Business processing method and device, computer device and storage medium

Abstract

Embodiments of the present application disclose a service processing method and device, computer equipment and a storage medium, wherein the method comprises: acquiring a reference image containing a target object, and acquiring a cropped image obtained by performing object cropping processing on the reference image; performing a target service related to the target object by using the cropped image; when there is an exception in the execution of the target service, performing cropping positioning processing on the cropped image according to the reference image to obtain a cropping positioning result; performing service exception detection processing on the cropped image according to the cropping positioning result to obtain a detection result, and generating a reason for generating a corresponding exception when the target service is executed according to the detection result, so that processing resources can be effectively saved under the condition of ensuring service execution safety.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a business processing method, apparatus, computer equipment, and storage medium. Background Technology

[0002] With the continuous development of computer technology, using computer technology to assist in the execution of business processes has become an industry trend. For example, in some business processes that require object recognition, after acquiring relevant object images, computer technology can be used to recognize and process these images, thereby triggering the execution of related business processes, which can effectively improve the efficiency of business execution. However, practice has shown that the current method of triggering business execution based on image recognition consumes a lot of processing resources. Therefore, how to save computer processing resources while ensuring the security of business execution has become a current research hotspot. Summary of the Invention

[0003] The present invention provides a business processing method, apparatus, computer equipment, and storage medium that can effectively save processing resources while ensuring the security of business execution.

[0004] On one hand, embodiments of the present invention provide a business processing method, including:

[0005] Obtain a reference image containing the target object, and obtain a cropped image obtained by cropping the reference image.

[0006] The cropped image is used to perform target services related to the target object;

[0007] When the execution of the target service is abnormal, the cropped image is cropped and located according to the reference image to obtain the cropping and location result;

[0008] Based on the cropping and positioning results, the cropped image is subjected to business anomaly detection processing to obtain detection results, and the reasons for the corresponding anomalies generated when the target business is executed are generated based on the detection results.

[0009] In another aspect, embodiments of the present invention provide a business processing apparatus, including:

[0010] The acquisition unit is used to acquire a reference image containing the target object, and to acquire a cropped image obtained by cropping the reference image.

[0011] The processing unit is configured to perform target services related to the target object based on the cropped image;

[0012] The processing unit is further configured to, when there is an anomaly in the execution of the target service, use the reference image to perform cropping and positioning processing on the cropped image to obtain a cropping and positioning result;

[0013] The processing unit is further configured to perform business anomaly detection processing on the cropped image based on the cropping positioning result, obtain the detection result, and generate the cause of the corresponding anomaly when executing the target business based on the detection result.

[0014] In another aspect, embodiments of the present invention provide a computer device, including a processor, an input device, an output device, and a memory, wherein the processor, input device, output device, and memory are interconnected, wherein the memory is used to store a computer program supporting the computer device in executing the above-described method, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute the following steps:

[0015] Obtain a reference image containing the target object, and obtain a cropped image obtained by cropping the reference image.

[0016] The cropped image is used to perform target services related to the target object;

[0017] When the execution of the target service is abnormal, the cropped image is cropped and located according to the reference image to obtain the cropping and location result;

[0018] Based on the cropping and positioning results, the cropped image is subjected to business anomaly detection processing to obtain detection results, and the reasons for the corresponding anomalies generated when the target business is executed are generated based on the detection results.

[0019] In another aspect, embodiments of the present invention provide a computer-readable storage medium storing program instructions, which, when executed by a processor, are used to perform the business processing method as described in the first aspect.

[0020] In this embodiment, after acquiring a reference image containing a target object, the computer device can crop the acquired reference image to obtain a cropped image. This cropped image is then used to execute target services related to the target object. This enables the computer device to perform corresponding services based on a small image, improving the efficiency of service execution and effectively saving computer processing resources. However, if an anomaly occurs when the computer device executes the target service based on the cropped image, in order to detect the cause of the anomaly, the computer device can use the original reference image containing the target object to perform service anomaly detection processing on the cropped image. This determines whether the anomaly is due to a cropping error, thus ensuring both the user experience of the target object during service execution and the security of service execution. Attached Figure Description

[0021] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0022] Figure 1a This is a schematic diagram of a business process provided by an embodiment of the present invention;

[0023] Figure 1b This is a schematic diagram of another business process provided by an embodiment of the present invention;

[0024] Figure 2 This is a schematic flowchart of a business processing method provided in an embodiment of the present invention;

[0025] Figure 3a This is a schematic diagram of an image cropping method provided in an embodiment of the present invention;

[0026] Figure 3b This is a schematic diagram of another business processing method provided by an embodiment of the present invention;

[0027] Figure 3c This is a flowchart of a business process provided in an embodiment of the present invention;

[0028] Figure 4 This is a schematic block diagram of a business processing device provided in an embodiment of the present invention;

[0029] Figure 5 This is a schematic block diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation

[0030] This application proposes a business processing method. When executing a target business, a computer device can use a cropped small image to perform the target business, thereby improving the efficiency of business processing. Since the computer device also stores the original image corresponding to the small image used to execute the target business, if an anomaly occurs when the computer device uses the cropped small image for business processing, the corresponding original image can be used to locate the cause of the problem. This achieves effective localization of business execution anomalies while improving the efficiency of business processing. In one embodiment, the target business may be, for example, an electronic resource transfer service or an access control authentication service. When executing the target business, an image acquisition device can be used to acquire an image of the target object, thereby obtaining a reference image containing the target object. When the computer device acquires an image of the target object using the image acquisition device, it will also acquire the overall environment containing the target object. That is, the reference image containing the target object acquired by the image acquisition device not only includes the target object but also the environmental image of the environment in which the target object is located. Therefore, it can be understood that if the environment in which the target object is located contains other objects, then the reference image will also contain other objects in that environment.

