Intelligent sorting and packaging system and sorting and packaging method thereof

Abstract

The invention provides an intelligent sorting and packaging system and a sorting and packaging method thereof, and relates to the technical field of sorting and packaging control, the intelligent sorting and packaging system comprises an information checking unit, the information checking unit checks information of goods which are wrongly sorted in a packaging area, and obtains the error rate of goods identification errors in the sorting errors; the algorithm simulation unit is used for establishing a simulation system model for machine recognition, selecting and adjusting a visual recognition algorithm for the goods, and controlling and updating a recognition mechanism in the simulation system model; and the simulation judgment unit is used for judging whether the use error condition of the adjusted visual identification algorithm in a simulation system model meets a preset requirement or not. According to the system and the method, virtualization simulation training can be carried out on an identification mechanism during sorting, and the visual identification mechanism is applied to the materialized processing equipment after the accurate visual identification mechanism is found, so that the sorting and packaging efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of sorting and packing control, and in particular to an intelligent sorting and packing system and a sorting and packing method thereof. Background Art

[0002] Logistics sorting and packaging is an important link in the cargo handling process, which mainly refers to the classification, sorting and packaging of cargo according to different specifications, categories and destinations.

[0003] When sorting and packing goods, an intelligent visual recognition mechanism is often used to determine the sorting area of ​​the current goods. However, since the goods are difficult to distinguish accurately due to the inaccurate recognition of the visual recognition mechanism, it is not convenient to quickly complete the redevelopment and application of the visual recognition mechanism for goods identification, which in turn affects the accuracy and efficiency of sorting and packing. Summary of the invention

[0004] In view of the shortcomings of the prior art described above, the purpose of the present invention is to provide an intelligent sorting and packaging system and a sorting and packaging method thereof, which are used to solve the problem in the prior art that if an identification error occurs during the sorting of goods, it is difficult to correct the error in a short time, thereby affecting the efficiency of intelligent sorting and packaging.

[0005] To achieve the above-mentioned purpose and other related purposes, the intelligent sorting and packaging system provided by the present invention includes: an information investigation unit, which investigates the information of the goods that are sorted incorrectly in the packaging area, and obtains the error rate of the goods identification errors in the sorting errors; an algorithm simulation unit, which establishes a simulation system model of machine recognition, selects and adjusts the visual recognition algorithm for the goods, and controls the update of the recognition mechanism in the simulation system model; a simulation judgment unit, which determines whether the error situation used by the adjusted visual recognition algorithm in the simulation system model meets the preset requirements, and if so, controls the output of the recognition mechanism corresponding to the adjusted visual recognition algorithm used in the simulation system model; if not, continues to control the adjustment of the visual recognition algorithm until the error situation meets the preset requirements; and an execution control unit, which obtains the adjusted recognition mechanism, and controls the adjustment to continue to execute the sorting and packaging of the goods using the recognition mechanism corresponding to the visual recognition algorithm.

[0006] In one embodiment of the present invention, the information investigation unit includes: an investigation and monitoring module, which scans the goods that pass through the sorting area and arrive at the packaging area, and confirms whether the arriving goods are wrong; and an error analysis module, which controls the inspection and analysis of the goods identification errors in the sorting errors based on the signal of the goods identification errors, and calculates the error rate of the goods identification errors.

[0007] In one embodiment of the present invention, the inspection and monitoring module includes: a cargo scanning module, which controls a secondary information scanning and analysis of the cargo delivered to the packaging area; and an inquiry and confirmation module, which analyzes and confirms whether the cargo corresponds to the corresponding packaging area. If not, it controls the output of a cargo error signal. If it does, it continues to analyze and confirm the next cargo.

[0008] In one embodiment of the present invention, the error analysis module includes: a monitoring and retrieval analysis module, which controls the retrieval of video data of the cargo allocation corresponding to the cargo error, and analyzes and determines whether an allocation error occurs in each link; and an error rate accumulation module, which collects control signals in which no allocation error occurs, and controls the classification of the cargo error as the cargo identification error.

