Logistics package picture recognition method, device and equipment and storage medium

Abstract

The invention discloses a logistics package picture identification method, and the method comprises the steps: obtaining a logistics package picture, putting a logistics package into distribution equipment, and carrying out the shooting of the logistics package, and obtaining the pictures of the logistics package passing through different distribution equipment; determining a to-be-recognized area, and determining a picture target recognition area by adopting an image processing technology; constructing Kafka serialization, and generating the logistics package information into a Kafka queue; carrying out feature extraction and classification on the pictures, and carrying out feature extraction and classification on the logistics package pictures; and judging whether the picture is qualified or not, comparing the extracted features with preset rule features, and judging whether the logistics package picture is qualified or not. According to the invention, goods pictures can be effectively acquired, identified, processed and transmitted, the shot pictures are matched with preset picture features, whether the pictures meet business requirements is judged, all links and equipment of logistics distribution are effectively monitored, and the logistics management level and the working efficiency are improved.

Description

Technical Field

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

[0002] With the development of e-commerce, the logistics industry is facing huge challenges. Traditional logistics management methods mainly rely on manual operations, which are inefficient and prone to errors. At present, there are still many problems in the identification and management of goods in the logistics distribution link. For example, the method of obtaining goods pictures is not efficient enough, the transmission and processing of goods pictures are not timely and accurate enough, and it is difficult to quickly and accurately determine whether the goods meet the requirements of logistics business. These problems not only affect the speed and quality of logistics, but also increase logistics costs and management difficulties.

[0003] In order to effectively obtain and identify cargo images in the logistics distribution link, accurately process and transmit cargo images, and thereby determine whether the cargo meets business requirements, it is necessary to invent a logistics package image recognition method, device, equipment and storage medium. Summary of the invention

[0004] The main purpose of the present invention is to solve the technical problem that the prior art cannot effectively obtain and identify cargo images in the logistics distribution link, and cannot accurately process and transmit cargo images to determine whether the cargo meets business requirements.

[0005] A first aspect of the present invention provides a method for identifying a logistics package image, the method comprising:

[0006] Obtaining pictures of logistics packages, placing the logistics packages into the distribution equipment and photographing the logistics packages, and obtaining pictures of the logistics packages passing through different distribution equipment;

[0007] Determine the area to be identified, and use image processing technology to determine the target identification area of ​​the image;

[0008] Build Kafka serialization, produce logistics package information into Kafka queue, and obtain logistics package information topic queue;

[0009] Extract and classify image features, use a deep image classification algorithm to extract features from the logistics package images, and classify them according to the extracted features;

[0010] To determine whether the image is qualified, the extracted features are compared with the preset rule features to determine whether the image scanned by the logistics package on each distribution device is qualified.

[0011] Optionally, in a first implementation of the first aspect of the present invention, the obtaining of the logistics package image, placing the logistics package into a distribution device and photographing the logistics package, and obtaining images of the logistics package passing through different distribution devices, includes:

[0012] Count the number of distribution equipment in the distribution center;

[0013] According to the working pipeline of the distribution center, the distribution equipment is encoded in sequence;

[0014] The logistics package passes through the distribution devices in sequence according to the codes of the distribution devices, and the distribution devices take photos of the logistics package to obtain pictures of the logistics package in each distribution device.

[0015] Optionally, in a second implementation of the first aspect of the present invention, the determining of the area to be identified, using image processing technology to determine the image target identification area, includes:

[0016] Performing a frame selection on the image to obtain a frame selection area of ​​the image;

[0017] Performing contour recognition processing on the framed area to obtain a contour recognition image of the picture;

[0018] Filtering the interference contours in the contour recognition image to obtain a filtered target contour image;

[0019] Using a target contour algorithm to estimate the filtered target contour image, and estimating the shape of the target contour image;

[0020] The category of the target contour image is preset, and the shape of the target contour image is classified.

[0021] Optionally, in a third implementation of the first aspect of the present invention, Kafka serialization is constructed, wherein the construction of Kafka serialization produces the logistics package information into a Kafka queue to obtain a logistics package information topic queue, including:

[0022] Create a topic in Kafka for storing the logistics package image storage database;

[0023] Predefine the message format of logistics package information;

[0024] Serializing the message of the logistics package information to obtain a logistics package information topic queue;

[0025] The logistics package information is divided into pieces according to a preset custom partitioner.