[0031] After a computer device acquires a reference image containing the target object, in order to improve the processing efficiency for the target service, it can perform object cropping on the acquired reference image to obtain a corresponding cropped image. This allows the computer device to use the cropped image to execute the target service. In one embodiment, the computer device may make a cropping error during object cropping. For example, instead of cropping the target object, it might crop other objects contained in the reference image, leading to a cropping error. If the computer device crops the reference image incorrectly, subsequent execution of the target service using the cropped image will encounter an anomaly. Therefore, in order to determine the cause of an anomaly when performing a target service using a cropped image and when such an anomaly occurs, the computer device can, after acquiring a reference image containing the target object, store the reference image (i.e., the large image corresponding to the cropped small image) to detect the cause of the anomaly when performing a target service using a cropped image of the reference image and an anomaly occurs. This allows the stored reference image to be used to analyze the cause of the anomaly when performing a target service using a cropped image and an anomaly occurs, thereby obtaining the cause of the anomaly.

[0032] In one embodiment, when a computer device analyzes a business processing anomaly caused by performing a target business using a cropped image, using a previously acquired reference image, in one implementation, the computer device can compare the cropped image and the reference image to determine whether the object contained in the cropped image is the target object included in the reference image required for performing the target business processing. If it is determined that the object contained in the cropped image is not the target object required for performing the target business, the computer device can determine that the abnormality in the target business execution is caused by cropping the reference image. Conversely, if the computer device determines that the object contained in the cropped image is the target object required for performing the target business, the computer device can determine that the abnormality in the target business execution is not caused by cropping the reference image. In other implementations, when analyzing situations that cause abnormal execution of the target service, the computer device can also perform cropping and positioning processing on the reference image based on the cropped image. Based on the results of the cropping and positioning processing, the abnormality generated when the target service is executed by the cropped image can be analyzed. Specifically, based on the cropping and positioning processing of the reference image by the cropped image, the cropping position of the cropped image in the reference image can be determined. Then, based on the determined cropping position, it can be determined whether the object contained in the cropped image is the target object required to execute the target service, thereby realizing the judgment of the cause of the abnormality of the target service.

[0033] In one embodiment, if the computer device determines that the cause of the abnormal execution of the target service is not due to cropping of the reference image, the computer device can send the reference image and the cropped image to other business analysis servers, or output them to business analysts to determine the specific cause of the abnormal execution of the target service. It should be noted that when the computer device acquires an image of the target object to obtain a reference image containing the target object, it does so only after requesting the target object's consent to the image acquisition and obtaining the target object's authorization. Furthermore, after acquiring the reference image of the target object, the computer device can display the acquired reference image to the target object and inform the target object that the acquired reference image will be used for subsequent target services. The computer device can only use the reference object for subsequent target service execution after obtaining authorization from the target object to use the reference image, thus effectively ensuring the object privacy and security of the target object when the computer device acquires images of the target object and when using the reference image containing the target object for related business execution.

[0034] When the target business is electronic resource transfer, the computer equipment can first acquire a reference image containing the target object. After obtaining the reference image, a corresponding cropped image can be obtained. Once the computer equipment obtains the cropped image, if it determines that it can accurately identify the object contained within the cropped image, it can directly... Figure 1a As shown, the corresponding electronic resource transfer process is initiated. In one embodiment, if the computer device cannot accurately identify the object contained in the cropped image, such as if the image of the target object acquired by the computer device is an image of object A, and object A has a similar-looking twin (assumed to be object B), then if the computer device, after obtaining the cropped image containing object A, cannot determine whether the object contained in the cropped image is object A or object B, then the computer device can proceed as follows: Figure 1b As shown, a verification request interface is first output, and then the verification information obtained from the verification request interface and the reference image are combined to perform recognition processing on the object in the reference image. In one embodiment, the verification information obtained by the computer device may be a unique identification code such as the corresponding object's phone number, ID card number, or random verification code. Similarly, the process of the computer device obtaining the verification information of the corresponding object is also executed after obtaining the authorization to use the corresponding information from the corresponding object.

[0035] Please see Figure 2 This is a schematic flowchart illustrating a business processing method provided in an embodiment of this application. This method can be executed by the aforementioned computer equipment, such as... Figure 2 As shown, the method may include:

[0036] S201, Obtain a reference image containing the target object, and obtain a cropped image obtained by cropping the reference image.

[0037] S202, using cropped images to perform target business related to the target object.

[0038] In steps S201 and S202, when the computer device detects that a target user needs to perform a trigger operation for a corresponding target service, it can guide the target object to perform image acquisition. Furthermore, it can call an image acquisition device to perform image acquisition processing on the target object, thereby obtaining an initial image set containing the target object. In one embodiment, the computer device can output a corresponding application interface and display the image acquisition process on the target object within the application interface, thus guiding the image acquisition process of the target object. Specifically, by calling the image acquisition device to perform one image acquisition on the target object, the computer device will obtain an image group containing one or more of color images, infrared images, and depth images. In this embodiment, the explanation mainly focuses on the computer device performing one image acquisition on the target object, resulting in an image group containing color images, infrared images, and depth images. Therefore, it can be understood that the initial image set obtained by the computer device, which calls the image acquisition device to perform image acquisition processing on the target object to obtain an initial image set containing the target object, includes one or more of the following initial images: color images, infrared images, and depth images, and the number of each type of initial image is one or more. Color images refer to color images captured by a color sensor under natural light, primarily used for image optimization and comparison recognition during the processing stage. Depth images, on the other hand, are obtained by capturing speckle-structured infrared light using an infrared sensor, and then resolving the speckle using depth units. In 3D computer graphics and computer vision, a depth image is an image or image channel containing information related to the distance from the surface of a scene object to the viewpoint. Each pixel in the depth image represents the vertical distance between the plane of the depth camera (i.e., the image acquisition device that performs depth image acquisition) and the plane of the object being photographed, typically represented by 16 bits in millimeters. In the processing stage, it is mainly used for object authenticity detection and auxiliary comparison recognition. Infrared images, on the other hand, are infrared images captured by an infrared sensor under diffuse infrared light, primarily used for object authenticity detection during the processing stage.

[0039] Based on the color, infrared, and depth images acquired by the computer device through the image acquisition device, since the computer device will perform multiple acquisitions when acquiring images of the target object, a large number of color, infrared, and depth images will be obtained. Therefore, after obtaining an initial image set containing a large number of color, infrared, and depth images, the computer device can perform optimization processing on the initial images to select preferred images. This optimization processing refers to selecting a set of color, depth, and infrared images that meet the preconditions of the object authenticity detection and comparison recognition algorithm. In other words, after the computer device calls upon multiple color, depth, and infrared images from the initial image set acquired by the image processing device and performs optimization processing, it will retain only a set of preferred images with obvious corresponding image features. After obtaining the selected preferred image set, the selected preferred images can be used to perform object authenticity detection on the target object, and the selected preferred images that pass the object authenticity detection are used as reference images containing the target object. In other words, the reference image acquired by the computer device includes at least one or more of color images, depth images, and infrared images. In this embodiment, the description mainly focuses on the fact that the reference image includes three images: color images, depth images, and infrared images.

