CN115463844A - Intelligent cargo sorting method and system based on dual recognition - Google Patents
Intelligent cargo sorting method and system based on dual recognition Download PDFInfo
- Publication number
- CN115463844A CN115463844A CN202211026261.1A CN202211026261A CN115463844A CN 115463844 A CN115463844 A CN 115463844A CN 202211026261 A CN202211026261 A CN 202211026261A CN 115463844 A CN115463844 A CN 115463844A
- Authority
- CN
- China
- Prior art keywords
- goods
- information
- sorting
- sorted
- image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C3/00—Sorting according to destination
- B07C3/10—Apparatus characterised by the means used for detection ofthe destination
Landscapes
- Sorting Of Articles (AREA)
Abstract
The invention discloses a cargo intelligent sorting method and system based on dual recognition, and relates to the field of cargo sorting, wherein the method comprises the following steps: in the process of conveying the goods to be sorted, a code scanning device is adopted to acquire and obtain identification code information on the goods to be sorted; inputting the identification code into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not; if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted; in the transmission process, the image acquisition device is adopted to acquire the image information of the goods to be sorted; inputting the image information into a pre-constructed image recognition model to obtain a second recognition result; and sorting the goods to be sorted according to the second recognition result. The accuracy and the degree of automation of improving goods identification are reached, and then technical effects such as quality and efficiency of goods sorting are improved.
Description
Technical Field
The invention relates to the field of goods sorting, in particular to a goods intelligent sorting method and system based on double recognition.
Background
A large number of goods need to be sorted every day, and the existing goods sorting processing method has high dependence on manpower, so that the efficiency is low and errors are easy to occur. Goods letter sorting also directly influences the work efficiency and the productivity effect of enterprise. A set of goods sorting method with low cost and good performance is urgently needed by a plurality of enterprises, and in order to meet the requirement, the method for optimizing goods sorting is designed, and has important practical significance.
In the prior art, the technical problems of insufficient accuracy and low automation degree of goods identification and low quality and efficiency of goods sorting are caused.
Disclosure of Invention
The application provides a goods intelligent sorting method and system based on double recognition. The problem of among the prior art to the accuracy of goods discernment not enough, degree of automation is not high, and then cause the quality and the not high technique of efficiency of goods letter sorting.
In view of the above problems, the present application provides a cargo intelligent sorting method and system based on dual recognition.
In a first aspect, the present application provides a cargo intelligent sorting method based on dual recognition, wherein the method is applied to a cargo intelligent sorting system based on dual recognition, and the method includes: conveying goods to be sorted; in the transmission process, the code scanning device is adopted to acquire and obtain identification code information on the goods to be sorted; inputting the identification code information into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not; if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted; in the transmission process, the image acquisition device is adopted to acquire and obtain the image information of the goods to be sorted; inputting the image information into a pre-constructed image recognition model to obtain a second recognition result; and sorting the goods to be sorted according to the second recognition result.
In a second aspect, the present application further provides a cargo intelligent sorting system based on dual recognition, wherein the system includes: the conveying module is used for conveying goods to be sorted; the code scanning module is used for acquiring identification code information on the goods to be sorted by adopting the code scanning device in the transmission process; the judgment module is used for inputting the identification code information into a pre-constructed identification code information space and judging whether a first identification result is obtained or not; the sorting and conveying selection module is used for sorting the goods to be sorted according to the first recognition result if the goods to be sorted are in the first recognition result, and continuously conveying the goods to be sorted if the goods to be sorted are not in the first recognition result; the image acquisition module is used for acquiring and acquiring image information of the goods to be sorted by adopting the image acquisition device in the transmission process; the input module is used for inputting the image information into a pre-constructed image recognition model to obtain a second recognition result; and the sorting module is used for sorting the goods to be sorted according to the second recognition result.
One or more technical solutions provided in the present application have at least the following technical effects or advantages:
acquiring identification code information on goods to be sorted through a code scanning device; inputting the identification code information into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not; if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted; in the transmission process, the image acquisition device is adopted to acquire and obtain the image information of the goods to be sorted; inputting the image information into a pre-constructed image recognition model to obtain a second recognition result; and sorting the goods to be sorted according to the sorting order. The accuracy and the degree of automation of goods identification are improved, and the quality and the efficiency of goods sorting are improved; simultaneously, improve intelligent, the scientificity of goods letter sorting, reduce the cost of goods letter sorting, it is extravagant to reduce the manpower that the goods letter sorting caused, high-efficient, quick completion goods letter sorting's technological effect.
