforementioned business processing operation evaluation method. In this embodiment, the target evaluation system can be used as the executing entity to perform the aforementioned business processing operation evaluation method.

[0033] The first target can be a staff member at a financial institution's offline branch who provides business services through gestures. The second target can be a user conducting business at the offline branch, and the second target must be a person with hearing impairment. The aforementioned gestures include, but are not limited to, sign language or other operations performed through hand movements. The gesture videos of the first and second targets conducting business are captured by cameras at the offline branch. The target assessment system can interact with the aforementioned cameras to obtain the gesture videos. Furthermore, when the second target conducts the target business on a terminal device at the offline branch, the target assessment system can interact with the aforementioned terminal device to obtain business information, including but not limited to the name of the target business.

[0034] Step S102: Based on the first standard gesture information matched with the business information and the gesture information in the gesture video, determine whether the gesture of the first object conforms to the initial standard.

[0035] In step S102, the target evaluation system can find the first standardized gesture information that matches the business information from the database based on the business information. The database pre-stores standardized gesture information that matches different businesses, and the standardized gesture information in the database can be indexed by the business information of the business. The first standardized gesture information is used to indicate the execution order of the gestures matched by the business information, and can also be used to indicate other information.

[0036] Furthermore, after acquiring the first standardized gesture information, the target evaluation system can extract the gesture information of the first object from the gesture video. The gesture information is information used to characterize the motion features of the gesture of the first object, including but not limited to motion time, posture, etc.

[0037] Furthermore, the target evaluation system can determine whether the gestures of the first object conform to the initial specifications by comparing the gesture information in the first specification gesture information and the gesture video.

[0038] It should be noted that by performing a preliminary assessment of the gestures of the first object based on the first standardized gesture information matched with business information, an effective assessment of the gesture accuracy of the first object in the business dimension is achieved.

[0039] Step S103: If the gesture of the first object conforms to the initial specification, determine whether the object type of the second object is the target type.

[0040] Because different users at offline outlets need to be provided with gesture services, and each user has different needs for gesture services. For example, the elderly need to use slower gestures so that they can understand the meaning of the gestures, while people with visual impairments need to use larger gestures so that they can see the gestures clearly.

[0041] Therefore, in this embodiment, pre-defined standardized gesture information matching different object types can be set. After the gesture of the first object passes the initial evaluation, it is determined whether the object type of the second object is the target type. The target type is the object type with matching standardized gesture information. Thus, even if the object type of the second object is determined to be the target type, it can be determined that other conditions must be met for the gesture of the first object.

[0042] When a user enters a service point, if they require gesture services, they will proactively seek assistance from staff or register information on the relevant interactive device. In this case, staff can identify the user's object type and upload it to the target evaluation system, or the target evaluation system can obtain the user's object type from the interactive device, thereby achieving effective acquisition of the object type.

[0043] Step S104: When the object type is the target type, target information is obtained based on the second specification gesture information matched by the target type and the gesture information in the gesture video. The target information is used to characterize whether the gesture of the first object conforms to the target specification.

[0044] Optionally, after determining that the object type is the target type, the target evaluation system can find the second standard gesture information that matches the target type. The second standard gesture information is used to indicate the information that needs to be met when handling business for the second object of the target type. For example, the second standard gesture information may be that the total execution time of the business-related gestures is greater than or equal to the preset time, or the second standard gesture information may be that the spatial execution range of the business-related gestures is not less than the preset range, etc.

[0045] Furthermore, the target evaluation system can determine whether the gestures of the second object conform to the target specifications by comparing the gesture information in the second specification gesture information and the gesture video.

[0046] It should be noted that by further evaluating the gestures of the first object based on the second-specification gesture information matched with the target type, an effective assessment of the gesture accuracy of the first object in a specific service requirement dimension is achieved. This enables an accurate assessment of the first object's business processing operations, avoiding the problem that assessing gesture accuracy solely based on business operations can lead to a lack of judgment on specific features of the gestures due to ignoring the object displaying the gestures, thus affecting the accuracy of business processing operation assessments.

