Parcel damage detection method and device, electronic equipment and storage medium

Abstract

The invention discloses a parcel damage detection method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a package image of a posted object and object information of the posted object; performing matching in a preset knowledge base based on the article information of the posted article, and determining a vulnerable label of the posted article; generating text information based on a damage judgment condition corresponding to the vulnerable tag; and inputting the text information and the package image of the posted object into a pre-trained damage detection model to obtain a damage detection result of the posted object, the damage detection result of the posted object including damage degree information of the posted object. According to the method, damage detection is carried out through the data of the two modes of the image and the text, and the damage detection accuracy is improved while the package damage detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and in particular to a package damage detection method, device, electronic equipment and storage medium. Background Art

[0002] During the transportation of logistics packages, the goods may be damaged due to various factors. When the operator on site finds suspected damage to the goods, the operator will report the abnormality in the system. For abnormal waybills, customer service can appease customers in advance through outbound calls and other means before the package is delivered to lower customer expectations, which can reduce or even avoid the negative experience brought to customers by abnormalities and increase customer stickiness. However, the current risk reporting of operators is relatively subjective. For example, even if the package is slightly damaged or has creases, it will also be reported. If customer service is involved in all of this, it will not only waste labor costs, but also cause a poor customer experience.

[0003] Currently, there are two methods for detecting damaged packages. One is that customer service staff judge the damage condition based on the product information of the consigned items and the photos reported by the operator, but this method consumes a lot of manpower. The other is to train a prediction model and use the prediction model for damage detection. However, the traditional prediction model is based on a large amount of historical data to build samples to train the model. The trained model is used to make inferences and predictions on waybills during the operation process. Although this method can make decisions based on multiple factors, relying only on a single modality of data will lose a lot of necessary information, affecting the accuracy of damage detection. Summary of the invention

[0004] The present invention provides a package damage detection method, device, electronic equipment and storage medium to improve the efficiency and accuracy of damage detection.

[0005] According to one aspect of the present invention, a package damage detection method is provided, which is characterized by comprising:

[0006] Acquire a package image of a consignment and item information of the consignment;

[0007] Based on the item information of the consignment, a match is performed in a preset knowledge base to determine a fragile label of the consignment;

[0008] Generate text information based on the damage determination condition corresponding to the fragile label;

[0009] The text information and the package image of the consignment are input into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0010] According to another aspect of the present invention, there is provided a package damage detection device, comprising:

[0011] An image and item information acquisition module, used to acquire a package image of a consignment and item information of the consignment;

[0012] A fragile label determination module, used for matching the item information of the consignment in a preset knowledge base to determine the fragile label of the consignment;

[0013] A fragile label text information generation module, used to generate text information based on damage determination conditions corresponding to the fragile label;

[0014] The damage detection module is used to input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0015] According to another aspect of the present invention, there is provided an electronic device, the electronic device comprising:

[0016] at least one processor; and

[0017] a memory communicatively connected to the at least one processor; wherein,

[0018] 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 package damage detection method described in any embodiment of the present invention.

[0019] According to another aspect of the present invention, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the package damage detection method described in any embodiment of the present invention when executed.

[0020] The technical solution of the embodiment of the present invention obtains the package image and item information of the consignment; matches the item information of the consignment in a preset knowledge base to determine the fragile label of the consignment; generates text information based on the damage judgment condition corresponding to the fragile label; inputs the text information and the package image of the consignment into a pre-trained damage detection model to obtain the damage detection result of the consignment. The damage detection model is constructed by using data of both image and text modalities, which improves the accuracy of damage detection while improving the efficiency of damage detection and reduces the consumption of human resources.