Drawings
Fig. 1 is a schematic flow chart of an intelligent cargo sorting method based on dual recognition according to the present application;
fig. 2 is a schematic flow chart illustrating the process of acquiring image information of goods to be sorted in the intelligent goods sorting method based on dual recognition according to the present application;
fig. 3 is a schematic flowchart of a process for constructing an image recognition model in the intelligent goods sorting method based on dual recognition according to the present application;
fig. 4 is a schematic structural diagram of an intelligent cargo sorting system based on dual recognition according to the present application.
Description of reference numerals: the system comprises a transmission module 11, a code scanning module 12, a judgment module 13, a sorting transmission selection module 14, an image acquisition module 15, an input module 16 and a sorting module 17.
Detailed Description
The application provides an intelligent cargo sorting method and system based on dual recognition. The technical problems that in the prior art, the accuracy of goods identification is insufficient, the automation degree is not high, and then the quality and the efficiency of goods sorting are not high are solved. The accuracy and the degree of automation of goods identification are improved, and the quality and the efficiency of goods sorting are improved; simultaneously, improve intelligent, the scientificity of goods letter sorting, reduce the cost of goods letter sorting, it is extravagant to reduce the manpower that the goods letter sorting caused, high-efficient, quick completion goods letter sorting's technological effect.
Example one
Referring to fig. 1, the present application provides a cargo intelligent sorting method based on dual recognition, wherein the method is applied to a cargo intelligent sorting system based on dual recognition, the system includes a code scanning device and an image acquisition device, and the method specifically includes the following steps:
step S100: conveying goods to be sorted;
step S200: in the transmission process, the code scanning device is adopted to acquire and obtain identification code information on the goods to be sorted;
particularly, treat through logistics transmission equipment such as conveyer belt and sort the goods and transmit, simultaneously, treat the in-process of sorting the goods at the transmission, treat through sweeping a yard device and sort the goods and discern, obtain identification code information. The goods to be sorted are any goods which are automatically sorted by using the intelligent goods sorting system based on dual recognition. And the surface of the goods to be sorted is provided with an identification code which corresponds to the goods to be sorted one by one. The identification code information comprises data information such as type information, sorting batch information and the like which are identified by the code scanning device of the identification code of the goods to be sorted. The method and the device have the advantages that the identification code information of the goods to be sorted is obtained through the code scanning device, the automation degree of identification code information acquisition is improved, and the problems of manpower waste caused by manpower acquisition of the identification code information, high identification code information acquisition cost, low efficiency and the like are solved; meanwhile, the technical effects of improving the efficiency and the accuracy of identification code information acquisition are achieved.
The code scanning device is included in the intelligent cargo sorting system based on double recognition. Illustratively, the code scanning device consists of five code scanning guns, sensors and the like. It can treat the letter sorting goods through this sweep a yard device and carry out five-sided scanning, when treating the letter sorting goods through sweeping the preceding sensor of a yard rifle, triggers to sweep a yard rifle and sweep the sign indicating number, five sweep a yard rifle and treat the letter sorting goods simultaneously and begin to sweep a yard, one of them sweeps a yard rifle and sweep a yard rifle for main, and main yard rifle of sweeping can be all the other four identification code information of sweeping a yard rifle and obtaining gather to reach all identification code information a goods intelligence letter sorting system based on dual discernment.
After installing sweeping a yard device, in order to guarantee the accuracy of the identification code information that obtains, prevent to produce the influence because of identification code information quality is not high to goods letter sorting, need sweep a yard rifle in a yard device and carry out static debugging, dynamic test usually. Static debugging means that a software Dataman provided by a code scanning gun supplier Cognex performs a series of parameter settings on the code scanning gun, wherein the parameter settings comprise the focal length, the aperture size, the exposure value size and the gain size of the code scanning gun and the selective setting of a main code scanning gun. Meanwhile, the parameter setting also comprises the steps of selecting the type and the number of the identification codes to be identified by the code scanning gun, and screening the digits of the identification codes to be identified and the contents to be displayed. In addition, static debugging still includes adjusting the mounted position of sweeping a yard rifle. The dynamic test refers to that the code scanning device which completes static debugging is used for collecting identification codes of goods being transmitted, identification code information of the goods is obtained, quality parameters such as definition, integrity and the like of the identification code information are judged, if the definition of the identification code information is not high and the integrity is not enough, the static debugging is carried out on the code scanning device again until the code scanning device achieves a better code scanning effect.