[0040] When a computer device performs optimization processing on the initial images contained in an initial image set, it can optimize the color images contained in the initial image set based on the display attributes of the target object contained in each color image and the clarity of each color image. The display attributes of the target object in the color image include one or more of the following: display angle, object size, and centering. If the target service requires facial features of the target object, then during the optimization processing of the color images, one or more of the target object's facial angle, facial size, and centering, as well as the clarity of each color image, can be combined to optimize multiple color images containing the target object. In a specific implementation, the computer device can determine the display score of the target object in the corresponding color image based on its display angle, object size, and centering, and determine the clarity score of each color image. Then, based on the display score and clarity score of each color image, a selection score can be determined for each color image, thereby selecting the optimal color image from multiple color images based on the selection score. It is understandable that when computer equipment performs optimal processing on color images, it expects to select images where the target object's face is facing the acquisition device, the face size is relatively large, and the centering is good.

[0041] Furthermore, when optimizing the depth and infrared images in the initial image set, the optimization can be performed based on the image brightness of each infrared image in the initial image set; and the optimization can also be performed based on the image integrity of each depth image in the initial image set. In other words, the optimized image obtained by the computer device after optimizing the acquired initial image set is an image group consisting of color images, depth images, and infrared images. Then, after performing object authenticity detection on the obtained optimized image group, the optimized image group that passes the authenticity detection can be used as a reference image. In one embodiment, when the computer device optimizes the initial images contained in the initial image set, it can obtain multiple optimized images, i.e., multiple image groups consisting of color images, depth images, and infrared images. Then, object authenticity detection is performed to obtain the final image group used as the reference image from the multiple optimized image groups based on the object authenticity detection. Even if the optimized image obtained by the computer device is only one optimized image group, it is still necessary to perform object authenticity detection on this optimized image group before using it as a reference image for subsequent processing. Among them, object authenticity detection, also known as object liveness detection, is used to detect whether the corresponding image is directly collected from the object, rather than being collected from the image of the object itself. This ensures that the image collection process for the object is communicated to the relevant object, thereby guaranteeing the accuracy and feasibility of subsequent business operations.

[0042] In one embodiment, when the computer device performs optimization processing on the initial images in the initial image set, it may also perform optimization processing on the color images, depth images and infrared images separately according to the above optimization processing method. Assuming that the optimized image obtained contains 5 color images, 2 depth images and 3 infrared images, then the optimized image is subjected to object authenticity detection processing, and an image group containing one color image, one depth image and one infrared image is obtained as a reference image.

[0043] In one embodiment, after acquiring a reference image, the computer device can perform object cropping processing on the reference image to obtain a corresponding cropped image. Specifically, the computer device can first acquire the object features required for executing the target service, and then perform feature recognition processing on the reference image based on the required object features to determine the location region of the required object features in the reference image. This allows the computer device to perform image cropping processing on the reference image based on the location region of the required object features to obtain the cropped image. The object features are related to the target service to be performed by the target object. For example, if the target service is an electronic resource transfer service, the object features could be facial or head features, while if the target service is an access control service, they could be hand or pupil features. In this embodiment, the required object feature is mainly described in detail as a facial feature.

[0044] Since the reference image is an image group composed of preferred color images, infrared images, and depth images, when determining the location region of the desired object feature in the reference image, the computer device can select any image from the image group corresponding to the reference image as the recognition image of the object feature, thereby determining the location region of the object feature in the recognition image. Then, when the computer device performs image cropping on the reference image based on the location region of the desired object feature in the reference image to obtain a cropped image, it can perform image cropping on the recognition image based on the location region of the object feature corresponding to the recognition image, obtaining the cropped image corresponding to the recognition image. Similarly, it can perform image cropping on other images in the image group according to the same location region, obtaining the cropped images corresponding to the other images in the image group. If the color image contained in the reference image acquired by the computer device is as follows... Figure 3a The image shown, the target object contained in the color image is as follows Figure 3a If an object is marked with 30, then the cropped image obtained by cropping the object can be as follows: Figure 3a The image marked with 31 in the middle can also be as follows: Figure 3a The image is shown in the figure marked with 32.

[0045] After the computer device crops the reference image containing the target object, the cropped image can be used to execute the target service. If the cropped image contains the object features required for the target object to execute the target service, the computer device can successfully execute the target service using the cropped image. However, if the cropped image does not contain (or only partially contains) the object features required for the target object to execute the target service, an error will occur when the computer device uses the cropped image to execute the target service. When an error occurs when the computer device uses the cropped image to execute the target service, step S203 can be executed to analyze the cause of the error using the reference image.

[0046] S203, When there is an anomaly in the execution of the target service, the cropped image is cropped and located based on the reference image to obtain the cropping and location result.

[0047] S204, perform business anomaly detection processing on the cropped image based on the cropping positioning result, obtain the detection result, and generate the reason for the corresponding anomaly when executing the target business based on the detection result.

[0048] In steps S203 and S204, when an anomaly occurs in the computer device when performing target business using the cropped image, the reference image can be used to perform business anomaly detection processing on the cropped image. In a specific implementation, the computer device can perform cropping and positioning processing on the cropped image based on the reference image to obtain a cropping and positioning result. Then, the cropping and positioning result can be used to perform business anomaly detection processing on the cropped image to obtain a detection result. In one embodiment, after obtaining the reference image, the computer device performs image compression processing on the reference image before storing it to reduce the device resources required for storing the reference image. Specifically, after performing image compression processing on the reference image and obtaining the corresponding compressed image, the computer device can directly store the compressed image, or it can send the compressed image to other servers (such as business servers) for storage. The other server can be an independent physical server or a server cluster consisting of multiple physical servers. The other server can also be a node device connected to a blockchain network, thereby enabling the compressed image to be sent to the blockchain network for storage.

