Container number recognition compensation system, individual identification and storage method, and device

Abstract

The application provides a container number identification compensation system, individual identification and storage method and equipment. The container number identification compensation system comprises an individual feature library part and an individual feature service part; the individual feature library part comprises a key-value database and a vector database; and the individual feature service part comprises an individual feature Web service module, an individual feature extraction service module, a vector service module and a key-value service module. The application can replace the original manual identification operation when the OCR container number identification fails, thereby reducing the work burden of the operator; the application has the advantages of high operation performance and second-level execution efficiency, can realize real-time automatic compensation of the OCR container number identification, improve the operation efficiency of the related business and reduce the operation cost; and the application can identify the individual features of the container, can improve the overall identification rate to between 99.0% and 99.9% and fully play the compensation role of the OCR container number identification technology.

Description

Technical Field

[0001] This invention relates to the field of image recognition technology, specifically to the field of container recognition technology, and more particularly to a container number recognition and compensation system, an individual recognition and storage method, and a device. Background Technology

[0002] Container identification technology is a crucial technology in modern logistics. It utilizes advanced image and character recognition techniques to achieve rapid container identification and is widely applied in scenarios such as automatic data entry for container operations and automatic confirmation of operational processes. This technology significantly improves logistics efficiency, reduces operating costs, and possesses broad application prospects and development potential.

[0003] The core of container identification technology is the identification of individual containers. The container number is the unique identifier of each container; by recognizing the printed container number, individual containers can be identified. Alternatively, identifying individual containers can also be achieved by extracting, labeling, storing, searching, and matching container features. The former is called the container number identification method, and the latter is called the container feature identification method.

[0004] The most common method for identifying container numbers is OCR (Optical Character Recognition). This technology captures images of the container surface using a high-definition camera and processes the images using advanced technologies such as deep learning to automatically identify and record the container number. However, this technology is highly dependent on image quality, and real-world scenarios are typically outdoors, where image quality is greatly affected by interference factors. For example, improper camera placement can cause lens distortion; changes in sunlight and shadows can cause glare and uneven lighting; and special weather conditions (heavy fog, heavy rain, snow, etc.) can cause blurred vision and obstruction. In particular, defects in the container itself (rust, dirt, peeling paint, etc.) can lead to blurred or damaged container numbers that are even difficult to identify with the naked eye. Blurred and damaged container numbers essentially limit the upper limit of OCR container number recognition rate; the older the container, the more difficult it is to directly identify the container number using OCR. There are two situations where OCR cannot recognize the container number: either the characters are completely unrecognizable, or only some characters are recognized. Therefore, when conditions permit, a compensation mechanism can be activated for these two situations to output an accurate container number as much as possible.

[0005] Traditional container number recognition compensation mechanisms and methods use OCR-recognized characters to perform fuzzy matching with a list of alternative containers provided by an external system, inferring the container number with the closest match. However, this fuzzy matching method is not ideal in actual production environments, as the error rate of characters recognized by OCR is inherently higher than that of normally recognized characters, resulting in insufficient overall confidence.

[0006] Because shipping companies set the service life of containers to be 10-20 years before they are retired, as time goes by, the containers are exposed to the sun and rain, and the number of loading and unloading operations increases, making the containers increasingly old. Once the container number is blurred or damaged, it is difficult to directly identify the container number through OCR, and the recognition rate is only between 85% and 90%. Summary of the Invention

[0007] The purpose of this invention is to solve the problem that is common in container OCR container number recognition technology: due to rust, dirt, paint peeling and other reasons on the container surface, the container becomes increasingly old, and once the container number is blurred or damaged, it is difficult to directly identify the container number through OCR.

[0008] To overcome the low recognition rate of existing technologies that directly identify container numbers using OCR, this invention provides a container number recognition compensation system, an individual identification and storage method, and equipment. The targeted compensation scheme proposed in this invention is highly applicable and effective; it can replace manual identification when OCR fails to recognize the container number, reducing the workload of operators; furthermore, it boasts high computing performance and second-level execution efficiency, enabling real-time automatic compensation for OCR container number recognition, improving operational efficiency and reducing operating costs; by identifying individual container features, the overall recognition rate can be increased to between 99.0% and 99.9%, effectively fulfilling the compensation function of OCR container number recognition.