Step S300: inputting the identification code information into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not;
further, step S300 of the present application further includes:
step S310: constructing the identification code information space;
further, step S310 of the present application further includes:
step S311: acquiring types of the multiple types of goods according to the identification codes of the multiple types of goods to acquire multiple types of information;
step S312: acquiring batches of the various goods according to the identification codes of the various goods to acquire a plurality of batch information;
step S313: obtaining a plurality of entity information according to the various goods;
step S314: obtaining a first attribute and a plurality of first attribute value information according to the plurality of types of information;
step S315: obtaining a second attribute and a plurality of second attribute value information according to the plurality of batch information;
step S316: and constructing the identification code information space based on the plurality of entity information, the first attribute, the plurality of first attribute value information, the second attribute and the plurality of second attribute value information and based on a knowledge graph.
Specifically, based on the identification codes of multiple types of goods, the goods intelligent sorting system based on dual recognition is used for collecting information of types and batches of the multiple types of goods to obtain multiple types of information and multiple batches of information. Further, a plurality of entity information is determined according to the plurality of types of goods. According to the plurality of types of information, a first attribute and a plurality of first attribute value information are obtained. And obtaining the second attribute and a plurality of second attribute value information according to the plurality of batch information. Based on the method, an identification code information space is constructed and obtained by combining the knowledge map thought. The identification codes of the various types of goods comprise a plurality of identification codes corresponding to various types of goods which are automatically sorted by using the intelligent goods sorting system based on dual recognition. The multiple types of information comprise multiple goods type information corresponding to the identification codes of multiple types of goods. The batch information comprises sorting batch information corresponding to identification codes of various types of goods. The sorting batch information comprises sorting line information corresponding to the identification codes of various types of goods. The plurality of entity information includes a plurality of types of goods. The first attribute is a type of multiple types of goods. The plurality of first attribute value information includes a plurality of type information. The second attribute is a batch of multiple types of goods. The plurality of second attribute value information includes a plurality of batch information. The knowledge graph is an expression mode of data information. The knowledge graph comprises a mode layer and a data layer. The data layer consists of a series of facts; the mode layer is constructed on the data layer and is mainly used for carrying out canonical expression on a series of facts of the data layer. The identification code information space includes a plurality of entity information, a first attribute, a plurality of first attribute value information, a second attribute, a plurality of second attribute value information, and a correspondence relationship therebetween. The technical effect that a reliable identification code information space is established through a plurality of types of information and a plurality of batches of information, and the accuracy of a subsequently obtained first identification result is improved is achieved.
Step S320: inputting the identification code information into the identification code information space for mapping, and judging whether a mapping result exists or not;
step S330: if so, taking the mapping result as the first identification result;
step S340: if not, the first identification result is not obtained.
Specifically, the obtained identification code information is input into the established identification code information space to be mapped correspondingly, and whether the identification code information has a mapping result is judged, that is, whether the identification code information has a corresponding relationship with the plurality of entity information, the first attribute, the plurality of first attribute value information, the second attribute and the plurality of second attribute value information in the identification code information space is judged. Further, if the identification code information has a mapping result, the mapping result is set as a first identification result. And if the identification code information does not have the mapping result, the first identification result cannot be obtained. The mapping result comprises entity information, a first attribute, first attribute value information, a second attribute and second attribute value information corresponding to the identification code information. The first recognition result is the mapping result. The technical effects that the identification code information is accurately mapped and judged through the identification code information space, the first identification result is adaptively obtained, and the goods sorting accuracy is improved are achieved.
Step S400: if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted;
specifically, when judging whether the first identification result is obtained, if the first identification result is obtained, sorting the goods to be sorted according to the first identification result. And if the first identification result is not obtained, continuing to convey the goods to be sorted. The technical effects that the goods to be sorted are sorted or transmitted adaptively according to whether the first recognition result is obtained or not, and the goods sorting accuracy is improved are achieved.
Step S500: in the transmission process, the image acquisition device is adopted to acquire and obtain the image information of the goods to be sorted;
further, as shown in fig. 2, step S500 of the present application further includes:
step S510: obtaining a plurality of preset angles;
step S520: acquiring and acquiring a plurality of angle image information of the goods to be sorted by the image acquisition device based on the plurality of preset angles;
step S530: and taking the angle image information as the image information.
