Method, apparatus, electronic device, and storage medium for extracting key information of express delivery

Abstract

The present invention discloses a method, apparatus, electronic device and storage medium for extracting key information of express parcels. The method includes: obtaining a waybill image of a current express parcel, and recognizing the waybill image to obtain a key information sequence of each key information of the express parcel in the waybill image; wherein, the key information sequence includes a text information sequence and a coordinate information sequence of each key information of the express parcel; extracting the key information of the current express parcel based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express parcel. Through the technical solution disclosed by the present invention, the problem in the prior art that the key information of different types of express parcel waybills cannot be directly recognized, resulting in a reduction in recognition efficiency, is solved, and the recognition effect of the key information of express parcels in different waybills is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of express logistics, and particularly to a method, device, electronic device and storage medium for extracting key information of express parcels. Background Art

[0002] With the rapid development of the Internet and the application of machine learning and deep learning in the industrial field, in the field of express delivery, the demand for automatic sorting of express parcels is increasing. When the express waybill information is in the electronic data read by the express sorting equipment, the automatic sorting equipment can be effectively used to realize the automatic sorting of express parcels, which greatly improves the efficiency.

[0003] Currently, the digitization of waybill information generally adopts the technology combining OCR and NLP. For a single type of waybill image, the traditional method is to use OCR detection and recognition in a specific area, and the text can be directly extracted according to the specific area category to obtain the content corresponding to the image. When the types of waybills increase, it is necessary to train an OCR detection and recognition model for each type of waybill image to extract the text content. Each single model is relatively simple, but when the scenario is complex, multiple waybills rely on multiple models for recognition, making the task of digitizing multiple waybill information complex and difficult to maintain. In the same business scenario, when multiple waybills exist simultaneously, if the type of a certain waybill is not known in advance, it is impossible to match a suitable model to correctly extract its key information, resulting in a decline in the recognition effect of the key information of the express parcel in the waybill. Summary of the Invention

[0004] The present invention provides a method, device, electronic device and storage medium for extracting key information of express parcels, so as to solve the problem in the prior art that the key information of different types of express waybills cannot be directly recognized, resulting in a reduction in recognition efficiency, and improve the recognition effect of the key information of express parcels in different waybills.

[0005] In a first aspect, an embodiment of the present invention provides a method for extracting key information of express parcels, the method comprising:

[0006] Obtain a waybill image of the current express parcel, and recognize the waybill image to obtain a key information sequence of each key information of the express parcel in the waybill image; wherein, the key information sequence includes a text information sequence and a coordinate information sequence of each key information of the express parcel.

[0007] Extract the key information of the current express parcel based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express parcel.

[0008] Optionally, the recognizing the waybill image to obtain a key information sequence of each key information of the express parcel in the waybill image includes:

[0009] Based on a pre-trained image sequence recognition model, perform image sequence recognition on the waybill image to obtain the initial waybill information sequence of each waybill sequence in the waybill image;

[0010] Obtain each text information sequence in the initial waybill information sequence, and based on the text information sequence and a preset key information sequence library, determine the key information sequence in the initial waybill information sequence.

[0011] Optionally, the express key information includes the express belonging unit and the express waybill number;

[0012] Correspondingly, the extracting the express key information of the current express based on the text information sequence and / or the coordinate information sequence to obtain the express key information of the current express includes:

[0013] Based on each coordinate information sequence and a preset sequence threshold range, determine the candidate key information sequence in the key information sequence;

[0014] Determine the name similarity between each text information sequence and each express unit name in a preset express unit name library, and based on each name similarity, determine the express belonging unit of the current express;

[0015] Determine the waybill number similarity between each text information sequence and the waybill number composition features preset, based on the waybill number similarity, determine the candidate waybill number information sequence in each candidate key information sequence, and based on the coordinate information sequence in the candidate waybill number information sequence and / or the express belonging unit of the current express, determine the express waybill number of the current express.

[0016] Optionally, the determining the express belonging unit of the current express based on each name similarity includes:

[0017] Obtain the maximum name similarity among each name similarity;

[0018] If the maximum name similarity is greater than a preset similarity threshold, determine the express unit corresponding to the maximum name similarity as the express belonging unit of the current express;

[0019] If the maximum name similarity is less than a preset similarity threshold, obtain the express waybill number of the current express, and based on the express waybill number, determine the express belonging unit of the current express.

[0020] Optionally, the express key information includes express three-section code information,

[0021] Correspondingly, the extracting the express key information of the current express based on the text information sequence and / or the coordinate information sequence to obtain the express key information of the current express includes:

[0022] Obtain the preset code composition features respectively, and extract information from the text information sequence based on the code composition features to obtain the three-segment code information of the current express delivery; wherein, the three-segment code information includes a province code, a city code, and a region code.

[0023] Optionally, the key express delivery information includes the recipient information and the sender information of the express delivery;

[0024] The extracting the key express delivery information of the current express delivery from the text information sequence and / or the coordinate information sequence to obtain the key express delivery information of the current express delivery includes:

[0025] Determine the first phone information sequence of the current express delivery based on the text information sequence, and determine the recipient information area of the current express delivery based on the first phone information sequence;

[0026] Determine the second phone information sequence or the second address information sequence of the current express delivery based on the text information sequence, and determine the sender information area of the current express delivery based on the second phone information sequence or the second address information sequence;

[0027] Determine the sequence area area of the key waybill sequence corresponding to each coordinate information sequence;

[0028] If the intersection ratio of the current sequence area area and the area of the recipient information area is greater than a preset intersection threshold, determine that the key express delivery information corresponding to the current sequence area area is the recipient information of the current express delivery;

[0029] If the intersection ratio of the current sequence area area and the area of the sender information area is greater than a preset intersection threshold, determine that the key express delivery information corresponding to the current sequence area area is the sender information of the current express delivery.