[0009] The present invention solves the above-mentioned technical problems through the following technical solutions:

[0010] This invention provides a container number identification and compensation system, which includes an individual feature database and an individual feature service.

[0011] The individual feature database includes: a vector database for storing individual feature vectors of different containers; and a key-value database for storing basic container information corresponding to the individual feature vectors of each container, wherein the basic container information includes the container number.

[0012] The individual feature service component includes: an individual feature web service module, used to respond to service requests initiated by an external system; an individual feature extraction service module, used to respond to a first service request initiated by the individual feature web service module; the first service request includes: converting container exterior photos contained in the incoming data from the external system into individual feature vectors to be processed; a vector service module, used to respond to a second service request initiated by the individual feature web service module, the second service request including: searching or storing the vector based on the individual feature vectors to be processed in conjunction with the vector database, and obtaining a vector ID; and a key-value service module, used to respond to a third service request initiated by the individual feature web service module, the third service request including: matching or storing the vector ID in conjunction with the key-value database, and obtaining the result.

[0013] In this invention, the vector database is a dedicated database for storing and querying vectors. The stored vectors are derived from the vectorization of text, speech, images, videos, etc. Compared to traditional databases, the vector database not only performs basic CRUD (add, read, query, update, delete) operations, but also enables faster vector similarity searches. Vector similarity search compares the query vector with the database to find the most similar and confident result vector. Vector similarity search uses Approximate Nearest Neighbor Search (ANN) algorithms to accelerate the search process. Common ANN algorithms include Locality Sensitive Hash (LSH), graph embedding methods (such as HNSW), and Vector Quantization (VQ), which optimize the search process in different ways and play an important role in the field of image recognition.

[0014] In this invention, the key-value database, also known as a key-value store, is a non-relational database (also called a NoSQL database). Each unique identifier is stored as a key with an associated value; this data pairing is called a "key-value" pair. Within a key-value pair, the key must be unique, and the value associated with the key can be accessed through the key. A single-node key-value database offers performance advantages in reading and writing data compared to more general-purpose relational databases, and when combined with a vector database, it can be used to search for labeled data bound to feature vectors, such as bin numbers.

[0015] In some implementations, the vector service module includes a vector search service module; the key-value service module includes a key-value matching service module; the vector search service module is used to search the vector database for the result vector that is most similar to the feature vector to be processed and has confidence, and obtain a first vector ID recorded in the vector database; the key-value matching service module is used to match the first vector ID from the key-value database to obtain basic container information.

[0016] In some implementations, the vector service module includes a vector storage service module; the key-value service module includes a key-value storage service module; the vector storage service module is used to store the feature vector to be processed into the vector database to obtain a second vector ID recorded in the vector database; the key-value storage service module is used to combine the second vector ID with the container basic information contained in the incoming data from the external system into a key-value pair and store it in the key-value database.

[0017] In some implementations, the individual characteristic service component is deployed in a distributed deployment or a microservice architecture; it provides public services externally and operates an individual characteristic database internally.

[0018] In some implementations, the individual characteristics web service module is deployed in the DMZ zone of the individual characteristics service portion.

[0019] In some implementations, the individual feature extraction service module is deployed on the intranet of the individual feature service section.

[0020] In this invention, the Value in the key-value database can store not only the container number, but also more information, such as the container type, load capacity, and other basic information about the container.

[0021] In some implementations, the key-value database includes an open-source key-value database.

[0022] In a specific implementation, the key-value database includes a Redis key-value database.

[0023] In some implementations, the key-value database is deployed in a distributed manner.

[0024] In some implementations, the vector database includes an open-source vector database.

[0025] In a specific implementation, the vector database includes the Milvus vector database.