[0009] In one embodiment of the present invention, the algorithm simulation unit includes: a simulation modeling module, which establishes a simulation system model of the sorting and packing processes corresponding to the site and the recognition mechanism of machine recognition according to the sorting and packing processes; and an algorithm update module, which collects update signals of the visual recognition algorithm of the cargo, replaces the visual recognition algorithm in the simulation system model, and obtains the updated recognition mechanism.

[0010] In one embodiment of the present invention, the simulation judgment unit includes: an operation data acquisition module, which collects the error rate of the adjusted visual recognition algorithm when it runs and recognizes in the simulation system model; an analysis control module, which collects the control signal when the error rate exceeds the set value, and controls the algorithm simulation unit to adjust and update the visual recognition algorithm again; and an output control module, which outputs the visual recognition algorithm corresponding to the error rate collected by the analysis control module that does not exceed the set value to the execution control unit.

[0011] In one embodiment of the present invention, the analysis control module includes: a parameter adjustment module, which controls the re-training of the enhanced model of the visual recognition algorithm to obtain the training model of the visual recognition algorithm after updating the parameters; and an algorithm replacement module, which controls the retrieval of other algorithms in the algorithm library to replace the visual recognition algorithm when the training model of the visual recognition algorithm after updating the parameters still does not meet the preset requirements.

[0012] In one embodiment of the present invention, the enhanced model training includes acquiring all image information corresponding to the cargo, and controlling and enhancing local enhanced training on features of the cargo that are difficult to visually distinguish.

[0013] In one embodiment of the present invention, the execution control unit includes: an algorithm import module, which controls the import of the adjusted visual recognition algorithm into the recognition mechanism of the physical control system; and a sorting control module, which continues to execute the sorting and recognition control of the goods according to the imported recognition mechanism.

[0014] The present invention also provides an intelligent sorting and packaging method, comprising the following steps:

[0015] S1: Checking the information of the parcels with sorting errors in the packing area, and obtaining the error rate of the parcel identification errors in the sorting errors;

[0016] S2: Establishing a simulation system model for machine recognition, selecting and adjusting a visual recognition algorithm for the cargo, and controlling and updating the recognition mechanism in the simulation system model;

[0017] S3: Determine whether the error condition of the adjusted visual recognition algorithm used in the simulation system model meets the preset requirements. If so, control the output of the recognition mechanism corresponding to the adjusted visual recognition algorithm used in the simulation system model; if not, continue to control the adjustment of the visual recognition algorithm until the error condition meets the preset requirements.

[0018] S4: Obtain the adjusted recognition mechanism, and control the adjustment to use the recognition mechanism corresponding to the visual recognition algorithm to continue sorting and packing the goods.

[0019] As described above, an intelligent sorting and packaging system and a sorting and packaging method of the present invention have the following beneficial effects: by monitoring and analyzing the error conditions of the goods after sorting, the sorting conditions of the goods with identification errors are determined. And when the goods are identified incorrectly, the identification of the goods can also be simulated by adjusting and strengthening the original visual recognition algorithm or replacing the visual recognition algorithm in the virtual system. In this way, the algorithm that can meet the recognition accuracy after optimization is determined and put into the physical sorting and packaging, so as to achieve the ability to quickly adjust the visual recognition algorithm according to the sorting and packaging requirements, so as to reduce or even avoid the difficulty in distinguishing goods due to the difficulty in correcting identification errors, which affects the accuracy of sorting and packaging. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is an architectural diagram of the sorting and packaging system of the present invention.

[0021] Figure 2 It is a flow chart of the sorting and packing method of the present invention. DETAILED DESCRIPTION

[0022] The following is an explanation of the embodiments of the present invention by specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the features in the following embodiments and the embodiments can be combined with each other without conflict. It should also be understood that the terms used in the embodiments of the present invention are intended to describe specific embodiments, rather than to limit the scope of protection of the present invention. The test methods for which specific conditions are not specified in the following examples are usually carried out under conventional conditions or according to the conditions recommended by the manufacturers.