[0047] Based on the scheme defined in steps S101 to S104 above, it can be understood that in this embodiment of the invention, a multi-dimensional evaluation method for business processing operations is adopted. This involves obtaining a gesture video of the first object processing business for the second object, and obtaining business information of the target business actually processed by the second object. Then, based on the first standard gesture information matched with the business information and the gesture information in the gesture video, it is determined whether the gesture of the first object conforms to the initial standard. Next, if it is determined that the gesture of the first object conforms to the initial standard, it is determined whether the object type of the second object is the target type. Thus, if the object type is the target type, target information is obtained based on the second standard gesture information matched with the target type and the gesture information in the gesture video. The target information is used to characterize whether the gesture of the first object conforms to the target standard.

[0048] It is noteworthy that in the above process, by performing a preliminary assessment of the first object's gestures in handling business based on the first standardized gesture information matched with business information, an effective assessment of the accuracy of the first object's gestures in the business dimension is achieved. By further assessing the first object's gestures based on the second standardized gesture information matched with the target type, an effective assessment of the accuracy of the first object's gestures in the specific service requirement dimension is achieved. Thus, an accurate assessment of the first object's business handling operations is achieved, avoiding the problem that when assessing gesture accuracy solely based on business needs, neglecting the object displaying the gesture leads to a lack of judgment on specific features of the gesture, thereby affecting the accuracy of the business handling operation assessment.

[0049] Therefore, the solution provided in this application achieves the goal of evaluating business processing operations from multiple dimensions, thereby improving the technical effect of evaluation accuracy and solving the technical problem of low evaluation accuracy in the prior art when evaluating the gesture operations of staff during business processing.

[0050] In an optional embodiment, during the process of determining whether the gesture of the first object conforms to the initial specification based on the first standardized gesture information matched with business information and the gesture information in the gesture video, the target evaluation system can identify the coordinates of bone points in multiple frames of images in the gesture video through a hand skeleton prediction model to obtain gesture information. Then, the gesture information is input into a gesture classification model and processed to obtain a gesture classification result. The gesture classification result is used to characterize the gesture categories corresponding to the multiple frames of images in sequence. Next, the gesture category of each gesture in the gesture process matched with business information is determined to obtain the first standardized gesture information. Thus, based on the gesture category in the gesture classification result and the gesture category in the first standardized gesture information, the initial information is determined. The initial information is used to characterize whether the gesture of the first object conforms to the initial specification.

[0051] Optionally, the method for extracting gesture information is described. In this embodiment, the hand skeleton prediction model can employ MediaPipe Hands, a machine learning framework used to detect and track hand gestures. It uses deep neural network technology to identify, locate, and track the human hand by inputting a video stream or image sequence. Specifically, the target evaluation system can input the gesture video into the hand skeleton prediction model. After acquiring the gesture video, the hand skeleton prediction model will first automatically extract multiple frames from the gesture video according to preset parameters, and then identify the coordinates of the bone points in each frame to obtain the gesture information. Optionally, in other embodiments, the gesture video can also be first decomposed into an image sequence composed of multiple frames, and then the image sequence can be input into the hand skeleton prediction model to obtain the gesture information.

[0052] The aforementioned hand skeleton prediction model can be trained using a large training sample set consisting of images and videos containing hand poses and movements.

[0053] Furthermore, after acquiring the gesture information, the target evaluation system can input the gesture information into the gesture classification model to obtain the gesture classification result. The gesture classification result is a sequence of gesture categories, where each gesture category corresponds to one frame in a multi-frame image sequence, or each gesture category corresponds to at least two frames in a multi-frame image sequence. When a gesture category in the sequence corresponds to at least two frames, these at least two frames are consecutive images, and the number of images corresponding to each gesture category is the same.

[0054] in, Figure 2 This is a schematic diagram illustrating the training of an optional gesture classification model according to an embodiment of the present invention, such as... Figure 2 As shown, the aforementioned gesture classification model can be trained using a training sample set containing multiple sample gesture information and real gesture category sequence labels. The sample gesture information is obtained by processing sample gesture videos using a hand skeleton prediction model. During training, a ten-fold cross-validation method can be used to train the gesture classification model to improve its accuracy and reliability.