[0026] In one embodiment, the Milvus vector database includes the HNSW algorithm; HNSW stands for Hierarchical Navigable Small World graphs, a graph-based approximate nearest neighbor search algorithm widely used in industry. It is well-suited for applications requiring high-precision ANN search (such as face recognition and semantic search) and can meet the confidence requirements for identifying individual container features in this technical solution. It constructs a multi-layered graph structure, with the number of nodes and edges gradually decreasing in each layer. In this way, during the top-down search process, the globally possible nearest node can be found first through long links, and then a local search is performed in the lower layers, thus achieving fast and accurate approximate nearest neighbor search.

[0027] In some implementations, the vector database is deployed in a distributed manner.

[0028] In a specific implementation, the individual feature extraction service module includes the SIFT algorithm, which is used to extract individual feature points of the container from the container surface photos contained in the input data, and generate the individual feature vector to be processed.

[0029] The SIFT (Scale-Invariant Feature Transform) algorithm is a classic computer vision algorithm used to extract feature points from images, enabling the identification of target objects in complex scenes. The SIFT algorithm aims to represent information in an image as features, ensuring that these features remain invariant under image scaling, rotation, brightness changes, and even to some extent, affine transformations. Essentially, SIFT extracts target feature points by statistically analyzing and encoding gradient information within the feature point's neighborhood, ultimately combining it into a descriptor vector to characterize the direction and relative intensity distribution of local image features. The SIFT algorithm is characterized by strong robustness and excellent anti-interference capabilities.

[0030] In this invention, the container's outline, dimensions, paintwork, and unique environmental traces left by natural erosion and loading / unloading operations are used as individual container features for extraction, marking, and storage in a container feature database. When the OCR cannot recognize the container number, recognition compensation is performed by searching and matching in the container feature database to obtain the correctly marked container number.

[0031] The present invention also provides a container individual identification method, which uses the container number identification compensation system described above to provide individual feature Web services and respond to public service requests initiated by external systems;

[0032] The container individual identification method includes the following steps:

[0033] S1. Receive incoming data from the external system; and determine the program to be executed based on the incoming data;

[0034] S2. The individual feature extraction service module converts the container exterior photos contained in the input data into individual feature vectors to be processed and performs a search process. The individual feature web service module calls the vector search service of the vector service module and the key-value matching service of the key-value service module to feed back the searched information to the external system based on the individual feature vectors to be processed.

[0035] In some implementations, the container individual identification method satisfies either condition I or condition II:

[0036] Condition I: S1, Receive incoming data from the external system, the incoming data does not contain photos of the container's exterior; The individual feature Web service module sends a feedback to the external system indicating that the call failed;

[0037] Condition II: S1, Receive incoming data from the external system, the incoming data containing only photos of the container's exterior, and determine to execute the search process;

[0038] S2. First, the individual feature Web service module calls the individual feature extraction service and uses the SIFT algorithm to extract individual feature points from the container surface photo to obtain the individual feature vector to be processed.

[0039] Secondly, the individual feature web service module calls the vector search service; using the individual feature vector to be processed as the query vector, it searches the vector database for the result vector that is most similar to the query vector and has confidence, and obtains the first vector ID recorded in the vector database; if the search fails, the individual feature web service module sends a feedback to the external system, indicating that no result was found.

[0040] Then, the individual feature web service module calls the key-value matching service; using the first vector ID as the search key value, it matches the container basic information from the key-value database, the container basic information including the container number; if the match is successful, the individual feature web service module feeds back the container basic information to the external system; if the match fails, the individual feature web service module feeds back to the external system, indicating that no results were found.

[0041] In this invention, the container photos include photos of five sides of the container, such as: left, right, front, back, and top.

[0042] The present invention also provides a method for individual container storage, which uses the container number identification and compensation system described above to provide individual characteristic Web services and respond to public service requests initiated by external systems;

[0043] The method for storing individual containers includes the following steps:

[0044] S1. Receive incoming data from the external system; and determine the program to be executed based on the incoming data;

[0045] S2. The individual feature extraction service module converts the container exterior photos contained in the input data into individual feature vectors to be processed, and executes a storage procedure. The individual feature web service module calls the vector storage service of the vector service module and the key-value storage service of the key-value service module to combine the individual feature vectors to be processed with the basic container information contained in the input data into key-value pairs and store them in the key-value database. The basic container information includes the container number.