[0030] Optionally, the obtaining the key express delivery information of the current express delivery includes:

[0031] Extract information from the recipient information based on a pre-trained information extraction model to obtain the recipient delivery information of the current express delivery;

[0032] Correspondingly, the obtaining the key express delivery information of the current express delivery includes:

[0033] Extract information from the recipient information based on a pre-trained information extraction model to obtain the sender and recipient information of the current express delivery. The recipient information includes the sender phone information, the sender name information, and the sender address information.

[0034] In a second aspect, an embodiment of the present invention further provides an express delivery key information extraction device, which is characterized by including:

[0035] A key information sequence acquisition module, configured to obtain a waybill image of the current express delivery, and identify the waybill image to obtain a key information sequence of each express delivery key information in the waybill image; wherein, the key information sequence includes a text information sequence and a coordinate information sequence of each express delivery key information;

[0036] An express delivery key information acquisition module, configured to perform express delivery key information extraction on the current express delivery based on the text information sequence and / or the coordinate information sequence to obtain the express delivery key information of the current express delivery.

[0037] In a third aspect, an embodiment of the present invention further provides an electronic device, including:

[0038] At least one processor; and

[0039] A memory communicatively connected to the at least one processor; wherein,

[0040] The memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the express delivery key information extraction method of any embodiment of the present invention.

[0041] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, which stores computer instructions for causing a processor to implement the express delivery key information extraction method of any embodiment of the present invention when executed.

[0042] The technical solution of this embodiment specifically includes obtaining a waybill image of the current express delivery, and identifying the waybill image to obtain a key information sequence of each express delivery key information in the waybill image; wherein, the key information sequence includes a text information sequence and a coordinate information sequence of each express delivery key information; performing express delivery key information extraction on the current express delivery based on the text information sequence and / or the coordinate information sequence to obtain the express delivery key information of the current express delivery. The above technical solution solves the technical problem in the prior art that the recognition of different types of waybill images is inaccurate, resulting in low delivery efficiency, by performing OCR recognition on the current express delivery waybill image, obtaining a key information sequence containing the express delivery key information of the delivered express delivery, and performing intelligent parsing on the key information sequence to obtain the express delivery key information of the current express delivery, achieving the improvement of the recognition accuracy and recognition rate of the waybill, and thus improving the efficiency of express delivery.

[0043] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it used to limit the scope of the present invention. Other features of the present invention will become easily understandable through the following description. Brief Description of the Drawings

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0045] Figure 1 It is a flowchart of a method for extracting key information of express deliveries provided in Embodiment 1 of the present invention;

[0046] Figure 2 It is a schematic structural diagram of a device for extracting key information of express deliveries provided in Embodiment 2 of the present invention;

[0047] Figure 3 It is a schematic structural diagram of an electronic device provided in Embodiment 3 of the present invention. Detailed Embodiments

[0048] In order to enable those skilled in the art to better understand the solutions of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0049] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that such used data can be interchanged under appropriate circumstances so that the embodiments of the present invention described here can be implemented in an order other than those illustrated or described here.

[0050] The names of the messages or information exchanged between multiple devices in the embodiments of the present disclosure are only for illustrative purposes, and are not used to limit the scope of these messages or information.

[0051] It can be understood that before using the technical solutions disclosed in the embodiments of the present disclosure, the types, usage scopes, usage scenarios, etc. of the personal information involved in the present disclosure should be informed to users and user authorization should be obtained in an appropriate manner in accordance with relevant laws and regulations.

[0052] For example, when responding to an active request from a user, a prompt message is sent to the user to clearly prompt the user that the operation requested by the user will require obtaining and using the user's personal information. Thus, the user can autonomously choose whether to provide personal information to software or hardware such as an electronic device, an application program, a server, or a storage medium that performs the operations of the present disclosure technical solution based on the prompt message.

[0053] As an optional but non-limiting implementation manner, the manner of sending a prompt message to the user in response to receiving an active request from the user may be, for example, a pop-up window manner, and the prompt message may be presented in text in the pop-up window. In addition, the pop-up window may also carry a selection control for the user to select "agree" or "disagree" to provide personal information to the electronic device.

[0054] It can be understood that the above notification and user authorization process is only illustrative and does not limit the implementation manner of the present disclosure. Other manners that meet relevant laws and regulations can also be applied to the implementation manner of the present disclosure.

[0055] It can be understood that the data involved in the present technical solution (including but not limited to the data itself, the acquisition or use of the data) should comply with the requirements of corresponding laws, regulations and related regulations.

[0056] Embodiment 1

[0057] Figure 1 The present invention provides a flowchart of a method for extracting key information of an express delivery in Embodiment 1. This embodiment is applicable to the situation of identifying an express delivery package. The method can be executed by an express delivery key information extraction device, and the express delivery key information extraction device can be implemented in the form of hardware and / or software. The express delivery key information extraction device can be configured in a server or an intelligent terminal. As Figure 1 shown, the method includes:

[0058] S110. Obtain a waybill image of the current express delivery, and identify the waybill image to obtain a key information sequence of each key information of the express delivery in the waybill image.

[0059] In the embodiments of the present invention, an express item can be understood as an express parcel delivered by a user through an express company. In order to quickly and accurately deliver the express item, the key information of the express item can be integrated into the express waybill, and the express waybill can be pasted on the express parcel, so that the key information of the current express item can be obtained by scanning the express waybill during the delivery process. In other words, the waybill image of the express item can be understood as an image storing the key information of the express item required for delivering the current express item. It should be noted that the waybill image of the express waybill not only contains the key information of the current express item, but also includes other peripheral information of the express item such as the welfare information of the current express company. The key information of the express item can be understood as the information indispensable during the delivery process of the express item, such as the ownership information of the express item, the three-section code information of the express item, and the receiving and sending information of the express item.