[0049] In one embodiment, the computer device can simultaneously compress the reference image while cropping it to obtain the cropped image, thereby improving the efficiency of its business execution. Specifically, the computer device can invoke a first thread to perform cropping on the reference image and obtain the cropped image. While the computer device is using the first thread to obtain the cropped image, a second thread can be invoked asynchronously to compress the reference image, obtaining a compressed image. The compressed image is added to an image transmission queue. After the computer device completes the target business using the cropped image, it can sequentially send the compressed images from the image transmission queue to the business server for storage. Specifically, the computer device can send the compressed images from the transmission queue to the business server for storage after completing the target business and when the corresponding processing resources are idle. This allows the computer device to ensure the target user experiences the corresponding business while using compressed images to guarantee the security of the business execution.

[0050] In one embodiment, the cropped image can also be sent to a business server for storage. When the cropped image is sent to the business server, the computer device can add a corresponding business identifier to the cropped image. The business identifier is used to indicate the target business. Similarly, when the computer device stores a compressed image of the reference image, it can also add a corresponding business identifier to the compressed image. Therefore, when the computer device stores the compressed image in the business server, it can associate and store the cropped image and the compressed image based on the business identifier. If the target business is an electronic resource transfer business, then the corresponding business identifier can be a resource transfer identifier, such as a unique resource transfer identifier (id) or an object identifier that performs the corresponding resource transfer. In one embodiment, when the computer device sends the cropped image to the business server for storage, it also stores it after packaging and compression. However, the computer device uses a low level of compression for the cropped image to enable effective identification of objects in the image later. On the other hand, the computer device uses a higher level of compression for the reference image. The compression of the reference image only needs to retain the distinguishability of a certain scene for anomaly localization. This allows the computer device to transmit the compressed image with minimal bandwidth while ensuring that the target object can perform the corresponding business, thus ensuring the security of business execution.

[0051] In one embodiment, to avoid risks to the execution of target services due to cropping of reference images and the inability to detect anomalies based on the reference images, it can be determined whether cropping is necessary before performing the image cropping operation. In a specific implementation, the computer device can perform scene recognition on the reference image. If the scene recognition result determines that there is a risk of anomalies in the execution of services, the reference image is not cropped, and the target service is executed directly using the reference image. In another embodiment, the computer device can also formulate a corresponding decision tree model based on factors such as network type, network speed, and environmental reliability. After obtaining a reference image containing the target object, the computer device can first determine whether to crop the reference image based on the above factors and the decision tree model. This allows the computer device to dynamically execute the image cropping function, achieving a balance between service execution risk and performance under the premise of dynamic traffic optimization and network speed optimization, facilitating the location of anomalies.

[0052] Since the reference image is stored after image compression processing, when the computer device performs cropping and positioning processing on the cropped image based on the reference image, it can first determine at least one feature point and the relative position of each feature point in the cropped image based on the image features of the cropped image. The at least one feature point determined by the computer device from the cropped image can be, for example, the image vertex and / or the image center point of the cropped image. After determining at least one feature point and the relative position of each feature point from the cropped image, the computer device can obtain the compressed image corresponding to the reference image and determine the matching point from the compressed image that matches the image features of any feature point. This allows the acquisition of the cropping shape corresponding to the cropped image. Then, based on the relative position, cropping shape, and matching point, the cropping boundary line can be determined from the compressed image of the reference image, and the cropping boundary is used as the cropping positioning result. The cropping location result can be used to indicate the cropping position of the cropped image in the reference image. When the computer device generates the cause of the corresponding abnormality when performing the target service based on the detection result, it can obtain the object features required by the target object when performing the target service, and judge the feature completeness of the object features contained in the cropped image according to the cropping position indicated by the cropping location result. When the computer device determines that the object features contained in the cropped image are incomplete, it determines that the corresponding abnormality occurs when performing the target service due to the incomplete features.

[0053] In one embodiment, the object features required for the target object to perform the target service are related to the service type of the target service performed by the target object. For example, when the target service is an electronic resource transfer service, the object features required to perform the electronic resource transfer service are the head features and / or facial features of the target object. Assuming that the object features required to perform the target service are the head features of the target object, if the computer device determines that the head features contained in the cropped image are only part of the head features of the target object, then it can be determined that the object features contained in the cropped image are incomplete. In other embodiments, the computer device can also determine whether the object features contained in the cropped image are complete based on the ratio between the object features contained in the cropped image and the complete object features. In a specific implementation, the computer device can consider the object features contained in the cropped image to be complete when the ratio between the object features contained in the cropped image and the complete object features exceeds 0.5. In one embodiment, the value of the ratio used to determine whether the object features contained in the cropped image are complete can be set by the user and adjusted based on the requirements of recognition accuracy. It can be understood that when the recognition accuracy requirement is higher, the value of the ratio can be set to a larger value, such as 0.8 or 0.9, while the recognition accuracy requirement is lower, the value of the ratio can be set to a smaller value, such as 0.5 or 0.4. In other embodiments, the computer device can also determine the corresponding value of the ratio of whether the object features contained in the cropped image are complete based on its own feature recognition capabilities. When the computer device has strong feature recognition capabilities, it means that the computer device can obtain more object features from the corresponding image. In this case, the ratio can be set to a smaller value, while the computer device can still obtain object features for object identification from the cropped image. When the computer device has weak feature recognition capabilities, it means that the computer device needs more object features from the corresponding image to identify the contained objects. In this case, the ratio can be set to a larger value to ensure that the computer device can accurately judge the objects contained in the cropped image. That is to say, the computer device can dynamically adjust the ratio used to determine whether the object features contained in the cropped image are complete based on its own object recognition capabilities, thereby achieving a match with the object recognition capabilities of the computer device.