[0046] In some implementations, the container individual storage method satisfies either condition I or condition II:

[0047] Condition I: S1, Receive incoming data from the external system, the incoming data does not contain photos of the container's exterior; The individual feature Web service module sends a feedback to the external system indicating that the call failed;

[0048] Condition II: S1, Receive incoming data from the external system, the incoming data includes a photo of the container's exterior and basic container information, the basic container information including the container number, and determine to execute the stored procedure;

[0049] S2. First, the individual feature Web service module calls the individual feature extraction service and uses the SIFT algorithm to extract individual feature points from the container surface photo to obtain the individual feature vector to be processed.

[0050] Secondly, the individual feature web service module calls the vector storage service; stores the individual feature vector to be processed in the vector database, and obtains the second vector ID of the vector database record;

[0051] Then, the individual feature web service module calls the key-value storage service; combines the second vector ID with the container basic information into a key-value pair and stores it in the key-value database; when there is the same key value in the key-value database, the old record is overwritten; and the individual feature web service module feeds back to the external system to confirm that the container's individual features have been stored.

[0052] In this invention, the container photos include photos of five sides of the container, such as: left, right, front, back, and top.

[0053] The present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the container individual identification method or the container individual storage method as described above.

[0054] The beneficial effects of this invention are as follows:

[0055] 1. The technical solution provided by this invention can be used in combination with OCR box number recognition technology, and the two complement each other; it can replace the original manual recognition work when OCR box number cannot be recognized, reducing the workload of operators.

[0056] 2. This invention has high computing performance and second-level execution efficiency, and can automatically compensate for OCR box number recognition in real time, thereby improving the operational efficiency of related businesses and reducing operating costs.

[0057] 3. For blurry or damaged container numbers, the OCR container number recognition rate is between 85% and 90%. After being supplemented by the container individual feature recognition method of the present invention, the overall recognition rate of such containers is significantly improved to between 99.0% and 99.9%.

[0058] 4. This invention can not only compensate for OCR container number recognition technology, but also be applied to business scenarios such as automatic entry of container operation data and automatic confirmation of operation links. This includes, but is not limited to, intelligent cargo handling, intelligent gates, yard loading and unloading, customs inspection, container unpacking and packing, empty container damage inspection, and safe production in container shipping, rail transport, land transport and warehousing, as well as other business and operation links that require individual container identification and individual information collection (such as container number recognition and container number information collection). Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the container number identification and compensation system according to Embodiment 1 of the present invention;

[0060] Figure 2 This is a flowchart illustrating the container individual identification method and individual storage method according to Embodiment 2 of the present invention;

[0061] Figure 3 This is a schematic diagram of the search process of the container individual identification method according to Embodiment 2 of the present invention;

[0062] Figure 4 This is a schematic diagram of the storage process of the container individual storage method according to Embodiment 2 of the present invention. Detailed Implementation

[0063] The identification system and method of the present invention are a container feature identification system and method for distinguishing individual containers. It utilizes the container's outline, size, and paintwork, as well as unique traces left by environmental factors such as natural erosion and loading / unloading operations, as individual container features for extraction, marking, storage, searching, and matching.

[0064] Specifically, SIFT or other free algorithms are used to extract individual feature points from five images of the container (left, right, front, back, and top) and generate feature descriptor vectors (i.e., feature vectors, referred to in this invention as container "individual feature vectors"). These vector vectors are stored in the Milvus open-source vector database. The vector ID obtained during storage (as the key value) and the manually labeled container number (as the value value, which may contain other basic container information) are stored together in the Redis open-source key-value database. This combination of vector database and key-value database is called the container "individual feature library".