[0021] It should be understood that the contents described in this section are not intended to identify the key or important features of the embodiments of the present invention, nor are they intended to limit the scope of the present invention. Other features of the present invention will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0023] Figure 1 is a flow chart of a package damage detection method provided by an embodiment of the present invention;

[0024] Figure 2 is a flow chart of a package damage detection method provided by an embodiment of the present invention;

[0025] Figure 3 is a flow chart of a package damage detection method provided by an embodiment of the present invention;

[0026] Figure 4 is a flow chart of a package damage detection method provided by an embodiment of the present invention;

[0027] Figure 5 is a flow chart of a package damage detection method provided by an embodiment of the present invention;

[0028] Figure 6 is a flow chart of a damage handling event provided by an embodiment of the present invention;

[0029] Figure 7 is a structural schematic diagram of a package damage detection device provided by an embodiment of the present invention;

[0030] Figure 8 It is a structural schematic diagram of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0031] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0032] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0033] Figure 1 This is a flowchart of a package damage detection method provided by an embodiment of the present invention. This embodiment is applicable to the case where a package is damaged during transportation. The method can be performed by a package damage detection device. The package damage detection device can be implemented in the form of hardware and / or software. The package damage detection device can be configured in electronic devices such as computers and servers. Figure 1 As shown, the method includes:

[0034] S110: Acquire a package image of the consignment and item information of the consignment.

[0035] The package image refers to the image of the consignment package taken by the staff of the consignment processing outlet. The item information of the consignment refers to the information related to the consignment, and specifically, the item information includes but is not limited to the item name, item type, item attribute, item quality, etc., which is not limited here.

[0036] On the basis of the above embodiment, optionally, the obtaining of the package image of the consignment and the item information of the consignment includes: obtaining the package image of the consignment and the item information of the consignment uploaded by the consignment processing outlet; determining the damage reporting portrait information of the consignment processing outlet based on the identification of the consignment processing outlet; and sorting the package image and item information of the consignment according to processing priority based on the damage reporting portrait information of the consignment processing outlet.

[0037] Among them, the identification of the consignment handling outlet is used to uniquely identify the consignment handling outlet. Specifically, the identification of the consignment handling outlet can be the outlet number, name, location information, etc. In this embodiment, the package image of the consignment and the item information of the consignment uploaded by the consignment handling outlet are obtained; the damage reporting portrait information of the consignment handling outlet is determined based on the identification of the consignment handling outlet; wherein, the damage reporting portrait information is constructed based on the damage rate in the historical report form of the consignment handling outlet. It can be understood that due to the natural differences in the items of different consignment handling outlets and the different reporting habits of the outlet personnel, the damage rate of the consignment handling outlet report form also has certain differences. For the report form with a higher damage rate, priority should be given to intervention. Specifically, the damage reporting portrait information of the consignment handling outlet can be divided into three levels of labels: "high", "medium" and "low". The higher the label level, the higher the processing priority. Further, the package image and item information of the consignment are sorted according to the processing priority based on the damage reporting portrait information of the consignment handling outlet. Specifically, the package images and item information of the consignment may be prioritized according to the label levels in the damage reporting portrait information of the consignment handling outlets, so as to give priority to the package images and item information uploaded by the consignment handling outlets with higher damage rates.

[0038] S120: Matching the item information of the consignment in a preset knowledge base to determine a fragile label of the consignment.

[0039] Among them, the preset knowledge base includes the item information of the consignment and the associated information of the fragile label of the consignment. The relevant information of the fragile label of the consignment may be the fragile label of the consignment itself, or the associated information related to the fragile label. The fragile label can be determined based on the associated information of the fragile label. In this embodiment, the fragile label of the consignment can be directly obtained by matching in the preset knowledge base based on the item information of the consignment. Optionally, the associated information of the fragile label of the consignment can also be obtained by matching in the preset knowledge base based on the item information of the consignment; further, the fragile label of the consignment can be determined based on the associated information of the fragile label of the consignment. Among them, the fragile label is a label defined in combination with the properties of the consignment, the frequency of damage to the consignment, and the cause. Specifically, fragile labels include fragile, perishable, and leaky.

[0040] S130: Generate text information based on the damage determination condition corresponding to the fragile label.

[0041] It should be noted that different fragile labels correspond to different damage determination conditions, which include damage state description information and damage degree information corresponding to the damage state description information; specifically, the damage state description information is used to describe the damage state of the consignment, such as: scratches on the outer packaging, cracks on the outer packaging, deformation of the outer packaging, etc. The damage degree information includes slight damage, moderate damage and severe damage, and the damage degree information corresponds to the damage state description information.