[0055] Optionally, during the construction of the training sample set, the hand skeleton prediction model's output gesture information can be normalized, and then the training sample set for the gesture classification model can be constructed based on the normalized gesture information. Furthermore, during the process of inputting gesture information into the gesture classification model and obtaining the gesture classification result, the target evaluation system can first normalize the gesture information before inputting the processed gesture information into the gesture classification model. Normalization can avoid the influence of the distance between the spatial location of the gesture and the shooting location on the gesture classification result.

[0056] Furthermore, after obtaining the gesture classification results, the target evaluation system can determine the gesture category of each gesture in the gesture flow matching the business information, thereby obtaining the first standardized gesture information. In this embodiment, the first standardized gesture information is a sequence of standardized gesture categories. Then, the target evaluation system can determine initial information based on the gesture categories in the gesture classification results and the gesture categories in the first standardized gesture information. For example, if the gesture categories in the gesture classification results include all gesture categories in the first standardized gesture information, the initial information is determined to be information indicating that the gestures of the first object conform to the initial standard; otherwise, the initial information is determined to be information indicating that the gestures of the first object do not conform to the initial standard. Another example is determining the initial information by combining the execution order of the gestures.

[0057] It should be noted that by determining the initial information based on the gesture category in the gesture classification result and the gesture category in the first standard gesture information, the initial information can be accurately determined.

[0058] In an optional embodiment, during the process of determining initial information based on the gesture categories in the gesture classification results and the gesture categories in the first standardized gesture information, the target evaluation system can determine the first gesture execution order based on the gesture categories in the gesture classification results, and then determine the second gesture execution order based on the gesture categories in the first standardized gesture information. Thus, if the first gesture execution order includes the second gesture execution order, the initial information is determined to be information indicating that the gestures of the first object conform to the initial standard; if the first gesture execution order does not include the second gesture execution order, the initial information is determined to be information indicating that the gestures of the first object do not conform to the initial standard.

[0059] Optionally, since the gesture classification result is a sequence of gesture categories, and the first standardized gesture information is a sequence of standardized gesture categories, the target evaluation system can determine the execution order of the first gesture based on the order of the gesture categories in the sequence. For example, if the gesture category sequence is "gesture 1-gesture 1-gesture 2-gesture 2-gesture 2-gesture 2-gesture 3-gesture 3", then the execution order of the first gesture is "gesture 1-gesture 2-gesture 3". Similarly, the target evaluation system can determine the execution order of the second gesture based on the order of the gesture categories in the standardized gesture category sequence, so this will not be elaborated further here.

[0060] Furthermore, since gesture videos may contain not only business-related gestures but also gestures used to express other content, the target evaluation system can determine that the gestures of the first object conform to the initial specification if the first gesture execution order includes the second gesture execution order, and determine that the gestures of the first object do not conform to the initial specification if the first gesture execution order does not include the second gesture execution order. For example, if the first gesture execution order is "gesture 1-gesture 2-gesture 3" and the second gesture execution order is "gesture 1-gesture 2" or "gesture 2-gesture 3", then it is determined that the first gesture execution order includes the second gesture execution order; if the first gesture execution order is "gesture 1-gesture 2-gesture 3" and the second gesture execution order is "gesture 1-gesture 3", then it is determined that the first gesture execution order does not include the second gesture execution order.

[0061] It should be noted that by determining the initial information based on the execution order of the first gesture and the second gesture, a more accurate determination of the initial information can be achieved.

[0062] In one optional embodiment, the gesture classification result is a gesture category sequence. Specifically, when the target type is elderly, during the process of obtaining target information based on the second standardized gesture information matched to the target type and the gesture information in the gesture video, the target evaluation system can extract a target gesture category sequence that conforms to the second gesture execution order from the gesture category sequence. Then, it determines the sequence length of the target gesture category sequence. Next, it determines the length of the first target sequence matching the target service from the second standardized gesture information. Based on the relationship between the first target sequence length and the sequence length, the target information is determined. The sequence length is correlated with the number of images corresponding to the target gesture category sequence, and the number of images is correlated with the time factor.