[0065] In this way, if the OCR cannot recognize the container number and it cannot be found in the individual feature database, this technical solution can be activated to store the container's individual feature vector and the manually identified container number together in this individual feature database. Then, the next time the OCR cannot recognize the container number, this technical solution can be activated to search for the container number in this individual feature database.

[0066] The specific method involves extracting individual feature vectors from the input container photos and using them as query vectors to perform vector similarity searches in a vector database, identifying similar records with confidence. Then, the result vector ID (Key value) of the searched record is matched against a key-value database to determine the corresponding container number (Value value), thus compensating for OCR container number recognition issues. Under normal circumstances, as long as the container is not old and the container number is clear, OCR can generally recognize it. In this case, the compensation mechanism is not necessary, but the individual feature vectors and the recognized container numbers can be saved to an individual feature database for future use if OCR fails to recognize the container.

[0067] The present invention is further illustrated below by way of embodiments, but the invention is not limited to the scope of the embodiments described herein. Experimental methods in the following embodiments that do not specify specific conditions were performed according to conventional methods and conditions, or as selected according to the product instructions.

[0068] Example 1

[0069] This embodiment discloses a container number identification and compensation system; Figure 1 This is a schematic diagram of the container number identification and compensation system in this embodiment.

[0070] The container number identification and compensation system consists of an individual feature service component and an individual feature database component.

[0071] The individual feature database consists of a key-value database and a vector database, providing persistent services for container "individual feature vectors" and "container number and other basic container information" for individual feature services; among them,

[0072] Key-value database: It adopts the Redis open-source key-value database and adopts a distributed deployment; it provides a vector database to record "vector ID" and "container number and other basic container information" key-value pairs for storage and matching for individual feature services.

[0073] Vector Database: Utilizing the Milvus open-source vector database, a distributed deployment is adopted. It provides storage and search functionality for containerized "individual feature vectors" for individual feature services, resulting in a record "vector ID". The search function employs the database's HNSW algorithm + L2 (Euclidean distance) index, with the confidence threshold adjusted to a suitable value based on the application scenario, such as between 0.85 and 0.92.

[0074] The individual feature service component consists of an individual feature web service module, an individual feature extraction service module, a key-value service module, and a vector service module. The key-value service module further includes a key-value storage service module and a key-value matching service module, while the vector service module includes a vector storage service module and a vector search service module.

[0075] Individual Characteristics Web Service Module: Responds to public service requests initiated by external systems. The data structure of the data passed in by the external system includes "five photos of the container's front, back, and top sides" and "container number and other basic container information," and must include container photos. If only container photos are passed in, a search process is executed; otherwise, a stored procedure is executed. In the former case, the caller is informed of the found "container number and other basic container information," or a message indicating that no results were found; in the latter case, the caller is informed that the individual characteristics have been stored.

[0076] Individual Feature Extraction Service Module: Responds to service requests initiated by internal system modules, receiving "photos of five container surfaces (left, right, front, back, and top)". It uses the SIFT algorithm to extract individual feature points from these images, generating an "individual feature vector" for each container, which is then returned to the caller.

[0077] Key-value storage service module: Responds to service requests initiated by internal modules of the system: The incoming data consists of vector database records "vector ID" and "container number and other basic container information", which are combined into key-value pairs and stored in the key-value database. If there is a record with the same key value, the old record will be overwritten.

[0078] Key-value matching service module: Responds to service requests initiated by internal modules of the system: The input data is the vector database record "vector ID", which is used as the search key value. It matches the "container number and other basic container information" from the key-value database and returns it to the caller.

[0079] Vector storage service module: Responds to service requests initiated by internal modules of the system: The incoming data is the "individual feature vector" of the container, which is stored in the vector database, the "vector ID" of the vector database record is obtained, and returned to the caller.

[0080] Vector Search Service Module: Responds to service requests initiated by internal modules of the system: The input data is the "individual feature vector" of the container, which is used as the query vector. It searches the vector database for the result vector that is most similar to the query vector and has confidence, obtains the "vector ID" record in the vector database and returns it to the caller, or indicates that no search can be found.