[0042] Exemplarily, the damage determination conditions corresponding to different fragile labels are shown in Table 1:

[0043] Table 1

[0044]

[0045] In this embodiment, text information is generated based on the damage judgment condition corresponding to the fragile label. The text information includes context information and the damage judgment condition corresponding to the fragile label. The context information is determined based on the item information or category attributes of the consignment. The context information can make the generated text information more fluent. Exemplarily, the text information may be: "Given a package image, the goods in the package are <wine>. Please infer the degree of damage of the goods based on the damage of the outer packaging. Please answer (a) severe, (b) moderate, (c) slight. Please answer the questions based on the following prompts: 1. It is slight if there are scratches on the outer packaging; 2. It is moderate if the outer packaging is cracked or deformed; 3. It is severe if it is wet and dirty."

[0046] S140: Input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0047] Among them, the damage detection model is used to detect the degree of damage of the consignment. Specifically, the damage detection model can be a large multimodal model, such as chatGPT-4V; it can also be a machine learning model, such as a neural network model. In this embodiment, the text information and the package image of the consignment are input into the pre-trained damage detection model to obtain the damage detection result of the consignment. Among them, the damage detection result indicates the degree of damage of the consignment. Specifically, the damage detection result of the consignment includes the degree of damage information of the consignment, and the degree of damage information includes severe, moderate and slight. Exemplarily, the damage detection result can be: "According to the package image, the outer packaging has obvious cracks, but there is no obvious wet and dirty baseline. According to the given prompts, the degree of damage of the goods is (b) moderate."

[0048] The technical solution of this embodiment obtains the package image and item information of the consignment; matches the item information of the consignment in a preset knowledge base to determine the fragile label of the consignment; generates text information based on the damage judgment condition corresponding to the fragile label; inputs the text information and the package image of the consignment into a pre-trained damage detection model to obtain the damage detection result of the consignment. The damage detection model is constructed by using data in both image and text modes, which improves the efficiency of damage detection while improving the accuracy of damage detection and reduces the consumption of human resources.

[0049] Figure 2 It is a flow chart of a package damage detection method provided by an embodiment of the present invention. In this embodiment, based on the above embodiment, optionally, the preset knowledge base includes item information of multiple items and fragile labels of items; based on the item information of the consignment, matching is performed in the preset knowledge base to determine the fragile label of the consignment, including: converting the item information of the consignment into a feature vector, performing similarity matching on the feature vector in the preset knowledge base, determining the item corresponding to the maximum similarity, and determining the fragile label of the item corresponding to the maximum similarity as the fragile label of the consignment, wherein the item information in the preset knowledge base is stored in the form of vectors.

[0050] like Figure 2 As shown, the method includes:

[0051] S210: Acquire a package image of the consignment and item information of the consignment.

[0052] S220, converting the item information of the consignment into a feature vector, performing similarity matching on the feature vector in the preset knowledge base, determining an item corresponding to the maximum similarity, and determining the fragile label of the item corresponding to the maximum similarity as the fragile label of the consignment, wherein the item information in the preset knowledge base is stored in the form of a vector.

[0053] The preset knowledge base includes item information of multiple items and fragile labels of items, and the item information in the preset knowledge base is stored in the form of vectors. In this embodiment, feature extraction can be performed on the item information of the consignment based on a preset feature extraction method, and the item information can be converted into a feature vector; wherein the preset feature extraction method can be word embedding, TF-IDF, etc. Further, based on a preset similarity matching method, the feature vector is similarity matched in the preset knowledge base, the item corresponding to the maximum similarity is determined, and the fragile label of the item corresponding to the maximum similarity is determined as the fragile label of the consignment. wherein the preset similarity matching method can be cosine similarity, Euclidean distance, etc.

[0054] S230: Generate text information based on the damage determination condition corresponding to the fragile label.

