Method, system, device and storage medium for providing image collection suggestions

Abstract

The present application discloses a method, system, device and storage medium for providing image collection suggestions, belonging to the field of artificial intelligence technology. The method includes obtaining an image field combination required to make the target business meet the information completeness requirements; according to the image field combination, the image type set available for collection, and the image fields contained in the image corresponding to the image type in the image type set, the image type of each image that respectively has at least one image field in the image field combination is determined from the image type set; based on the determined image type, the type of image recommended for collection is determined. Starting from meeting the information completeness requirements of the target business, an image field combination is obtained, and from the image type set available for collection, the image type of each image containing the image field combination is determined, so as to provide image collection suggestions, which can objectively and quickly provide image type collection suggestions that meet the current business processing needs.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence, and in particular to a method, system, device and storage medium for providing image collection suggestions. Background Art

[0002] When handling business, business operators usually require users to provide imaging materials to prove that the user meets the requirements of the target business to be handled. For example, when handling medical insurance reimbursement business, users are required to provide various types of images such as identity cards, medical insurance cards, disease diagnosis certificates, outpatient medical records, examination result reports, and fee receipts issued by medical institutions.

[0003] At present, for the collection requirements of image types, especially for the types of images that users need to provide in special business scenarios, business handlers often provide users with image collection suggestions based on business scenarios and experience. For example, in the case of death claims, the images that users need to provide include but are not limited to documents proving the relationship between the beneficiary of the death insurance and the insured. The specific type of image to be selected as the proof document needs to be determined by the salesperson based on the actual relationship between the beneficiary and the insured, whether the beneficiary is deceased, and other conditions.

[0004] According to the above method, it often happens that the image materials provided by the user once cannot meet the needs of handling the target business and need to be provided again; or the user is required to provide too many unnecessary types of image materials, which affects the user experience. In short, from the perspective of collecting images, this method has the problem of low image collection efficiency. Summary of the invention

[0005] The embodiments of the present application provide a method, system, device and storage medium for providing image collection suggestions, which are used to solve the problem in the prior art that human collection suggestions are required for images required to handle target business, resulting in low image collection efficiency.

[0006] The present application embodiment adopts the following technical solutions:

[0007] A method for providing an image collection suggestion includes:

[0008] Obtaining a combination of image fields required to make the target business meet the information completeness requirement; the image field is a field contained in the image;

[0009] According to the image field combination, the image type set available for collection, and the image fields contained in the images corresponding to the image types in the image type set, determine, from the image type set, the image type of each image respectively having at least one image field in the image field combination;

[0010] Based on the determined image type, the type of image recommended for collection is determined.

[0011] A system for providing image collection suggestions, comprising:

[0012] A field combination acquisition unit is used to acquire an image field combination required to make the target business meet the information completeness requirement; the image field is a field contained in the image;

[0013] an image processing unit, configured to determine, from the image type set, image types of each image having at least one image field in the image field combination, based on the image field combination, the image type set available for collection, and the image fields contained in the images corresponding to the image types in the image type set;

[0014] The collection suggestion determination unit determines the type of images recommended for collection based on the determined image type.

[0015] A computing device, comprising: a memory and a processor, wherein:

[0016] The memory is used to store computer programs;

[0017] The processor is coupled to the memory, and is configured to execute the computer program stored in the memory to perform the method.

[0018] A computer-readable storage medium storing a computer program, wherein the computer program can implement the method when executed by a computer.

[0019] At least one of the above technical solutions adopted in the embodiments of the present application can achieve the following beneficial effects:

[0020] Starting from the image fields included in the information completeness requirements of the target business, an image field combination that meets the information completeness requirements is obtained. Based on the image field combination, the image type of each image that includes the image field combination is determined from the set of image types available for collection, thereby providing image collection suggestions. Since the embodiment of the present application shows image collection suggestions obtained based on image fields that meet the information completeness requirements, the obtained image collection suggestions meet the processing requirements of the target business. The entire process does not require human participation, and can objectively and quickly provide image type collection suggestions that meet the current business processing requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 A specific implementation flow chart of a method for providing image collection suggestions provided in an embodiment of the present application;

[0023] Figure 2 A schematic diagram of a decision tree obtained in one implementation of a method for providing image collection suggestions provided in an embodiment of the present application;