[0026] Optionally, in a fourth implementation of the first aspect of the present invention, the extracting and classifying the image features, using a deep image classification algorithm to extract features from the logistics package images, and classifying according to the extracted features, includes:

[0027] Determine the classification basis of the logistics package image;

[0028] Using a deep image classification algorithm, extracting image features of the logistics package image from the logistics package image according to the determined classification basis;

[0029] The logistics package images are classified according to the extracted image features.

[0030] Optionally, in a fifth implementation of the first aspect of the present invention, the step of determining whether the image is qualified, comparing the extracted features with preset rule features, and determining whether the image scanned by the logistics package on each distribution device is qualified includes:

[0031] Preset rule features and feature thresholds;

[0032] Matching the image features of the logistics package image with the rule features, and if the feature matching degree is greater than the feature threshold, judging that the image scanned on the distribution device is unqualified;

[0033] If the feature matching degree is less than or equal to the feature threshold, it is determined that the image scanned on the distribution device is qualified.

[0034] Optionally, in a sixth implementation of the first aspect of the present invention, the image features of the logistics package image are matched with the rule features by feature points.

[0035] Including: using scale-invariant features, accelerated robust features, color features or structural edge features to match feature points of two images to be matched.

[0036] A second aspect of the present invention provides a logistics package image recognition device, comprising:

[0037] A module for obtaining pictures of logistics packages is used to place logistics packages into distribution equipment and take pictures of the logistics packages to obtain pictures of the logistics packages passing through different distribution equipment;

[0038] A module for determining an area to be identified is used to determine the target identification area of ​​the image by using image processing technology;

[0039] Build a Kafka serialization module to produce logistics package information into the Kafka queue and obtain the logistics package information topic queue;

[0040] A module for extracting and classifying image features, which is used to extract features from the logistics package images using a deep image classification algorithm, and classify the images according to the extracted features;

[0041] The module for judging whether the image is qualified is used to compare the extracted features with the preset rule features to judge whether the image scanned by the logistics package on each distribution device is qualified.

[0042] A third aspect of the present invention provides a computer device, the computer device comprising a memory and at least one processor, the memory storing instructions;

[0043] The at least one processor calls the instructions in the memory so that the computer device executes the logistics package image recognition method as described above.

[0044] A fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is loaded by a processor to execute the logistics package image recognition method as described above.

[0045] The logistics package image recognition method of the present invention can effectively obtain, recognize, process and transmit goods images in the logistics distribution link, match the images taken by each distribution device with preset image features and judge whether the images meet business requirements, effectively monitor each link and equipment of logistics distribution, reduce logistics costs, and improve logistics management level and work efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flow chart of a logistics package image recognition method provided by an embodiment of the present invention;

[0047] Figure 2 A flow chart of obtaining a logistics package image in the logistics package image recognition method provided in an embodiment of the present invention;

[0048] Figure 3 A flow chart of determining the area to be identified in the logistics package image recognition method provided by an embodiment of the present invention;

[0049] Figure 4 A flowchart for constructing Kafka serialization in the logistics package image recognition method provided in an embodiment of the present invention;

[0050] Figure 5 A flow chart of extracting and classifying image features in a logistics package image recognition method provided by an embodiment of the present invention;

[0051] Figure 6 A flow chart for determining whether an image is qualified in the logistics package image recognition method provided by an embodiment of the present invention;

[0052] Figure 7 A schematic diagram of the structure of a logistics package image recognition device provided by an embodiment of the present invention;

[0053] Figure 8 Another structural schematic diagram of a logistics package image recognition device provided by an embodiment of the present invention;

[0054] Fig. 9 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0055] The present invention provides a method for identifying a logistics package image, the method comprising:

[0056] Obtaining pictures of logistics packages, placing the logistics packages into the distribution equipment and photographing the logistics packages, and obtaining pictures of the logistics packages passing through different distribution equipment;

[0057] Determine the area to be identified, and use image processing technology to determine the target identification area of ​​the image;

[0058] Build Kafka serialization, produce logistics package information into Kafka queue, and obtain logistics package information topic queue;

[0059] Extract and classify image features, use a deep image classification algorithm to extract features from the logistics package images, and classify them according to the extracted features;

[0060] To determine whether the image is qualified, the extracted features are compared with the preset rule features to determine whether the image scanned by the logistics package on each distribution device is qualified.