Specifically, for the goods to be sorted which do not obtain the first recognition result, in the process of continuously transmitting the goods to be sorted, according to a plurality of preset angles, the image acquisition device is used for acquiring image information of a plurality of angles of the goods to be sorted, and the image information of the goods to be sorted is obtained. Wherein the plurality of preset angles comprise a plurality of image acquisition angle information of the image acquisition device. The preset angles are preset and determined by the intelligent cargo sorting system based on dual recognition. The image information of the goods to be sorted comprises a plurality of angle image information. The image information at the plurality of angles comprises image data information of goods to be sorted corresponding to a plurality of preset angles of the image acquisition device. The technical effects that the goods to be sorted which do not obtain the first recognition result are subjected to image acquisition according to a plurality of preset angles, the image information of the goods to be sorted is obtained, and a foundation is laid for subsequently obtaining the second recognition result are achieved.
The image acquisition device is included in the intelligent cargo sorting system based on dual recognition. Exemplarily, the image acquisition device comprises a camera and a sensor. In order to improve the definition of the obtained image information of the goods to be sorted, a reflector and an LED lamp tube can be arranged near the goods to be sorted, the brightness of the goods to be sorted is improved, the obtained image information of the goods to be sorted is clearer, and the image recognition is easy. When goods to be sorted pass through the sensor behind the camera, the camera is triggered to take pictures, and the camera can take pictures of the top and the side of the goods to be sorted. The installation position, the focal length, the aperture, the exposure value and the RGB value of the camera can be adjusted, so that the image information of the goods to be sorted is clearly visible in the middle of the visual field.
Step S600: inputting the image information into a pre-constructed image recognition model to obtain a second recognition result;
further, step S600 of the present application further includes:
step S610: constructing the image recognition model;
further, as shown in fig. 3, step S610 of the present application further includes:
step S611: acquiring image information of a plurality of types of cargos to obtain a sample image information set;
step S612: acquiring sorting category information of the multiple types of cargos to obtain a sample sorting category information set;
step S613: dividing and combining the sample image information set and the sample sorting category information set to obtain a first data set and a second data set;
step S614: constructing an image recognition model based on a deep convolutional neural network;
step S615: dividing the first data set and carrying out data identification to obtain a training set, a verification set and a test set;
step S616: carrying out supervision training on the image recognition model by adopting the training set until convergence or accuracy reaches a preset threshold;
step S617: and verifying and testing the image recognition model by adopting the verification set and the test set, and if the accuracy rate meets the preset threshold value, obtaining the image recognition model.
Specifically, the intelligent cargo sorting system based on dual recognition is used for collecting image information and sorting category information of multiple types of cargos to obtain a sample image information set and a sample sorting category information set, and dividing and combining the sample image information set and the sample sorting category information set to obtain a first data set and a second data set. Further, the first data set is divided and data identification is carried out, and a training set, a verification set and a test set are obtained. And then, performing supervision training on the image recognition model by using the training set until convergence or the accuracy reaches a preset threshold, inputting the verification set and the test set into the image recognition model respectively for verification and test, and if the accuracy meets the preset threshold, obtaining the image recognition model. Wherein the sample image information set includes image information of multiple categories of goods. The sample sorting category information set comprises sorting category information of multiple types of goods. The first data set comprises partial data information of the sample image information set and partial data information of the sample sorting category information set. The second data set comprises partial data information of the sample image information set and partial data information of the sample sorting category information set. And, the first data set is different from the second data set. The image recognition model is a deep convolution neural network model. The accuracy rate is parameter information used for representing the similarity between the sorting category information output by the image recognition model and the sorting category information of the input training set, verification set and test set in the sample sorting category information set, and the higher the similarity between the sorting category information and the input training set, the verification set and the test set is, the higher the corresponding accuracy rate is. The preset threshold value is determined by the intelligent cargo sorting system based on dual recognition in a self-adaptive setting mode according to the accuracy requirement of the image recognition model. The technical effects of constructing the image recognition model and providing data support for subsequently obtaining a second recognition result are achieved.
Step S620: verifying the stability of the image recognition model, and putting the image recognition model into use if the stability meets the preset requirement;
further, step S620 in the present application further includes:
step S621: dividing and data identifying the second data set to obtain a first stability verification set and a second stability verification set;
step S622: inputting the first stability verification set and the second stability verification set into the image recognition model respectively, obtaining a plurality of first recognition results and a plurality of second recognition results respectively, and obtaining a first sorting category set and a second sorting category set;
step S623: calculating a stability of the image recognition model from the first and second sorted set of categories by:
wherein, P i Is the proportion of the ith sort category information in the first sort category set, Q i And n is the proportion of the ith sorting category information in the second sorting category set, and is the category number of the sorting category information.