[0060] It should be noted that there may be various types of waybill images of express items. For example, the waybill image types of different express companies are different, and for another example, the simple image types of different delivery services are different. This embodiment does not limit this. The positions and display forms of the key information of the express item in different types of waybill images are different. In order to solve the technical problem in the prior art that the recognition of different types of waybill images is inaccurate, resulting in low delivery efficiency, the technical solution in this embodiment first recognizes the obtained waybill image to obtain the key sequence information of the waybill image, and extracts the key information of the express item according to the key sequence information, so as to improve the fast and accurate recognition of different types of waybill images during the delivery process, thereby improving the delivery efficiency of the express item.

[0061] Specifically, in this embodiment, the method for obtaining the waybill image of the current express item may include but is not limited to scanning the current express item with a preset image acquisition device to obtain the waybill image of the current express item. On this basis, information recognition is performed on the obtained waybill image to obtain the key information sequence of each key information of the express item in the waybill image.

[0062] It should be explained that the information sequence can be understood as a sequence used to represent each information in the current express item. Optionally, the method for recognizing the key information sequence of each key information of the express item in the waybill image by recognizing the waybill image may include: performing image sequence recognition on the waybill image based on a pre-trained image sequence recognition model to obtain the initial waybill information sequence of each waybill sequence in the waybill image; obtaining each text information sequence in the initial waybill information sequence, and determining the key information sequence in the initial waybill information sequence based on the text information sequence and a preset key information sequence library.

[0063] Specifically, the obtained waybill image is input into a pre-trained image sequence recognition model to obtain an initial waybill information sequence output by the image sequence recognition model. Exemplarily, the image sequence recognition model in this embodiment can be recognized by OCR (Optical Character Recognition), and of course, other network models can also be used for recognition, which is not limited in this embodiment. Optionally, the initial waybill information sequence may include, but is not limited to, the text information sequence and the coordinate information sequence of each express delivery information, such as Figure 2 As shown, the information identification, information name, and position coordinates of each information in the initial waybill in the waybill image, etc. Correspondingly, the initial waybill information sequence includes both the key information sequence of the express delivery key information, such as key information like "Yang *" and "152***", and also includes the secondary information sequence of the express delivery peripheral information such as "red envelope" and "scan the code to draw".

[0064] On the basis of the above implementation manner, a preset key information sequence library is obtained, and the key information sequence library stores the text information sequences of each express delivery key information of the express delivery. Specifically, each text information sequence in the initial waybill information sequence is obtained, and each text information sequence is respectively matched with the text information sequences of each express delivery key information in the key information sequence library. Optionally, if the match is successful, it means that the express delivery information corresponding to the current text information sequence is the express delivery key information of the current express delivery; otherwise, if the match is not successful, it means that the express delivery information corresponding to the current text sequence is the express delivery peripheral information of the current express delivery.

[0065] It should be noted that in order to facilitate the determination of the key information sequence in the initial waybill information sequence, when the initial waybill information sequence is obtained, it can be sorted in a preset format to obtain the sorted initial waybill information sequence, for example, sorted into a sequence consistent with the sequence information in the key information sequence, so as to improve the recognition rate of the key information sequence, thereby further improving the information extraction of the express delivery key information in the waybill image.

[0066] S120. Extract the express delivery key information of the current express delivery based on the text information sequence and / or the coordinate information sequence to obtain the express delivery key information of the current express delivery.

[0067] In the embodiment of the present invention, the express delivery key information includes, but is not limited to, the express delivery attribution information, the express delivery three-section code information, and the express delivery receipt and delivery information of the current express delivery. The express delivery key information sequence includes the text information sequence and the coordinate information sequence of the express delivery key information in the current express delivery.

[0068] Optionally, the key express delivery information includes the express delivery affiliated unit and the express delivery order number; correspondingly, the method for extracting the key express delivery information of the current express delivery based on the text information sequence and / or the coordinate information sequence may include: determining the candidate key information sequence in the key information sequence based on each coordinate information sequence and the preset sequence threshold range; determining the name similarity between each text information sequence and each express delivery unit name in the preset express delivery unit name library, and determining the express delivery affiliated unit of the current express delivery based on each name similarity; determining the order number similarity between each text information sequence and the preset order number composition feature, determining the candidate order number information sequence in each candidate key information sequence based on the order number similarity, and determining the express delivery order number of the current express delivery based on the coordinate information sequence in the candidate order number information sequence and / or the express delivery affiliated unit of the current express delivery.

[0069] Specifically, on the basis of determining the key information sequence according to the above embodiment, the coordinate information sequence in each sequence is obtained. Among them, the coordinate information sequence can be understood as the information sequence used to reflect the position coordinates of the key express delivery information in the waybill image. Optionally, for any key information sequence, the sequence width of the key information sequence is determined based on the coordinate information sequence in the key information sequence; further, the preset sequence threshold range is obtained, and the obtained sequence width is compared with the sequence threshold range; if the comparison result meets the requirements, it means that the current key information sequence is the candidate key information sequence in the key information sequence. It should be explained that the candidate key information sequence is the candidate sequence for determining the express delivery attribution information. In other words, the express delivery attribution information of the current express delivery can be determined based on the determined candidate key information sequence. Further, based on the above embodiment, each candidate key information sequence in the key information sequence is determined.