[0063] Optionally, when the target type is an elderly person, the first target needs to use a relatively slower speed to display gestures so that the second target can understand the information conveyed by the gestures. Therefore, in this embodiment, the target assessment system can analyze the first target's business-related gestures in terms of time factors to determine the target information. The age range of the elderly person can be preset by the staff.

[0064] Specifically, the target evaluation system can first extract the target gesture category sequence that matches the second gesture execution order from the gesture category sequence, that is, the gesture execution order corresponding to the target gesture category sequence is the same as the second gesture execution order.

[0065] Furthermore, the target evaluation system determines the sequence length of the target gesture category sequence. For example, if the target gesture category sequence is "gesture 1-gesture 1-gesture 2-gesture 2-gesture 2", the sequence length is 5; if the target gesture category sequence is "gesture 1-gesture 1-gesture 2-gesture 2", the sequence length is 4.

[0066] It is important to emphasize that when the hand skeleton prediction model processes gesture videos, it extracts multiple frames of images from the video according to preset parameters, such as 30 frames per second, with each frame having the same time interval. Therefore, the longer the gesture video, the more images are extracted. Furthermore, since the number of images matching each gesture category in the gesture category sequence is a preset fixed value, the sequence length of the target gesture category sequence can indirectly characterize the execution duration of the gesture corresponding to the target gesture category sequence.

[0067] Furthermore, the target evaluation system can determine the first target sequence length matching the target service from the second standardized gesture information. The first target sequence length represents the sequence length that the standardized gesture category sequence corresponding to the target service needs to satisfy, which indirectly represents the execution duration that the gesture corresponding to the target service needs to satisfy. Then, the target evaluation system can determine the target information based on the relationship between the first target sequence length and the sequence length. For example, it can determine the difference between the first target sequence length and the sequence length, compare the difference with a preset value, and thus determine the target information based on the comparison result. The difference can be positive or negative.

[0068] It should be noted that, in the case of an elderly person, by determining the sequence length of the target gesture category sequence, the execution duration of the gestures related to the business of the first object is effectively determined. By determining the target information based on the relationship between the first target sequence length and the sequence length, the target information is determined based on the execution duration of the gesture, thereby achieving accurate determination of the target information.

[0069] In an optional embodiment, during the process of determining target information based on the relationship between the first target sequence length and the sequence length, the target evaluation system can determine the target information as information indicating that the gesture of the first object conforms to the target specification when the sequence length is greater than or equal to the first target sequence length, and determine the target information as information indicating that the gesture of the first object does not conform to the target specification when the sequence length is less than the first target sequence length.

[0070] Specifically, if the sequence length is greater than or equal to the first target sequence length, the execution duration of the gesture related to the first object and the business is determined to meet the required duration; conversely, if the sequence length is less than the first target sequence length, the execution duration of the gesture related to the first object and the business is determined to not meet the required duration.

[0071] It should be noted that by determining the target information based on whether the sequence length is the same as the length of the first target sequence, the target information is accurately determined.

[0072] In an optional embodiment, during the process of determining that the target information is information representing that the gesture of the first object conforms to the target specification, the target evaluation system can determine the key gesture category and the second target sequence length matching the key gesture category from the second specification gesture information, and then determine the maximum sequence length of the sequence in which the key gesture category appears consecutively in the target gesture category sequence. Thus, if the maximum sequence length is greater than or equal to the second target sequence length, the target information is determined to be information representing that the gesture of the first object conforms to the target specification.

[0073] Optionally, during the process of processing business for the second party, there may be some key gestures. If these key gestures are problematic or not clearly expressed, the second party may enter incorrect information, thereby affecting business processing efficiency and user experience. Therefore, in this embodiment, the target evaluation system can analyze the key gestures of the first party related to the business in terms of time factors to more accurately determine the target information.

[0074] Specifically, the target evaluation system can determine the key gesture category and the second target sequence length matching the key gesture category from the second standard gesture information. The second target sequence length represents the sequence length that the key gesture category corresponding to the target business needs to satisfy, which indirectly represents the execution time that the key gesture corresponding to the target business needs to satisfy.