[0081] Example 2

[0082] This embodiment discloses a method for identifying and storing individual containers, which uses the container number identification and compensation system of Embodiment 1; Figure 2 This is a flowchart illustrating the container individual identification method and individual storage method of this embodiment.

[0083] The container individual identification and individual storage method includes the following steps:

[0084] S1. Provide web services based on individual characteristics and respond to public service requests initiated by external systems;

[0085] S1.1 Receive incoming data from an external system. This incoming data includes "photos of the five sides of the container (left, right, front, back, and top)" and "container number and other basic container information".

[0086] S1.2 The data input from the external system must include five photos of the container front; otherwise, the individual feature web service module will report back to the external system that the call failed. The provided container front photos are "five photos of the container front, back, left, right, front, and top".

[0087] S1.3 If the data input from the external system only contains box surface photos, then execute the search process. Figure 3 This is a schematic diagram of the search process of the container individual identification method in this embodiment. The specific search process is as follows:

[0088] S1.3.1 The individual feature Web service module calls the individual feature extraction service, inputs "five container surface photos (left, right, front, back, and top)," and uses the SIFT algorithm to extract the container's individual feature points to obtain the container's "individual feature vector."

[0089] S1.3.2 The Individual Feature Web Service module calls the Vector Search Service, passing in the container "Individual Feature Vector" as the query vector. It searches the vector database for the result vector that is most similar to the query vector and has confidence, and obtains the vector database record "Vector ID". If no result is found, the Individual Feature Web Service will report back to the external system, indicating that no result was found.

[0090] S1.3.3 The Individual Feature Web Service module calls the key-value matching service, passes in the "vector ID" record in the vector database as the search key value, and obtains the "container number and other basic container information" from the key-value database. If no match is found, the Individual Feature Web Service will report back to the external system, indicating that no search was found.

[0091] S1.3.4 The Individual Characteristics Web Service Module will send the retrieved "container number and other basic container information" back to the external system;

[0092] S1.4 If the data input from the external system includes "container number and other basic container information" in addition to "container front view photo", then execute the stored procedure. Figure 4 This is a schematic diagram of the storage process of the container individual storage method in this embodiment. The specific storage process is as follows:

[0093] S1.4.1 The individual feature Web service module calls the individual feature extraction service, inputs "five container surface photos (left, right, front, back, and top)," and uses the SIFT algorithm to extract the container's individual feature points to obtain the container's "individual feature vector."

[0094] S1.4.2 The individual feature web service module calls the vector storage service, passes in the container "individual feature vector" to store in the vector database, and obtains the vector database record "vector ID";

[0095] S1.4.3 The individual characteristics web service module calls the key-value storage service, passes in the vector database record "vector ID" and "container number and other basic container information", combines them into a Key-Value pair, and stores it in the key-value database. If there is a record with the same key value, the old record will be overwritten.

[0096] S1.4.4 The individual characteristics web service module feeds back to the external system to confirm that the individual characteristics have been stored.

[0097] The above steps S1.2, S1.3, and S1.4 are different procedures performed based on the input data, which can play a compensating role in OCR box number recognition technology.

[0098] This embodiment can replace the original manual identification work, reducing the workload of operators; it also has high computing performance and second-level execution efficiency, and can complete the automatic compensation of OCR container number recognition in real time, improving the operational efficiency of related businesses and reducing operating costs; furthermore, the method of this invention can significantly improve the overall recognition rate of such containers, reaching between 99.0% and 99.9%.

[0099] Example 3

[0100] This embodiment discloses an electronic device including a memory, a processor, and a computer program stored in the memory and executed on the processor. When the processor executes the computer program, it implements the container individual identification method and individual storage method provided in Embodiment 2 above. This electronic device is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments disclosed herein.

[0101] Electronic devices can take the form of general-purpose computing devices, such as server devices. Components of an electronic device may include, but are not limited to: at least one processor, at least one memory, and buses connecting different system components (including memory and processor).

[0102] The bus includes a data bus, an address bus, and a control bus.

[0103] The memory may include volatile memory, such as random access memory (RAM) and / or cache memory, and may further include read-only memory (ROM).

[0104] The memory may also include program tools (or utilities) having a set (at least one) of program modules, including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include an implementation of a network environment.

[0105] The processor executes various functional applications and data processing by running computer programs stored in memory, such as the container individual identification method and individual storage method provided in Embodiment 2 above.

[0106] Electronic devices can also communicate with one or more external devices (such as keyboards, pointing devices, etc.). This communication can be achieved through input / output (I / O) interfaces. Furthermore, electronic devices can communicate with one or more networks (such as local area networks (LANs), wide area networks (WANs), and / or public networks, such as the Internet) via network adapters. As shown in the figure, the network adapter communicates with other modules of the electronic device via a bus. It should be understood that other hardware and / or software modules can be used in conjunction with the electronic device, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID (disk array) systems, tape drives, and data backup storage systems.

[0107] It should be noted that although several units / modules or sub-units / modules of the electronic device have been mentioned in the detailed description above, this division is merely exemplary and not mandatory. In fact, according to embodiments of this disclosure, the features and functions of two or more units / modules described above can be embodied in one unit / module. Conversely, the features and functions of one unit / module described above can be further divided and embodied by multiple units / modules.

Claims

1. A container number identification and compensation system, characterized in that, It includes an individual feature database and an individual feature service. The individual feature database includes: A vector database is used to store individual feature vectors of different containers; A key-value database is used to store basic information of containers corresponding to individual feature vectors of each container, the basic information of which includes container number; The individual characteristic service component includes: The Individual Characteristics Web Service Module is used to respond to service requests initiated by external systems; The individual feature extraction service module is used to respond to a first service request initiated by the individual feature Web service module; the first service request includes: converting the container exterior photos contained in the incoming data from the external system into individual feature vectors to be processed; The individual feature extraction service module includes the SIFT algorithm, which is used to extract individual feature points of the container from the container surface photos contained in the input data, and generate the individual feature vector to be processed. The vector service module is used to respond to a second service request initiated by the individual feature web service module. The second service request includes: searching or storing the vector based on the individual feature vector to be processed in the vector database, and obtaining the vector ID. The key-value service module is used to respond to third service requests initiated by the individual feature web service module. The third service request includes: matching or storing the vector ID in conjunction with the key-value database and obtaining the result. The container number identification and compensation system meets one or more of the following conditions: i. The deployment method of the individual feature service part includes distributed deployment or microservice architecture. It provides public services to the outside world and operates the individual feature database internally. ii. The individual characteristic web service module is deployed in the DMZ area of ​​the individual characteristic service section; iii. The individual feature extraction service module is deployed on the intranet of the individual feature service section; iv. The key-value database includes an open-source key-value database; v. The deployment methods of the key-value database include distributed deployment; vi. The vector database includes an open-source vector database; vii. The deployment methods of the vector database include distributed deployment; The container number identification and compensation system meets one or more of the following conditions: i. The key-value database includes a Redis key-value database, which is deployed in a distributed manner and provides key-value pair storage and matching functions for individual feature services, recording "vector ID" and "container number and other basic container information" in the vector database. ii. The vector database includes the Milvus vector database, which is deployed in a distributed manner to provide storage and search functions for containerized "individual feature vectors" for individual feature services. The search function uses the database's HNSW algorithm + L2 index. The SIFT algorithm is used to extract individual feature points of the container from five photos of the container front and generate individual feature vectors, which are then stored in the Milvus open-source vector database. The vector ID obtained during storage is stored together with the manually labeled container number in the Redis open-source key-value database.