[0070] Optionally, for the text information sequence of any sequence in the candidate key information sequence, the sequence length of the current text information sequence is determined. If the sequence length meets the conditions, the name similarity between the current text information sequence and each express delivery unit name in the preset express delivery unit name library is determined; further, based on the above embodiment, the name similarity of each candidate key information sequence is determined. At the same time, the preset name similarity threshold is obtained, and the express delivery affiliated unit of the current express delivery is determined based on each name similarity and the above preset name similarity threshold. Optionally, the maximum name similarity among each name similarity is obtained; if the maximum name similarity is greater than the preset similarity threshold, the express delivery unit corresponding to the maximum name similarity is determined as the express delivery affiliated unit of the current express delivery.

[0071] Optionally, for the text information sequence of any sequence in the candidate key information sequences, obtain the preset single number composition feature, match the current text information sequence with the single number composition feature, and determine the single number similarity between the current text information sequence and the preset single number composition feature; at the same time, obtain the preset single number similarity threshold, and determine the candidate single number information sequences in the candidate key information sequences based on the single number similarity and the single number similarity threshold. Further, based on the above implementation, determine the candidate single number information sequences in the candidate key information sequences. At the same time, on the basis of determining the express delivery belonging unit of the current express delivery, match the relevance between the single number information in each candidate single number information sequence and the express delivery belonging unit, and determine the single number information in the candidate single number information sequence that meets the matching conditions as the express delivery number of the current express delivery; conversely, if each relevance does not meet the matching conditions, determine the highest relevance among the relevances; and determine the single number information in the candidate single number information sequence corresponding to the highest relevance as the express delivery number of the current express delivery; of course, the single number information of any one in the candidate single number information sequence can also be determined as the express delivery number of the current express delivery, and this embodiment does not limit this.

[0072] On the basis of the above implementation, in the process of determining the express delivery belonging unit of the current express delivery based on the name similarity, if the maximum name similarity is less than the preset similarity threshold, then on the basis of determining the express delivery number of the current express delivery, determine the express delivery belonging unit of the express delivery number, and determine the express delivery belonging unit as the express delivery belonging unit of the current express delivery.

[0073] Optionally, the express delivery key information includes the express delivery three-segment code information. Correspondingly, the method for extracting the express delivery key information of the current express delivery based on the text information sequence and / or the coordinate information sequence may include: respectively obtain the preset code composition feature, and perform information extraction on the text information sequence based on the code composition feature to obtain the three-segment code information of the current express delivery; where the three-segment code information includes the province code, the city code, and the region code.

[0074] Specifically, obtain the preset code composition feature. Specifically, the code composition features respectively include the first-segment code composition feature, the second-segment code composition feature, and the third-segment code composition feature; further, respectively match each text information sequence in the key information sequence with the preset first-segment code composition feature. If the match is successful, then further match each text information sequence in the successfully matched key information sequence with the preset second-segment code composition feature; if the match is successful again, then further match each text information sequence in the successfully matched key information sequence with the preset third-segment code composition feature, and determine the province code information, the city code information, and the region code information of the current express delivery based on the key information corresponding to the text information sequence that is successfully matched again.

[0075] Optionally, the key express delivery information includes the recipient information and the sender information of the express delivery; correspondingly, the method for extracting the key express delivery information of the current express delivery based on the text information sequence and / or the coordinate information sequence may include: determining the first phone information sequence of the current express delivery based on the text information sequence, and determining the recipient information area of the current express delivery based on the first phone information sequence; determining the second phone information sequence or the second address information sequence of the current express delivery based on the text information sequence, and determining the sender information area of the current express delivery based on the second phone information sequence or the second address information sequence; determining the sequence area area of the waybill key sequence corresponding to each coordinate information sequence; if the intersection ratio of the current sequence area area and the area of the recipient information area is greater than the preset intersection threshold, determining that the key express delivery information corresponding to the current sequence area area is the recipient information of the current express delivery; if the intersection ratio of the current sequence area area and the area of the sender information area is greater than the preset intersection threshold, determining that the key express delivery information corresponding to the current sequence area area is the sender information of the current express delivery.

[0076] Specifically, traverse the text information sequence in the key information sequence to determine the first phone information sequence of the current express delivery. In this embodiment, it is defaulted that the recipient information of the express delivery is in the front position in the waybill image, so the first phone information sequence traversed is determined as the phone information of the recipient information. Optionally, the method for determining the first phone information sequence may include that if two phone information sequences are traversed and the interval between the two sequences is greater than 1, the phone information sequence at the first position in the key information sequence is determined as the phone information sequence of the recipient information. Further, determine the coordinate information sequence corresponding to the first phone information sequence, and based on this coordinate information sequence and the preset coordinate range as the recipient information area, and the key information sequence within the recipient information area is the recipient information sequence of the current express delivery.

[0077] On this basis, the phone information sequence at the second position in the sender and recipient information sequence is determined as the phone information sequence of the sender information. Further, determine the coordinate information sequence corresponding to the second phone information sequence, and based on this coordinate information sequence and the preset coordinate range as the sender information area, and the key information sequence within the sender information area is the sender information sequence of the current express delivery.

[0078] Based on the above embodiments, if only one telephone information sequence is matched, it is determined as the telephone information sequence of the recipient information, and the recipient information area is determined based on its telephone information sequence. Further, traverse the address information sequences in the key information sequences, and correspond the first address information sequence traversed to the address information sequence of the recipient information, determine the second address information sequence traversed as the address information sequence of the sender information, determine the coordinate information sequence corresponding to the second address sequence, and determine the sender information area based on this coordinate information sequence and the preset coordinate range. Of course, the recipient information area and the sender information area can also be determined based on the name information sequence, and this embodiment does not limit this.

[0079] Based on the above embodiments, the sequence area areas of each key information sequence are determined respectively based on the coordinate information sequences of each key information sequence; on this basis, taking any key information sequence as an example, taking the sequence area area of this sequence as an example, if the intersection ratio of the current sequence area area and the area of the recipient information area is greater than the preset intersection threshold, it is determined that the express key information corresponding to the current sequence area area is the recipient information of the current express; if the intersection ratio of the current sequence area area and the area of the sender information area is greater than the preset intersection threshold, it is determined that the express key information corresponding to the current sequence area area is the sender information of the current express.

[0080] Optionally, the recipient delivery information includes recipient telephone information, recipient name information, and recipient address information; based on the above embodiments, the method for obtaining the express key information of the current express can include: based on the pre-trained information extraction model, extracting information from the recipient information to obtain the recipient telephone information, recipient name information, and recipient address information of the current express.

[0081] Optionally, the sender delivery information includes sender telephone information, sender name information, and sender address information; correspondingly, the method for obtaining the express key information of the current express can include: based on the pre-trained information extraction model, extracting information from the recipient information to obtain the sender telephone information, sender name information, and sender address information of the current express. Of course, other methods can also be used to extract the above information, and this embodiment does not limit this.

[0082] It should be noted that the technical solution of this embodiment also includes that when it is detected that the express key information of the current express obtained based on the waybill image is completed in the delivery of the current express, the express key information of the current express will be deleted and its express key information will not be stored for a long time. The effect is that on the one hand, it can ensure the security of the express information of the current express, and on the other hand, it can reduce the storage memory of the current memory and improve the processing speed of the processor.

[0083] Based on the above embodiments, this embodiment further provides a preferred embodiment, which introduces another method for extracting key express information; specifically, the technical solution of this embodiment includes the following steps: obtaining the label image of the current express, performing OCR recognition on the label image, and obtaining an information sequence containing text content and coordinate information; and performing intelligent parsing on the information sequence to obtain the key express information of the current express. Specifically, its processing flow is to first preprocess the information sequence, remove the sequences determined to be meaningless, and parse the remaining key information sequences to obtain the key express information.

[0084] Specifically, the parsing of the key express information in this embodiment mainly includes three parsing modules: First, the express company and express order number parsing module; Second, the three-segment code parsing module; Third, the sender and recipient information parsing module.

[0085] Optionally, regarding the parsing of the express order number and express company parsing module: The express order number is the unique number of each express label, consisting of numbers and letters. The method for parsing the express order number and express company proposed in this embodiment is as follows: (1) Calculate the width of the sequence according to the coordinate points in the information sequence, and filter out the sequences with a width greater than 100; (2) Match the text in the first 5 sequences. If the text length is less than 6 and reaches the similarity threshold with the name of a certain express company, the express company is successfully matched; (3) According to the composition rules of the express order number, including letters, numbers, and length constraints, extract the potential list of express order numbers; (4) Comprehensively judge the results of (2) and (3) to determine the final express order number and express company. If the express company is missing, the affiliated express company can be inferred according to the mapping relationship between the express order number and the express company.

[0086] Optionally, regarding the parsing of the three-segment code parsing module: The three-segment code is three segments of numbers on the express label. The first segment represents the province code, the second segment represents the city code, and the third segment represents the area code of the courier. Summarize the combined characteristics of the three-segment code, and sequentially judge whether the category of the text in the text box belongs to the three-segment code, and extract the sequences identified as the three-segment code category. If two or more sequences are judged to be the three-segment code, then extract again according to the summarized three-segment code rules. Format the three-segment code on the label into a format such as the first segment % the second segment % the third segment. For example: 800H S04-00 402 is formatted as 800H%S04-00%402. Since the three-segment code formats of the labels of different express companies are different, and the delimiters used on the labels are also different, for the convenience of applying the extracted information, it is uniformly formatted and separated by %.

[0087] Specifically, regarding the parsing of the sender and recipient information parsing module: The recipient information and the sender information are the most important data in the waybill. The parsing of the sender and recipient information mainly includes two steps: 1. Determination of the area box of the sender and recipient information; 2. Parsing of the name, phone number, and address.

[0088] Optionally, regarding the determination of the area box of the sender and recipient information:

[0089] Step 1: Match the position of the phone number sequence in the key information sequence. By default, the recipient information is in the front and the sender information is in the back for all express waybills. If two phone number sequence positions are matched and the interval is greater than 1, the first position is the recipient phone number, and take the previous sequence rece_start_points. The second position is the sender phone number send_tel_points, and take the previous position sequence coordinates as send_start_points.

[0090] Step 2: If the sender phone number coordinates are not empty, then analyze the subsequent sequences in turn. When meaningless characters such as the "signature column" are matched or stop at the third sequence after it, record the previous sequence send_end_points of the current sequence;

[0091] The method of determining the possible coordinates of the end position of the recipient area box may include:

[0092] rece_end_points = send_tel_points - (send_tel_points[3][1] - send_tel_points[0][1])

[0093] Furthermore, determine the four-point coordinates according to the upper left vertex coordinates rece_start_points and the lower right vertex coordinates rece_end_points of the recipient area box. Similarly, determine the four-point coordinates of the sender area box.

[0094] Step 3: If the sender phone number coordinates are empty, then match the end rece_end_points of the recipient address after the recipient phone number, and the subsequent sequence coordinates are the starting coordinates send_start_points of the sender area; or match "sender" as the starting position of the sender area, and the previous sequence is the end coordinates of the recipient area; determine the four-point coordinates according to the upper left vertex coordinates rece_start_points and the lower right vertex coordinates rece_end_points of the recipient area box. Similarly, determine the four-point coordinates of the sender area box.

[0095] Step 4: If neither the recipient phone number nor the sender phone number is matched, then match the address information. The first address information matched is recorded as the recipient address information, and the second address is the sender address information, otherwise it is empty.

[0096] Optionally, the parsing of name information, telephone information, and address information:

[0097] Step 1: Use the sequence area intersection ratio to determine the information sequences included in the recipient area box and the sender area box; when the sequence area / recipient area box area > 0.6, it is considered that the sequence belongs to the corresponding area box, and all sequences that meet this condition form the recipient information sequence; when the sequence area / sender area box area > 0.6, it is considered that the sequence belongs to the corresponding area box, and all sequences that meet this condition form the sender information sequence.

[0098] Step 2: Calculate the slope of the recipient area box and the sender area box. If the slope exceeds 0.2, perform a rotation correction on the sequences it contains; the slope is represented by the slope of the first sequence.

[0099] k = (points[1][1] - points[0][1]) / (points[1][0] - points[0][0])

[0100] Step 3: Calculate the row and column information of the information sequences in the recipient area and the sender area respectively;

[0101] The sequence position is represented by the four-point coordinates, and the row and column of each sequence are determined according to the relative position relationship of the center points. Assume that the first sequence is the first row, and if the distance difference between the center point position of the subsequent sequence and the center point position of the current row in the vertical coordinate is less than or equal to 0.4 * row width, it is considered to belong to the current row, otherwise a new row is added.

[0102] The center point coordinates of the current sequence are as follows:

[0103] x_center, y_center = (points[0][0] + points[1][0]) / 2, (points[0][1] + points[3][1]) / 2

[0104] The center position of the current row is determined by the coordinates of the first sequence it contains, as follows: g_x_center, g_y_center = (g_points[0][0][0] + g_points[0][1][0]) / 2, (g_points[0][0][1] + g_points[0][3][1]) / 2

[0105] The row width calculation formula is as follows:

[0106] height = points[3][1] - points[0][1]

[0107] Then, sort the sequences within each row according to the center point of the abscissa. Thus, the calculation and marking of the row and column information of the information sequence in the region box are completed. Extract the text content row by row and store it in an array, where each row in the array represents the information of each row in the waybill.

[0108] Step 4: Classify and splice the sequences in the recipient region box. The categories include name, phone number, and address. The classification method is not limited to rules or machine learning models. Especially for the address category, when multiple text contents are recognized as address categories, adjacent contents need to be spliced and combined.

[0109] The technical solution of this embodiment specifically includes obtaining the waybill image of the current express delivery, and recognizing the waybill image to obtain the key information sequence of each express delivery key information in the waybill image; among them, the key information sequence includes the text information sequence and the coordinate information sequence of each express delivery key information; based on the text information sequence and / or the coordinate information sequence, extract the express delivery key information of the current express delivery to obtain the express delivery key information of the current express delivery. The above technical solution solves the technical problem of inaccurate recognition of different types of waybill images in the prior art, resulting in low delivery efficiency, by performing OCR recognition on the waybill image of the current express delivery, obtaining the key information sequence containing the express delivery key information of the delivered express delivery, and performing intelligent analysis on the key information sequence to obtain the express delivery key information of the current express delivery, realizing the improvement of the recognition accuracy and recognition rate of the waybill, and thus improving the efficiency of express delivery.

[0110] Embodiment Two

[0111] Figure 2 It is a schematic structural diagram of an express delivery key information extraction device provided for Embodiment Three of the present invention. As Figure 2 shown, the device includes: a key information sequence obtaining module 210 and an express delivery key information obtaining module 220; among them,

[0112] The key information sequence obtaining module 210 is used to obtain the waybill image of the current express delivery, and recognize the waybill image to obtain the key information sequence of each express delivery key information in the waybill image; among them, the key information sequence includes the text information sequence and the coordinate information sequence of each express delivery key information;

[0113] The express delivery key information obtaining module 220 is used to extract the express delivery key information of the current express delivery based on the text information sequence and / or the coordinate information sequence to obtain the express delivery key information of the current express delivery.

[0114] On the basis of the above embodiments, optionally, the key information sequence obtaining module 210 includes:

[0115] An initial waybill information sequence obtaining unit, configured to perform image sequence recognition on a waybill image based on a pre-trained image sequence recognition model, so as to obtain an initial waybill information sequence of each waybill sequence in the waybill image;

[0116] A key information sequence obtaining unit, configured to obtain each text information sequence in the initial waybill information sequence, and determine a key information sequence in the initial waybill information sequence based on the text information sequence and a preset key information sequence library.

[0117] Based on the above embodiments, optionally, the express delivery key information includes the express delivery belonging unit and the express delivery number;

[0118] Correspondingly, the express delivery key information obtaining module 220 includes:

[0119] A candidate key information sequence determining unit, configured to determine a candidate key information sequence in the key information sequence based on each coordinate information sequence and a preset sequence threshold range;

[0120] An express delivery belonging unit determining unit, configured to determine the name similarity between each text information sequence and each express delivery unit name in a preset express delivery unit name library, and determine the express delivery belonging unit of the current express delivery based on each name similarity;

[0121] An express delivery number determining unit, configured to determine the number similarity between each text information sequence and a preset single-number composition feature, determine a candidate single-number information sequence in each candidate key information sequence based on the number similarity, and determine the express delivery number of the current express delivery based on the coordinate information sequence in the candidate single-number information sequence and / or the express delivery belonging unit of the current express delivery.