Specifically, after the obtained second data set is divided and data identification is performed, a first stability verification set and a second stability verification set are obtained. And further, respectively taking a first stability verification set and a second stability verification set as input information, inputting the input information into the image recognition model, performing predictive recognition on sorting categories to obtain a plurality of first recognition results and a plurality of second recognition results, determining a first sorting category set and a second sorting category set based on the first recognition results and the second recognition results, calculating by combining the stability calculation formula to obtain the stability of the image recognition model, judging whether the stability meets a preset requirement, and putting the image recognition model into use if the stability meets the preset requirement. The first stability verification set and the second stability verification set are included in the second data set. The plurality of first identification results comprise a plurality of sorting category information output by the image identification model corresponding to the first stability verification set. The second recognition results comprise a plurality of sorting category information output by the image recognition model corresponding to the second stability verification set. The first sorting category set comprises a plurality of first recognition results. The second sorting category set includes a plurality of second recognition results. The preset requirement comprises a preset stability threshold value which is preset and determined by the intelligent cargo sorting system based on the dual recognition.
Specifically, since the first stability verification set and the second stability verification set are obtained by random division, the image information ratios of the goods in the same sort category in the first stability verification set and the second stability verification set should be similar, and therefore the image recognition model stability is calculated based on the above equation in the first sorting category set and the second sorting category set obtained by predictive recognition. The preset requirement of the stability is preferably 0.3, and if the stability R value of the image recognition model obtained through calculation is smaller than 0.3, the preset requirement is met, and the image recognition model can be put into use.
The technical effects that the stability of the constructed image recognition model is verified by utilizing the second data set, the image recognition model with the stability meeting the preset requirement is obtained, and the accuracy of the subsequently obtained second recognition result is improved are achieved.
Step S630: and inputting the image information into the image recognition model which is put into use to obtain the second recognition result.
Step S700: and sorting the goods to be sorted according to the second recognition result.
Specifically, the image information of the goods to be sorted is used as input information, the image recognition model is input, a second recognition result is obtained, and the goods to be sorted are sorted according to the second recognition result. And the second identification result comprises sorting category information corresponding to the image information of the goods to be sorted. The technical effects of obtaining an accurate second recognition result through the image recognition model and further improving the quality and the efficiency of goods sorting are achieved.
In summary, the intelligent cargo sorting method based on dual recognition provided by the application has the following technical effects:
1. acquiring and acquiring identification code information on goods to be sorted through a code scanning device; inputting the identification code information into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not; if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted; in the transmission process, the image acquisition device is adopted to acquire and obtain the image information of the goods to be sorted; inputting the image information into a pre-constructed image recognition model to obtain a second recognition result; and sorting the goods to be sorted according to the sorting order. The accuracy and the degree of automation of goods identification are improved, and the quality and the efficiency of goods sorting are improved; simultaneously, improve intelligent, the scientificity of goods letter sorting, reduce the cost of goods letter sorting, it is extravagant to reduce the manpower that the goods letter sorting caused, high-efficient, quick completion goods letter sorting's technological effect.
2. The identification code information of the goods to be sorted is obtained through the code scanning device, so that the automation degree of identification code information acquisition is improved, and the problems of manpower waste caused by manpower acquisition of the identification code information, high identification code information acquisition cost, low efficiency and the like are solved; meanwhile, the efficiency and the accuracy of identification code information acquisition are improved.
Example two
Based on the same inventive concept as the intelligent cargo sorting method based on dual recognition in the foregoing embodiment, the present invention further provides an intelligent cargo sorting system based on dual recognition, referring to fig. 4, where the system includes:
the conveying module 11 is used for conveying the goods to be sorted;
the code scanning module 12 is used for acquiring identification code information on the goods to be sorted by adopting the code scanning device in the transmission process;
the judging module 13 is used for inputting the identification code information into a pre-constructed identification code information space and judging whether a first identification result is obtained or not;
the sorting, transmitting and selecting module 14, the sorting, transmitting and selecting module 14 is configured to, if yes, sort the goods to be sorted according to the first recognition result, and if not, continue to transmit the goods to be sorted;
the image acquisition module 15 is used for acquiring and acquiring image information of the goods to be sorted by adopting the image acquisition device in the transmission process;
the input module 16 is used for inputting the image information into a pre-constructed image recognition model to obtain a second recognition result;
and the sorting module 17 is used for sorting the goods to be sorted according to the second recognition result.