[0061] The logistics package image recognition method of the present invention can effectively obtain, recognize, process and transmit goods images in the logistics distribution link, match the images taken by each distribution device with preset image features and judge whether the images meet business requirements, effectively monitor each link and equipment of logistics distribution, reduce logistics costs, and improve logistics management level and work efficiency.

[0062] The terms "first", "second", "third", "fourth", etc. (if any) 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 described herein can be implemented in an order other than that illustrated or described herein. In addition, the terms "including" or "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.

[0063] For ease of understanding, the specific process of the embodiment of the present invention is described below. Figure 1 , an embodiment of the present invention provides a logistics package image recognition method, the method comprising:

[0064] 101. Obtain a picture of a logistics package, put the logistics package into a distribution device and take a picture of the logistics package, and obtain pictures of the logistics package passing through different distribution devices;

[0065] 102. Determine the area to be identified, and use image processing technology to determine the target identification area of ​​the image;

[0066] 103. Build Kafka serialization, produce logistics package information into Kafka queue, and obtain logistics package information topic queue;

[0067] 104. Extract and classify image features, use a deep image classification algorithm to extract features from the logistics package image, and classify according to the extracted features;

[0068] 105. Determine whether the image is qualified, compare the extracted features with the preset rule features, and determine whether the image scanned by the logistics package on each distribution device is qualified.

[0069] The logistics package image recognition method of the present invention can effectively obtain, recognize, process and transmit goods images in the logistics distribution link, match the images taken by each distribution device with preset image features and judge whether the images meet business requirements, effectively monitor each link and equipment of logistics distribution, reduce logistics costs, and improve logistics management level and work efficiency.

[0070] See also Figure 2The second embodiment of the logistics package image recognition method in the embodiment of the present invention is based on the above embodiment, wherein the logistics package image is obtained, the logistics package is placed into the distribution device and photographed, and the images of the logistics package passing through different distribution devices are obtained, including:

[0071] 201. Count the number of distribution equipment in the distribution center;

[0072] In some embodiments, the distribution equipment includes unloading, weighing, and loading equipment, which are used to scan the logistics packages and take photos or scan the logistics packages when they pass through the distribution equipment.

[0073] In some embodiments, the number of distribution devices in the distribution center is counted and classified in turn.

[0074] 202. Encode the distribution equipment in sequence according to the working line of the distribution center;

[0075] In some embodiments, according to the workflow of logistics packages to the distribution center, a corresponding assembly line is set up and the distribution equipment is encoded in sequence.

[0076] 203. The logistics package passes through the distribution devices in sequence according to the codes of the distribution devices. The distribution devices take photos of the logistics package to obtain pictures of the logistics package in each distribution device.

[0077] The embodiment of the present invention counts and codes each device in the distribution center and photographs the logistics packages when passing through each distribution device, so as to effectively and comprehensively obtain the pictures of the logistics packages photographed or scanned by each distribution device, thereby providing a strong foundation and guarantee for the subsequent logistics management and assessment of the business work of the distribution center.

[0078] See also Figure 3 In a third embodiment of the method for recognizing a logistics package image in an embodiment of the present invention, the step of determining the area to be recognized, using image processing technology to determine the image target recognition area, includes:

[0079] 301. Frame the image to obtain a framed area of ​​the image.

[0080] 302. Perform contour recognition processing on the framed area to obtain a contour recognition image of the picture;

[0081] 303. Filter the interference contour in the contour recognition image to obtain a filtered target contour image;

[0082] 304. Estimating the filtered target contour image using a target contour algorithm to estimate the shape of the target contour image;

[0083] 305. Preset a category of the target contour image and classify the shape of the target contour image.

