Damage identification method and apparatus for goods inside package, electronic device and storage medium

Abstract

The application discloses a damage identification method and apparatus for goods inside a package, an electronic device and a storage medium. The damage identification method for goods in a package comprises: obtaining a current X-ray image of a package; inputting the current X-ray image into an object detection model, so as to obtain a detection result. The object detection model is obtained on the basis of historical damaged goods X-ray images and a target detection model. The detection result comprises a first detection result used for representing goods being damaged in the package and a second detection result used for representing goods not being damaged in the package. In the present method, damage identification is performed on goods inside a package on the basis of the object detection model, such that the identification accuracy of damage identification for goods in a package is improved.

Description

Method, device, electronic device and storage medium for identifying damaged goods in a package Technical Field

[0001] The present application belongs to the field of logistics technology, and in particular relates to a method, device, electronic device and storage medium for identifying damaged goods in a package. Background Art

[0002] With the rapid development of e-commerce, the volume of express delivery business has gradually increased, and more and more goods are packaged into parcels for transportation. Parcels are often squeezed or collided during transportation.

[0003] However, when the goods in the package are fragile items, when the package is slightly impacted, the package packaging is intact, but the goods in the package are damaged. The inspectors may mistakenly judge that the goods in the package are not damaged, thereby reducing the accuracy of damage identification of the goods in the package.

[0004] Summary of the Invention

[0005] In view of this, embodiments of the present application provide a method, device, electronic device, and storage medium for identifying damaged goods in a package to overcome or at least partially solve the problems described in the above background technology.

[0006] In a first aspect, an embodiment of the present application provides a method for identifying damaged goods in a package, comprising: obtaining a current X-ray image of the package; inputting the current X-ray image into an object detection model to obtain a detection result, wherein the object detection model is obtained based on historical X-ray images of damaged goods and a target detection model, and the detection result includes a first detection result for characterizing that the goods in the package are damaged, and a second detection result for characterizing that the goods in the package are not damaged.

[0007] In a second aspect, embodiments of the present application provide a device for identifying damaged goods within a package, comprising an image acquisition module and an image input module. The image acquisition module is configured to acquire a current X-ray image of the package; the image input module is configured to input the current X-ray image into an object detection model to obtain a detection result. The object detection model is derived based on historical X-ray images of damaged goods and a target detection model. The detection results include a first detection result indicating that the goods within the package are damaged, and a second detection result indicating that the goods within the package are intact.

[0008] In a third aspect, an embodiment of the present application provides an electronic device comprising a memory; one or more processors coupled to the memory; and one or more applications, wherein the one or more applications are stored in the memory and configured to be executed by the one or more processors, and the one or more applications are configured to execute the method for identifying damaged goods in a package provided in the first aspect above.

[0009] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which program code is stored. The program code can be called by a processor to execute the method for identifying damaged goods in a package as provided in the first aspect above.

[0010] In a fifth aspect, an embodiment of the present application provides a computer program product. When the computer program product is run on a computer device, the computer device executes the method for identifying damaged goods in a package provided in the first aspect above.

[0011] The solution provided in the present application obtains a detection result by acquiring a current X-ray image of a package and inputting the current X-ray image into an object detection model. The object detection model is obtained based on historical X-ray images of damaged goods and a target detection model. The detection result includes a first detection result for characterizing that the goods in the package are damaged, and a second detection result for characterizing that the goods in the package are not damaged. This realizes damage identification of the goods in the package based on the object detection model, thereby improving the recognition accuracy of damage identification of the goods in the package. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] FIG1 shows a schematic diagram of a scenario of a system for identifying damaged goods in a package provided in an embodiment of the present application.

[0013] FIG2 shows a flow chart of a method for identifying damaged goods in a package provided in an embodiment of the present application.

[0014] FIG3 shows another flow chart of the method for identifying damaged goods in a package provided in an embodiment of the present application.

[0015] FIG4 shows a structural block diagram of a device for identifying damaged goods in a package provided in an embodiment of the present application.

[0016] FIG5 shows a functional block diagram of an electronic device provided in an embodiment of the present application.

[0017] FIG6 shows a functional block diagram of a computer-readable storage medium provided in an embodiment of the present application.