[0024] Figure 3 Another implementation flow chart of a method for providing image collection suggestions provided in an embodiment of the present application;

[0025] Figure 4 A flowchart of another implementation of a method for providing image collection suggestions provided in an embodiment of the present application;

[0026] Figure 5 A schematic diagram of the structure of a system for providing image collection suggestions provided in an embodiment of the present application;

[0027] Figure 6 A schematic diagram of a model training and configuration interface in a system for providing image collection suggestions provided in an embodiment of the present application;

[0028] Figure 7 It is a schematic diagram of a new model training interface in a system for providing image collection suggestions provided in an embodiment of the present application;

[0029] Figure 8 It is a schematic diagram of a training model interface in a system for providing image collection suggestions provided in an embodiment of the present application;

[0030] Fig. 9 It is a schematic diagram of a model viewing interface in a system for providing image collection suggestions provided in an embodiment of the present application;

[0031] Fig.10 It is a schematic diagram of an interface for viewing completeness review rules in a system for providing image collection suggestions provided in an embodiment of the present application;

[0032] Fig.11 It is a schematic diagram of an interface for viewing image combination and association analysis of a model in a system for providing image collection suggestions provided in an embodiment of the present application;

[0033] Fig.12 A schematic diagram of the specific structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0034] In order to make the purpose, technical solution and advantages of the present application clearer, the technical solution of the present application will be clearly and completely described below in combination with the specific embodiments of the present application and the corresponding drawings. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present application.

[0035] Those skilled in the art may appreciate that, with the development of technology and the emergence of new scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0036] The terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and need not be used to describe a specific order or sequential order. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances, which is only to describe the distinction mode adopted by the objects of the same attributes when describing in the embodiments of the present application. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, so that the process, method, system, product or equipment comprising a series of units need not be limited to those units, but may include other units that are not clearly listed or inherent to these processes, methods, products or equipment.

[0037] Example 1

[0038] Embodiment 1 of the present application provides a method for providing image collection suggestions, which is used to solve the problem in the prior art that human collection suggestions are required for images required for handling target business, resulting in low image collection efficiency.

[0039] The executor of the method may be any computing device that can implement the method, such as a server, a mobile phone, a personal computer, a smart wearable device, a smart robot, and the like.

[0040] The different steps of the method may be implemented by the same execution subject or by different execution subjects. The embodiment of the present application does not limit the execution subject used to implement the method.

[0041] In addition, the execution order of different steps is not limited in the embodiment of the present application. When using the method provided in the embodiment of the present application, the execution order of different steps can be adjusted according to actual needs.

[0042] For ease of description, the method provided in the embodiment of the present application is described in detail below by taking a device that provides image collection suggestions as an example of the execution subject of the method.

[0043] like Figure 1As shown, a method for providing image collection suggestions provided in an embodiment of the present application includes:

[0044] Step 101, obtaining a combination of image fields required to make the target business meet the information completeness requirement; the image field is a field contained in the image.

[0045] In one implementation, a classification model is used to train a set of image types that can be collected to obtain an image field combination that meets information completeness requirements.

[0046] In this implementation, the classification model uses a decision tree model to construct an information completeness decision tree, trains the image type set available for collection, and obtains an image field combination that meets the information completeness requirements. Rule mining is performed on the image type data collected historically for the target business, and after calculating the collected images, the image field combination required to meet the completeness requirements is obtained. In actual situations, when the business scenario is complex and there are many fields to be collected, a variety of image documents may be collected, and not every transaction will provide the same document or the same field. Therefore, for a target business, multiple image field combinations that meet the information completeness requirements may be obtained.

[0047] The decision tree model has a tree structure. In classification problems, it represents the process of classifying data based on features. It can be considered as a collection of if-then rules. Each internal node represents a test on an attribute, each branch represents a test output, and each leaf node represents a category.

[0048] The implementation principle of the decision tree algorithm is (divided into the following three steps):

[0049] Step 1: Feature selection; feature selection determines which features are used for judgment. The role of feature selection is to screen out features that are highly correlated with the classification results, that is, features with strong classification capabilities. The criterion commonly used in feature selection is information gain. This application case uses the CART algorithm and uses the Gini coefficient instead of the information entropy model. The Gini index is also an important criterion for measuring purity. It reflects the probability that two samples randomly selected from a data set have inconsistent category labels. The smaller the Gini coefficient, the higher the purity.