[0084] The embodiment of the present invention can further obtain an accurate contour recognition image of the logistics package image based on the images taken by the above-mentioned distribution devices, and then determine the target recognition area of ​​the image.

[0085] See also Figure 4 , the fourth embodiment of the logistics package image recognition method in the embodiment of the present invention, the construction of Kafka serialization, the production of logistics package information to the Kafka queue, and the acquisition of the logistics package information topic queue include:

[0086] 401. Create a topic in Kafka for storing the logistics package image storage database;

[0087] 402. Predefine the message format of logistics package information;

[0088] In some embodiments, a topic named "cloudimage_express_info" may be created and the following message format may be predefined:

[0089] Ship ID: used to uniquely identify a logistics order;

[0090] Distribution code (centreCode): indicates the location of the current logistics node;

[0091] Scan time (scan Tm): records the timestamp of each scan;

[0092] Device Code: identifies the specific scanning device;

[0093] Platform Code: Indicates the platform where the cargo is located;

[0094] Image scene type identifier (clip): distinguishes different image types (such as thumbnail, full image or panorama).

[0095] 403. Serialize the message of the logistics package information to obtain a logistics package information topic queue;

[0096] In some embodiments, the data of the logistics package information message can be converted into JSON format.

[0097] 404. According to a preset custom partitioner, the logistics package information is divided into pieces.

[0098] The embodiment of the present invention adopts Kafka's message transmission mechanism to transmit these pictures to the queue for identification. The preset custom partitioner can evenly distribute the logistics package information in different partitions to achieve efficient storage and meet the requirements of real-time update of logistics package pictures.

[0099] See also Figure 5 In the fifth embodiment of the method for recognizing a logistics package image in the embodiment of the present invention, the step of extracting and classifying image features, extracting features of the logistics package image using a deep image classification algorithm, and classifying the features according to the extracted features, comprises:

[0100] 501. Determine the classification basis of the logistics package image;

[0101] 502. Using a deep image classification algorithm, extracting image features of the logistics package image from the logistics package image according to the determined classification basis;

[0102] 503. Classify the logistics package images according to the extracted image features.

[0103] The embodiment of the present invention can obtain better classification effect by extracting the image features of the image wrapped in the extraction flow and classifying it, and occupies less memory and does not require a large amount of computing resources.

[0104] See also Figure 6 In the sixth embodiment of the method for recognizing a logistics package image in the embodiment of the present invention, the step of judging whether the image is qualified, comparing the extracted features with the preset rule features, and judging whether the image scanned by the logistics package on each distribution device is qualified includes:

[0105] 601. Preset rule features and feature thresholds;

[0106] 602. Matching feature points of the image features of the logistics package image with the rule features;

[0107] 603. If the feature matching degree is greater than the feature threshold, the image scanned on the distribution device is judged to be unqualified; if the feature matching degree is less than or equal to the feature threshold, the image scanned on the distribution device is judged to be qualified.

[0108] In some embodiments, if it is determined that the image scanned on the distribution device is unqualified, the unqualified image is marked and the marking result is fed back to the distribution device.

[0109] In some embodiments, if the image scanned on the distribution device is determined to be qualified, the qualified image is screened out and displayed.

[0110] The embodiment of the present invention can match the extracted features with the preset rule features, and compare the feature matching degree with the feature threshold to determine whether the picture taken or scanned by the distribution device is qualified. It can effectively identify the working process and results of each distribution device, and can effectively perform statistics and feedback on the distribution center.

[0111] The above describes the logistics package image recognition method in the embodiment of the present invention. The following describes the logistics package image recognition device in the embodiment of the present invention. Figure 7 , an embodiment of the logistics package image recognition device in the embodiment of the present invention includes:

[0112] The module 701 for obtaining pictures of logistics packages is used to put the logistics packages into the distribution equipment and take pictures of the logistics packages to obtain pictures of the logistics packages passing through different distribution equipments;

[0113] The module 702 for determining the area to be identified is used to determine the target identification area of ​​the image by using image processing technology;

[0114] Construct a Kafka serialization module 703 to produce the logistics package information into the Kafka queue to obtain the logistics package information topic queue;

[0115] The image feature extraction and classification module 704 is used to extract features from the logistics package image using a deep image classification algorithm and classify the image according to the extracted features;

[0116] The module 705 for judging whether the image is qualified is used to compare the extracted features with the preset rule features to judge whether the image of the logistics package scanned on each distribution device is qualified.