[0023] See also Figure 1 to Figure 2 . It should be noted that the structures, proportions, sizes, etc. illustrated in the drawings in this specification are only used to match the contents disclosed in the specification for people familiar with this technology to understand and read, and are not used to limit the limiting conditions for the implementation of the present invention, so they have no substantial technical significance. Any modification of the structure, change in the proportional relationship, or adjustment of the size should still fall within the scope of the technical content disclosed by the present invention without affecting the effects and purposes that can be achieved by the present invention. At the same time, the terms such as "upper", "lower", "left", "right", "middle" and "one" quoted in this specification are only for the convenience of description, and are not used to limit the scope of the implementation of the present invention. Changes or adjustments in their relative relationships should also be regarded as the scope of the implementation of the present invention without substantially changing the technical content.

[0024] See also Figure 1 The intelligent sorting and packaging system provided by the present invention includes: an information checking unit, which checks the information of the goods with sorting errors in the packaging area and obtains the error rate of the goods identification errors in the sorting errors; an algorithm simulation unit, which establishes a simulation system model of machine recognition, selects and adjusts the visual recognition algorithm for the goods, and controls the update of the recognition mechanism in the simulation system model; a simulation judgment unit, which judges whether the error situation used by the adjusted visual recognition algorithm in the simulation system model meets the preset requirements, and if so, controls the output of the recognition mechanism corresponding to the adjusted visual recognition algorithm used in the simulation system model; if not, continues to control the adjustment of the visual recognition algorithm until the error situation meets the preset requirements; and an execution control unit, which obtains the adjusted recognition mechanism, and controls the adjustment to continue to execute the sorting and packaging of the goods using the recognition mechanism corresponding to the visual recognition algorithm.

[0025] It is not difficult to find from the above content that in the process of intelligent sorting and packaging, the information of the goods that pass through the packaging area and arrive at the sorting area can be analyzed through the information investigation unit, and then the information of the goods that are sorted incorrectly can be determined. And according to the information of the goods that are sorted incorrectly, the situation of the goods identification error that occurs during the identification process of the goods is determined. And the proportion of the number of times the goods identification error occurs during the sorting of the entire type of goods is evaluated, that is, the error rate of the goods identification error. Through the algorithm simulation unit, a virtual system is established to simulate the visual recognition algorithm of the machine identification in the goods identification process, and an identification mechanism and simulation system model based on the visual recognition algorithm are established to perform a virtual simulation of the sorting and packaging process of the goods. Through the simulation judgment unit, in the virtual simulation process, it is judged whether the error rate of the goods identification error that still exists after the visual recognition algorithm is adjusted meets the preset requirements. If it is satisfied, the adjusted visual recognition algorithm is transmitted to the execution control unit, and the visual control logic of the physical device is adjusted by the execution control unit to ensure the optimized sorting and packaging monitoring of the goods. If it is not satisfied, the control will continue to adjust the visual recognition algorithm until the adjusted visual recognition algorithm can make the error situation meet the preset requirements.

[0026] Furthermore, the information investigation unit includes: an investigation and monitoring module, which scans the goods that pass through the sorting area and arrive at the packaging area, and confirms whether the arriving goods are wrong; and an error analysis module, which controls the inspection and analysis of the goods identification errors in the sorting errors based on the signals of the goods identification errors, and calculates the error rate of the goods identification errors.

[0027] In one embodiment of the present invention, during the process of identifying errors, the information checking unit scans and processes the parcels that have passed through the sorting area and arrived at the packaging area through the checking monitoring module, thereby finding the parcels with sorting errors. The error analysis module is used to find the parcels with identification errors, and the error rate of identification errors is obtained according to the proportion of parcel identification errors to the total sorted parcels.