[0050] Step 2: Decision tree generation: After selecting the features, trigger from the root node, calculate the Gini index of all features for the node, select the feature with the largest Gini index as the node feature, and establish child nodes according to different values ​​of the feature; use the same method to generate new child nodes for each child node until the Gini index is very small or there is no feature to select.

[0051] Step 3: Decision tree pruning: The main purpose of pruning is to combat "overfitting" and reduce the risk of overfitting by actively removing some branches. This case uses post-pruning processing. First, a complete decision tree is generated from the training set, and then the non-leaf nodes are examined from the bottom up. If replacing the subtree corresponding to the leaf node with a leaf node can improve the generalization performance of the decision tree, then the subtree is replaced with a leaf node. This case uses the holdout method, that is, a portion of the data is set aside as a validation set for performance evaluation. The final decision tree is obtained.

[0052] like Figure 2 As shown, a partial decision tree diagram obtained by taking medical insurance as the target business in the embodiment of the present application is shown. Figure 2 , the image field combinations that meet the completeness requirements are (ID number, consultation serial number, admission time, total medical expenses, total payment from the pooling fund, total out-of-pocket expenses, name of the disease diagnosed at discharge, name of the medical structure, medical category, bank account number) and (ID number, consultation serial number, admission time, total medical expenses, total payment from the pooling fund, total out-of-pocket expenses, name of the disease diagnosed at discharge, name of the disease diagnosed at admission, name of the medical institution, medical category, bank account number).

[0053] In one implementation, before obtaining the image field combination required for the target service to meet the information completeness requirement, the method further includes:

[0054] Through the classification model, the image type set available for collection is trained to obtain the image field combination corresponding to the target business meeting the information completeness requirements;

[0055] The corresponding relationship between the target service and the image field combination is stored in a preset database.

[0056] Step 102, according to the image field combination, the image type set available for collection, and the image fields contained in the images corresponding to the image types in the image type set, determine the image type of each image having at least one image field in the image field combination from the image type set.

[0057] The set of image types available for collection provides historically collected image types corresponding to different target services. Therefore, one image type may appear multiple times in the image types available for collection.

[0058] Step 103: Determine the type of images recommended for collection based on the determined image type.

[0059] Example 2

[0060] Embodiment 2 of the present application provides a method for providing image collection suggestions, which is used to solve the problem in the prior art that human collection suggestions are required for images required to handle target business, resulting in low image collection efficiency.

[0061] The executor of the method may be any computing device that can implement the method, such as a server, a mobile phone, a personal computer, a smart wearable device, a smart robot, and the like.

[0062] The different steps of the method may be implemented by the same execution subject or by different execution subjects. The embodiment of the present application does not limit the execution subject used to implement the method.

[0063] In addition, the execution order of different steps is not limited in the embodiment of the present application. When using the method provided in the embodiment of the present application, the execution order of different steps can be adjusted according to actual needs.

[0064] For ease of description, the method provided in the embodiment of the present application is described in detail below by taking a device that provides image collection suggestions as an example of the execution subject of the method.

[0065] like Figure 3 As shown, a method for providing image collection suggestions provided in an embodiment of the present application includes:

[0066] Step 301, obtaining a combination of image fields required to make the target business meet the information completeness requirement; the image field is a field contained in the image.

[0067] In one implementation, there are multiple image field combinations that enable the target business to meet the information completeness requirements, and the method also includes: determining the type of recommended image collection corresponding to each image field combination; calculating the union of the types of recommended image collection corresponding to each image field combination to obtain an image collection recommendation combination; and using the image collection recommendation combination as the image collection recommendation for the target business.

[0068] Step 302: Select each image type having a common field with the image field combination from the image type set available for collection.

[0069] In one implementation, selecting each image type having a common field with the image field combination from the image type set available for collection specifically includes:

[0070] Put the fields included in the image field combination into a field set, put the fields included in each image type in the image type set into a type set, and obtain multiple type sets corresponding to multiple image types;

[0071] The field set is intersected with each type set to obtain a plurality of type sets that have intersections with the field set, and the corresponding image types are obtained according to the plurality of type sets.