[0117] The logistics package image recognition device of the present invention can effectively obtain, recognize, process and transmit cargo images in the logistics distribution link, match the images taken by each distribution device with preset image features and determine whether the images meet business requirements, effectively monitor each link and equipment of logistics distribution, reduce logistics costs, and improve logistics management level and work efficiency.

[0118] See also Figure 8 Another embodiment of the logistics package image recognition device in the embodiment of the present invention includes:

[0119] The module 701 for obtaining pictures of logistics packages is used to put the logistics packages into the distribution equipment and take pictures of the logistics packages to obtain pictures of the logistics packages passing through different distribution equipments;

[0120] In some embodiments, the module 701 for obtaining a logistics package image includes:

[0121] The statistical unit 7011 is used to count the number of distribution devices in the distribution center;

[0122] In some embodiments, the distribution equipment includes unloading, weighing, and loading equipment, which are used to scan the logistics packages and take photos or scan the logistics packages when they pass through the distribution equipment.

[0123] In some embodiments, the number of distribution devices in the distribution center is counted and classified in turn.

[0124] The encoding unit 7012 is used to encode the distribution equipment in sequence according to the working pipeline of the distribution center;

[0125] In some embodiments, according to the workflow of logistics packages to the distribution center, a corresponding assembly line is set up and the distribution equipment is encoded in sequence.

[0126] The photographing unit 7013 is used for the logistics package to pass through the distribution devices in sequence according to the codes of the distribution devices, and the distribution devices photograph the logistics package to obtain pictures of the logistics package in each distribution device.

[0127] The embodiment of the present invention counts and codes each device in the distribution center and photographs the logistics packages when passing through each distribution device, so as to effectively and comprehensively obtain the pictures of the logistics packages photographed or scanned by each distribution device, thereby providing a strong foundation and guarantee for the subsequent logistics management and assessment of the business work of the distribution center.

[0128] The module 702 for determining the area to be identified is used to determine the target identification area of ​​the image by using image processing technology;

[0129] In some embodiments, the module 702 for determining the area to be identified includes:

[0130] A frame selection unit 7021, configured to frame the image to obtain a frame selection area of ​​the image;

[0131] The contour recognition unit 7022 is used to perform contour recognition processing on the framed area to obtain a contour recognition image of the picture;

[0132] A filtering unit 7023, used to filter the interference contours in the contour recognition image to obtain a filtered target contour image;

[0133] An estimation unit 7024 is used to estimate the filtered target contour image using a target contour algorithm to estimate the shape of the target contour image;

[0134] The preset classification unit 7025 is used to preset the category of the target contour image and classify the shape of the target contour image.

[0135] The embodiment of the present invention can further obtain an accurate contour recognition image of the logistics package image based on the images taken by the above-mentioned distribution devices, and then determine the target recognition area of ​​the image.

[0136] Construct a Kafka serialization module 703 to produce the logistics package information into the Kafka queue to obtain the logistics package information topic queue;

[0137] In some embodiments, the construction of the Kafka serialization module 703 includes:

[0138] A creation unit 7031 is used to create a topic in Kafka for storing the logistics package image storage database;

[0139] A pre-defined unit 7032 is used to pre-define the message format of the logistics package information;

[0140] In some embodiments, a topic named "cloudimage_express_info" may be created and the following message format may be predefined:

[0141] Ship ID: used to uniquely identify a logistics order;

[0142] Distribution code (centreCode): indicates the location of the current logistics node;

[0143] Scan time (scan Tm): records the timestamp of each scan;

[0144] Device Code: identifies the specific scanning device;

[0145] Platform Code: Indicates the platform where the cargo is located;

[0146] Image scene type identifier (clip): distinguishes different image types (such as thumbnail, full image or panorama).

[0147] A serialization unit 7033 is used to serialize the message of the logistics package information to obtain a logistics package information topic queue;

[0148] In some embodiments, the data of the logistics package information message can be converted into JSON format.

[0149] The partition unit 7034 is used to divide the logistics package information into slices according to a preset custom partitioner.

[0150] The embodiment of the present invention adopts Kafka's message transmission mechanism to transmit these pictures to the queue for identification. The preset custom partitioner can evenly distribute the logistics package information in different partitions to achieve efficient storage and meet the requirements of real-time update of logistics package pictures.

[0151] The image feature extraction and classification module 704 is used to extract features from the logistics package image using a deep image classification algorithm and classify the image according to the extracted features;

[0152] In some embodiments, the image feature extraction and classification module 704 includes:

[0153] A determination unit 7041, used to determine the classification basis of the logistics package image;

[0154] An extraction unit 7042 is used to extract image features of the logistics package image from the logistics package image according to the determined classification basis by using a deep image classification algorithm;

[0155] The classification unit 7043 is used to classify the logistics package image according to the extracted image features.

[0156] The embodiment of the present invention can obtain better classification effect by extracting the image features of the image wrapped in the extraction flow and classifying it, and occupies less memory and does not require a large amount of computing resources.

[0157] The module 705 for judging whether the image is qualified is used to compare the extracted features with the preset rule features to judge whether the image of the logistics package scanned on each distribution device is qualified.

[0158] In some embodiments, the module 705 for determining whether the image is qualified includes:

[0159] A preset feature unit 7051 is used to preset rule features and feature thresholds;

[0160] A matching unit 7052, used for matching the image features of the logistics package image with the rule features for feature point matching;

[0161] The judgment unit 7053 is used to judge that the image scanned on the distribution device is unqualified if the feature matching degree is greater than the feature threshold; if the feature matching degree is less than or equal to the feature threshold, judge that the image scanned on the distribution device is qualified.

[0162] In some embodiments, if it is determined that the image scanned on the distribution device is unqualified, the unqualified image is marked and the marking result is fed back to the distribution device.

[0163] In some embodiments, if the image scanned on the distribution device is determined to be qualified, the qualified image is screened out and displayed.

[0164] The embodiment of the present invention can match the extracted features with the preset rule features, and compare the feature matching degree with the feature threshold to determine whether the picture taken or scanned by the distribution device is qualified. It can effectively identify the working process and results of each distribution device, and can effectively perform statistics and feedback on the distribution center.

[0165] above Figure 7 and Figure 8 The logistics package image recognition method in the embodiment of the present invention is described in detail from the perspective of modular functional entities, and the electronic device in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0166] Fig. 9 800 may have relatively large differences due to different configurations or performances, and may include one or more processors (central processing units, CPU) 810 (for example, one or more processors) and a memory 820, and one or more storage media 830 (for example, one or more mass storage devices) storing application programs 833 or data 832. Among them, the memory 820 and the storage medium 830 may be temporary storage or permanent storage. The program stored in the storage medium 830 may include one or more upper modules (not shown in the figure) for vehicle dispatching at a distribution center, and each module may include a series of instruction operations in the electronic device 800. Furthermore, the processor 810 may be configured to communicate with the storage medium 830 to execute a series of instruction operations in the storage medium 830 on the electronic device 800.

[0167] The electronic device 800 may also include one or more power supplies 840, one or more wired or wireless network interfaces 850, one or more input and output interfaces 850, and / or one or more operating systems 831, such as Windows Serve, Mac OS X, Unix, Linux, FreeBSD, etc. It will be appreciated by those skilled in the art that Fig. 9 The structure of the electronic device shown does not constitute a limitation on the electronic device, and may include more or less components than shown in the figure, or combine some components, or arrange the components differently.

[0168] The present invention also provides a computer-readable storage medium, which may be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. Instructions are stored in the computer-readable storage medium. When the instructions are executed on a computer, the computer executes the steps of the distribution center vehicle dispatching method.

[0169] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described system, device, or unit can refer to the corresponding process in the aforementioned method embodiment and will not be repeated here.

[0170] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk and other media that can store program codes.