[0075] Furthermore, the target evaluation system can determine the maximum sequence length of a sequence in which the key gesture category appears consecutively in the target gesture category sequence. For example, if the target gesture category sequence is “gesture 1-gesture 1-gesture 2-gesture 2-gesture 3-gesture 3-gesture 2-gesture 4”, and the key gesture is gesture 1, then the maximum sequence length is 2.

[0076] Furthermore, the target evaluation system can determine the target information as information indicating that the gesture of the first object conforms to the target specification if the aforementioned maximum sequence length is greater than or equal to the second target sequence length; otherwise, if the aforementioned maximum sequence length is less than the second target sequence length, the target information is determined as information indicating that the gesture of the first object does not conform to the target specification.

[0077] It should be noted that by combining the execution duration of key gestures to determine target information, a more accurate determination of target information is achieved.

[0078] In an optional embodiment, after determining whether the object type of the second object is the target type, the target evaluation system can determine that the gesture of the first object conforms to the target specification if the object type is not the target type.

[0079] Optionally, if the object type of the first object is not the target type, it can be determined that the gesture of the first object does not need to meet any additional specific requirements, and thus it can be directly determined that the gesture of the first object conforms to the target specification. This achieves effective evaluation of the gesture of the first object.

[0080] Therefore, the solution provided in this application achieves the goal of evaluating business processing operations from multiple dimensions, thereby improving the technical effect of evaluation accuracy and solving the technical problem of low evaluation accuracy in the prior art when evaluating the gesture operations of staff during business processing.

[0081] Example 2

[0082] According to an embodiment of the present invention, an embodiment of an evaluation device for business processing operations is provided, wherein, Figure 3 This is a schematic diagram of an optional business processing operation evaluation device according to an embodiment of the present invention, such as... Figure 3 As shown, the device includes:

[0083] The acquisition module 301 is used to acquire a video of the gestures of the first object when it handles business for the second object, and to acquire business information of the target business actually handled by the second object.

[0084] The first judgment module 302 is used to determine whether the gesture of the first object conforms to the initial specification based on the first standard gesture information matched by the business information and the gesture information in the gesture video.

[0085] The second judgment module 303 is used to determine whether the object type of the second object is the target type when the gesture of the first object conforms to the initial specification.

[0086] The processing module 304 is used to obtain target information based on the second specification gesture information matched by the target type and the gesture information in the gesture video when the object type is the target type. The target information is used to characterize whether the gesture of the first object conforms to the target specification.

[0087] It should be noted that the above-mentioned acquisition module 301, first judgment module 302, second judgment module 303 and processing module 304 correspond to steps S101 to S104 in the above embodiments. The four modules and the corresponding steps implement the same examples and application scenarios, but are not limited to the content disclosed in the above embodiment 1.

[0088] Optionally, the first judgment module 302 includes: a recognition submodule, used to recognize the coordinates of bone points in multiple frames of images in a gesture video through a hand skeleton prediction model to obtain gesture information; a processing submodule, used to input the gesture information into a gesture classification model and process it to obtain a gesture classification result, wherein the gesture classification result is used to characterize the gesture categories corresponding to the multiple frames of images in sequence; a first determination submodule, used to determine the gesture category of each gesture in the gesture process of business information matching to obtain first standardized gesture information; and a second determination submodule, used to determine initial information based on the gesture category in the gesture classification result and the gesture category in the first standardized gesture information, wherein the initial information is used to characterize whether the gesture of the first object conforms to the initial standard.

[0089] Optionally, the second determining submodule further includes: a first determining unit, configured to determine the first gesture execution order based on the gesture category in the gesture classification result; a second determining unit, configured to determine the second gesture execution order based on the gesture category in the first standardized gesture information; a third determining unit, configured to determine the initial information as information indicating that the gestures of the first object conform to the initial standard when the first gesture execution order includes the second gesture execution order; and a fourth determining unit, configured to determine the initial information as information indicating that the gestures of the first object do not conform to the initial standard when the first gesture execution order does not include the second gesture execution order.