[0054] In one embodiment, when the computer device detects the cause of the corresponding anomaly when executing the target service based on the detection result obtained from the service anomaly detection processing of the cropped image using the reference image, it can also determine whether the object contained in the cropped image is the target object based on the cropping position indicated by the cropping positioning result. If the cropped image does not contain the target object, it is determined that the anomaly in the execution of the target service is caused by an error in locating the target object in the reference image. In a specific implementation, if the target object in the reference image obtained by the computer device is object A, and the reference image also contains object B, and if the cropped image obtained by the computer device based on the cropping processing of the reference image contains object B, then the computer device determines that the object contained in the obtained cropped image is not the target object, and it can be determined that the computer device encountered an anomaly in object positioning (i.e., object recognition) of the target object in the reference image during the cropping processing. If the computer device determines that the cropped image is caused by an abnormal positioning of an object contained in the reference image, resulting in an abnormal execution of subsequent business operations, the computer device can reposition the target object in the reference image and, based on the positioning result of the repositioning process, re-crop the reference image to obtain a new cropped image. Then, the new cropped image can be used to execute the target business and obtain the execution result for the target business, thereby ensuring the smooth execution of the target business by the computer device.

[0055] The following describes the business processing flow when the target service is a resource transfer service, and the computer device is a terminal device held by the target object. In one embodiment, when the target service is a resource transfer service, the terminal device held by the target object can also be called the front screen, and the terminal device used to receive the electronic resources transferred by the target object can be called the back screen. Then, when the target object uses its held terminal device to perform electronic resource transfer, the display interfaces of the front screen and the back screen can be as follows: Figure 3b As shown, where, Figure 3b The interface shown above is a schematic diagram of the front screen of the target object transferring electronic resources using its own terminal device. The interface shown below is a schematic diagram of the back screen of the device receiving the transferred electronic resources from the target object. Figure 3b As shown, when a target object uses its own terminal device to transfer electronic resources, the terminal device receiving the transferred electronic resources can confirm and view the operations performed by the target object on its own terminal device in real time to ensure the security of the electronic resource transfer process. Similarly, the terminal receiving the electronic resource transfer can only view the electronic resource transfer process performed by the target object on its own terminal device after obtaining the corresponding viewing authorization from the target object.

[0056] The process of detecting the cause of anomalies in the execution of a target service using a corresponding terminal device, and when such anomalies occur, can be specifically described as follows: Figure 3c As shown, the specific steps can be performed as follows:

[0057] ① The target object triggers the terminal device to launch the application corresponding to the target service, and collects the object image based on the application to guide the acquisition of the initial image of the target object (including multiple color images, infrared images and depth images of the target object). Then, the acquired initial image is subjected to image optimization processing and object authenticity detection.

[0058] ② Obtain the processing results of optimization processing and object authenticity detection, and obtain the color image, infrared image and depth image that have passed object authenticity detection and optimization processing as reference images;

[0059] ③ Crop the face area in the selected color image, infrared image, and depth image to obtain the cropped image; at the same time, start another thread (i.e., the second thread) to perform compression operations on the color image, infrared image, and depth image (the compression here mainly changes the image clarity. Since the images here are mainly used for the location process when payment problems occur, they do not need to be too high-definition; they only need to have a certain degree of resolution).

[0060] ④ The cropped image is packaged and sent directly to the backend (i.e., the business server) for final identity verification, and a payment code (resource transfer voucher) is returned to indicate successful electronic resource transfer; the sub-thread (i.e., the second thread) adds the compressed image to the queue of the upload backend after receiving it. This thread should have low priority;

[0061] In other words, the computer device can generate and store a resource transfer certificate when the execution result of the target service indicates that the target service has been successfully executed. The resource transfer certificate can be used to record the amount of electronic resources transferred by the target user and the resource receiving user who received the electronic resources transferred by the target user.

[0062] ⑤ When the idle time arrives, the second thread uploads the compressed images in the queue to the background. When the compressed image is uploaded to the background, it carries a resource transfer ID. This resource transfer ID is used to uniquely identify a certain resource transfer process. The background can match the corresponding compressed image with a certain executed resource transfer based on the resource transfer ID to perform subsequent detection of the cause of the anomaly.

[0063] In this embodiment, after acquiring a reference image containing a target object, the computer device can crop the acquired reference image to obtain a cropped image. This cropped image is then used to execute target services related to the target object. This enables the computer device to perform corresponding services based on a small image, improving the efficiency of service execution and effectively saving computer processing resources. However, if an anomaly occurs when the computer device executes the target service based on the cropped image, in order to detect the cause of the anomaly, the computer device can use the original reference image containing the target object to perform service anomaly detection processing on the cropped image. This determines whether the anomaly is due to a cropping error, thus ensuring both the user experience of the target object during service execution and the security of service execution.

[0064] Based on the description of the above-described business processing method embodiments, this invention also proposes a business processing apparatus, which can be a computer program (including program code) running on the aforementioned computer device. This business processing apparatus can be used to execute, for example... Figure 2 For the aforementioned business processing method, please refer to [link / reference]. Figure 4 The business processing device includes an acquisition unit 401 and a processing unit 402.

[0065] The acquisition unit 401 is used to acquire a reference image containing the target object, and to acquire a cropped image obtained by cropping the reference image.

[0066] Processing unit 402 is used to perform target business related to the target object using the cropped image;

[0067] The processing unit 402 is further configured to perform cropping and positioning processing on the cropped image based on the reference image when there is an abnormality in the execution of the target service, and obtain a cropping and positioning result.

[0068] The processing unit 402 is further configured to perform business anomaly detection processing on the cropped image based on the cropping positioning result, obtain the detection result, and generate the cause of the corresponding anomaly when executing the target business based on the detection result.

[0069] In one embodiment, the acquisition unit 401 is specifically used for:

[0070] An image acquisition device is invoked to acquire and process images of a target object, resulting in an initial image set containing the target object. The initial images in the initial image set include one or more of the following: color images, infrared images, and depth images, and the number of each type of initial image is one or more.

[0071] The initial images contained in the initial image set are subjected to optimization processing to select preferred images from the initial image set;

[0072] The selected preferred images are used to perform object authenticity detection on the target object, and the selected preferred images that pass the object authenticity detection are used as reference images containing the target object.

[0073] In one embodiment, the acquisition unit 401 is specifically used for:

[0074] Based on the display attributes of the target objects contained in each color image in the initial image set, and the clarity of each color image, the color images contained in the initial image set are optimized.

[0075] Based on the image brightness of each infrared image in the initial image set, the infrared images contained in the initial image set are optimized.

[0076] Based on the image integrity of each depth image in the initial image set, the depth images contained in the initial image set are subjected to optimization processing.