2. The container number identification and compensation system as described in claim 1, characterized in that, The container number recognition compensation system is used to automatically compensate for OCR container number recognition. The vector service module includes a vector search service module and a storage service module; the key-value service module includes a key-value matching service module and a key-value storage service module. The vector search service module is used to search the vector database for the result vector that is most similar to the feature vector to be processed and has a confidence level, and obtain the first vector ID recorded in the vector database; the key-value matching service module is used to match the first vector ID from the key-value database to obtain the basic information of the container. The vector storage service module is used to store the feature vector to be processed into the vector database to obtain the second vector ID of the vector database record; the key-value storage service module is used to combine the second vector ID with the container basic information contained in the data input from the external system into a key-value pair and store it into the key-value database.

3. A method for identifying individual containers, characterized in that, It employs the container number identification and compensation system as described in claim 1 to provide individual characteristic Web services and respond to public service requests initiated by external systems; The container individual identification method includes the following steps: S1. Receive incoming data from the external system; and determine the program to be executed based on the incoming data; S2. The individual feature extraction service module converts the container exterior photos contained in the input data into individual feature vectors to be processed and performs a search process. The individual feature web service module calls the vector search service of the vector service module and the key-value matching service of the key-value service module to feed back the searched information to the external system based on the individual feature vectors to be processed.

4. The container individual identification method as described in claim 3, characterized in that, The container individual identification method satisfies either condition I or condition II: Condition I: S1, Receive incoming data from the external system, the incoming data does not contain photos of the container's exterior; The individual feature Web service module sends a feedback to the external system indicating that the call failed; Condition II: S1, Receive incoming data from the external system, the incoming data containing only photos of the container's exterior, and determine to execute the search process; S2. First, the individual feature Web service module calls the individual feature extraction service and uses the SIFT algorithm to extract individual feature points from the container surface photo to obtain the individual feature vector to be processed. Secondly, the individual feature web service module calls the vector search service; using the individual feature vector to be processed as the query vector, it searches the vector database for the result vector that is most similar to the query vector and has confidence, and obtains the first vector ID recorded in the vector database; if the search fails, the individual feature web service module sends a feedback to the external system, indicating that no result was found. Then, the individual feature web service module calls the key-value matching service; using the first vector ID as the search key value, it matches the container basic information from the key-value database, the container basic information including the container number; if the match is successful, the individual feature web service module feeds back the container basic information to the external system; if the match fails, the individual feature web service module feeds back to the external system, indicating that no results were found.

5. A method for storing individual containers, characterized in that, It employs the container number identification and compensation system as described in claim 1 to provide individual characteristic Web services and respond to public service requests initiated by external systems; The method for storing individual containers includes the following steps: S1. Receive incoming data from the external system; and determine the program to be executed based on the incoming data; S2. The individual feature extraction service module converts the container exterior photos contained in the input data into individual feature vectors to be processed and executes a storage procedure. The individual feature web service module calls the vector storage service of the vector service module and the key-value storage service of the key-value service module to combine the individual feature vectors to be processed with the basic container information contained in the input data into key-value pairs and store them in the key-value database. The basic container information includes the container number.

6. The method for individual container storage as described in claim 5, characterized in that, The method for individual container storage satisfies either condition I or condition II: Condition I: S1, Receive incoming data from the external system, the incoming data does not contain photos of the container's exterior; The individual feature Web service module sends a feedback to the external system indicating that the call failed; Condition II: S1, Receive incoming data from the external system, the incoming data includes a photo of the container's exterior and basic container information, the basic container information including the container number, and determine to execute the stored procedure; S2. First, the individual feature Web service module calls the individual feature extraction service and uses the SIFT algorithm to extract individual feature points from the container surface photo to obtain the individual feature vector to be processed. Secondly, the individual feature web service module calls the vector storage service; stores the individual feature vector to be processed in the vector database, and obtains the second vector ID of the vector database record; Then, the individual feature web service module calls the key-value storage service; combines the second vector ID with the container basic information into a key-value pair, and stores it in the key-value database; When the same key value exists in the key-value database, the old record is overwritten; and the individual characteristics web service module feeds back to the external system to confirm that the individual characteristics of the container have been stored.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the container individual identification method as described in claim 3 or 4, or the container individual storage method as described in claim 5 or 6.

Images