Further, the system further comprises:
the identification code information space construction module is used for constructing the identification code information space;
the judging module is used for inputting the identification code information into the identification code information space for mapping and corresponding, and judging whether a mapping result exists or not;
a first identification result obtaining module, configured to, if the mapping result is the first identification result, take the mapping result as the first identification result;
and the first execution module is used for not obtaining the first identification result if the first identification result is not obtained.
Further, the system further comprises:
the type information determining module is used for acquiring types of the multiple types of goods according to identification codes of the multiple types of goods to obtain multiple types of information;
the batch information determining module is used for acquiring batches of the multiple types of goods according to the identification codes of the multiple types of goods to acquire multiple pieces of batch information;
the entity information determining module is used for acquiring a plurality of entity information according to the goods of the plurality of types;
the first attribute information determining module is used for obtaining a first attribute and a plurality of first attribute value information according to the plurality of types of information;
the second attribute determining module is used for obtaining a second attribute and a plurality of second attribute value information according to the plurality of batch information;
an identification code information space determination module to construct the identification code information space based on the plurality of entity information, the first attribute, the plurality of first attribute value information, the second attribute, and the plurality of second attribute value information, based on a knowledge graph.
Further, the system further comprises:
a preset angle determination module for obtaining a plurality of preset angles;
the angle image information determining module is used for acquiring and acquiring a plurality of angle image information of the goods to be sorted through the image acquisition device based on the plurality of preset angles;
an image information determination module to take the plurality of angle image information as the image information.
Further, the system further comprises:
a second execution module to construct the image recognition model;
the stability verification module is used for verifying the stability of the image recognition model, and if the stability meets the preset requirement, the image recognition model is put into use;
and the third execution module is used for inputting the image information into the image recognition model which is put into use to obtain the second recognition result.
Further, the system further comprises:
the system comprises a sample image information set determining module, a sample image information set determining module and a data processing module, wherein the sample image information set determining module is used for acquiring image information of multiple types of cargos and acquiring a sample image information set;
the sample sorting category information set determining module is used for acquiring and acquiring sorting category information of the multiple types of goods to acquire a sample sorting category information set;
a data set determining module, configured to divide and combine the sample image information set and the sample sorting category information set to obtain a first data set and a second data set;
a fourth execution module for constructing an image recognition model based on a deep convolutional neural network;
the first division identification module is used for dividing the first data set and identifying data to obtain a training set, a verification set and a test set;
the supervised training module is used for carrying out supervised training on the image recognition model by adopting the training set until convergence or accuracy reaches a preset threshold value;
and the verification test module is used for verifying and testing the image identification model by adopting the verification set and the test set, and if the accuracy rate meets the preset threshold value, the image identification model is obtained.
Further, the system further comprises:
the second division identification module is used for dividing and identifying the second data set to obtain a first stability verification set and a second stability verification set;
a sorting category set determination module, configured to input the first stability verification set and the second stability verification set into the image recognition model, respectively, obtain a plurality of first recognition results and a plurality of second recognition results, respectively, and obtain a first sorting category set and a second sorting category set;
a fifth execution module to calculate a stability of the image recognition model from the first and second sets of sort categories by:
wherein, P i Is the proportion, Q, of the ith sort category information in the first sort category set i And n is the proportion of the ith sorting category information in the second sorting category set, and is the category number of the sorting category information.
The application provides a cargo intelligent sorting method based on dual recognition, wherein the method is applied to a cargo intelligent sorting system based on dual recognition, and the method comprises the following steps: acquiring identification code information on goods to be sorted through a code scanning device; inputting the identification code information into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not; if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted; in the transmission process, the image acquisition device is adopted to acquire and obtain the image information of the goods to be sorted; inputting the image information into a pre-constructed image recognition model to obtain a second recognition result; and sorting the goods to be sorted according to the sorting order. The technical problems that in the prior art, the accuracy of goods identification is insufficient, the automation degree is not high, and then the quality and the efficiency of goods sorting are not high are solved. The accuracy and the degree of automation of goods identification are improved, and the quality and the efficiency of goods sorting are improved; simultaneously, improve intelligent, the scientificity of goods letter sorting, reduce the cost of goods letter sorting, it is extravagant to reduce the manpower that the goods letter sorting caused, high-efficient, quick completion goods letter sorting's technological effect.