[0072] Step 303, selecting the image type with the highest frequency of occurrence in the set of image types available for collection, searching a preset database for an associated image type corresponding to the image type with the highest frequency of occurrence, and combining the image type with the associated image type to form an image combination.

[0073] In one implementation, before obtaining the image field combination required for the target business to meet the information completeness requirements, the method also includes: training the image types available for collection through an association analysis algorithm to obtain associated image types corresponding to each image type; and storing each image type and the associated image type corresponding to each image type in a preset database.

[0074] The association analysis algorithm is used to discover implicit relationships and rules between objects from large-scale data, and to find the associations between items in the data set. When both the confidence and lift reach the corresponding preset thresholds, the rule can be considered valid. The important concepts involved are as follows:

[0075] 1. Items and Itemsets

[0076] An item refers to an object in the analysis data; an item set is a set of items consisting of several items. For example, the set {ID card, invoice, medical settlement form, expense list, investigation report} is a 5-item set.

[0077] 2. Support

[0078] The probability of an item set appearing in a data set. That is, the number of times an item set appears in a record divided by the number of all records in the data set. Support reflects the frequency of an item set. Only when the support of an item set reaches a certain level, it is necessary to study the item set.

[0079] 3. Confidence

[0080] The probability that item set B will occur if item set A occurs. In the association rule {A->B}, the number of times A and B appear at the same time, divided by the number of times A appears.

[0081] Confidence reflects the reliability of the association rule. If the confidence of the association rule {A->B} is high, it means that when A occurs, B is likely to occur as well.

[0082] 4. Improvement

[0083] In the association rule {A->B}, lift refers to the confidence of {A->B} divided by the support of B.

[0084] The lift reflects the ratio of combination (applying association rules) to no combination (not applying association rules). If the lift is greater than 1, it means that applying the association rule is valuable. The greater the lift, the better the effect of applying the association rule.

[0085] 5. Frequent Itemsets

[0086] If the support of item set I meets the predefined minimum support threshold, then I is a frequent item set. If a frequent item set contains K elements, it is called a frequent K item set.

[0087] 6. Minimum support

[0088] A user- or expert-defined threshold for measuring support that represents the minimum statistical importance of an item set.

[0089] 7. Minimum Confidence

[0090] A user- or expert-defined threshold for measuring confidence, indicating the minimum reliability of an association rule.

[0091] A rule that satisfies both the minimum support threshold and the minimum confidence threshold is considered a strong rule. When setting the minimum support and minimum confidence, the initial values ​​are mostly set based on business experience, and then after multiple adjustments, the association rule results that match the business are obtained. In this application case, the parameters can be configured on the page, assuming that they are set to 0.6 and 0.4 respectively.

[0092] In this implementation, the image types available for collection are trained by the association analysis algorithm to obtain the associated image types corresponding to each image type, specifically including:

[0093] Traverse the combinations of various image types in the image type set available for collection, and each combination constitutes an item set;

[0094] Calculate the support of each item set, and select the item sets whose support is greater than the preset support threshold as frequent item sets;

[0095] For a frequent item set, it is split into two non-empty subsets to form an association rule between a non-empty subset and another non-empty subset. If the association rule is greater than a preset confidence threshold, the image type corresponding to the current non-empty subset is the associated image type of the image type corresponding to the other non-empty subset.

[0096] There are many ways to split a frequent item set into two non-empty subsets, and each different possibility should be considered. For example, the frequent item set {ID card, invoice, medical settlement form, expense list} can be split into: {ID card->invoice, medical settlement form, expense list}, {ID card, invoice->medical settlement form, expense list}, {medical settlement form->ID card, invoice, expense list} and other combinations. For each association rule, its confidence needs to be calculated separately, and only the association rules that meet the set minimum confidence are retained.

[0097] For example, the model judgment results are shown in the following table:

[0098]

[0099] It can be seen that only the second rule (ID card, invoice -> medical settlement form, expense list) meets the support and confidence thresholds. Its business meaning is: if the ID card and invoice can be obtained, the medical settlement form and expense list can also be obtained according to the rule.

[0100] In one implementation, searching a preset database for an associated image type corresponding to the image type with the highest frequency of occurrence specifically includes:

[0101] Searching a preset database for at least one associated image type of the image type with the highest appearance frequency as a first associated image type;

[0102] selecting at least one of the first associated image type and the image type with the highest appearance frequency to form a combined image, and searching for an associated image type of the combined image from a preset database;

[0103] The first associated image type and the associated image type of the combined image are used as the associated image type corresponding to the image type with the highest appearance frequency.

[0104] Step 304: If the fields included in the image combination cover all the fields of the image field combination, the image combination is used as the determined image type.

[0105] In one implementation, the fields included in the image combination cover all fields of the image field combination, specifically including:

[0106] Put the fields included in the image field combination into the field set, and put the fields included in the image combination into the image set; calculate the difference between the field set and the image set. If the calculation result is an empty set, the fields included in the image combination cover all the fields of the image field combination.

[0107] In one implementation, if the fields included in the image combination do not cover all fields of the image field combination, the method further includes:

[0108] The uncovered fields are taken as a new image field combination, and the image type set available for collection is excluding the image types corresponding to the image combination to serve as a new image type set; based on the new image field combination, the new image type set, and the new image fields contained in the images corresponding to the image types in the new image type set, new image types of each image having at least one image field in the new image field combination are determined from the new image type set; and the combination of the determined image type and the new image type is used as an image collection suggestion.

[0109] Step 305: Determine the type of images recommended for collection based on the determined image type.

[0110] Example 3

[0111] Embodiment 2 of the present application provides a method for providing image collection suggestions, which is used to solve the problem in the prior art that human collection suggestions are required for images required to handle target business, resulting in low image collection efficiency.

[0112] The executor of the method may be any computing device that can implement the method, such as a server, a mobile phone, a personal computer, a smart wearable device, a smart robot, and the like.

[0113] The different steps of the method may be implemented by the same execution subject or by different execution subjects. The embodiment of the present application does not limit the execution subject used to implement the method.

[0114] In addition, the embodiments of the present application do not limit the execution order of different steps. When using the method provided in the embodiments of the present application, the execution order of different steps can be adjusted according to actual needs.

[0115] For ease of description, the method provided in the embodiment of the present application is described in detail below by taking a device that provides image collection suggestions as an example of the execution subject of the method.

[0116] like Figure 4 As shown, a method for providing image collection suggestions provided in an embodiment of the present application includes:

[0117] Step 401: Obtain at least one image field combination that has passed the information completeness review based on the decision tree classification model, and put the image field combination into a set G. Each approved path corresponds to an image field combination, and the business meaning is the minimum field that needs to be provided to meet the current target business requirements.

[0118] Step 402, when set G is not an empty set, extract an image field combination from set G, put the fields included in the image field combination into set 1, and put the fields included in all image types (such as invoices and settlement sheets) available for collection into sets a, b, c, etc. one by one.

[0119] Step 403, intersect set 1 with set a, set b, set c, ... respectively. If any field is matched, there is an intersection, and all image types with intersection are found.

[0120] Step 404: Select the image type m with the highest probability of occurrence from all the image types with intersections, and put the image type m into set A.

[0121] It should be noted that the entire set of images available for collection includes the historically collected image type sets of different target services. Therefore, an image type x may be in the historically collected image type sets of multiple target services. Furthermore, the image type x may appear multiple times in the entire set of images available for collection. Therefore, in step 404, the image type with the highest probability of occurrence (or the image type with the highest frequency of occurrence) is selected. The significance of this is that the image type x is easy to obtain.

[0122] Step 405: put all associated image types of all image type combinations in set A into set A.

[0123] It should be noted that in this step, the image type initially existing in set A is m. If the associated image type of image type m is found to be n from the preset database, n is put into set A; then the associated image type of image type combination m and n is continued to be searched from the preset database. If the associated image type q still exists, image type q is put into set A, and the associated image types of image type combinations m, n, q, m, q and n, q are searched from the preset database respectively until there are no new associated image types.

[0124] Step 406: put the fields contained in all image types of set A into set 2

[0125] Step 407, calculate the difference between set 1 and set 2, that is, subtract the fields included in set 1 from the fields included in set 2.

[0126] Step 408, if the calculation result of step 407 is a non-empty set, put the image field corresponding to the calculation result into a new set 1, and put all images except the image types of set A into new sets a, b, c... Go to step 403, and execute steps 403 to 407 in sequence until the calculation result of step 407 is an empty set.

[0127] Step 409: If the calculation result of step 407 is an empty set, then obtain an image type collection combination suggestion (corresponding to set A), store the image types contained in set A into set R, and remove set 1 from set G. Determine whether set G is an empty set. If not, go to step 402 and execute steps 402 to 409 in sequence until set G is an empty set.

[0128] Step 410: Combining the image types corresponding to the set R as image type collection suggestions.

[0129] Example 4

[0130] In order to solve the problem that the existing technology requires manual collection suggestions for the images required for handling the target business, and the image collection efficiency is low. Based on the same inventive concept as the above embodiments of the present application, embodiment 4 of the present application provides a system for providing image collection suggestions, such as Figure 5 As shown, the functional units of the system include:

[0131] The field combination acquisition unit 501 is used to acquire the image field combination required to make the target business meet the information completeness requirement; the image field is a field contained in the image;

[0132] The image processing unit 502 is used to determine, from the image type set, the image type of each image that respectively has at least one image field in the image field combination according to the image field combination, the image type set available for collection, and the image fields contained in the images corresponding to the image types in the image type set;

[0133] The collection suggestion determination unit 503 determines the type of images recommended for collection based on the determined image types.

[0134] The field combination acquisition unit 501 acquires the image field combination from the preset database. The image field combination required for the target business to meet the information completeness requirement is obtained by training the model on the image type set available for collection, such as Figure 6 to Figure 11 A schematic diagram of a portion of the system interface for model training using the system for providing image collection suggestions described in this embodiment is provided.

[0135] like Figure 6 As shown in the figure, the model training and configuration interface is shown. For example, for the model with ID sys1, on the model rule query page, click Add to perform basic configuration of the model, including various parameters for training the model; click Train Model to start a model training; click Query Model to view the model training results and manually debug the rules.

[0136] like Figure 7As shown in the figure, a schematic diagram of the new model training interface is added, in which new parameters for configuring the training model are added, including the association rule support threshold, the association rule confidence threshold, and the cosine similarity threshold of the word vector.

[0137] like Figure 8 As shown in the figure, the training model interface diagram, click model training, upload the labeled training transaction data set, and the system performs asynchronous model training.

[0138] For trained model rules, this system supports click-to-view and manual debugging operations.

[0139] like Fig. 9 As shown in the figure, view the model interface diagram, from which you can see the image classification model, completeness review rule model, and image combination and association analysis model. The specific training result data set can be dynamically displayed on the page.

[0140] like Fig.10 As shown in the figure, a schematic diagram of the interface for viewing completeness review rules is shown, which shows the trained decision tree model.

[0141] like Fig.11 As shown in the figure, a schematic diagram of the interface for viewing the image combination and association analysis in the model is shown. The training results are not shown in the figure.

[0142] The image processing unit 502 selects the image type with the highest frequency of occurrence in the set of image types available for collection, searches for the associated image type corresponding to the image type with the highest frequency of occurrence from a preset database, and combines the image type with the image type with the associated image type into an image combination.

[0143] In one implementation, the system is also used to: train the image types available for collection through an association analysis algorithm to obtain associated image types corresponding to each image type; and store each image type and the associated image types corresponding to each image type in a preset database.

[0144] In one implementation, searching a preset database for an associated image type corresponding to the image type with the highest frequency of occurrence is specifically used for:

[0145] Search a preset database for at least one associated image type of the image type with the highest frequency of occurrence as a first associated image type; select the first associated image type and at least one of the image type with the highest frequency of occurrence to form a combined image, and search a preset database for associated image types of the combined image; and use the first associated image type and the associated image type of the combined image as the associated image type corresponding to the image type with the highest frequency of occurrence.

[0146] If the fields included in the image combination cover all the fields of the image field combination, the image combination is used as the determined image type.

[0147] In one implementation, the fields included in the image combination cover all fields of the image field combination, specifically used for:

[0148] Put the fields included in the image field combination into the field set, and put the fields included in the image combination into the image set; calculate the difference between the field set and the image set. If the calculation result is an empty set, the fields included in the image combination cover all the fields of the image field combination.

[0149] In one implementation, if the fields included in the image combination do not cover all fields of the image field combination, the system is further configured to:

[0150] The uncovered fields are taken as a new image field combination, and the image type set available for collection is excluding the image types corresponding to the image combination to serve as a new image type set; based on the new image field combination, the new image type set, and the new image fields contained in the images corresponding to the image types in the new image type set, new image types of each image having at least one image field in the new image field combination are determined from the new image type set; and the combination of the determined image type and the new image type is used as an image collection suggestion.

[0151] Example 5

[0152] Based on the same inventive concept as the aforementioned embodiments of the present application, embodiment 5 of the present application provides an electronic device to solve the problem in the prior art that images required for handling target business require manual collection suggestions, resulting in low image collection efficiency.

[0153] like Fig.12 As shown, the computing device includes: a memory 1201 and a processor 1202. The memory 1201 can be configured to store various other data to support operations on the electronic device. Examples of these data include instructions for any application or method for operating on the electronic device. The memory 1201 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic memory, a flash memory, a magnetic disk or an optical disk.

[0154] The processor 1202 is coupled to the memory 1201 and is used to execute the computer program stored in the memory 1201 to execute a method for providing image collection suggestions described in any one of Embodiments 1 to 3 of the present application.

[0155] When the processor 1202 executes the computer program in the memory 1101, in addition to the above functions, it can also realize other functions, and the details can be found in the description of the previous embodiments.

[0156] Further, such as Fig.12 As shown, the computing device also includes: a display 1204, a communication component 1203, a power component 1205, an audio component 1206 and other components. Fig.12 Only some components are shown schematically, and it does not mean that the computing device only includes Fig.12 Components shown.

[0157] Accordingly, an embodiment of the present application further provides a computer-readable storage medium storing a computer program, and when the computer program is executed by a computer, the methods provided in the above embodiments can be implemented.

[0158] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0159] Through the description of the above implementation methods, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0160] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for providing image collection suggestions, characterized in that: include: Obtaining a combination of image fields required to make the target business meet the information completeness requirement; the image field is a field contained in the image; According to the image field combination, the image type set available for collection, and the image fields contained in the images corresponding to the image types in the image type set, determine, from the image type set, the image type of each image respectively having at least one image field in the image field combination; Based on the determined image type, determining the type of image that is recommended to be collected; The determining of the image type of each image having at least one image field in the image field combination specifically includes: Selecting, from the set of image types available for collection, each image type having a common field with the image field combination; Selecting the image type with the highest frequency of occurrence among the image types available for collection, searching a preset database for an associated image type corresponding to the image type with the highest frequency of occurrence, and combining the image type with the associated image type to form an image combination; If the fields included in the image combination cover all the fields of the image field combination, the image combination is used as the determined image type; Associated image type: For a frequent item set, split it into two non-empty subsets, and form an association rule between a non-empty subset and another non-empty subset. If the association rule is greater than a preset confidence threshold, the image type corresponding to the current non-empty subset is the associated image type of the image type corresponding to the other non-empty subset; Selecting each image type having a common field with the image field combination from the image type set available for collection, specifically including: Put the fields included in the image field combination into a field set, put the fields included in each image type in the image type set into a type set, and obtain multiple type sets corresponding to multiple image types; Taking the intersection of the field set and each type set respectively, obtaining a plurality of type sets having intersections with the field set, and obtaining the corresponding image types according to the plurality of type sets; The step of searching the preset database for the associated image type corresponding to the image type with the highest appearance frequency specifically includes: Searching a preset database for at least one associated image type of the image type with the highest appearance frequency as a first associated image type; selecting at least one of the first associated image type and the image type with the highest appearance frequency to form a combined image, and searching for an associated image type of the combined image from a preset database; The first associated image type and the associated image type of the combined image are used as the associated image type corresponding to the image type with the highest appearance frequency.

2. The method according to claim 1, characterized in that The fields included in the image combination cover all the fields of the image field combination, including: Put the fields contained in the image field combination into the field collection, and put the fields contained in the image combination into the image collection; The difference between the field set and the image set is calculated. If the calculated result is an empty set, the fields included in the image combination cover all the fields of the image field combination.