[0171] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A logistics package image recognition method, characterized in that: The method comprises: Obtaining pictures of logistics packages, placing the logistics packages into the distribution equipment and photographing the logistics packages, and obtaining pictures of the logistics packages passing through different distribution equipment; Determine the area to be identified, and use image processing technology to determine the target identification area of ​​the image; Build Kafka serialization, produce logistics package information into Kafka queue, and obtain logistics package information topic queue; Extract and classify image features, use a deep image classification algorithm to extract features from the logistics package images, and classify them according to the extracted features; To determine whether the image is qualified, the extracted features are compared with the preset rule features to determine whether the image scanned by the logistics package on each distribution device is qualified.

2. The logistics package image recognition method according to claim 1 is characterized in that: The obtaining of the logistics package image, placing the logistics package into the distribution equipment and photographing the logistics package, and obtaining the images of the logistics package passing through different distribution equipment, includes: Count the number of distribution equipment in the distribution center; According to the working pipeline of the distribution center, the distribution equipment is encoded in sequence; The logistics package passes through the distribution devices in sequence according to the codes of the distribution devices, and the distribution devices take photos of the logistics package to obtain pictures of the logistics package in each distribution device.

3. The logistics package image recognition method according to claim 2 is characterized in that: The step of determining the area to be identified, using image processing technology to determine the image target identification area, includes: Performing a frame selection on the image to obtain a frame selection area of ​​the image; Performing contour recognition processing on the framed area to obtain a contour recognition image of the picture; Filtering the interference contours in the contour recognition image to obtain a filtered target contour image; Using a target contour algorithm to estimate the filtered target contour image, and estimating the shape of the target contour image; The category of the target contour image is preset, and the shape of the target contour image is classified.

4. The logistics package image recognition method according to claim 3 is characterized in that: The construction of Kafka serialization produces the logistics package information into the Kafka queue to obtain the logistics package information topic queue, including: Create a topic in Kafka for storing the logistics package image storage database; Predefine the message format of logistics package information; Serializing the message of the logistics package information to obtain a logistics package information topic queue; The logistics package information is divided into pieces according to a preset custom partitioner.

5. The logistics package image recognition method according to claim 2 is characterized in that: The extracting and classifying the image features, using a deep image classification algorithm to extract features from the logistics package images, and classifying according to the extracted features, includes: Determine the classification basis of the logistics package image; Using a deep image classification algorithm, extracting image features of the logistics package image from the logistics package image according to the determined classification basis; The logistics package images are classified according to the extracted image features.

6. The method for identifying logistics package images according to claim 5, characterized in that: The process of judging whether the image is qualified, comparing the extracted features with the preset rule features, and judging whether the image scanned by the logistics package on each distribution device is qualified includes: Preset rule features and feature thresholds; Matching the image features of the logistics package image with the rule features, and if the feature matching degree is greater than the feature threshold, judging that the image scanned on the distribution device is unqualified; If the feature matching degree is less than or equal to the feature threshold, it is determined that the image scanned on the distribution device is qualified.

7. The method for identifying logistics package images according to claim 6, characterized in that: The image features of the logistics package image are matched with the rule features for feature points, Including: using scale-invariant features, accelerated robust features, color features or structural edge features to match feature points of two images to be matched.

8. A logistics package image recognition device, characterized in that: include: A module for obtaining pictures of logistics packages is used to place logistics packages into distribution equipment and take pictures of the logistics packages to obtain pictures of the logistics packages passing through different distribution equipment; A module for determining an area to be identified is used to determine the target identification area of ​​the image by using image processing technology; Build a Kafka serialization module to produce logistics package information into the Kafka queue and obtain the logistics package information topic queue; A module for extracting and classifying image features, which is used to extract features from the logistics package images using a deep image classification algorithm, and classify the images according to the extracted features; The module for judging whether the image is qualified is used to compare the extracted features with the preset rule features to judge whether the image scanned by the logistics package on each distribution device is qualified.

9. A computer device, characterized in that: The computer device includes a memory and at least one processor, wherein instructions are stored in the memory; The at least one processor calls the instructions in the memory so that the computer device executes the logistics package image recognition method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that: A computer program is stored thereon, and the computer program is loaded by a processor to execute the logistics package image recognition method described in any one of claims 1-7.

Images