[0090] Optionally, the gesture classification result is a gesture category sequence. In the case where the target type is elderly, the processing module 304 includes: an extraction submodule for extracting a target gesture category sequence that conforms to the second gesture execution order from the gesture category sequence; a third determination submodule for determining the sequence length of the target gesture category sequence, wherein the sequence length is related to the number of images corresponding to the target gesture category sequence, and the number of images is related to the time factor; a fourth determination submodule for determining the first target sequence length matching the target business from the second standardized gesture information; and a fifth determination submodule for determining the target information based on the relationship between the first target sequence length and the sequence length.

[0091] Optionally, the fifth determining submodule further includes: a fifth determining unit, used to determine that the target information is information indicating that the gesture of the first object conforms to the target specification when the sequence length is greater than or equal to the first target sequence length; and a sixth determining unit, used to determine that the target information is information indicating that the gesture of the first object does not conform to the target specification when the sequence length is less than the first target sequence length.

[0092] Optionally, the fifth determining unit further includes: a first determining subunit, used to determine the key gesture category and the second target sequence length matching the key gesture category from the second standard gesture information; a second determining subunit, used to determine the maximum sequence length of the sequence in which the key gesture category appears consecutively in the target gesture category sequence; and a third determining subunit, used to determine that the target information is information representing that the gesture of the first object conforms to the target standard when the maximum sequence length is greater than or equal to the second target sequence length.

[0093] Optionally, the evaluation device for business processing operations also includes: a determination module, used to determine that the gesture of the first object conforms to the target specification when the object type is not the target type.

[0094] Example 3

[0095] According to another aspect of the present invention, a computer-readable storage medium is also provided, wherein a computer program is stored in the computer-readable storage medium, and the computer program is configured to perform the above-described evaluation method for business processing operations when it is run.

[0096] Example 4

[0097] According to another aspect of the present invention, an electronic device is also provided, wherein, Figure 4 This is a schematic diagram of an optional electronic device according to an embodiment of the present invention, such as... Figure 4 As shown, the electronic device includes one or more processors; and a memory for storing one or more programs, which, when executed by the one or more processors, enable the one or more processors to run the programs, wherein the programs are configured to perform the aforementioned evaluation method for business processing operations during runtime.

[0098] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0099] In the above embodiments of the present invention, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0100] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0101] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0102] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0103] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0104] The above are merely preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principle of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A method for evaluating business processing operations, characterized in that, include: Obtain a video of the gestures of the first object when handling business for the second object, and obtain the business information of the target business actually handled by the second object; Based on the first standardized gesture information matched by the business information and the gesture information in the gesture video, determining whether the gesture of the first object conforms to the initial standard includes: identifying the coordinates of bone points in multiple frames of the gesture video using a hand skeleton prediction model to obtain the gesture information; inputting the gesture information into a gesture classification model to obtain a gesture classification result, wherein the gesture classification result is used to characterize the gesture categories corresponding to the multiple frames in sequence; determining the gesture category of each gesture in the gesture process matched by the business information to obtain the first standardized gesture information; and determining initial information based on the gesture categories in the gesture classification result and the gesture categories in the first standardized gesture information, wherein the initial information is used to characterize whether the gesture of the first object conforms to the initial standard. Specifically, determining initial information based on the gesture categories in the gesture classification results and the gesture categories in the first standardized gesture information includes: determining a first gesture execution order based on the gesture categories in the gesture classification results; determining a second gesture execution order based on the gesture categories in the first standardized gesture information; if the first gesture execution order includes the second gesture execution order, determining the initial information as information indicating that the gestures of the first object conform to the initial standard; if the first gesture execution order does not include the second gesture execution order, determining the initial information as information indicating that the gestures of the first object do not conform to the initial standard. If the gesture of the first object conforms to the initial specification, it is determined whether the object type of the second object is the target type; When the object type is the target type, target information is obtained based on the second specification gesture information matched by the target type and the gesture information in the gesture video, wherein the target information is used to characterize whether the gesture of the first object conforms to the target specification; The gesture classification result is a gesture category sequence. In the case where the target type is elderly, target information is obtained based on the second standardized gesture information matching the target type and the gesture information in the gesture video. This includes: extracting a target gesture category sequence that conforms to the second gesture execution order from the gesture category sequence; determining the sequence length of the target gesture category sequence, wherein the sequence length is related to the number of images corresponding to the target gesture category sequence, and the number of images is related to a time factor; determining the first target sequence length matching the target service from the second standardized gesture information; and determining the target information based on the relationship between the first target sequence length and the sequence length.