[0122] Based on the above embodiments, optionally, the express delivery belonging unit determining unit includes:

[0123] A maximum name similarity obtaining subunit, configured to obtain the maximum name similarity among each name similarity;

[0124] A first express delivery belonging unit determining unit, configured to, if the maximum name similarity is greater than a preset similarity threshold, determine the express delivery unit corresponding to the maximum name similarity as the express delivery belonging unit of the current express delivery;

[0125] A second express delivery belonging unit determining unit, configured to, if the maximum name similarity is less than the preset similarity threshold, obtain the express delivery number of the current express delivery, and determine the express delivery belonging unit of the current express delivery based on the express delivery number.

[0126] Based on the above embodiments, optionally, the express delivery key information includes express delivery three-section code information, and the express delivery key information obtaining module 220 includes:

[0127] The three - segment code information extraction unit is used to respectively obtain the preset code composition features, and based on the code composition features, extract information from the text information sequence to obtain the three - segment code information of the current express delivery; among them, the three - segment code information includes the province code, the city code, and the regional code.

[0128] Based on the above - mentioned various embodiments, optionally, the key express delivery information includes the recipient information and the sender information of the express delivery;

[0129] Correspondingly, the key express delivery information obtaining module 220 includes:

[0130] The recipient information area determination unit is used to determine the first phone information sequence of the current express delivery based on the text information sequence, and determine the recipient information area of the current express delivery based on the first phone information sequence;

[0131] The sender information area determination unit is used to determine the second phone information sequence, or the second address information sequence of the current express delivery based on the text information sequence, and determine the sender information area of the current express delivery based on the second phone information sequence, or the second address information sequence;

[0132] The sequence area area determination unit is used to determine the sequence area area of the key waybill sequence corresponding to each coordinate information sequence;

[0133] The first recipient information determination unit is used to determine that the key express delivery information corresponding to the current sequence area area is the recipient information of the current express delivery if the intersection ratio of the current sequence area area and the area of the recipient information area is greater than the preset intersection threshold;

[0134] The second sender information determination unit is used to determine that the key express delivery information corresponding to the current sequence area area is the sender information of the current express delivery if the intersection ratio of the current sequence area area and the area of the sender information area is greater than the preset intersection threshold.

[0135] Based on the above - mentioned various embodiments, optionally, the recipient information includes the recipient phone information, the recipient name information, and the recipient address information;

[0136] Correspondingly, the key express delivery information obtaining module 220 includes:

[0137] The first recipient information determination unit is used to extract information from the recipient information based on the pre - trained information extraction model to obtain the recipient phone information, the recipient name information, and the recipient address information of the current express delivery;

[0138] The sender information includes the sender phone information, the sender name information, and the sender address information;

[0139] Correspondingly, the key express delivery information obtaining module 220 includes:

[0140] The second sender information determination unit is configured to perform information extraction on the recipient information based on a pre-trained information extraction model, so as to obtain the sender's phone number information, sender's name information, and sender's address information of the current express item.

[0141] The express key information extraction device provided by the embodiments of the present invention can execute the express key information extraction method provided by any embodiment of the present invention, and has corresponding functional modules and beneficial effects for executing the method.

[0142] Embodiment III

[0143] Figure 3 FIG. shows a schematic structural diagram of an electronic device 10 that can be used to implement the embodiments of the present invention. The electronic device is intended to represent various forms of digital computers, such as, for example, a laptop computer, a desktop computer, a workbench, a personal digital assistant, a server, a blade server, a mainframe computer, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, a personal digital processor, a cellular phone, a smart phone, a wearable device (such as a helmet, glasses, a watch, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0144] As Figure 3 shown, the electronic device 10 includes at least one processor 11, and a memory communicatively connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc. Among them, the memory stores a computer program executable by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. In the RAM 13, various programs and data required for the operation of the electronic device 10 can also be stored. The processor 11, the ROM 12, and the RAM 13 are connected to each other through a bus 14. The input / output (I / O) interface 15 is also connected to the bus 14.

[0145] A plurality of components in the electronic device 10 are connected to the I / O interface 15, including: an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices through a computer network such as the Internet and / or various telecommunication networks.

[0146] The processor 11 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the express key information extraction method.

[0147] In some embodiments, the express key information extraction method can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed onto the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the express key information extraction method described above can be executed. Alternatively, in other embodiments, the processor 11 can be configured to execute the express key information extraction method by any other suitable means (e.g., by means of firmware).

[0148] Various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), system-on-chip systems (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs, which can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special or general programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0149] The computer program for implementing the method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowchart and / or block diagram are implemented. The computer program can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0150] In the context of the present invention, a computer-readable storage medium can be a tangible medium that can contain or store a computer program for use by or in connection with an instruction execution system, apparatus, or device. The computer-readable storage medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. Alternatively, the computer-readable storage medium can be a machine-readable signal medium. More specific examples of the machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0151] In order to provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the electronic device. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, speech input, or tactile input).

[0152] The systems and techniques described herein can be implemented in a computing system that includes backend components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system that includes any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: local area network (LAN), wide area network (WAN), blockchain network, and the Internet.

[0153] A computing system may include a client and a server. The client and the server are generally far from each other and usually interact via a communication network. The relationship between the client and the server is created by computer programs running on respective computers and having a client-server relationship with each other. The server may be a cloud server, also known as a cloud computing server or a cloud host, which is a host product in the cloud computing service system, solving the defects of difficult management and weak business scalability existing in traditional physical hosts and VPS services.

[0154] It should be understood that various forms of the processes shown above can be used, with steps reordered, added, or deleted. For example, the steps recited in the present invention can be executed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved, and no limitation is imposed herein.