[0055] S240: Input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0056] The technical solution of this embodiment is to obtain the package image of the consignment and the item information of the consignment; convert the item information of the consignment into a feature vector, perform similarity matching on the feature vector in a preset knowledge base, determine the item corresponding to the maximum similarity, and determine the fragile label of the item corresponding to the maximum similarity as the fragile label of the consignment; generate text information based on the damage judgment condition corresponding to the fragile label; input the text information and the package image of the consignment into a pre-trained damage detection model to obtain the damage detection result of the consignment. The damage detection model is constructed by using data in both image and text modes, which improves the accuracy of damage detection while improving the efficiency of damage detection and reduces the consumption of human resources.

[0057] Figure 3 It is a flow chart of a package damage detection method provided by an embodiment of the present invention. Based on the above embodiment, optionally, the preset knowledge base includes item information and category attributes of multiple items, and different category attributes correspond to different fragile labels; matching is performed in the preset knowledge base based on the item information of the consignment to determine the fragile label of the consignment, including: matching is performed in the preset knowledge base based on the item information of the consignment to determine the category attribute of the consignment, and determining the fragile label of the consignment based on the category attribute of the consignment.

[0058] like Figure 3 As shown, the method includes:

[0059] S310: Acquire a package image of the consignment and item information of the consignment.

[0060] S320: Matching the item information of the consignment in a preset knowledge base to determine the category attribute of the consignment, and determining a fragile label of the consignment based on the category attribute of the consignment.

[0061] Among them, the category attribute refers to the type attribute of the consignment. Specifically, the category attribute includes but is not limited to wine, vegetables, fruits, glassware, etc., which are not limited here. The preset knowledge base includes the item information and category attributes of a variety of items. Different category attributes correspond to different fragile labels, and the association relationship between the category attributes and the fragile labels is pre-established. In this embodiment, the category attribute of the consignment is determined based on the item information of the consignment in the preset knowledge base, and the fragile label of the consignment is determined based on the category attribute of the consignment and the pre-established association relationship between the category attribute and the fragile label.

[0062] S330: Generate text information based on the damage determination condition corresponding to the fragile label.

[0063] S340: Input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0064] The technical solution of this embodiment is to obtain the package image and item information of the consignment; match the item information of the consignment in a preset knowledge base to determine the category attributes of the consignment, and determine the fragile label of the consignment based on the category attributes of the consignment; generate text information based on the damage judgment condition corresponding to the fragile label; input the text information and the package image of the consignment into a pre-trained damage detection model to obtain the damage detection result of the consignment. The damage detection model is performed by using data in both image and text modes, which improves the efficiency of damage detection while improving the accuracy of damage detection and reduces the consumption of human resources.

[0065] Figure 4 It is a flow chart of a package damage detection method provided by an embodiment of the present invention. Based on the above embodiment, this embodiment optionally further includes: determining the quality of damage reporting to the consignment processing point based on the damage detection result of the consignment and the consignment processing point corresponding to the consignment; and adjusting the load quantity of the consignment processing point based on the damage reporting quality of the consignment processing point.

[0066] like Figure 4 As shown, the method includes:

[0067] S410: Acquire a package image of the consignment and item information of the consignment.

[0068] S420: Matching the item information of the consignment in a preset knowledge base to determine a fragile label of the consignment.

[0069] S430: Generate text information based on the damage determination condition corresponding to the fragile label.

[0070] S440: Input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0071] S450. Determine the quality of damage reporting to the consignment processing point based on the damage detection result of the consignment and the consignment processing point corresponding to the consignment; and adjust the load quantity of the consignment processing point based on the damage reporting quality of the consignment processing point.

[0072] Among them, the damage reporting quality is used to characterize the quality of the damage detection results of the consigned items reported by the consignment processing outlets. Specifically, at least one damage reporting quality evaluation index can be set, and the damage reporting quality of the consignment processing outlets is evaluated based on the damage reporting quality evaluation index to obtain the damage reporting quality of the consignment processing outlets. Among them, the damage reporting quality evaluation index can be the accuracy of the reported damage detection results, the timeliness rate of the damage detection results reported, etc. It is understandable that for slightly damaged consignments, no processing is required. Therefore, if the reported damage detection results are inaccurate, it may lead to the processing of consignments that originally do not need to be processed, wasting resources.

[0073] Among them, the load quantity refers to the number of parcels handled by the parcel handling outlet. It can be understood that each parcel handling outlet is set with a certain load quantity. The larger the load quantity, the more parcels are allocated to the parcel handling outlet.

[0074] In this embodiment, the quality of the damage detection results of the consigned items reported by the consignment processing outlets corresponding to the consigned items can be evaluated based on the damage reporting quality evaluation index to obtain the damage reporting quality of the consignment processing outlets; the damage reporting quality of each consignment processing outlet is ranked to obtain the ranking result of the damage reporting quality of each consignment processing outlet, and the load quantity of the consignment processing outlets is allocated based on the preset load quantity allocation ratio. The higher the damage reporting quality of the consignment processing outlet is ranked, the more load quantity the consignment processing outlets will be allocated. Among them, the preset load quantity allocation ratio is set by a person skilled in the art and is not limited here.

[0075] The technical solution of this embodiment determines the quality of damage reporting to the consignment processing outlets based on the damage detection results of the consignments and the consignment processing outlets corresponding to the consignments; and adjusts the load quantity of the consignment processing outlets based on the damage reporting quality of the consignment processing outlets. Timely adjustments are made to each consignment processing outlet to optimize the load distribution plan.

[0076] Figure 5It is a flowchart of a package damage detection method provided by an embodiment of the present invention. Based on the above embodiment, this embodiment optionally further includes: when the damage degree information of the consignment meets the damage processing conditions, the item information of the consignment is sent to the processing device end, so that the processing device end generates a damage processing event and executes the damage processing event.

[0077] like Figure 5 As shown, the method includes:

[0078] S510: Acquire a package image of the consignment and item information of the consignment.

[0079] S520: Match the item information of the consignment in a preset knowledge base to determine a fragile label of the consignment.

[0080] S530: Generate text information based on the damage determination condition corresponding to the fragile label.

[0081] S540: Input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0082] S550: When the damage degree information of the consigned item meets the damage processing conditions, the item information of the consigned item is sent to the processing device end, so that the processing device end generates a damage processing event and executes the damage processing event.

[0083] Among them, the damage handling condition refers to the condition for carrying out damage handling on the consignment. Specifically, the damage handling condition may be that the degree of damage of the consignment reaches severe or moderate. It can be understood that if the degree of damage of the consignment is slight, there is no need to carry out damage handling on the consignment. If the degree of damage of the consignment is severe or moderate, the consignment shall be carried out damage handling. In this embodiment, when the damage degree information of the consignment meets the damage handling condition, the item information of the consignment is sent to the processing device end; the processing device end receives the item information of the consignment, generates a damage handling event based on the item information of the consignment, and executes the damage handling event. Among them, the processing device end is used to handle the damage handling event, and the processing device end may be a customer service system.

[0084] For example, Figure 6 is a flowchart of a device end processing a damage processing event provided by Embodiment 5 of the present invention. The processing flow of the damage processing event is as follows Figure 6 shown.

[0085] The technical solution of this embodiment is to send the item information of the consigned item to the processing device end when the damage degree information of the consigned item meets the damage processing conditions, so that the processing device end generates a damage processing event and executes the damage processing event. The damaged package can be processed in time, improving the user experience.

[0086] Figure 7 Schematic diagram of the structure of a package damage detection device provided by an embodiment of the present invention. Figure 7 As shown, the device comprises:

[0087] An image and item information acquisition module 710 is used to acquire a package image of a consignment and item information of the consignment;

[0088] A fragile label determination module 720, configured to determine a fragile label of the consignment by matching the item information of the consignment in a preset knowledge base;

[0089] A fragile label text information generating module 730, configured to generate text information based on a damage determination condition corresponding to the fragile label;

[0090] The damage detection module 740 is used to input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

[0091] The technical solution of this embodiment obtains the package image and item information of the consignment; matches the item information of the consignment in a preset knowledge base to determine the fragile label of the consignment; generates text information based on the damage judgment condition corresponding to the fragile label; inputs the text information and the package image of the consignment into a pre-trained damage detection model to obtain the damage detection result of the consignment. The damage detection model is constructed by using data in both image and text modes, which improves the efficiency of damage detection while improving the accuracy of damage detection and reduces the consumption of human resources.

[0092] On the basis of the above embodiment, optionally, the preset knowledge base includes item information of multiple items and fragile labels of items; the fragile label determination module 720 is used to convert the item information of the consignment into a feature vector, perform similarity matching on the feature vector in the preset knowledge base, determine the item corresponding to the maximum similarity, and determine the fragile label of the item corresponding to the maximum similarity as the fragile label of the consignment, wherein the item information in the preset knowledge base is stored in vector form.

[0093] On the basis of the above embodiment, optionally, the preset knowledge base includes item information and category attributes of multiple items, and different category attributes correspond to different fragile labels; the fragile label determination module 720 is used to match the item information of the consignment in the preset knowledge base, determine the category attributes of the consignment, and determine the fragile label of the consignment based on the category attributes of the consignment.

[0094] Based on the above embodiment, optionally, different fragile labels correspond to different damage judgment conditions, and the damage judgment conditions include damage status description information and damage degree information corresponding to the damage status description information; the text information includes context information and the damage judgment condition corresponding to the fragile label, wherein the context information is determined based on the item information or category attributes of the consignment.

[0095] On the basis of the above embodiment, optionally, the image and item information acquisition module 710 is specifically used to obtain the package image of the consignment and the item information of the consignment uploaded by the consignment processing outlet; determine the damage reporting portrait information of the consignment processing outlet based on the identification of the consignment processing outlet; and sort the processing priority of the package image and item information of the consignment based on the damage reporting portrait information of the consignment processing outlet.

[0096] On the basis of the above embodiment, optionally, the device also includes a load quantity adjustment module, which is used to determine the quality of damage reporting to the consignment processing point based on the damage detection result of the consignment and the consignment processing point corresponding to the consignment; and adjust the load quantity of the consignment processing point based on the damage reporting quality of the consignment processing point.

[0097] On the basis of the above embodiment, optionally, the device also includes a damage processing module, which is used to send the item information of the consignment to the processing device end when the damage degree information of the consignment meets the damage processing conditions, so that the processing device end generates a damage processing event and executes the damage processing event.

[0098] The package damage detection device provided in the embodiment of the present invention can execute the package damage detection method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0099] Figure 81 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. The electronic device 10 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device may also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or required herein.

[0100] like Figure 8 As shown, the electronic device 10 includes at least one processor 11, and a memory connected to the at least one processor 11, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., wherein the memory stores a computer program that can be executed by at least one processor, and 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 to 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. An input / output (I / O) interface 15 is also connected to the bus 14.

[0101] A number 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 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.

[0102] The processor 11 may be a variety of general and / or special 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 special artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any appropriate processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as a package damage detection method.

[0103] In some embodiments, the package damage detection method may be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as a storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on 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 package damage detection method described above may be performed. Alternatively, in other embodiments, the processor 11 may be configured to execute the package damage detection method in any other appropriate manner (e.g., by means of firmware).

[0104] Various implementations of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), load programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various implementations can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0105] The computer program for implementing the package damage detection method of the present invention can be written in any combination of one or more programming languages. These computer programs can be provided to a processor of a general-purpose computer, a special-purpose computer or other programmable data processing device, so that when the computer program is executed by the processor, the functions / operations specified in the flow chart and / or block diagram are implemented. The computer program can be executed completely on the machine, partially on the machine, partially on the machine as an independent software package and partially on a remote machine, or completely on a remote machine or server.

[0106] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause a processor to execute a package damage detection method, the method comprising:

[0107] Acquire a package image of the consignment and item information of the consignment; perform matching in a preset knowledge base based on the item information of the consignment to determine a fragile label of the consignment; generate text information based on damage determination conditions corresponding to the fragile label; input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes information on the degree of damage of the consignment.

[0108] In the context of the present invention, a computer-readable storage medium may be a tangible medium that may contain or store a computer program for use by or in combination with an instruction execution system, device or equipment. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. A more specific example of a machine-readable storage medium may include an electrical connection based on one or more lines, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0109] To provide interaction with a user, the systems and techniques described herein may 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 trackball) through which the user can provide input to the electronic device. Other types of devices may also be used to provide interaction with the user; for example, the feedback provided to the user may be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user may be received in any form (including acoustic input, voice input, or tactile input).

[0110] The systems and techniques described herein may 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 with a graphical user interface or a web browser through which a user can interact with implementations 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 may be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0111] A computing system may include a client and a server. The client and the server are generally remote from each other and usually interact through a communication network. The client and server relationship is generated by computer programs running on the corresponding 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 cloud host, which is a host product in the cloud computing service system to solve the defects of difficult management and weak business scalability in traditional physical hosts and VPS services.

[0112] It should be understood that the various forms of processes shown above can be used to reorder, add or delete steps. For example, the steps described in the present invention can be executed in parallel, sequentially or in different orders, as long as the desired results of the technical solution of the present invention can be achieved, and this document does not limit this.

[0113] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions can be made according to design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A package damage detection method, characterized in that: include: Acquire a package image of a consignment and item information of the consignment; Based on the item information of the consignment, a match is performed in a preset knowledge base to determine a fragile label of the consignment; Generate text information based on the damage determination condition corresponding to the fragile label; The text information and the package image of the consignment are input into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

2. The method according to claim 1, characterized in that The preset knowledge base includes item information of multiple items and fragile labels of items; The method of determining a fragile label of the consignment based on the item information of the consignment in a preset knowledge base includes: The item information of the consignment is converted into a feature vector, and the feature vector is similarity matched in the preset knowledge base to determine the item corresponding to the maximum similarity, and the fragile label of the item corresponding to the maximum similarity is determined as the fragile label of the consignment, wherein the item information in the preset knowledge base is stored in the form of a vector.

3. The method according to claim 1, characterized in that The preset knowledge base includes item information and category attributes of multiple items, and different category attributes correspond to different fragile labels; The method of determining a fragile label of the consignment based on the item information of the consignment in a preset knowledge base includes: The item information of the consignment is matched in a preset knowledge base to determine the category attribute of the consignment, and a fragile label of the consignment is determined based on the category attribute of the consignment.

4. The method according to claim 1, characterized in that: Different fragile labels correspond to different damage determination conditions, and the damage determination conditions include damage state description information and damage degree information corresponding to the damage state description information; The text information includes context information and a damage determination condition corresponding to the fragile label, wherein the context information is determined based on item information or category attributes of the consignment.

5. The method according to any one of claims 1 to 4, characterized in that: The obtaining of the package image of the consignment and the item information of the consignment includes: Obtaining a package image of the consignment and item information of the consignment uploaded by a consignment processing network point; Determining damage reporting portrait information of the consignment handling outlet based on the identifier of the consignment handling outlet; The package images and item information of the consignment are sorted by processing priority based on the damaged image information reported by the consignment processing outlet.

6. The method according to claim 5, characterized in that The method further comprises: Determining the quality of damage reporting to the consignment handling outlet based on the damage detection result of the consignment and the consignment handling outlet corresponding to the consignment; The load quantity of the luggage processing network point is adjusted based on the quality of damage reports of the luggage processing network point.

7. The method according to claim 1, characterized in that The method further comprises: When the damage degree information of the consignment meets the damage processing conditions, the item information of the consignment is sent to the processing device end, so that the processing device end generates a damage processing event and executes the damage processing event.

8. A package damage detection device, characterized in that: include: An image and item information acquisition module, used to acquire a package image of a consignment and item information of the consignment; A fragile label determination module, used for matching the item information of the consignment in a preset knowledge base to determine the fragile label of the consignment; A fragile label text information generation module, used to generate text information based on damage determination conditions corresponding to the fragile label; The damage detection module is used to input the text information and the package image of the consignment into a pre-trained damage detection model to obtain a damage detection result of the consignment, wherein the damage detection result of the consignment includes damage degree information of the consignment.

9. An electronic device, characterized in that: The electronic device comprises: 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 package damage detection method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and the computer instructions are used to enable a processor to implement the package damage detection method according to any one of claims 1 to 7 when executed.

Images