2. The method according to claim 1, characterized in that, Determining the target information based on the relationship between the first target sequence length and the sequence length includes: If the sequence length is greater than or equal to the first target sequence length, the target information is determined to be information representing that the gesture of the first object conforms to the target specification; If the sequence length is less than the first target sequence length, the target information is determined to be information indicating that the gesture of the first object does not conform to the target specification.

3. The method according to claim 2, characterized in that, Determining the target information as information representing that the gesture of the first object conforms to the target specification includes: Determine the key gesture category and the length of the second target sequence matching the key gesture category from the second standardized gesture information; Determine the maximum sequence length of the sequence in which the key gesture category appears consecutively in the target gesture category sequence; If the maximum sequence length is greater than or equal to the second target sequence length, the target information is determined to be information representing that the gesture of the first object conforms to the target specification.

4. The method according to claim 1, characterized in that, After determining whether the object type of the second object is the target type, the method further includes: If the object type is not the target type, it is determined that the gesture of the first object conforms to the target specification.

5. An evaluation device for business processing operations, characterized in that, include: The acquisition module is used to acquire a video of the gestures of the first object when it is handling business for the second object, and to acquire business information of the target business actually handled by the second object. The first judgment module is used to determine whether the gesture of the first object conforms to the initial specification based on the first standard gesture information matched by the business information and the gesture information in the gesture video. The first judgment module includes: an identification submodule, used to identify the coordinates of bone points in multiple frames of images in the gesture video through a hand skeleton prediction model to obtain the gesture information; a processing submodule, used to input the gesture information into a gesture classification model and process it to obtain a gesture classification result, wherein the gesture classification result is used to characterize the gesture categories corresponding to the multiple frames of images in sequence; a first determination submodule, used to determine the gesture category of each gesture in the gesture process of matching business information to obtain the first standardized gesture information; and a second determination submodule, used to determine initial information based on the gesture category in the gesture classification result and the gesture category in the first standardized gesture information, wherein the initial information is used to characterize whether the gesture of the first object conforms to the initial standard. The second determining submodule includes: a first determining unit, configured to determine a first gesture execution order based on the gesture category in the gesture classification result; a second determining unit, configured to determine a second gesture execution order based on the gesture category in the first standardized gesture information; a third determining unit, configured to determine that the initial information is information indicating that the gestures of the first object conform to the initial standard when the first gesture execution order includes the second gesture execution order; and a fourth determining unit, configured to determine that the initial information is information indicating that the gestures of the first object do not conform to the initial standard when the first gesture execution order does not include the second gesture execution order. The second judgment module is used to determine whether the object type of the second object is the target type when it is determined that the gesture of the first object conforms to the initial specification; The processing module is configured to, when the object type is the target type, obtain target information based on the second specification gesture information matched by the target type and the gesture information in the gesture video, wherein the target information is used to characterize whether the gesture of the first object conforms to the target specification; The gesture classification result is a gesture category sequence. When the target type is elderly, the processing module includes: an extraction submodule for extracting a target gesture category sequence from the gesture category sequence that conforms to the second gesture execution order; a third determination submodule for determining the sequence length of the target gesture category sequence, wherein the sequence length is related to the number of images corresponding to the target gesture category sequence, and the number of images is related to a time factor; a fourth determination submodule for determining the first target sequence length matching the target service from the second standardized gesture information; and a fifth determination submodule for determining the target information based on the relationship between the first target sequence length and the sequence length.

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein the computer program is configured to perform the evaluation method for the business processing operation as described in any one of claims 1 to 4 when it is run.

7. An electronic device, characterized in that, The electronic device includes one or more processors; A memory for storing one or more programs, which, when executed by one or more processors, cause the one or more processors to be configured to run the programs, wherein the programs are configured to perform the evaluation method for the business processing operation as described in any one of claims 1 to 4 at runtime.

Images