All possible combinations of the technical features in the above embodiments may not be described for the sake of brevity, but should be considered as being within the scope of the present disclosure as long as there is no contradiction between the combinations of the technical features.
The specification and drawings are merely illustrative of the present application, and it is intended that the present invention covers the modifications and variations of this invention provided they come within the scope of the invention and its equivalents.
Claims (8)
1. The intelligent cargo sorting method based on double recognition is applied to an intelligent cargo sorting system based on double recognition, the system comprises a code scanning device and an image acquisition device, and the method comprises the following steps:
conveying goods to be sorted;
in the transmission process, the code scanning device is adopted to acquire and obtain identification code information on the goods to be sorted;
inputting the identification code information into a pre-constructed identification code information space, and judging whether a first identification result is obtained or not;
if so, sorting the goods to be sorted according to the first recognition result, and if not, continuously transmitting the goods to be sorted;
in the transmission process, the image acquisition device is adopted to acquire and obtain the image information of the goods to be sorted;
inputting the image information into a pre-constructed image recognition model to obtain a second recognition result;
and sorting the goods to be sorted according to the second recognition result.
2. The method of claim 1, wherein inputting the identification code information into a pre-constructed identification code information space, and determining whether a first recognition result is obtained comprises:
constructing the identification code information space;
inputting the identification code information into the identification code information space for mapping correspondence, and judging whether a mapping result exists or not;
if so, taking the mapping result as the first identification result;
if not, the first identification result is not obtained.
3. The method of claim 2, wherein constructing the identification code information space comprises:
acquiring types of the multiple types of goods according to identification codes of the multiple types of goods to acquire multiple types of information;
acquiring batches of the various goods according to the identification codes of the various goods to acquire a plurality of batches of information;
obtaining a plurality of entity information according to the goods of the plurality of types;
obtaining a first attribute and a plurality of first attribute value information according to the plurality of types of information;
obtaining a second attribute and a plurality of second attribute value information according to the plurality of batch information;
and constructing the identification code information space based on the plurality of entity information, the first attribute, the plurality of first attribute value information, the second attribute and the plurality of second attribute value information and based on a knowledge graph.
4. The method according to claim 1, wherein the step of acquiring the image information of the goods to be sorted by using the image acquisition device comprises the steps of:
obtaining a plurality of preset angles;
acquiring and acquiring a plurality of angle image information of the goods to be sorted by the image acquisition device based on the plurality of preset angles;
and taking the angle image information as the image information.
5. The method of claim 1, wherein inputting the image information into a pre-constructed image recognition model comprises:
constructing the image recognition model;
verifying the stability of the image recognition model, and if the stability meets the preset requirement, putting the image recognition model into use;
and inputting the image information into the image recognition model which is put into use to obtain the second recognition result.
6. The method of claim 5, wherein constructing the image recognition model comprises:
acquiring image information of multiple types of cargos to obtain a sample image information set;
collecting and obtaining sorting category information of the multiple types of goods to obtain a sample sorting category information set;
dividing and combining the sample image information set and the sample sorting category information set to obtain a first data set and a second data set;
constructing an image recognition model based on a deep convolutional neural network;
dividing the first data set and carrying out data identification to obtain a training set, a verification set and a test set;
carrying out supervision training on the image recognition model by adopting the training set until convergence or accuracy reaches a preset threshold;
and verifying and testing the image recognition model by adopting the verification set and the test set, and if the accuracy meets the preset threshold, obtaining the image recognition model.
7. The method of claim 6, wherein verifying the stability of the image recognition model comprises:
dividing and data identifying the second data set to obtain a first stability verification set and a second stability verification set;
inputting the first stability verification set and the second stability verification set into the image recognition model respectively, obtaining a plurality of first recognition results and a plurality of second recognition results respectively, and obtaining a first sorting category set and a second sorting category set;
calculating a stability of the image recognition model from the first and second sorted set of categories by:
wherein, P i Is the proportion of the ith sort category information in the first sort category set, Q i And n is the proportion of the ith sorting category information in the second sorting category set, and is the category number of the sorting category information.