[0155] The above specific embodiments do not constitute a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principle of the present invention shall be included within the protection scope of the present invention.

Claims

1. A method for extracting key information of express parcels, characterized in that, it includes: Obtain the label image of the current express parcel, and identify the label image to obtain the key information sequence of each key information of the express parcel in the label image; wherein, the key information sequence includes the text information sequence and the coordinate information sequence of each key information of the express parcel; Extract the key information of the current express parcel based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express parcel; wherein, the key information of the express parcel includes the recipient information and the sender information of the express parcel; Correspondingly, the extracting the key information of the current express parcel based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express parcel includes: Determine the first phone information sequence of the current express parcel based on the text information sequence, and determine the recipient information area of the current express parcel based on the first phone information sequence; Determine the second phone information sequence or the second address information sequence of the current express parcel based on the text information sequence, and determine the sender information area of the current express parcel based on the second phone information sequence or the second address information sequence; Determine the sequence area area of each coordinate information sequence corresponding to the label key sequence; If the intersection ratio of the current sequence area area and the area of the recipient information area is greater than a preset intersection threshold, determine that the key information corresponding to the current sequence area area is the recipient information of the current express parcel; If the intersection ratio of the current sequence area area and the area of the sender information area is greater than a preset intersection threshold, determine that the key information corresponding to the current sequence area area is the sender information of the current express parcel.

2. The method according to claim 1, characterized in that, the identifying the label image to obtain the key information sequence of each key information of the express parcel in the label image includes: Based on a pre-trained image sequence recognition model, perform image sequence recognition on the label image to obtain the initial label information sequence of each label sequence in the label image; Obtain each text information sequence in the initial label information sequence, and determine the key information sequence in the initial label information sequence based on the text information sequence and a preset key information sequence library.

3. The method according to claim 1, characterized in that, the key information of the express parcel includes the affiliated unit of the express parcel and the express parcel number; Correspondingly, the extracting the key information of the current express parcel based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express parcel includes: Determine the candidate key information sequence in the label key information sequence based on each coordinate information sequence and a preset sequence threshold range; Determine the name similarity between each text information sequence and each express parcel unit name in a preset express parcel unit name library, and determine the affiliated unit of the current express parcel based on each name similarity; Determine the waybill number similarity between each of the text information sequences and the preset waybill number composition features, determine the candidate waybill number information sequences in each of the candidate key information sequences based on the waybill number similarity, and determine the waybill number of the current express delivery based on the coordinate information sequence in the candidate waybill number information sequence and / or the express delivery belonging unit of the current express delivery.

4. The method according to claim 3, wherein, the determining the express delivery belonging unit of the current express delivery based on each of the name similarities includes: obtaining the maximum name similarity among each of the name similarities; if the maximum name similarity is greater than the preset similarity threshold, determining the express delivery unit corresponding to the maximum name similarity as the express delivery belonging unit of the current express delivery; if the maximum name similarity is less than the preset similarity threshold, obtaining the waybill number of the current express delivery, and determining the express delivery belonging unit of the current express delivery based on the waybill number.

5. The method according to claim 1, wherein, the key information of the express delivery includes the three-segment code information of the express delivery, correspondingly, the extracting the key information of the current express delivery based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express delivery includes: respectively obtaining the preset code composition features, and performing information extraction on the text information sequence based on the code composition features to obtain the three-segment code information of the current express delivery; wherein, the three-segment code information includes the province code, the city code, and the region code.

6. The method according to claim 1, wherein, the obtaining the key information of the current express delivery includes: performing information extraction on the recipient information based on a pre-trained information extraction model to obtain the recipient delivery information of the current express delivery; the obtaining the key information of the current express delivery includes: performing information extraction on the sender information based on a pre-trained information extraction model to obtain the sender delivery information of the current express delivery.

7. An apparatus for extracting key information of an express delivery, wherein, it includes: a key information sequence obtaining module, configured to obtain the label image of the current express delivery, and identify the label image to obtain the key information sequences of each key information of the express delivery in the label image; wherein, the key information sequences include the text information sequences and the coordinate information sequences of each key information of the express delivery; a key information obtaining module of the express delivery, configured to perform key information extraction on the current express delivery based on the text information sequence and / or the coordinate information sequence to obtain the key information of the current express delivery; wherein, the key information of the express delivery includes the recipient information of the express delivery and the sender information of the express delivery; the key information obtaining module of the express delivery includes: a recipient information area determining unit, configured to determine the first phone information sequence of the current express delivery based on the text information sequence, and determine the recipient information area of the current express delivery based on the first phone information sequence; A sender information area determination unit, configured to determine a second telephone information sequence or a second address information sequence of the current express item based on the text information sequence, and determine the sender information area of the current express item based on the second telephone information sequence or the second address information sequence; A sequence area area determination unit, configured to determine the sequence area area of the waybill key sequence corresponding to each coordinate information sequence; A first recipient information determination unit, configured to determine that the express item key information corresponding to the current sequence area area is the recipient information of the current express item if the intersection ratio of the current sequence area area and the area of the recipient information area is greater than a preset intersection threshold; A second sender information determination unit, configured to determine that the express item key information corresponding to the current sequence area area is the sender information of the current express item if the intersection ratio of the current sequence area area and the area of the sender information area is greater than a preset intersection threshold.

8. An electronic device, characterized in that the electronic device includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores a computer program executable by the at least one processor, and the computer program is executed by the at least one processor so that the at least one processor can execute the express item key information extraction method according to any one of claims 1-6.

9. A computer-readable storage medium, characterized in that the computer-readable storage medium stores computer instructions for causing a processor to implement the express item key information extraction method according to any one of claims 1-6 when executed.

Images