[0028] Furthermore, the investigation and monitoring module includes: a cargo scanning module, which controls a secondary information scanning and analysis of the cargo delivered to the packaging area; and an inquiry and confirmation module, which analyzes and confirms whether the cargo corresponds to the corresponding packaging area. If not, it controls the output of a cargo error signal. If it does, it continues to analyze and confirm the next cargo.

[0029] In one embodiment of the present invention, during the inspection and monitoring, the cargo scanning module is used to scan the information of the cargo from the sorting area to the packaging area, and the query confirmation module is used to confirm whether the cargo arriving at the packaging area is accurate. If it is inaccurate, a signal of cargo sorting error will be generated.

[0030] Specifically, the error analysis module includes: a monitoring and retrieval analysis module, which controls the retrieval of video data of the cargo allocation corresponding to the cargo error, and analyzes and determines whether an allocation error occurs in each link; and an error rate accumulation module, which collects control signals when no allocation error occurs, and controls the classification of the cargo error as the cargo identification error.

[0031] In one embodiment of the present invention, when analyzing the sorting of wrong shipments, the error analysis module retrieves the video surveillance data through the monitoring retrieval analysis module, and analyzes the video surveillance data for whether there are errors in the procedures during the transportation and manual processing of the shipments, which leads to the sorting errors of the shipments. If not, it means that there is an error in the identification of the wrong shipment during sorting. And the error rate accumulation module is used to classify the errors that are not caused by program errors as shipment identification errors. Then the error rate of the proportion of shipment identification errors to the total identification of this type of shipments is calculated.

[0032] Among them, the algorithm simulation unit includes: a simulation modeling module, which establishes a simulation system model of the sorting and packing processes corresponding to the site and the recognition mechanism of machine recognition according to the sorting and packing processes; and an algorithm updating module, which collects the update signal of the visual recognition algorithm of the cargo, replaces the visual recognition algorithm in the simulation system model, so as to obtain the updated recognition mechanism.

[0033] In one embodiment of the present invention, when performing algorithm simulation, the algorithm simulation unit uses the simulation modeling module to establish a virtual simulation system corresponding to the real device according to the visual recognition algorithm, that is, a simulation system model. The algorithm update module imports the adjusted visual recognition algorithm into the simulation system model as needed to simulate the identification of the goods, so as to find a visual recognition algorithm that can meet the needs.

[0034] Specifically, the simulation judgment unit includes: an operation data acquisition module, which collects the error rate of the adjusted visual recognition algorithm when it runs and recognizes in the simulation system model; an analysis control module, which collects the control signal when the error rate exceeds the set value, and controls the algorithm simulation unit to adjust and update the visual recognition algorithm again; and an output control module, which outputs the visual recognition algorithm corresponding to the error rate collected by the analysis control module that does not exceed the set value to the execution control unit.

[0035] In one embodiment of the present invention, during the simulation judgment process, the error rate of sorting and recognition errors is acquired and counted by running the data acquisition module to run the goods in the virtual system. The control module is used to collect the control signal when the error rate exceeds the set value, and control the visual recognition algorithm to be updated again until the set value position is met. And the visual recognition algorithm with an error rate that meets the set value is output to the execution control unit for execution through the output control module.

[0036] Preferably, the analysis control module includes: a parameter adjustment module, which controls the re-training of the enhanced model of the visual recognition algorithm to obtain the training model of the visual recognition algorithm after updating the parameters; and an algorithm replacement module, which controls the retrieval of other algorithms in the algorithm library to replace the visual recognition algorithm when the training model of the visual recognition algorithm after updating the parameters still does not meet the preset requirements.

[0037] In one embodiment of the present invention, when the analysis and control module performs analysis and control, the parameter adjustment module implements enhanced model training of the original visual recognition algorithm according to actual conditions to obtain a training model that can improve the recognition accuracy. Similarly, the algorithm replacement module can also be used to directly replace the original visual recognition algorithm so that the recognition accuracy meets the requirements.

[0038] The enhanced model training includes acquiring all image information corresponding to the cargo, and controlling and enhancing local enhanced training on features of the cargo that are difficult to visually distinguish.

[0039] In addition, the execution control unit includes: an algorithm import module, which controls the import of the adjusted visual recognition algorithm into the recognition mechanism of the physical control system; and a sorting control module, which continues to execute the sorting and recognition control of the goods according to the imported recognition mechanism.

[0040] In one embodiment of the present invention, when executing visual recognition algorithm control, the algorithm import module imports the adjusted and accurate visual recognition algorithm into the physical control system to re-establish the recognition mechanism. The sorting control module is used to implement the sorting and packaging recognition of the goods again.

[0041] like Figure 2 As shown, the present invention also provides an intelligent sorting and packaging method, comprising the following steps:

[0042] S1: Checking the information of the parcels with sorting errors in the packing area, and obtaining the error rate of the parcel identification errors in the sorting errors;

[0043] S2: Establishing a simulation system model for machine recognition, selecting and adjusting a visual recognition algorithm for the cargo, and controlling and updating the recognition mechanism in the simulation system model;

[0044] S3: Determine whether the error condition of the adjusted visual recognition algorithm used in the simulation system model meets the preset requirements. If so, control the output of the recognition mechanism corresponding to the adjusted visual recognition algorithm used in the simulation system model; if not, continue to control the adjustment of the visual recognition algorithm until the error condition meets the preset requirements.

[0045] S4: Obtain the adjusted recognition mechanism, and control the adjustment to use the recognition mechanism corresponding to the visual recognition algorithm to continue sorting and packing the goods.

[0046] In summary, the present invention monitors and analyzes the errors of the goods after sorting, thereby determining the sorting of goods with identification errors. And when the goods are identified incorrectly, it is also possible to simulate the identification of the goods by adjusting and strengthening the original visual recognition algorithm or replacing the visual recognition algorithm in the virtual system. Thereby, the algorithm that can meet the recognition accuracy after optimization is determined, and is put into the physical sorting and packaging, so that the visual recognition algorithm can be quickly adjusted according to the sorting and packaging needs, so as to reduce or even avoid the difficulty in distinguishing goods due to the difficulty in correcting identification errors, which affects the accuracy of sorting and packaging. Therefore, the present invention effectively overcomes the various shortcomings of the prior art and has a high industrial utilization value.

[0047] The above embodiments are merely illustrative of the principles and effects of the present invention, and are not intended to limit the present invention. Anyone familiar with the art may modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by a person of ordinary skill in the art without departing from the spirit and technical concept disclosed by the present invention shall still be covered by the claims of the present invention.

Claims

1. An intelligent sorting and packaging system, characterized in that: include: An information checking unit, wherein the information checking unit checks the information of the goods with sorting errors in the packing area, and obtains the error rate of the goods identification errors in the sorting errors; An algorithm simulation unit, which establishes a simulation system model for machine recognition, selects and adjusts a visual recognition algorithm for cargo, and controls and updates the recognition mechanism in the simulation system model; A simulation judgment unit, wherein the simulation judgment unit judges whether the error condition of the adjusted visual recognition algorithm used in the simulation system model meets the preset requirements, and if so, controls the output of the recognition mechanism corresponding to the adjusted visual recognition algorithm used in the simulation system model; if not, continues to control and adjust the visual recognition algorithm until the error condition meets the preset requirements; as well as An execution control unit is configured to obtain the adjusted recognition mechanism and control the adjustment to continue sorting and packing the goods using the recognition mechanism corresponding to the visual recognition algorithm.