[0018] FIG7 shows a functional block diagram of a computer program product provided in an embodiment of the present application. DETAILED DESCRIPTION

[0019] As described in the background technology, with the rapid development of e-commerce, the volume of express delivery business has gradually increased, and more and more goods are packaged into parcels for transportation. During transportation, parcels are often damaged due to squeezing or collision, which in turn causes damage to the goods inside the parcels. In order to prevent customers from receiving parcels with damaged goods, logistics companies often identify the damage of the goods inside the parcels before delivery, and intercept the parcels when it is identified that the goods inside the parcels are damaged.

[0020] Currently, damage identification of goods within a package is primarily performed by sorters visually observing the package packaging for damage. If the sorter observes that the package packaging is intact, the goods within the package are deemed intact. If the sorter observes that the package packaging is damaged, the goods within the package are deemed broken.

[0021] However, when the goods in the package are fragile items, when the package is slightly impacted, the package packaging is intact, but the goods in the package are damaged. The inspector observes that the package packaging is not damaged, so he mistakenly judges that the goods in the package are not damaged, which reduces the accuracy of damage identification of the goods in the package.

[0022] In response to the above problems, the inventors have conducted a long period of research and proposed the method, device, electronic device and storage medium for identifying damaged goods in packages provided in the embodiments of the present application. The detection results are obtained by acquiring the current X-ray image of the package and inputting the current X-ray image into an object detection model. The object detection model is obtained based on historical X-ray images of damaged goods and a target detection model. The detection results include a first detection result for characterizing that the goods in the package are damaged, and a second detection result for characterizing that the goods in the package are not damaged. This realizes the identification of damaged goods in the package based on the object detection model, thereby improving the recognition accuracy of damaged goods in the package.

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application.

[0024] Please refer to Figure 1, which shows a schematic diagram of an application scenario of the system for identifying damaged goods in a package provided in an embodiment of the present application, which may include a security inspection machine 100 and a processing device 200. The security inspection machine 100 is communicatively connected to the processing device 200 and exchanges data with the processing device 200.

[0025] The security inspection machine 100 can emit X-rays, capture images of the package through X-rays, obtain X-ray images, and send the X-ray images to the processing device 200.

[0026] The processing device 200 can be used to receive the X-ray image sent by the security inspection machine 100 and identify damage to the goods in the package based on the X-ray image.

[0027] The processing device 200 may be a terminal device or a server, etc. The type of the processing device 200 is not limited here and can be specifically configured according to actual needs.

[0028] The terminal device can be a mobile terminal device (for example, any one of a mobile phone, a personal digital assistant (PDA), a tablet personal computer (Tablet PC), a laptop computer, a smart watch, a smart bracelet or a wearable device, etc.), or a fixed terminal device (a desktop computer or a smart panel, etc.).

[0029] The server can be an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides any of the basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (CDNs), big data or artificial intelligence platforms.

[0030] In some embodiments, the security inspection machine 100 can be integrated with the processing device 200 to form an integrated device. For example, the processing device 200 can be embedded in the security inspection machine 100 and form an integrated device with the security inspection machine 100. The integrated device can be used to capture images of packages, obtain X-ray images, and identify damage to the goods in the packages based on the X-ray images.

[0031] Please refer to Figure 2, which shows a flow chart of a method for identifying damaged goods within a package, according to one embodiment of the present application. This method can be applied to processing device 200 in the system for identifying damaged goods within a package, as shown in Figure 1. The following describes the process shown in Figure 2 in detail, using processing device 200 as an example. The method can include steps S110 through S120.

[0032] Step S110: Acquire the current X-ray image of the package.

[0033] In an exemplary embodiment, when a sorter needs to identify damage to goods in a package, he or she may send an identification instruction to a processing device. The processing device receives and responds to the identification instruction and obtains a current X-ray image of the package.

[0034] For example, when a sorter needs to identify damage to goods in a package, he or she can send an image acquisition instruction to the security inspection machine. The security inspection machine receives and responds to the image acquisition instruction, performs an X-ray scan on the package, obtains the current X-ray image, and sends the current X-ray image to the processing device. The processing device receives the current X-ray image returned by the security inspection machine.

[0035] X-ray imaging works by exploiting the varying degrees of absorption of X-rays by materials of varying densities when they penetrate an object, revealing information about the object's internal structure. Due to the varying densities of different materials, X-rays are attenuated to varying degrees upon penetration, resulting in different signals being detected by the receiver. Based on the strength of these signals, the security inspection machine then renders them in different colors for easier detection. Generally speaking, textiles appear light yellow, liquids appear orange-yellow, metals appear blue, and composites such as glass and ceramics appear green. Objects impenetrable to X-rays appear black.

[0036] In an exemplary embodiment, the processing device may include an input panel. When the sorter needs to identify damage to goods in a package, the sorter may input an identification instruction on the input panel, and the processing device receives the identification instruction through the input panel.

[0037] In an exemplary embodiment, the processing device may include a voice recognition module. When the sorter needs to identify damage to the goods in the package, he can send a voice message carrying a recognition instruction within the voice collection range of the voice recognition module. The processing device collects the voice information through the voice recognition module, performs voice recognition on the voice information, and obtains the recognition instruction.

[0038] In an exemplary embodiment, the system for identifying damaged goods in a package may further include a client, which is connected to the processing device via a network and exchanges data with the processing device via the network.

[0039] When the sorter needs to identify damaged goods in a package, he can send an identification instruction to the client. The client receives and responds to the identification instruction and forwards the identification instruction to the processing device through the network. The processing device receives the identification instruction forwarded by the client.

[0040] The client can be any one of a mobile client (for example, a mobile phone client, a personal digital assistant (PDA) client, a tablet personal computer (Tablet PC) client, a laptop client, a smart watch client, a smart bracelet client or a wearable client, etc.) or a fixed client (for example, a desktop computer client or a smart panel client, etc.). The type of client is not limited here and can be set according to actual needs.

[0041] The network can be any one of a ZigBee network, a Bluetooth (BT) network, a Wireless Fidelity (Wi-Fi) network, a Thread network, a Long Range Radio (LoRa) network, a Low-Power Wide-Area Network (LPWAN), an infrared network, a Narrow Band Internet of Things (NB-IoT), a Controller Area Network (CAN), a Digital Living Network Alliance (DLNA) network, a Wide Area Network (WAN), a Local Area Network (LAN), a Metropolitan Area Network (MAN) or a Wireless Personal Area Network (WPAN), etc. The type of network is not limited here and can be set according to actual needs.

[0042] Step S120: Input the current X-ray image into the object detection model to obtain the detection result.

[0043] In an exemplary embodiment, after acquiring a current X-ray image of a package, the processing device may input the current X-ray image into an object detection model. The object detection model receives and responds to the current X-ray image, identifies damage to the cargo within the package based on the current X-ray image, obtains a detection result, and outputs the detection result to the processing device, which then receives the detection result output by the object detection model. The object detection model may be derived based on historical X-ray images of damaged cargo and an object detection model. The detection results may include a first detection result indicating that the cargo within the package is damaged, and a second detection result indicating that the cargo within the package is intact. This enables damage identification of the cargo within the package based on the object detection model, thereby improving the accuracy of damage identification of the cargo within the package.

[0044] The object detection model can be obtained by training a target detection model based on historical X-ray images of damaged goods. The target detection model can be any of the following models: the YOLO-v3 model, the Faster-RCNN model, or the YOLOX model. The type of target detection model is not limited here and can be set according to actual needs.

[0045] In an exemplary embodiment, after obtaining the current X-ray image of the package, the processing device can obtain a training set and a test set based on historical X-ray images of damaged goods, and input the training set into the target detection model. The target detection model receives and responds to the training set, performs training based on the training set, obtains an object detection model, and determines whether the object detection model has converged based on the test set. When it is determined that the object detection model has converged based on the test set, the current X-ray image is input into the object detection model. The object detection model receives and responds to the current X-ray image, identifies damage to the goods in the package based on the current X-ray image, obtains a detection result, and outputs the detection result to the processing device. The processing device receives the detection result output by the object detection model, and realizes damage identification of the goods in the package based on the object detection model when it is determined that the object detection model has converged, thereby ensuring the stability of the object detection model and improving the recognition reliability of damage identification of the goods in the package.

[0046] In an exemplary embodiment, the processing device can obtain historical X-ray images of damaged goods, annotate the historical X-ray images of damaged goods, obtain annotated X-ray images of damaged goods, divide the annotated X-ray images of damaged goods according to preset division rules, obtain a training set and a test set, and input the training set into the target detection model. The target detection model receives and responds to the training set, performs training according to the training set, and obtains an object detection model.

[0047] The preset partitioning rule may be a manual partitioning rule. For example, the preset partitioning rule may be a training set:test set = 9:2 manual partitioning rule. If the number of X-ray images of damaged goods is 22,000, the training set is 18,000 and the test set is 4,000. The preset partitioning rule may be a training set:test set = 3:1 manual partitioning rule. If the number of X-ray images of damaged goods is 40,000, the training set is 30,000 and the test set is 10,000. The partitioning method of the preset partitioning rule is not limited here and can be set according to actual needs.

[0048] In one application scenario, the target detection model is a YOLOX model. The processing device can obtain historical X-ray images of damaged goods, label the historical X-ray images of damaged goods, obtain labeled X-ray images of historical damaged goods, and divide the labeled X-ray images of historical damaged goods according to preset division rules to obtain a training set and a test set, and input the training set into the YOLOX model. The YOLOX model receives and responds to the training set, performs training based on the training set, and obtains an object detection model.

[0049] In an exemplary embodiment, the processing device can obtain historical X-ray images of damaged goods, and annotate the historical X-ray images of damaged goods to obtain annotated X-ray images of damaged goods, and divide the annotated X-ray images of damaged goods according to preset division rules to obtain a training set and a test set, and perform data enhancement processing on the training set according to a preset data enhancement strategy to obtain an enhanced training set, and input the enhanced training set into a target detection model. The target detection model receives and responds to the enhanced training set, and is trained according to the enhanced training set to obtain an object detection model, thereby realizing the training of the target detection model according to the enhanced training set, avoiding the low robustness of the object detection model obtained by training the target detection model using fewer historical X-ray images of damaged goods due to the small amount of historical X-ray image data of damaged goods, and increasing the robustness of the object detection model.

[0050] The preset data enhancement strategy may be a brightness change processing strategy, for example, the brightness of the historical damaged cargo X-ray image is changed to [0.7, 1.3] times the initial brightness of the historical damaged cargo X-ray image.

[0051] The preset data enhancement strategy may also be a contrast change processing strategy, for example, the contrast of the historical damaged cargo X-ray image is changed to [0.7, 1.3] times the initial contrast of the historical damaged cargo X-ray image.

[0052] The preset data augmentation strategy can also be an image scaling strategy. For example, for each side of each historical X-ray image of damaged goods, a number x% is randomly selected from -5% to 5%. If the number is greater than zero, the side is expanded and the expanded area is filled with x% of the pixels multiplied by the side length; if the number is less than zero, the side is cropped with a crop size of x% of the pixels multiplied by the side length; if the number is zero, the side length remains unchanged.

[0053] The preset data enhancement strategy can also be an image flipping processing strategy. For example, according to a preset percentage (e.g., 50%) probability of the historical X-ray images of damaged goods, the historical X-ray images of damaged goods are flipped up and down in a clockwise or counterclockwise direction, and the up and down flipping angle is 0-360°.

[0054] The preset data enhancement strategy can also be a transparency overlay processing strategy. For example, two images to be overlaid are randomly selected from the historical damage X-ray images, and the two images to be overlaid are overlaid with a transparency of 40%-60%, and the damage detection frame in the image is also overlaid.

[0055] The preset data enhancement strategy can also be other processing strategies. The type of the preset data enhancement strategy is not limited here and can be set according to actual needs.

[0056] In an exemplary embodiment, the processing device may input a test set into the object detection model, the object detection model receives and responds to the test set, performs testing based on the test set, obtains test results, and determines a test accuracy of the test results based on the test results and the test set, and determines whether the object detection model has converged based on the test accuracy.

[0057] When the test accuracy is greater than or equal to the preset accuracy threshold, it is determined that the object detection model has converged; when the test accuracy is less than the preset accuracy threshold, it is determined that the object detection model has not converged.

[0058] The preset accuracy threshold can be used to represent the minimum accuracy value when the object detection model has converged. The preset accuracy threshold can be an accuracy value pre-set by the sorter. The preset accuracy threshold can also be an accuracy value automatically generated by the processing equipment based on the recognition process of multiple damage identification of goods in the package, etc., which is not limited here.

[0059] The solution provided in this embodiment obtains a detection result by acquiring a current X-ray image of a package and inputting the current X-ray image into an object detection model. The object detection model is obtained based on historical X-ray images of damaged goods and a target detection model. The detection result includes a first detection result for indicating that the goods in the package are damaged, and a second detection result for indicating that the goods in the package are not damaged. This implements damage identification of the goods in the package based on the object detection model, thereby improving the accuracy of identification of damage to the goods in the package.

[0060] Please refer to Figure 3, which shows a flow chart of a method for identifying damaged goods within a package, according to another embodiment of the present application. This method can be applied to processing device 200 in the system for identifying damaged goods within a package, as shown in Figure 1. The following describes the process shown in Figure 3 in detail, using processing device 200 as an example. The method can include steps S210 through S240.

[0061] Step S210: Acquire the current X-ray image of the package.

[0062] Step S220: Input the current X-ray image into the object detection model to obtain the detection result.

[0063] The details of step S210 and step S220 can refer to the contents of step S110 and step S120 in the aforementioned embodiment, and will not be repeated here.

[0064] Step S230: Upon receiving the first detection result output by the object detection model, obtain the package number of the package.

[0065] In an exemplary embodiment, the processing device receives a first detection result output by the object detection model, which indicates that the goods in the package are damaged, and the package tracking number of the package can be obtained.

[0066] In an exemplary embodiment, the system for identifying damaged goods in a package may further include a six-sided barcode scanner, which is connected to the processing device via a network and exchanges data with the processing device via the network.

[0067] When the processing device receives the first detection result output by the object detection model, it can send an order number collection instruction to the six-sided barcode scanner via the network. The six-sided barcode scanner receives and responds to the order number collection instruction, performs six-sided barcode scanning on the package, obtains the package order number, and sends the package order number to the processing device via the network. The processing device receives the package order number returned by the six-sided barcode scanner.

[0068] In an exemplary embodiment, when the processing device receives the first detection result output by the object detection model, a first package number input prompt message can be generated. The first package number input prompt message is used to prompt the sorter to input the package number of the package to the processing device, and when the sorter inputs the package number to the processing device, the processing device receives the package number input by the sorter.

[0069] The first package number input prompt information can be at least any one of a first text prompt information, a first sound prompt information, a first light prompt information, etc., which is not limited here.

[0070] In an exemplary embodiment, the system for identifying damaged goods in a package may further include a client. When the processing device receives the first detection result output by the object detection model, it may send a tracking number acquisition instruction to the client via the network. The client receives the tracking number acquisition instruction sent by the processing device and generates a second package tracking number input prompt information. The second package tracking number input prompt information is used to prompt the sorter to enter the package tracking number of the package to the client. When the sorter enters the package tracking number to the client, the client receives the package tracking number entered by the sorter and sends the package tracking number to the processing device via the network. The processing device receives the package tracking number returned by the client.

[0071] The second package number input prompt information can be at least any one of a second text prompt information, a second sound prompt information, or a second light prompt information, etc., which is not limited here.

[0072] Step S240: Generate interception prompt information including the package number.

[0073] In an exemplary embodiment, after the processing device obtains the parcel tracking number of the package, it can generate an interception prompt message carrying the parcel tracking number. The interception prompt message can be used to prompt the sorter to intercept the package according to the parcel tracking number, thereby preventing customers from receiving damaged goods and improving the customer's logistics experience.

[0074] The interception prompt information can be at least any one of text interception prompt information, sound interception prompt information or light interception prompt information, etc. The type of interception prompt information is not limited here and can be set according to actual needs.

[0075] In an exemplary embodiment, after the processing device generates interception prompt information including the package number, it can generate and store the package interception record, which facilitates the logistics controller to track the historical package management records and improves the management experience of the package.

[0076] The solution provided in this embodiment obtains a current X-ray image of the package, inputs the current X-ray image into an object detection model, obtains a detection result, obtains the package number of the package upon receiving the first detection result output by the object detection model, and generates interception prompt information including the package number. This achieves damage identification of goods in the package based on the object detection model, thereby improving the accuracy of damage identification of goods in the package.

[0077] Furthermore, when it is determined based on the current X-ray image that the goods in the package are damaged, the sorter is reminded to intercept the damaged goods package, thereby preventing customers from receiving damaged goods packages and improving the customer's logistics experience.

[0078] Please refer to Figure 4, which shows a device 300 for identifying damaged goods in a package provided by one embodiment of the present application. The device 300 can be applied to the processing device 200 in the system for identifying damaged goods in a package as shown in Figure 1. The device 300 shown in Figure 4 will be described in detail below using the processing device 200 as an example. The device 300 can include an image acquisition module 310 and an image input module 320.

[0079] The image acquisition module 310 can be used to obtain the current X-ray image of the package; the image input module 320 can be used to input the current X-ray image into the object detection model to obtain a detection result. The object detection model can be obtained based on historical X-ray images of damaged goods and the target detection model. The detection result may include a first detection result used to characterize that the goods in the package are damaged, and a second detection result used to characterize that the goods in the package are not damaged.

[0080] In an exemplary embodiment, the device 300 for identifying damaged goods in a package may further include a tracking number acquisition module and a generation module.

[0081] The order number acquisition module can be used to obtain the parcel order number of the package when receiving the first detection result output by the object detection model; the generation module can be used to generate interception prompt information including the parcel order number, and the interception prompt information can be used to prompt the sorter to intercept the package according to the parcel order number.

[0082] In an exemplary embodiment, the order number obtaining module may include a sending unit and a receiving unit.

[0083] The sending unit can be used to send a tracking number collection instruction to the six-sided barcode scanner when receiving the first detection result output by the object detection model. The tracking number collection instruction can be used to instruct the six-sided barcode scanner to scan the six sides of the package to obtain the package tracking number; the receiving unit can be used to receive the package tracking number returned by the six-sided barcode scanner.

[0084] In an exemplary embodiment, the order number acquisition module may include a generating unit and a receiving unit.

[0085] The generation unit can be used to generate a first package number input prompt message upon receiving a first detection result output by the object detection model. The first package number input prompt message is used to prompt the sorter to enter the package number of the package; the receiving unit can be used to receive the package number entered by the sorter.

[0086] In an exemplary embodiment, the order number obtaining module may include a sending unit and a receiving unit.

[0087] The sending unit can be used to send a tracking number acquisition instruction to the client upon receiving the first detection result output by the object detection model. The tracking number acquisition instruction is used to instruct the client to generate a second package tracking number input prompt information. The second package tracking number input prompt information is used to prompt the sorter to enter the package tracking number of the package to the client; the receiving unit can be used to receive the package tracking number returned by the client.

[0088] In an exemplary embodiment, the device 300 for identifying damaged goods in a package may further include a training set and a test set acquisition module, a training set input module, and a determination module.

[0089] The training set and test set acquisition module can be used for the image input module 320 to input the current X-ray image into the object detection model. Before obtaining the detection results, the training set and test set are obtained based on the historical X-ray images of damaged goods; the training set input module can be used to input the training set into the target detection model for training to obtain the object detection model; the determination module can be used to determine whether the object detection model has converged based on the test set.

[0090] In an exemplary embodiment, the image input module 320 may include a first input unit.

[0091] The first input unit can be used to input the current X-ray image into the object detection model to obtain a detection result when it is determined that the object detection model has converged according to the test set.

[0092] In an exemplary embodiment, the device 300 for identifying damaged goods in a package may further include an enhanced processing module.

[0093] The enhancement processing module can be used to input the training set into the target detection model for training. Before obtaining the object detection model, the training set is enhanced according to the preset data enhancement strategy to obtain an enhanced training set.

[0094] In an exemplary embodiment, the training set input module may include a second input unit.

[0095] The second input unit can be used to input the enhanced training set into the target detection model for training to obtain the object detection model.

[0096] In an exemplary embodiment, the preset data enhancement strategy includes at least one of the following processing strategies: a brightness change processing strategy, a contrast change processing strategy, an image scaling processing strategy, an image flipping processing strategy, or a transparency overlay processing strategy.

[0097] In an exemplary embodiment, the determination module may include a third input unit, a first determination unit, and a second determination unit.

[0098] The third input unit can be used to input the test set into the object detection model for testing to obtain the test results; the first determination unit can be used to determine the test accuracy of the test results based on the test results and the test set; the second determination unit can be used to determine whether the object detection model converges based on the test accuracy.

[0099] In an exemplary embodiment, the first determining unit may include a first determining subunit and a second determining subunit.

[0100] The first determination subunit can be used to determine that the object detection model has converged when the test accuracy is greater than or equal to a preset accuracy threshold; the second determination subunit can be used to determine that the object detection model has not converged when the test accuracy is less than the preset accuracy threshold.

[0101] It should be noted that the various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to in detail. For the device embodiments, since they are basically similar to the method embodiments, the description is relatively simple. For the relevant parts, refer to the partial description of the method embodiments. Any processing method described in the method embodiment can be implemented by the corresponding processing module in the device embodiment, and will not be repeated in detail in the device embodiment.

[0102] In addition, the functional modules in the various embodiments of the present application may be integrated into a processing module, or each module may exist physically separately, or two or more modules may be integrated into a single module. The above-mentioned integrated modules may be implemented in the form of hardware or software functional modules.

[0103] In addition, FIG4 shows only one embodiment of the device for identifying damaged goods in a package, rather than all embodiments. All other embodiments obtained by those skilled in the art based on the embodiment of the device for identifying damaged goods in a package in this application without further creative work shall fall within the scope of protection of this application.

[0104] Please refer to Figure 5, which shows a functional block diagram of an electronic device 400 provided by an embodiment of the present application. The electronic device 400 may include: a memory 410, one or more processors 420, and one or more applications. The one or more applications may be stored in the memory 410 and configured to be executed by one or more processors 420. The one or more applications are configured to execute the method described in the aforementioned method embodiment.

[0105] The memory 410 may include a random access memory (RAM) or a read-only memory (ROM). The memory 410 may be used to store instructions, programs, codes, code sets, or instruction sets. The memory 410 may include a program storage area and a data storage area. The program storage area may store instructions for implementing an operating system, instructions for implementing at least one function (such as obtaining a current X-ray image, inputting a current X-ray image, obtaining a detection result, receiving a first detection result, outputting a first detection result, obtaining a package number, generating an interception prompt message, prompting a sorter, intercepting a package, sending a number collection instruction, six-sided code scanning and recognition, obtaining a package number, returning a package number, receiving a package number, obtaining a training set, obtaining a test set, inputting a training set, performing training, obtaining an object detection model, determining whether the object detection model converges, determining whether the object detection model converges, data enhancement processing, obtaining an enhanced training set, inputting an enhanced training set, inputting a test set, performing testing, obtaining a test result, determining a test accuracy rate, or determining that the object detection model has not converged, etc.) or instructions for implementing the above-mentioned various method embodiments. The data storage area can also store data created by the electronic device 400 during use (such as packages, current X-ray images, object detection models, detection results, historical X-ray images of damaged goods, target detection models, first detection results, second detection results, package numbers, interception prompt information, sorters, number collection instructions, six-sided barcode scanners, training sets, test sets, preset data enhancement strategies, enhanced training sets, test results, test accuracy or preset accuracy thresholds), etc.

[0106] The processor 420 may include one or more processing cores. The processor 420 utilizes various interfaces and circuits to connect the various components within the electronic device 400. The processor 420 performs various functions and data processing for the electronic device 400 by running or executing instructions, programs, code sets, or instruction sets stored in the memory 410, as well as accessing data stored in the memory 410. Optionally, the processor 420 may be implemented using at least one hardware form factor of a digital signal processing (DSP), a field-programmable gate array (FPGA), or a programmable logic array (PLA). The processor 420 may integrate one or a combination of a central processing unit (CPU), a graphics processing unit (GPU), or a modem. The CPU primarily processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into the processor 420 and may be implemented separately via a communication chip.

[0107] In addition, FIG5 shows only one embodiment of the electronic device, not all embodiments. All other embodiments obtained by those skilled in the art based on the electronic device embodiment in this application without any creative work shall fall within the scope of protection of this application.

[0108] Please refer to Figure 6, which shows a block diagram of a computer-readable storage medium provided in an embodiment of the present application. The computer-readable storage medium 500 stores program code 510, which can be called by a processor to execute the method described in the above method embodiment.

[0109] The computer-readable storage medium 500 can be an electronic memory such as a flash memory, an EEPROM (Electrically Erasable Programmable Read-Only Memory), an EPROM, a hard disk, or a ROM. Alternatively, the computer-readable storage medium 500 includes a non-transitory computer-readable storage medium. The computer-readable storage medium 500 has storage space for program code 510 for executing any of the method steps described above. These program codes can be read from or written to one or more computer program products. The program code 510 can be compressed, for example, in a suitable form.

[0110] In addition, FIG6 shows only one embodiment of a computer-readable storage medium, not all embodiments. All other embodiments obtained by those skilled in the art based on the computer-readable storage medium embodiment in this application without any creative work shall fall within the scope of protection of this application.

[0111] Please refer to Figure 7, which shows a block diagram of a computer program product 600 provided in an embodiment of the present application. Computer program product 600 includes a computer program / instructions 610, which is stored in a computer-readable storage medium of a computer device. When computer program product 600 is executed on a computer device, a processor of the computer device reads computer program / instructions 610 from the computer-readable storage medium and executes computer program / instructions 610, causing the computer device to perform the method described in the above method embodiment.

[0112] In addition, FIG7 shows only one embodiment of a computer program product, not all embodiments. All other embodiments obtained by those skilled in the art based on the computer program product embodiment in this application without any creative work shall fall within the scope of protection of this application.

[0113] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for identifying damaged goods in a package, characterized in that: include: Get the current X-ray image of the package; The current X-ray image is input into an object detection model to obtain a detection result, wherein the object detection model is obtained based on a historical X-ray image of damaged goods and a target detection model, and the detection result includes a first detection result for characterizing that the goods in the package are damaged, and a second detection result for characterizing that the goods in the package are not damaged.

2. The method for identifying damaged goods in a package according to claim 1, characterized in that: Also includes: Upon receiving the first detection result output by the object detection model, obtaining a parcel number of the parcel; Generate interception prompt information including the package number, and the interception prompt information is used to prompt the sorter to intercept the package according to the package number.

3. The method for identifying damaged goods in a package according to claim 2, characterized in that: The step of obtaining the parcel number of the parcel upon receiving the first detection result output by the object detection model comprises: Upon receiving the first detection result output by the object detection model, sending a tracking number collection instruction to the six-sided barcode scanner, wherein the tracking number collection instruction is used to instruct the six-sided barcode scanner to scan and identify the six-sided barcodes of the package to obtain the tracking number of the package; Receive the parcel number returned by the six-sided barcode scanner.

4. The method for identifying damaged goods in a package according to claim 2, characterized in that: The step of obtaining the parcel number of the parcel upon receiving the first detection result output by the object detection model comprises: Upon receiving the first detection result output by the object detection model, generating first package number input prompt information, wherein the first package number input prompt information is used to prompt a sorter to input the package number of the package; Receive the parcel number input by the sorter.

5. The method for identifying damaged goods in a package according to claim 2, characterized in that: The step of obtaining the parcel number of the parcel upon receiving the first detection result output by the object detection model comprises: Upon receiving the first detection result output by the object detection model, sending a tracking number acquisition instruction to the client, wherein the tracking number acquisition instruction is used to instruct the client to generate second package tracking number input prompt information, and the second package tracking number input prompt information is used to prompt the sorter to input the package tracking number of the package to the client; Receive the parcel number returned by the client.

6. The method for identifying damaged goods in a package according to any one of claims 1 to 5, characterized in that: Also includes: Obtain training and test sets based on historical X-ray images of damaged goods; Inputting the training set into the target detection model for training to obtain the object detection model; Determining whether the object detection model converges according to the test set; The step of inputting the current X-ray image to the object detection model to obtain the detection result includes: When it is determined that the object detection model has converged according to the test set, the current X-ray image is input into the object detection model to obtain the detection result.

7. The method for identifying damaged goods in a package according to claim 6, characterized in that: Also includes: According to a preset data enhancement strategy, data enhancement processing is performed on the training set to obtain an enhanced training set; The step of inputting the training set into the target detection model for training to obtain the object detection model includes: The enhanced training set is input into the target detection model for training to obtain the object detection model.

8. The method for identifying damaged goods in a package according to claim 7, characterized in that: The preset data enhancement strategy includes at least one of the following processing strategies: a brightness change processing strategy, a contrast change processing strategy, an image scaling processing strategy, an image flipping processing strategy or a transparency overlay processing strategy.

9. The method for identifying damaged goods in a package according to any one of claims 6 to 8, characterized in that: The determining whether the object detection model converges according to the test set includes: Input the test set to the object detection model for testing to obtain a test result; Determining the test accuracy of the test result according to the test result and the test set; Determine whether the object detection model converges according to the test accuracy.

10. The method for identifying damaged goods in a package according to claim 9, characterized in that: The determining whether the object detection model converges according to the test accuracy includes: When the test accuracy is greater than or equal to a preset accuracy threshold, determining that the object detection model has converged; When the test accuracy is less than the preset accuracy threshold, it is determined that the object detection model has not converged.

11. A device for identifying damaged goods in a package, characterized in that: include: An image acquisition module, used to acquire a current X-ray image of the package; An image input module is used to input the current X-ray image into an object detection model to obtain a detection result, wherein the object detection model is obtained based on historical X-ray images of damaged goods and a target detection model, and the detection result includes a first detection result for characterizing that the goods in the package are damaged, and a second detection result for characterizing that the goods in the package are not damaged.

12. An electronic device, characterized in that: include: Memory; One or more processors coupled to the memory; One or more applications, wherein the one or more applications are stored in the memory and are configured to be executed by one or more processors, and the one or more applications are configured to execute the method for identifying damaged goods in a package as described in any one of claims 1 to 10.

13. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores program codes, and the program codes can be called by a processor to execute the method for identifying damaged goods in a package as described in any one of claims 1 to 10.

Images