3. The method according to claim 1, characterized in that If the fields included in the image combination do not cover all the fields of the image field combination, the method further includes: The uncovered fields are used as a new image field combination, and the image type set available for collection is reduced by removing the image type corresponding to the image combination to form a new image type set; According to the new image field combination, the new image type set, and the new image fields contained in the images corresponding to the image types in the new image type set, determining from the new image type set new image types of each image respectively having at least one image field in the new image field combination; The determined image type and the new image type are combined as an image collection suggestion.

4. The method according to claim 1, characterized in that: There are multiple image field combinations that enable the target service to meet the information completeness requirement, and the method further includes: Determine the type of imagery that is recommended to be collected for each image field combination; Calculating the union of the types of images recommended for collection corresponding to each image field combination to obtain an image collection recommendation combination; The image collection suggestion combination is used as an image collection suggestion for a target business.

5. The method according to claim 1, characterized in that: Before obtaining the image field combination required for the target service to meet the information completeness requirement, the method further includes: Through the classification model, the image type set available for collection is trained to obtain the image field combination corresponding to the target business meeting the information completeness requirements; The corresponding relationship between the target service and the image field combination is stored in a preset database.

6. The method according to claim 5, characterized in that The classification model is a decision tree model.

7. The method according to claim 1, characterized in that Before obtaining the image field combination required for the target service to meet the information completeness requirement, the method further includes: By using an association analysis algorithm, the image types available for collection are trained to obtain associated image types corresponding to each image type; The image types and the associated image types corresponding to the image types are stored in a preset database.

8. The method according to claim 7, characterized in that The image types available for collection are trained by the association analysis algorithm to obtain the associated image types corresponding to each image type, specifically including: Traverse the combinations of various image types in the image type set available for collection, and each combination constitutes an item set; Calculate the support of each item set, and select the item sets whose support is greater than the preset support threshold as frequent item sets; For a frequent item set, it is split into two non-empty subsets to form an association rule between a non-empty subset and another non-empty subset. If the association rule is greater than a preset confidence threshold, the image type corresponding to the current non-empty subset is the associated image type of the image type corresponding to the other non-empty subset.

9. A system for providing image collection suggestions, characterized in that: include: A field combination acquisition unit is used to acquire an image field combination required to make the target business meet the information completeness requirement; the image field is a field contained in the image; an image processing unit, configured to determine, from the image type set, image types of each image having at least one image field in the image field combination, based on the image field combination, the image type set available for collection, and the image fields contained in the images corresponding to the image types in the image type set; A collection suggestion determination unit determines the type of image that is recommended to be collected based on the determined image type; The determining of the image type of each image having at least one image field in the image field combination specifically includes: Selecting, from the set of image types available for collection, each image type having a common field with the image field combination; Selecting the image type with the highest frequency of occurrence among the image types available for collection, searching a preset database for an associated image type corresponding to the image type with the highest frequency of occurrence, and combining the image type with the associated image type to form an image combination; If the fields included in the image combination cover all the fields of the image field combination, the image combination is used as the determined image type; Association influence type: For a frequent item set, split it into two non-empty subsets, and form an association rule between a non-empty subset and another non-empty subset. If the association rule is greater than a preset confidence threshold, the image type corresponding to the current non-empty subset is the associated image type of the image type corresponding to the other non-empty subset; Selecting each image type having a common field with the image field combination from the image type set available for collection, specifically including: Put the fields included in the image field combination into a field set, put the fields included in each image type in the image type set into a type set, and obtain multiple type sets corresponding to multiple image types; Taking the intersection of the field set and each type set respectively, obtaining a plurality of type sets having intersections with the field set, and obtaining the corresponding image types according to the plurality of type sets; The step of searching the preset database for the associated image type corresponding to the image type with the highest appearance frequency specifically includes: Searching a preset database for at least one associated image type of the image type with the highest appearance frequency as a first associated image type; selecting at least one of the first associated image type and the image type with the highest appearance frequency to form a combined image, and searching for an associated image type of the combined image from a preset database; The first associated image type and the associated image type of the combined image are used as the associated image type corresponding to the image type with the highest appearance frequency.

10. A computing device, characterized in that include: Memory and processor, wherein: The memory is used to store computer programs; The processor is coupled to the memory, and is used to execute the computer program stored in the memory, so as to execute the method according to any one of claims 1 to 8.

11. A computer-readable storage medium storing a computer program, wherein the computer program can implement the method according to any one of claims 1 to 8 when executed by a computer.

Images