2. The intelligent sorting and packaging system according to claim 1 is characterized in that: The information checking unit comprises: A checking and monitoring module, which scans the goods that have passed through the sorting area and arrived at the packaging area, and confirms whether the arriving goods are wrong; and The error analysis module controls the inspection and analysis of the shipment identification errors in the sorting errors according to the signals of the shipment identification errors, and calculates the error rate of the shipment identification errors.

3. The intelligent sorting and packaging system according to claim 2 is characterized in that: The investigation and monitoring module comprises: A cargo scanning module, which controls a secondary information scanning and analysis of the cargo delivered to the packaging area; and The query confirmation module analyzes and confirms whether the shipment corresponds to the corresponding packaging area. If not, the module controls the output of a shipment error signal. If so, the module continues to analyze and confirm the next shipment.

4. The intelligent sorting and packaging system according to claim 2 is characterized in that: The error analysis module comprises: A monitoring, retrieval and analysis module, which controls the retrieval of video data of shipment allocation corresponding to shipment errors, and analyzes and determines whether allocation errors occur in each link; and The error rate accumulation module collects control signals indicating that no deployment errors have occurred, and controls the classification of the shipment errors as shipment identification errors.

5. The intelligent sorting and packaging system according to claim 1 is characterized in that: The algorithm simulation unit comprises: A simulation modeling module, wherein the simulation modeling module establishes a simulation system model of each sorting and packing process corresponding to the site and an identification mechanism of machine identification according to each sorting and packing process; and An algorithm updating module collects an update signal of a visual recognition algorithm for cargo and replaces the visual recognition algorithm in the simulation system model to obtain an updated recognition mechanism.

6. The intelligent sorting and packaging system according to claim 1 is characterized in that: The simulation judgment unit comprises: An operation data acquisition module, wherein the operation data acquisition module collects error rates when the adjusted visual recognition algorithm is run in the simulation system model; An analysis control module, the analysis control module collects control signals whose error rate exceeds a set value, and controls the algorithm simulation unit to adjust and update the visual recognition algorithm again; and An output control module, wherein the output control module outputs the visual recognition algorithm corresponding to the error rate collected by the analysis control module and does not exceed the set value to the execution control unit.

7. The intelligent sorting and packaging system according to claim 6, characterized in that: The analysis control module comprises: A parameter adjustment module, the parameter adjustment module controls the enhanced model training of the visual recognition algorithm to be repeated again to obtain a training model of the visual recognition algorithm after updating parameters; and An algorithm replacement module controls the calling of other algorithms in the algorithm library to replace the visual recognition algorithm when the training model of the visual recognition algorithm after updating the parameters still does not meet the preset requirements.

8. The intelligent sorting and packaging system according to claim 7 is characterized in that: The enhanced model training includes acquiring all image information corresponding to the cargo, and controlling and enhancing local enhanced training on features of the cargo that are difficult to visually distinguish.

9. The intelligent sorting and packaging system according to claim 1, characterized in that: The execution control unit comprises: an algorithm importing module, the algorithm importing module controls the importing of the adjusted visual recognition algorithm into the recognition mechanism of the materialized control system; and A sorting control module, wherein the sorting control module continues to perform sorting and identification control of the goods according to the imported identification mechanism.

10. Intelligent sorting and packaging method, characterized in that: The steps include: S1: Checking the information of the parcels with sorting errors in the packing area, and obtaining the error rate of the parcel identification errors in the sorting errors; S2: Establishing a simulation system model for machine recognition, selecting and adjusting a visual recognition algorithm for the cargo, and controlling and updating the recognition mechanism in the simulation system model; S3: Determine whether the error condition of the adjusted visual recognition algorithm used in the simulation system model meets the preset requirements. If so, control the output of the recognition mechanism corresponding to the adjusted visual recognition algorithm used in the simulation system model; if not, continue to control the adjustment of the visual recognition algorithm until the error condition meets the preset requirements. S4: Obtain the adjusted recognition mechanism, and control the adjustment to use the recognition mechanism corresponding to the visual recognition algorithm to continue sorting and packing the goods.

Images