[0077] In one embodiment, the processing unit 402 is specifically used for:

[0078] Obtain the object features required when performing the target service, and perform feature recognition processing on the reference image based on the required object features to determine the location region of the required object features in the reference image;

[0079] Based on the location region of the desired object features in the reference image, the reference image is cropped to obtain a cropped image.

[0080] In one embodiment, the reference image is an image group consisting of a preferred color image, an infrared image, and a depth image. The processing unit 402 is specifically used to: select any image from the image group corresponding to the reference image as an identification image of the object feature, and determine the location region of the object feature in the identification image.

[0081] The processing unit 402 is specifically used to: perform image cropping processing on the recognition image according to the position area of ​​the object feature corresponding to the recognition image to obtain the cropped image corresponding to the recognition image; and perform image cropping processing on other images in the image group according to the same position area to obtain the cropped images corresponding to the other images in the image group.

[0082] In one embodiment, the cropped image of the reference image is obtained by calling a first thread; the processing unit 402 is further configured to call a second thread to asynchronously perform image compression processing on the reference image during the process of calling the first thread to obtain the cropped image, thereby obtaining a compressed image of the reference image; wherein, the compressed image is added to an image transmission queue;

[0083] The processing unit 402 is further configured to, after completing the target service using the cropped image, sequentially send the compressed images in the image transmission queue to the service server and store the compressed images in the service server.

[0084] In one embodiment, the cropped image is also sent to the business server for storage. When the cropped image is sent to the business server, a business identifier is added to the cropped image, which is used to indicate the target business.

[0085] The processing unit 402 is specifically used for: storing the compressed image in the business server, including: associating and storing the cropped image and the compressed image in the business server based on the business identifier.

[0086] In one embodiment, the reference image is stored after being compressed to obtain a corresponding compressed image; the processing unit 402 is specifically used for:

[0087] Based on the image features of the cropped image, at least one feature point is determined from the cropped image, and the relative position of each feature point in the cropped image is determined.

[0088] Obtain the compressed image corresponding to the reference image, and determine the matching point in the compressed image that matches the image features of any feature point;

[0089] Obtain the cropping shape corresponding to the cropped image obtained;

[0090] In the compressed image of the reference image, the cropping boundary line is determined from the compressed image based on the relative position, the cropping shape, and the matching point, and the cropping boundary is used as the cropping positioning result.

[0091] In one embodiment, the cropping positioning result is used to indicate the cropping position of the cropped image in the reference image; the processing unit 402 is specifically used for:

[0092] Obtain the object features required by the target object when performing the target business, and judge the feature completeness of the object features contained in the cropped image according to the cropping position indicated by the cropping positioning result;

[0093] When it is determined that the object features contained in the cropped image are incomplete, it is determined that the incomplete features caused a corresponding abnormal situation when executing the target service.

[0094] In one embodiment, the cropping positioning result is used to indicate the cropping position of the cropped image in the reference image; the processing unit 402 is specifically used for:

[0095] Based on the cropping position indicated by the cropping positioning result, determine whether the object contained in the cropped image is the target object;

[0096] If the cropped image does not contain the target object, it is determined that the abnormal situation in executing the target service is caused by the incorrect positioning of the target object in the reference image.

[0097] In one embodiment, the processing unit 402 is further configured to perform relocation processing on the target object in the reference image, and re-crop the reference image according to the relocation result to obtain a new cropped image.

[0098] The processing unit 402 is further configured to perform the target service using a new cropped image to obtain an execution result for the target service.

[0099] In one embodiment, the target service includes a resource transfer service; the processing unit 402 is further configured to generate a resource transfer certificate if the execution result of the target service indicates that the target service was successfully executed, the resource transfer certificate being used to record the quantity of electronic resources transferred by the target user, and the resource receiving user who received the electronic resources transferred by the target user;

[0100] The processing unit 402 is also used to store the resource transfer certificate.

[0101] In this embodiment, after the acquisition unit 401 acquires a reference image containing the target object, the processing unit 402 can crop the acquired reference image to obtain a cropped image. The cropped image is then used to execute the target service related to the target object, realizing the implementation of the corresponding service based on the small image and improving the efficiency of service execution, thereby effectively saving processing resources. When an abnormal situation occurs when the processing unit 402 executes the target service based on the cropped small image, in order to detect the cause of the abnormality, the processing unit 402 can use the original reference image containing the target object to perform service abnormality detection processing on the cropped image, thereby determining whether the abnormality in executing the target service is caused by a cropping error. This ensures both the user experience of the target object in executing the target service and the security of service execution.

[0102] Please see Figure 5 This is a schematic block diagram of the structure of a computer device provided in an embodiment of the present invention. Figure 5 The computer device shown in this embodiment may include: one or more processors 501; one or more input devices 502; one or more output devices 503; and a memory 504. The processors 501, input devices 502, output devices 503, and memory 504 are connected via a bus 505. The memory 504 stores a computer program, which includes program instructions, and the processor 501 executes the program instructions stored in the memory 504.

[0103] The memory 504 may include volatile memory, such as random-access memory (RAM); the memory 504 may also include non-volatile memory, such as flash memory, solid-state drive (SSD), etc.; the memory 504 may also include a combination of the above types of memory.

[0104] The processor 501 may be a central processing unit (CPU). The processor 501 may further include hardware chips. These hardware chips may be application-specific integrated circuits (ASICs), programmable logic devices (PLDs), etc. The PLD may be a field-programmable gate array (FPGA), generic array logic (GAL), etc. The processor 501 may also be a combination of the above structures.

[0105] In this embodiment of the invention, the memory 504 is used to store a computer program, the computer program including program instructions, and the processor 501 is used to execute the program instructions stored in the memory 504 to implement the above-mentioned... Figure 2 The steps of the corresponding method.

[0106] In one embodiment, the processor 501 is configured to invoke the program instructions to execute:

[0107] Obtain a reference image containing the target object, and obtain a cropped image obtained by cropping the reference image.

[0108] The cropped image is used to perform target services related to the target object;

[0109] When the execution of the target service is abnormal, the cropped image is cropped and located according to the reference image to obtain the cropping and location result;

[0110] Based on the cropping and positioning results, the cropped image is subjected to business anomaly detection processing to obtain detection results, and the reasons for the corresponding anomalies generated when the target business is executed are generated based on the detection results.

[0111] In one embodiment, the processor 501 is configured to invoke the program instructions to execute:

[0112] An image acquisition device is invoked to acquire and process images of a target object, resulting in an initial image set containing the target object. The initial images in the initial image set include one or more of the following: color images, infrared images, and depth images, and the number of each type of initial image is one or more.

[0113] The initial images contained in the initial image set are subjected to optimization processing to select preferred images from the initial image set;

[0114] The selected preferred images are used to perform object authenticity detection on the target object, and the selected preferred images that pass the object authenticity detection are used as reference images containing the target object.

[0115] In one embodiment, the processor 501 is configured to invoke the program instructions to execute:

[0116] Based on the display attributes of the target objects contained in each color image in the initial image set, and the clarity of each color image, the color images contained in the initial image set are optimized.

[0117] Based on the image brightness of each infrared image in the initial image set, the infrared images contained in the initial image set are optimized.

[0118] Based on the image integrity of each depth image in the initial image set, the depth images contained in the initial image set are subjected to optimization processing.

[0119] In one embodiment, the processor 501 is configured to invoke the program instructions to execute:

[0120] Obtain the object features required when performing the target service, and perform feature recognition processing on the reference image based on the required object features to determine the location region of the required object features in the reference image;

[0121] Based on the location region of the desired object features in the reference image, the reference image is cropped to obtain a cropped image.

[0122] In one embodiment, the reference image is an image group consisting of a preferred color image, an infrared image, and a depth image. The processor 501 is configured to call the program instructions to perform: selecting any image from the image group corresponding to the reference image as an identification image of the object feature, and determining the location region of the object feature in the identification image.

[0123] The processor 501 is configured to call the program instructions to perform: image cropping processing on the recognition image according to the position region of the object feature corresponding to the recognition image to obtain a cropped image corresponding to the recognition image; and image cropping processing on other images in the image group according to the same position region to obtain cropped images corresponding to other images in the image group.

[0124] In one embodiment, the cropped image of the reference image is obtained by calling a first thread; the processor 501 is configured to call the program instructions to execute:

[0125] During the process of obtaining the cropped image by calling the first thread, the second thread is called asynchronously to perform image compression processing on the reference image to obtain a compressed image of the reference image; wherein, the compressed image is added to the image transmission queue;

[0126] After the target service is completed using the cropped image, the compressed images in the image transmission queue are sequentially sent to the service server and stored in the service server.

[0127] In one embodiment, the cropped image is also sent to the business server for storage. When the cropped image is sent to the business server, a business identifier is added to the cropped image, which is used to indicate the target business. The processor 501 is configured to call the program instructions to execute:

[0128] The cropped image and the compressed image are associated and stored in the business server based on the business identifier.

[0129] In one embodiment, the reference image is stored after being compressed to obtain a corresponding compressed image; the processor 501 is configured to invoke the program instructions to execute:

[0130] Based on the image features of the cropped image, at least one feature point is determined from the cropped image, and the relative position of each feature point in the cropped image is determined.

[0131] Obtain the compressed image corresponding to the reference image, and determine the matching point in the compressed image that matches the image features of any feature point;

[0132] Obtain the cropping shape corresponding to the cropped image obtained;

[0133] In the compressed image of the reference image, the cropping boundary line is determined from the compressed image based on the relative position, the cropping shape, and the matching point, and the cropping boundary is used as the cropping positioning result.

[0134] In one embodiment, the cropping positioning result is used to indicate the cropping position of the cropped image in the reference image; the processor 501 is configured to call the program instructions to execute:

[0135] Obtain the object features required by the target object when performing the target business, and judge the feature completeness of the object features contained in the cropped image according to the cropping position indicated by the cropping positioning result;

[0136] When it is determined that the object features contained in the cropped image are incomplete, it is determined that the incomplete features caused a corresponding abnormal situation when executing the target service.

[0137] In one embodiment, the cropping positioning result is used to indicate the cropping position of the cropped image in the reference image; the processor 501 is configured to call the program instructions to execute:

[0138] Based on the cropping position indicated by the cropping positioning result, determine whether the object contained in the cropped image is the target object;

[0139] If the cropped image does not contain the target object, it is determined that the abnormal situation in executing the target service is caused by the incorrect positioning of the target object in the reference image.

[0140] In one embodiment, the processor 501 is configured to invoke the program instructions to execute:

[0141] The target object in the reference image is repositioned, and based on the repositioning result, the reference image is cropped again to obtain a new cropped image.

[0142] The target service is executed using the new cropped image to obtain the execution result for the target service.

[0143] In one embodiment, the target service includes a resource transfer service; the processor 501 is configured to invoke the program instructions to execute:

[0144] If the execution result of the target service indicates that the target service was successfully executed, a resource transfer certificate is generated. The resource transfer certificate is used to record the quantity of electronic resources transferred by the target user and the resource receiving user who received the electronic resources transferred by the target user.

[0145] Store the resource transfer certificate.

[0146] This invention provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the aforementioned actions. Figure 2 The method embodiment shown. The computer-readable storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0147] The above-disclosed embodiments are merely partial examples of the present invention and should not be construed as limiting the scope of the invention. Those skilled in the art will understand that all or part of the processes for implementing the above embodiments, and equivalent variations made in accordance with the claims of the present invention, still fall within the scope of the invention.

Claims

1. A business processing method, characterized in that, include: Obtain a reference image containing the target object, and obtain a cropped image obtained by cropping the reference image. The cropped image is used to perform target services related to the target object; When the execution of the target service is abnormal, the cropped image is cropped and located according to the reference image to obtain the cropping and location result; Based on the cropping positioning result, the cropped image is subjected to business anomaly detection processing to obtain a detection result, and a reason for the corresponding anomaly generated when executing the target business is generated based on the detection result; wherein, the detection result is used to indicate whether the cropped image contains the target object; when the detection result indicates that the cropped image contains the target object, the generated reason is that the execution anomaly of the target business is caused by the cropping of the reference image, and when the detection result indicates that the cropped image does not contain the target object, the generated reason is that the execution anomaly of the target business is not caused by the cropping of the reference image.

2. The method as described in claim 1, characterized in that, The step of obtaining a reference image containing the target object includes: An image acquisition device is invoked to acquire and process images of a target object, resulting in an initial image set containing the target object. The initial images in the initial image set include one or more of the following: color images, infrared images, and depth images, and the number of each type of initial image is one or more. The initial images contained in the initial image set are subjected to optimization processing to select preferred images from the initial image set; The selected preferred images are used to perform object authenticity detection on the target object, and the selected preferred images that pass the object authenticity detection are used as reference images containing the target object.

3. The method as described in claim 2, characterized in that, The optimization process for the initial images contained in the initial image set includes: Based on the display attributes of the target objects contained in each color image in the initial image set, and the clarity of each color image, the color images contained in the initial image set are optimized. Based on the image brightness of each infrared image in the initial image set, the infrared images contained in the initial image set are optimized. Based on the image integrity of each depth image in the initial image set, the depth images contained in the initial image set are subjected to optimization processing.

4. The method as described in claim 1, characterized in that, The step of obtaining the cropped image obtained by cropping the reference image includes: Obtain the object features required when performing the target service, and perform feature recognition processing on the reference image based on the required object features to determine the location region of the required object features in the reference image; Based on the location region of the desired object features in the reference image, the reference image is cropped to obtain a cropped image.

5. The method as described in claim 4, characterized in that, The reference image is an image group consisting of a preferred color image, an infrared image, and a depth image. Determining the location region of the desired object feature in the reference image includes: selecting any image from the image group corresponding to the reference image as the recognition image of the object feature, and determining the location region of the object feature in the recognition image. The step of cropping the reference image according to the position region of the required object feature in the reference image to obtain a cropped image includes: cropping the recognition image according to the position region of the object feature corresponding to the recognition image to obtain a cropped image corresponding to the recognition image; and cropping other images in the image group according to the same position region to obtain cropped images corresponding to the other images in the image group.

6. The method as described in claim 1, characterized in that, The cropped image of the reference image is obtained by calling the first thread; the method further includes: During the process of obtaining the cropped image by calling the first thread, the second thread is called asynchronously to perform image compression processing on the reference image to obtain a compressed image of the reference image; wherein, the compressed image is added to the image transmission queue; After the target service is completed using the cropped image, the compressed images in the image transmission queue are sequentially sent to the service server and stored in the service server.

7. The method as described in claim 6, characterized in that, The cropped image is also sent to the business server for storage. When the cropped image is sent to the business server, a business identifier is added to the cropped image. The business identifier is used to indicate the target business. Storing the compressed image in the business server includes: associating and storing the cropped image and the compressed image in the business server based on the business identifier.

8. The method as described in claim 1, characterized in that, The reference image is stored after being compressed to obtain a corresponding compressed image; the cropping and positioning process based on the reference image to obtain the cropping and positioning result includes: Based on the image features of the cropped image, at least one feature point is determined from the cropped image, and the relative position of each feature point in the cropped image is determined. Obtain the compressed image corresponding to the reference image, and determine the matching point in the compressed image that matches the image features of any feature point; Obtain the cropping shape corresponding to the cropped image obtained; In the compressed image of the reference image, the cropping boundary line is determined from the compressed image based on the relative position, the cropping shape, and the matching point, and the cropping boundary is used as the cropping positioning result.

9. The method as described in claim 1, characterized in that, The cropping positioning result is used to indicate the cropping position of the cropped image in the reference image; the step of generating the cause of the corresponding anomaly when executing the target service based on the detection result includes: Obtain the object features required by the target object when performing the target business, and judge the feature completeness of the object features contained in the cropped image according to the cropping position indicated by the cropping positioning result; When it is determined that the object features contained in the cropped image are incomplete, it is determined that the incomplete features caused a corresponding abnormal situation when executing the target service.

10. The method as described in claim 1, characterized in that, The cropping positioning result is used to indicate the cropping position of the cropped image in the reference image; The step of generating the cause of the corresponding anomaly during the execution of the target service based on the detection results includes: Based on the cropping position indicated by the cropping positioning result, determine whether the object contained in the cropped image is the target object; If the cropped image does not contain the target object, it is determined that the abnormal situation in executing the target service is caused by the incorrect positioning of the target object in the reference image.

11. The method as described in claim 10, characterized in that, The method further includes: The target object in the reference image is repositioned, and based on the repositioning result, the reference image is cropped again to obtain a new cropped image. The target service is executed using the new cropped image to obtain the execution result for the target service.

12. The method as described in claim 1, characterized in that, The target service includes a resource transfer service; after executing the target service using the cropped image to obtain the execution result for the target service, the method further includes: If the execution result of the target service indicates that the target service was successfully executed, a resource transfer certificate is generated. The resource transfer certificate is used to record the quantity of electronic resources transferred by the target user and the resource receiving user who received the electronic resources transferred by the target user. Store the resource transfer certificate.

13. A business processing apparatus, characterized in that, include: The acquisition unit is used to acquire a reference image containing the target object, and to acquire a cropped image obtained by cropping the reference image. The processing unit is used to perform target business related to the target object using the cropped image; The processing unit is further configured to perform cropping and positioning processing on the cropped image based on the reference image when there is an abnormality in the execution of the target service, so as to obtain the cropping and positioning result; The processing unit is further configured to perform business anomaly detection processing on the cropped image based on the cropping positioning result, obtain a detection result, and generate a reason for the corresponding anomaly when executing the target business based on the detection result; wherein, the detection result is used to indicate whether the cropped image contains the target object; when the detection result indicates that the cropped image contains the target object, the generated reason is that the execution anomaly of the target business is caused by the cropping of the reference image, and when the detection result indicates that the cropped image does not contain the target object, the generated reason is that the execution anomaly of the target business is not caused by the cropping of the reference image.

14. A computer device, characterized in that, The device includes a processor, an input device, an output device, and a memory, wherein the processor, the input device, the output device, and the memory are interconnected, wherein the memory is used to store a computer program, the computer program including program instructions, and the processor is configured to invoke the program instructions to execute the method as described in any one of claims 1 to 12.

15. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, the computer program including program instructions that, when executed by a processor, cause the processor to perform the method as described in any one of claims 1 to 12.

16. A computer product, characterized in that, The computer product includes a computer program adapted to be loaded by a processor and executed as described in any one of claims 1 to 12.

Images