8. The utility model provides a goods intelligence letter sorting system based on dual discernment which characterized in that, the system is including sweeping yard device and image acquisition device, the system still includes:
the conveying module is used for conveying goods to be sorted;
the code scanning module is used for acquiring and acquiring identification code information on the goods to be sorted by adopting the code scanning device in the transmission process;
the judging module is used for inputting the identification code information into a pre-constructed identification code information space and judging whether a first identification result is obtained or not;
the sorting and conveying selection module is used for sorting the goods to be sorted according to the first recognition result if the goods to be sorted are in the first recognition result, and continuously conveying the goods to be sorted if the goods to be sorted are not in the first recognition result;
the image acquisition module is used for acquiring and acquiring image information of the goods to be sorted by adopting the image acquisition device in the transmission process;
the input module is used for inputting the image information into a pre-constructed image recognition model to obtain a second recognition result;
and the sorting module is used for sorting the goods to be sorted according to the second recognition result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211026261.1A CN115463844A (en) | 2022-08-25 | 2022-08-25 | Intelligent cargo sorting method and system based on dual recognition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211026261.1A CN115463844A (en) | 2022-08-25 | 2022-08-25 | Intelligent cargo sorting method and system based on dual recognition |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115463844A true CN115463844A (en) | 2022-12-13 |
Family
ID=84369481
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211026261.1A Pending CN115463844A (en) | 2022-08-25 | 2022-08-25 | Intelligent cargo sorting method and system based on dual recognition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115463844A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115953635A (en) * | 2023-03-10 | 2023-04-11 | 中国邮电器材集团有限公司 | Multi-category target object sorting method, AR glasses and system |
| CN117772616A (en) * | 2024-02-23 | 2024-03-29 | 苏州卓晟裕智能科技有限公司 | Automatic sorting method and system of intelligent logistics robot |
-
2022
- 2022-08-25 CN CN202211026261.1A patent/CN115463844A/en active Pending
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115953635A (en) * | 2023-03-10 | 2023-04-11 | 中国邮电器材集团有限公司 | Multi-category target object sorting method, AR glasses and system |
| CN117772616A (en) * | 2024-02-23 | 2024-03-29 | 苏州卓晟裕智能科技有限公司 | Automatic sorting method and system of intelligent logistics robot |
| CN117772616B (en) * | 2024-02-23 | 2024-05-14 | 苏州卓晟裕智能科技有限公司 | Automatic sorting method and system of intelligent logistics robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115463844A (en) | Intelligent cargo sorting method and system based on dual recognition | |
| CN109949290A (en) | Pavement crack detection method, device, equipment and storage medium | |
| CN111488874A (en) | Method and system for correcting inclination of pointer instrument | |
| CN112801094B (en) | Pointer instrument image inclination correction method | |
| CN111598827A (en) | Appearance flaw detection method, electronic device and storage medium | |
| CN116343228A (en) | Intelligent reading method and system for water meter | |
| CN107038444A (en) | A kind of image-recognizing method of pointer dial plate | |
| CN117372515B (en) | Self-adaptive deviation rectifying control system | |
| CN110852164A (en) | YOLOv 3-based method and system for automatically detecting illegal building | |
| CN110935646A (en) | Full-automatic crab grading system based on image recognition | |
| CN102054161B (en) | Fingerprint image inspection method | |
| CN116403223A (en) | Pointer instrument reading identification method and system based on machine learning | |
| CN112802018B (en) | Integrity detection method, device and equipment for segmented circular workpiece and storage medium | |
| CN115512343A (en) | Method for correcting and recognizing reading of circular pointer instrument | |
| CN110672631B (en) | Panel defect photographing method and panel defect photographing device | |
| CN207181307U (en) | A kind of stereoscopic image acquisition device | |
| CN112001388A (en) | Method for detecting circular target in PCB based on YOLOv3 improved model | |
| CN118411576B (en) | Carton classification method and device based on data processing | |
| CN110246133B (en) | Corn kernel classification method, device, medium and equipment | |
| CN116109080B (en) | Building integrated management platform based on BIM and AR | |
| CN117349734B (en) | Water meter equipment identification method and device, electronic equipment and storage medium | |
| CN118429728B (en) | Chemical fertilizer particle detection method based on small sample learning | |
| CN116524004B (en) | Method and system for detecting size of steel bar based on HoughLines algorithm | |
| CN117788020B (en) | Intelligent agriculture traceability management cloud platform based on Internet of things | |
| CN117583281B (en) | Plate robot sorting and stacking optimization method, control system and sorting production line |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |