CN115965900A - Express item identification and search system and method - Google Patents
Express item identification and search system and method Download PDFInfo
- Publication number
- CN115965900A CN115965900A CN202310220757.0A CN202310220757A CN115965900A CN 115965900 A CN115965900 A CN 115965900A CN 202310220757 A CN202310220757 A CN 202310220757A CN 115965900 A CN115965900 A CN 115965900A
- Authority
- CN
- China
- Prior art keywords
- express
- image
- identification
- color
- data
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 64
- 238000002372 labelling Methods 0.000 claims abstract description 29
- 238000004364 calculation method Methods 0.000 claims abstract description 28
- 230000008569 process Effects 0.000 claims abstract description 24
- 238000012544 monitoring process Methods 0.000 claims description 46
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000005457 optimization Methods 0.000 claims description 11
- 239000003086 colorant Substances 0.000 claims description 4
- 238000012546 transfer Methods 0.000 abstract description 3
- 238000013527 convolutional neural network Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000013139 quantization Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 230000032258 transport Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a system and a method for identifying and searching express mails, which comprises the following steps: the scanning labeling module scans each express to obtain a characteristic image and pastes a characteristic label; the shelving module transfers each express to a shelf layer to distribute corresponding shelf information; the image acquisition unit acquires the overall image of each express according to the shelf information; the first processing unit analyzes and obtains the identification image and the label image of each express, and processes to obtain a color characteristic index, and the identification image is output when the color characteristic index is larger than a first threshold value; the second processing unit inputs each identification image and the characteristic image into the similarity calculation model to obtain image similarity, and marks the identification image as a difference image when the image similarity is greater than a second threshold value; and the express identification unit shoots the express at a short distance to obtain express label information, processes the express label information to obtain an information similarity value, and outputs the difference image when the information similarity value is greater than a third threshold value. The invention improves the efficiency and convenience of searching express items in the express post house.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a system and a method for identifying and searching express mails.
Background
The express courier station is a station for express parcel pickup/collection and delivery, and the main business is to provide express pickup and collection business. When the client does not have time or is inconvenient to take the express delivery by himself, the express delivery post station can be used for collection, the express delivery post station can be helped to be stored for a period of time, and the express delivery post station can be taken when the client is empty. The inside a plurality of goods shelves that are provided with of express delivery post house divide into a plurality of goods shelves layers that are used for placing the express mail on the goods shelves. When the courier puts the express on the goods shelf, the express can be distributed with information on putting on the shelf, the express is registered into a code scanning system and put in storage, and the information on putting on the shelf comprises a goods shelf number, a goods shelf layer number and a code number. The express company sends the information on the shelf to the mobile phone of the receiver, and the receiver gets the express according to the information on the shelf to the express post station. And after finding the express according to the received position information of the express, the addressee places the express on an express code recognition machine for recognition and takes away the express. However, due to the fact that the express on the goods shelves is many, even if the addressees receive information and are informed of which serial number on which goods shelf and which layer, the goods shelf layer is required to be turned over for searching when the addressees find the express on the goods shelf layer, the searching process is very difficult, the searching efficiency is extremely low, and a large amount of time and energy of the person who takes the express are wasted.
Disclosure of Invention
Aiming at the defects in the prior art, the invention aims to provide a express item identification and search system and a express item identification and search method, which are used for improving the efficiency and convenience of an express item taker in searching express items in an express post station.
In order to achieve the purpose, the invention provides the following technical scheme: an express item identification and search system comprising:
the scanning and labeling module is used for scanning to obtain a characteristic image before each express item enters an express post house and pasting a characteristic label on each express item according to the characteristic image;
the shelf loading module is connected with the scanning and labeling module and used for distributing shelf information for each express and transferring each express to a shelf layer;
the express mail search module is connected the module of putting on the shelf includes:
the image acquisition unit is used for acquiring the overall image of each express item according to the shelf information; the first processing unit is connected with the image acquisition unit and used for analyzing and obtaining the identification image and the corresponding label image of each express according to the overall image, processing and obtaining a color characteristic index according to the label image and outputting the identification image when the color characteristic index is larger than a first threshold value;
the second processing unit is connected with the first processing unit and used for inputting each identification image and the characteristic image into a similarity calculation model to obtain image similarity, and when the image similarity is larger than a second threshold value, the identification images are marked as difference images;
and the express identification unit is connected with the second processing unit and used for shooting and analyzing the express close range corresponding to each difference image to obtain express label information, processing the express label information and standard express information to obtain an information similarity score, and outputting the difference image when the information similarity score is larger than a third threshold value.
Further, the scanning labeling module comprises:
the scanning unit is used for scanning the color, the shape and the volume of each express, so as to obtain color data, shape data and volume data to form the characteristic image;
the identification unit is connected with the scanning unit and used for identifying the color data in the characteristic image and inputting the color data into a background database, and a plurality of historical color data are stored in a plurality of storage sections of the background database in advance;
the color difference calculating unit is connected with the identifying unit and is used for analyzing and obtaining the colorimetric values of the color data and the colorimetric values of the historical color data, and calculating the color difference between the colorimetric values of the color data and the colorimetric values of the historical color data according to a preset color difference calculating formula to obtain corresponding color difference values;
and the labeling unit is connected with the color difference calculating unit and used for storing the color data into the storage interval corresponding to the historical color data when the color difference value is smaller than a preset lower limit threshold value, pasting the feature label corresponding to the historical color data in the feature identification area of the express, storing the color data into the newly-opened storage interval in the background database when the color difference value is larger than a preset upper limit threshold value, and pasting the new feature label in the feature identification area of the express.
Further, the color difference calculation formula is configured to:
wherein ,
for representing the color difference value;
Furthermore, express labels are pre-pasted on the express items, and the shelving module places the characteristic identification areas of the express items on the same shelf layer on the same side in the process of transferring the express items to the shelf layer of the shelf.
Further, the first processing unit includes:
the first identification subunit is used for identifying each express mail in the overall image according to a preset image identification algorithm to obtain each identification image;
the second identification subunit is connected with the first identification subunit and used for identifying the colors of the label images contained in the identification images to obtain chromatic values of the label images, and then processing the chromatic values of the label images according to a preset feature processing strategy to obtain corresponding color feature indexes;
and the first comparison subunit is connected with the second identification subunit and is used for comparing each color feature index with the first threshold respectively, outputting the identification image as a final image when the color feature index is greater than the first threshold, and sending each identification image to the second processing unit when the color feature index is not greater than the first threshold.
Further, the feature processing policy includes: distributing corresponding color characteristic indexes to a plurality of initial chromatic value ranges, wherein each initial chromatic value range forms a complete color gamut, and distributing corresponding color characteristic indexes to the chromatic values of the label images when the chromatic values fall into the initial chromatic value ranges;
the first processing unit further includes an adjusting subunit, connected to the second identifying subunit, and configured to dynamically adjust a ratio between the initial chromaticity value ranges according to the number of the plurality of historical color data in the plurality of storage intervals of the background database to obtain a plurality of optimized chromaticity value ranges, where a ratio between the optimized chromaticity value ranges is consistent with a ratio between the number of the plurality of historical color data in the plurality of storage intervals.
Furthermore, a plurality of historical shape data and a plurality of historical volume data are stored in a plurality of storage intervals of the background database, the historical color data, the historical shape data and the historical volume data of the same express are associated with each other, and the second processing unit comprises:
a first calculating subunit, configured to respectively allocate corresponding initial coefficients to the color data, the shape data, and the volume data in the feature image, and input each initial coefficient into a preset threshold calculation formula to obtain the second threshold;
a third identifying subunit, connected to the first calculating subunit, configured to perform corresponding adjustment on each initial coefficient according to a ratio of the color data, the shape data, and the volume data in the feature image in the historical color data, the historical shape data, and the historical volume data, respectively, to obtain a corresponding optimization coefficient, where a ratio of each optimization coefficient is inversely proportional to a ratio of the color data, the shape data, and the volume data in the historical color data, the historical shape data, and the historical volume data, respectively, and then dynamically update the second threshold according to each optimization coefficient;
and the processing subunit is connected with the third identifying subunit, and is configured to input each of the identified images and the feature image into the similarity calculation model to obtain the image similarity, compare the image similarity with the dynamically updated second threshold, and mark the identified image as the difference image when the image similarity is greater than the dynamically updated second threshold.
Further, the standard express delivery information includes a plurality of standard express delivery data, and the express delivery identification unit includes:
the shooting subunit is used for performing short-distance shooting on the express labels on the express mails corresponding to the difference images to obtain corresponding express label images;
the information identification subunit is connected with the shooting subunit and is used for identifying the text content on each express label image to obtain the express label information, and the express label information comprises a plurality of express data;
the second calculating subunit is connected with the information identifying subunit and used for calculating text similarity between each express data and the corresponding standard express data according to a preset text similarity calculation method and further calculating the information similarity score according to each text similarity;
the second comparison subunit is connected with the second calculation subunit and is used for comparing the information similarity score with the third threshold, marking the difference image as the final image to be output when the information similarity score is larger than the third threshold, and generating a monitoring calling instruction when the information similarity score is not larger than the third threshold;
the monitoring analysis subunit is connected with the second comparison subunit and used for calling a first monitoring video which is over against the shelf layer in the process of transferring the express mail to the shelf layer according to the monitoring calling instruction and marking a placement area of the express mail on the shelf layer in a plurality of continuous image frames of the first monitoring video;
and the marking subunit is connected with the monitoring and analyzing subunit and is used for introducing an initial marking model, taking a plurality of continuous image frames as input, taking the placement area in each continuous image frame as output, retraining the initial marking model to obtain an area marking model, and further inputting the overall image into the area marking model to obtain a predicted placement area.
Furthermore, the express identification unit further comprises a ground monitoring subunit, which is respectively connected to the second comparison subunit and is used for calling a second monitoring video facing the ground and a third monitoring video facing the tops of the shelf layers in the process of transporting the express to the shelf layers according to the monitoring calling instruction, respectively subtracting the first appearance moment of the express on the ground in the data frame of the second monitoring video and the falling moment of the express in the data frame of the third monitoring video to obtain a falling time difference, and associating the position area of the express on the ground with the predicted falling area on the corresponding shelf layer when the falling time difference is smaller than a preset time threshold.
An express mail identification and search method is applied to the express mail identification and search system, and comprises the following steps:
s1, scanning each express item by a scanning and labeling module before the express item enters an express post station to obtain a characteristic image of each express item, and pasting a characteristic label on each express item according to the characteristic image;
s2, distributing corresponding shelf information for each express item by a shelf loading module and transferring each express item to a shelf layer of a shelf, wherein the shelf information is associated with the corresponding characteristic image and standard express information;
s3, an image acquisition unit acquires the overall image of each express item placed on the corresponding shelf layer according to the shelf information;
s4, a first processing unit analyzes the overall image to obtain an identification image of each express and a corresponding label image, processes the identification image according to the label image to obtain a color characteristic index, and outputs the identification image as a final image when the color characteristic index is larger than a first threshold value;
step S5, when the color characteristic index is not larger than a first threshold value, a second processing unit inputs each identification image and the characteristic image into a preset similarity calculation model to obtain image similarity, and when the image similarity is larger than a second threshold value, the identification images are marked as difference images;
and S6, performing short-distance shooting analysis on the express corresponding to each difference image by an express identification unit to obtain express label information, processing the express label information and standard express information to obtain information similarity values, and marking the difference images as final images to be output when the information similarity values are larger than a third threshold value.
The invention has the beneficial effects that:
according to the express post station, express is scanned to obtain the characteristic image before the express enters the express post station, and meanwhile, the characteristic label is pasted on the express, so that subsequent identification is facilitated; the express item searching method comprises the steps of transferring each express item to a shelf layer, collecting an overall image, analyzing the overall image to obtain an identification image and a label image, processing the label image to obtain a color label index, outputting the identification image of the express item as a final image when the color label index is larger than a first threshold value, and realizing fast searching of the express item only according to the label image of the express item, so that the express item searching efficiency is improved;
meanwhile, when the color label index is not larger than the first threshold value, the image similarity between the characteristic image and the identification image of the express is calculated, when the image similarity is larger than the second threshold value, the characteristic label is shot in a short distance mode, the text content is identified to obtain express label information, the information similarity value between the express label information and the standard express label information is obtained through processing, and finally when the information similarity value is larger than the third threshold value, the difference image is output as a final image, so that the express identification accuracy is guaranteed, the express searching efficiency and convenience are improved, and the popularization is facilitated.
Drawings
FIG. 1 is a schematic diagram of a system for identifying and searching for express items according to the present invention;
FIG. 2 is a diagram of a method for identifying and searching express items according to the present invention.
Reference numerals: 1. scanning and labeling the module; 11. a scanning unit; 12. a color recognition unit; 13. a color difference calculation unit; 14. a labeling unit; 2. a racking module; 3. an express item searching module; 31. an image acquisition unit; 32. a first processing unit; 321. a first identification subunit; 322. a second identifier subunit; 323. a first comparison subunit; 324. an adjustment subunit; 33. a second processing unit; 331. a first calculation subunit; 332. a third identifier subunit; 333. a processing subunit; 34. an express item identification unit; 341. a shooting subunit; 342. an information identification subunit; 343. a second calculation subunit; 344. a second comparison subunit; 345. a monitoring and analyzing subunit; 346. a tagging subunit; 347. a ground monitoring subunit.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples. In which like parts are designated by like reference numerals. It should be noted that the terms "front," "back," "left," "right," "upper" and "lower" used in the following description refer to directions in the drawings, and the terms "bottom" and "top," "inner" and "outer" refer to directions toward and away from, respectively, the geometric center of a particular component.
As shown in fig. 1, an express item identification and search system of the present embodiment includes:
the scanning and labeling module 1 is used for scanning each express item before the express item enters the express post station to obtain a characteristic image of each express item, and pasting a characteristic label on each express item according to the characteristic image;
the shelving module 2 is connected with the scanning and labeling module 1 and used for distributing corresponding shelf information for each express and transferring each express to a shelf layer of a shelf, and the shelf information is associated with corresponding characteristic images and standard express information;
express mail searching module 3 connects upper bracket module 2, includes:
the image acquisition unit 31 is used for acquiring the overall image of each express item placed on the corresponding shelf layer according to the shelf information;
the first processing unit 32 is connected with the image acquisition unit 31, and is used for obtaining the identification image of each express and the label image of the corresponding characteristic label according to the overall image analysis, obtaining a color characteristic index according to the label image processing, and outputting the identification image as a final image when the color characteristic index is larger than a first threshold value;
the second processing unit 33 is connected to the first processing unit 32, and is configured to input each recognition image and the feature image into a preset similarity calculation model to obtain an image similarity when the color feature index is not greater than the first threshold, and mark the recognition image as a difference image when the image similarity is greater than the second threshold;
and the express identification unit 34 is connected to the second processing unit 33, and is configured to perform short-distance shooting and analysis on the express corresponding to each difference image to obtain express label information, process the express label information and the standard express information to obtain an information similarity score, and mark the difference image as a final image when the information similarity score is greater than a third threshold value.
In particular, in an embodiment, the scanning and labeling module 1 may be a device integrated with both the 3D scanning and imaging function and the labeling function. The scanning labeling module 1 carries out 3D scanning on each express item before the express item enters an express post station to obtain a characteristic image, wherein the characteristic image is a three-dimensional image containing color data, shape data and volume data. And simultaneously, the scanning labeling module 1 pastes corresponding characteristic labels on the express mails according to the color data in the characteristic images. Preferably, express labels are pre-pasted on the express items, the shelf loading module 2 places feature identification areas of the express items on the same shelf layer at the same side in the process of transferring the express items to the shelf layer of the shelf, the image acquisition unit 31 can conveniently acquire label images in the image acquisition process, and the problem that the express item searching efficiency is influenced because label images of individual express items are not acquired is avoided. The racking module 2 may be a mechanical conveying device, such as a conveyor belt, or a transfer robot. After the scanning labeling module 1 finishes pasting the characteristic labels on the express mails, the shelving module 2 firstly distributes shelf information for each express mail, wherein the shelf information comprises a shelf number, a shelf layer number and an express code number, and the shelving module 2 transports and places the express mails on the shelf layers of the corresponding shelves according to the shelf information.
The goods shelf information is pre-associated with the characteristic image of the same express and standard express information, and the standard express information is obtained when the express is scanned and labeled by recognizing the text content on the express label pasted on the express. The express item searching module 3 may be a mobile terminal of the person taking the item, and the mobile terminal includes, but is not limited to, a mobile phone, a notebook computer, and a tablet computer. And the mobile terminal is pre-provided with express item searching software. And the express item searching software is used for receiving the shelf information sent by the shelf loading module 2 and calling the image acquisition unit 31 on the mobile terminal to acquire the overall image of the corresponding shelf layer according to the shelf information. Wherein the image acquisition unit 31 may be a camera module on the mobile terminal. The first processing unit 32, the second processing unit 33 and the express identification unit 34 may be processing chips on the mobile terminal. The first processing unit 32 invokes an image recognition algorithm to analyze the overall image to obtain a recognition image of each express and a label image of the feature label pasted on the express, and then obtains a color feature index according to the label image processing, when the color feature index is greater than a first threshold, it indicates that the color of the label image is very rare, and the first threshold may be 70. In a particular embodiment, the color of the package is red, the characteristic index of the color of red is particularly high at 99, indicating that the color is scarce or never appeared, so that a characteristic label, which is also red, is regenerated from the color of the package, the color of the characteristic label corresponding to the color of the express mail. Because the color development is very rare, the color can be directly used as the basis of express mail identity identification, and then the identification image is used as a final image to be output in a green color on the quick search software, so that the express mail is quickly searched at present, and the express mail search efficiency is improved. When the color feature index is not greater than the first threshold value, it indicates that the color of the feature image is not scarce. In a specific embodiment, the color of the express package is khaki, so that the color of the corresponding characteristic label is khaki, and the color characteristic index of khaki is very small and 10, so that the khaki cannot be used as a basis for quickly identifying the identity of the express. At this time, the characteristic image can be further analyzed to determine the identity of the express: the second processing unit 33 inputs the feature images of the express items associated with the identification images and the shelf information into the similarity calculation model, so that the similarity calculation model outputs the image similarity, marks the corresponding identification image as a red difference image when the image similarity is greater than a second threshold, and does not mark the identification image when the image similarity is not greater than the second threshold. For the difference image marked with red color, the express identification unit 34 performs short-distance shooting on the express label on the express corresponding to the difference image, and identifies the text content on the express label to obtain the information of the express label. And then calculating the similarity between each express label information and the standard express information to obtain a corresponding information similarity score. And finally, comparing the information similarity score with a third threshold value, and when the information similarity score is larger than the third threshold value, indicating that the similarity between the express label information of the express and the standard express information of the scanned express is extremely high, so that the final image with the difference image corresponding to the express marked as green is output. The express item taking person takes the corresponding express item according to the final image displayed on the display screen of the mobile terminal, the technical scheme realizes the layer-by-layer progressive comparison and analysis of the identification image and the label image of the express item, improves the efficiency and convenience of express item searching while ensuring the express item identification accuracy, and is favorable for popularization.
Preferably, the scanning labelling module 1 comprises:
the scanning unit 11 is used for scanning the color, shape and volume of each express to obtain color data, shape data and volume data so as to form a characteristic image;
the color identification unit 12 is connected with the scanning unit 11 and is used for identifying color data in the characteristic image and recording the color data into a background database, and a plurality of historical color data are stored in a plurality of storage sections of the background database in advance;
the color difference calculating unit 13 is connected to the color identifying unit 12, and is configured to analyze the chrominance values of the obtained color data and the chrominance values of the historical color data, and calculate a color difference between the chrominance values of the color data and the chrominance values of the historical color data according to a preset color difference calculation formula to obtain a corresponding color difference value;
and the labeling unit 14 is connected with the color difference calculating unit 13 and is used for storing the color data into a storage interval to which corresponding historical color data belongs when the color difference value is smaller than a preset lower limit threshold value, pasting a feature label corresponding to the corresponding historical color data in the feature identification area of the express, storing the color data into a newly-arranged storage interval in the background database when the color difference value is larger than a preset upper limit threshold value, and pasting a new feature label in the feature identification area of the express.
Specifically, in this embodiment, the scanning unit 11 scans the color, shape, and volume of each express item entering the courier post to obtain a three-dimensional characteristic image including color data, shape data, and volume data. The color identification unit 12 is configured to identify color data in the feature image according to an orb algorithm in opencv, and record each identified color data into a storage interval in the background database to serve as newly warehoused color data, where the newly warehoused color data will become historical color data after being stored for a preset time period, where the preset time period may be one hour. The color difference calculating unit 13 calculates the difference between the colorimetric values of the newly stored color data and the colorimetric values of the historical color data in the storage sections to obtain a plurality of color difference values. The lower threshold may be 0.1 and the upper threshold may be 0.8. And when the color difference value is less than 0.1, storing the color data into a storage interval corresponding to the historical color data subjected to color difference calculation, and pasting a feature label corresponding to the color of the historical color data in a feature identification area of the express. The characteristic identification area comprises the intersection of two adjacent surfaces of the express mail, or an arc transition area, or a breakage area, or a crease area. The difference between the pixel points of the area is larger than that of the area on the plane, so that the image characteristics are more obvious and the identification is easy. When the color difference value is greater than 0.8, it indicates that the color difference between the current color data and each historical color data in the background database is large, so a new storage interval needs to be set for the current color data, and meanwhile, the labeling unit 14 regenerates a new feature label according to the color data and pastes the new feature label into the feature identification area of the express, and the color of the new feature label is consistent with the color of the color data.
Preferably, the color difference calculation formula is configured to:
wherein ,
for representing a color difference value;
Specifically, in this embodiment, the first chromaticity value may be a black-and-white chromaticity value, the red-green chromaticity value may be a second chromaticity value, and the yellow-blue chromaticity value may be a third chromaticity value.
Preferably, the first processing unit 32 comprises:
a first identification subunit 321, configured to identify each express item in the overall image according to a preset image identification algorithm to obtain each identification image;
the second identifying subunit 322 is connected to the first identifying subunit 321, and is configured to identify colors of the label images included in the identification images to obtain chromatic values of the label images, and further process the chromatic values of the label images according to a preset feature processing policy to obtain corresponding color feature indexes;
the first comparing subunit 323, connected to the second identifying subunit 322, is configured to compare each color feature index with the first threshold, respectively, and output the identification image as a final image when the color feature index is greater than the first threshold, and send each identification image to the second processing unit 33 when the color feature index is not greater than the first threshold.
Specifically, in this embodiment, the first identifying subunit 321 identifies the overall image according to the convolutional neural network model trained in advance, inputs the overall image into the convolutional neural network model, and outputs the overall image in which the identification image of the express in the overall image is labeled. The second identifying subunit 322 obtains the marked identifying image from the overall image, and performs color value identification on the label image according to the orb algorithm in opencv, so as to obtain a corresponding color feature index through feature processing policy matching.
Preferably, the feature processing policy includes: distributing corresponding color characteristic indexes to a plurality of initial chromatic value ranges, wherein each initial chromatic value range forms a complete color gamut, and distributing corresponding color characteristic indexes to the chromatic values of the label images when the chromatic values fall into the initial chromatic value ranges;
the first processing unit 32 further includes an adjusting subunit 324, connected to the second identifying subunit 322, and configured to dynamically adjust a ratio between the initial chromaticity value ranges according to the number of the plurality of historical color data in the plurality of storage intervals of the background database to obtain a plurality of optimized chromaticity value ranges, where a ratio between the optimized chromaticity value ranges is consistent with a ratio between the number of the plurality of historical color data in the plurality of storage intervals.
Specifically, in this embodiment, the adjusting subunit 324 is configured to perform quantity statistics on the historical color data in each storage interval, and further adjust the ratio between the initial colorimetric value ranges according to the quantity of the historical color data in each storage interval to perform dynamic adjustment, so that each initial colorimetric value range is converted into a plurality of optimized colorimetric value ranges. The historical color data in each storage interval of the background database are continuously updated, so that the proportion between the optimized colorimetric value ranges is updated along with the change of the quantity of the historical color data in the storage intervals, the feature processing strategy is further updated, the color feature indexes corresponding to the optimized colorimetric value ranges are consistent with the color feature indexes corresponding to the initial colorimetric value ranges before optimization in the updating process, and the matching accuracy of the colorimetric values of the label image is continuously improved along with the adjustment of the optimized colorimetric value ranges because the optimized colorimetric value ranges are continuously adjusted along with the increase of the historical color data.
Preferably, a plurality of storage sections of the background database further store a plurality of historical shape data and a plurality of historical volume data, the historical color data, the historical shape data, and the historical volume data of the same express are associated with each other, and the second processing unit 33 includes:
the first calculating subunit 331, configured to respectively allocate corresponding initial coefficients to the color data, the shape data, and the volume data in the feature image, and input each initial coefficient into a preset threshold calculation formula to obtain a second threshold;
a third identifying subunit 332, connected to the first calculating subunit 331, configured to correspondingly adjust each initial coefficient according to the ratio of the color data, the shape data, and the volume data in the feature image to the historical color data, the historical shape data, and the historical volume data, respectively, to obtain a corresponding optimization coefficient, where the ratio of each optimization coefficient is inversely proportional to the ratio of the color data, the shape data, and the volume data to the historical color data, the historical shape data, and the historical volume data, respectively, and further dynamically update the second threshold according to each optimization coefficient;
the processing subunit 333 is connected to the third identifying subunit 332, and is configured to input each of the identified images and the feature image into the similarity calculation model to obtain an image similarity, compare the image similarity with the dynamically updated second threshold, and mark the identified image as a difference image when the image similarity is greater than the dynamically updated second threshold.
Specifically, in this embodiment, the threshold calculation formula is configured as follows:
wherein ,
for representing a second threshold value>
For representing color data, is greater than or equal to>
An initial coefficient for representing color data; />
For representing shape data->
Initial coefficients for representing shape data; />
For the purpose of representing the volume data,
initial coefficients for representing the volumetric data. The color data is a data value obtained by quantizing colors through a color quantization algorithm, the shape data is a data value obtained by analyzing and quantizing the shape in the image through the shape analysis plug-in, and the volume data is a data value obtained by quantizing and calculating the length, width, height and depth in the characteristic image through a volume algorithm. The first calculating subunit 331 brings the color data, the shape data, the volume data, and the corresponding initial coefficients into a threshold value calculation formula to obtain a second threshold value. The third identifying subunit 332 adjusts the initial coefficients of the color data, the shape data, and the volume data by adjusting the ratio of each of the color data, the shape data, and the volume data in the storage section, thereby adjusting the second threshold. With the continuous scanning of the scanning unit 11, the historical color data, the historical shape data and the historical volume data are changed continuously, so that the proportion of the color data, the shape data and the volume data in the historical color data, the historical shape data and the historical volume data respectively is changed continuously, and the initial coefficients are dynamically adjusted according to the proportion of the historical color data, the historical shape data and the historical volume data to obtain corresponding optimal valuesThe coefficient is changed, dynamic optimization adjustment of the second threshold value is achieved according to continuous iteration updating of color data, shape data and volume data in the background database, the second threshold value can be used as an image similarity judgment threshold value, the height and the accuracy are higher, and the accuracy of express item searching in the technical scheme is improved.
Preferably, the standard express delivery information includes a plurality of standard express delivery data, and the express item identification unit 34 includes:
the shooting subunit 341 is configured to perform short-distance shooting on the express labels on the express items corresponding to the difference images to obtain corresponding express label images;
the information identification subunit 342 is connected to the shooting subunit 341, and is configured to identify text content on each express label image to obtain express label information, where the express label information includes a plurality of express data;
the second calculating subunit 343, which is connected to the information identifying subunit 342, is configured to calculate, according to a preset text similarity algorithm, a text similarity between each piece of express delivery data and the corresponding standard express delivery data, and further calculate, according to each text similarity, an information similarity score;
a second comparing subunit 344, connected to the second calculating subunit 343, configured to compare the information similarity score with a third threshold, mark the difference image as a final image for output when the information similarity score is greater than the third threshold, and generate a monitoring call instruction when the information similarity score is not greater than the third threshold;
the monitoring analysis subunit 345 is connected to the second comparison subunit 344, and is configured to call, according to the monitoring call instruction, the first monitoring video that is over against the shelf layer in the process of transferring the express to the shelf layer, and mark a placement area of the express on the shelf layer in a plurality of continuous image frames of the first monitoring video;
the labeling subunit 346, connected to the monitoring and analyzing subunit 345, is configured to introduce an initial labeling model, take a plurality of consecutive image frames as input, take the placement area in each consecutive image frame as output, retrain the initial labeling model to obtain an area labeling model, and input the total image into the area labeling model to obtain a predicted placement area.
Specifically, in this embodiment, the information identification subunit 342 identifies the text content on the express label image according to an OCR character identification algorithm to obtain the express label information, where the express data in the express label information includes a sender name, a sender mobile phone number, a sender address, a recipient name, a recipient mobile phone number, a recipient address, and the like, and the standard express data is consistent with the content in the express data. And the second analysis subunit calculates the similarity between the express data and the standard express data according to a DSSM algorithm to obtain an information similarity score. The second comparing subunit 344 compares the information similarity score with a third threshold, and when the information similarity score is greater than the third threshold, it indicates that the express data distinguishing the currently-photographed express from the scanned express is extremely similar, and meanwhile, because the image similarity is also extremely high, the two express can be determined to be the same express, that is, the express can be marked green on the quick searching software, so that the express can be quickly searched and taken by the express taker. When the information similarity score is not greater than the third threshold, the express may be pushed by other express and cannot be seen, and at this time, the second comparing unit generates a monitoring call instruction, so that the monitoring analysis subunit 345 calls the first monitoring video of the express in the transfer process, and at the same time, manually marks and stores the placement area in the continuous image frame in which the express is placed in the first monitoring video. And finally, taking the continuous image frames marked with the placement area as data of a training set and a verification set, retraining the introduced FCN full convolution neural network model to obtain an area marking model, inputting the unmarked overall image into the area marking model to obtain a predicted placement area, displaying the area as blue on quick search software, facilitating a pick-up person to find the area, and improving the express mail search efficiency. In the process of finding, the second processing unit 33 continuously compares the image similarity with the second threshold, the express identification unit 34 continuously compares the information similarity with the third threshold after the image similarity is greater than the second threshold, and finally marks the difference image of the express as a final image after the information similarity is greater than the third threshold, so that the express can be quickly found.
Preferably, the express item identification unit 34 further includes a ground monitoring subunit 347, which is respectively connected to the second comparing subunit 344, and is configured to call, according to the monitoring call instruction, a second monitoring video directly facing the ground and a third monitoring video facing the tops of the shelf layers in the process of transferring the express item to the shelf layers, and respectively perform a difference between a time when the express item first appears on the ground in a data frame of the second monitoring video and a time when the express item drops in a data frame of the third monitoring video to obtain a drop time difference, and when the drop time difference is smaller than a preset time threshold, associate a position area of the express item on the ground with a predicted drop area on the corresponding shelf layer.
Specifically, in this embodiment, when the information similarity score is not greater than the third threshold, the express mail may also fall on the ground, so the ground monitoring subunit 347 is required to call the second monitoring video directly facing the ground and the third monitoring video facing the tops of the shelf layers in the process of transferring the express mail to the shelf layers, a drop time difference between a time when the express mail first appears on the ground and a time when the express mail falls in a data frame of the third monitoring video is small to a certain extent, and when the drop time difference is smaller than the time threshold, because there is a case where express mails at different positions simultaneously fall, it is further required to eliminate an influence of such a case, and a position area of the express mail on the ground is associated with a predicted drop area on the corresponding shelf layer, so when express mails on different shelf layers simultaneously fall on the ground, a distinction can be made through distances between the ground position of the express mail and the shelf layers, and a default association is made between the position area of the express mail on the ground and the predicted drop area of the shelf layers closest to the shelf layers. When the express item taking person takes the express item, the express item is not identified and is located in the predicted falling area on the shelf layer corresponding to the shelf information of the express item, and meanwhile, the express item is not pressed, so that the position area of the express item, which is related to the predicted falling area, on the ground can be highlighted on the quick searching software, the express item taking person can search quickly, and the express item searching efficiency is improved.
A method for identifying and searching a quick dispatch, applied to the system for identifying and searching a quick dispatch, as shown in fig. 2, includes:
step S1, scanning and labeling a module 1 to scan each express item before the express item enters an express post station to obtain a characteristic image of each express item, and pasting a characteristic label on each express item according to the characteristic image;
s2, distributing corresponding shelf information for each express and transferring each express to a shelf layer of a shelf by the shelf loading module 2, wherein the shelf information is associated with corresponding characteristic images and standard express information;
s3, the image acquisition unit 31 acquires the overall image of each express item placed on the corresponding shelf layer according to the shelf information;
step S4, the first processing unit 32 analyzes the overall image to obtain an identification image of each express and a corresponding label image, processes the identification image according to the label image to obtain a color characteristic index, and outputs the identification image as a final image when the color characteristic index is larger than a first threshold value;
step S5, when the color characteristic index is not larger than a first threshold value, the second processing unit 33 inputs each identification image and the characteristic image into a preset similarity calculation model to obtain image similarity, and when the image similarity is larger than a second threshold value, the identification image is marked as a difference image;
and S6, the express identification unit 34 performs short-distance shooting analysis on the express corresponding to each difference image to obtain express label information, processes the express label information and standard express information to obtain an information similarity score, and marks the difference image as a final image to be output when the information similarity score is larger than a third threshold value.
The above are only preferred embodiments of the present invention, and the scope of the present invention is not limited to the above examples, and all technical solutions that fall under the spirit of the present invention belong to the scope of the present invention. It should be noted that modifications and adaptations to those skilled in the art without departing from the principles of the present invention should also be considered as within the scope of the present invention.
Claims (10)
1. An express mail identification and search system, comprising:
the scanning and labeling module (1) is used for scanning to obtain a characteristic image before each express item enters an express post, and pasting a characteristic label on each express item according to the characteristic image;
the shelving module (2) is connected with the scanning and labeling module (1) and is used for distributing shelf information for each express and transferring each express to a shelf layer;
express mail search module (3), connect put on shelf module (2), include:
the image acquisition unit (31) is used for acquiring the overall image of each express according to the shelf information; the first processing unit (32) is connected with the image acquisition unit (31) and used for analyzing and obtaining the identification image and the corresponding label image of each express according to the overall image, processing and obtaining a color characteristic index according to the label image and outputting the identification image when the color characteristic index is larger than a first threshold value;
the second processing unit (33) is connected with the first processing unit (32) and used for inputting each identification image and the characteristic image into a similarity calculation model to obtain image similarity, and when the image similarity is larger than a second threshold value, the identification images are marked as difference images;
and the express identification unit (34) is connected with the second processing unit (33) and is used for performing close-range shooting and analysis on the express corresponding to each difference image to obtain express label information, processing the express label information and standard express information to obtain an information similarity score, and outputting the difference image when the information similarity score is larger than a third threshold value.
2. An express mail identification search system as claimed in claim 1, wherein: the scanning labeling module (1) comprises:
the scanning unit (11) is used for scanning the color, the shape and the volume of each express to obtain color data, shape data and volume data so as to form the characteristic image;
the color identification unit (12) is connected with the scanning unit (11) and is used for identifying the color data in the characteristic image and recording the color data into a background database, and a plurality of historical color data are stored in a plurality of storage sections of the background database in advance;
the color difference calculating unit (13) is connected with the color identifying unit (12) and is used for analyzing and obtaining the colorimetric values of the color data and the colorimetric values of the historical color data, and calculating the color difference between the colorimetric values of the color data and the colorimetric values of the historical color data according to a preset color difference calculating formula to obtain corresponding color difference values;
and the labeling unit (14) is connected with the color difference calculating unit (13) and is used for storing the color data into the storage interval corresponding to the historical color data when the color difference value is smaller than a preset lower limit threshold value, pasting the feature label corresponding to the historical color data in the feature identification area of the express mail, storing the color data into the storage interval newly arranged in the background database when the color difference value is larger than a preset upper limit threshold value, and pasting the new feature label in the feature identification area of the express mail.
3. An express item identification search system as claimed in claim 2, wherein the color difference calculation formula is configured to:
wherein ,
for representing a color difference value;
4. The express item identification search system of claim 2, wherein: express labels are pre-pasted on the express items, and the shelf loading module (2) places the characteristic identification areas of the express items on the same shelf layer on the same side in the process of transferring the express items to the shelf layer of the shelf.
5. The express item identification search system of claim 4, wherein: the first processing unit (32) comprises:
a first identification subunit (321) for identifying each express mail in the overall image according to a preset image identification algorithm to obtain each identification image;
a second identifying subunit (322), connected to the first identifying subunit (321), configured to identify colors of the label images included in each of the identification images to obtain chromatic values of the label images, and further process the chromatic values of the label images according to a preset feature processing policy to obtain corresponding color feature indexes;
a first comparison subunit (323) connected to the second identification subunit (322), configured to compare each of the color feature indexes with the first threshold, respectively, and output the identification image as a final image when the color feature index is greater than the first threshold, and send each of the identification images to the second processing unit (33) when the color feature index is not greater than the first threshold.
6. The express item identification search system of claim 5, wherein: the feature processing strategy comprises: distributing corresponding color characteristic indexes to a plurality of initial chromatic value ranges, wherein each initial chromatic value range forms a complete color gamut, and distributing corresponding color characteristic indexes to the chromatic values of the label images when the chromatic values fall into the initial chromatic value ranges;
the first processing unit (32) further comprises an adjusting subunit (324), connected to the second identifying subunit (322), configured to dynamically adjust a ratio between the initial chromaticity value ranges according to a number of the historical color data in a plurality of storage intervals of the background database to obtain a plurality of optimized chromaticity value ranges, where a ratio between the optimized chromaticity value ranges is consistent with a ratio between a number of the historical color data in a plurality of storage intervals.
7. An express mail identification search system as claimed in claim 2, wherein: a plurality of historical shape data and a plurality of historical volume data are also stored in a plurality of storage intervals of the background database, the historical color data, the historical shape data and the historical volume data of the same express are associated with one another, and the second processing unit (33) comprises:
a first calculating subunit (331) configured to allocate corresponding initial coefficients to the color data, the shape data, and the volume data in the feature image, respectively, and input each of the initial coefficients into a preset threshold calculation formula to obtain the second threshold;
a third identification subunit (332), connected to the first calculation subunit (331), configured to perform corresponding adjustment on each initial coefficient according to a ratio of the color data, the shape data, and the volume data in the feature image to the historical color data, the historical shape data, and the historical volume data, respectively, to obtain a corresponding optimization coefficient, where a ratio of each optimization coefficient is inversely proportional to a ratio of the color data, the shape data, and the volume data to the historical color data, the historical shape data, and the historical volume data, respectively, and further dynamically update the second threshold according to each optimization coefficient;
and the processing subunit (333) is connected to the third identifying subunit (332), and is configured to input each of the identified images and the feature image into the similarity calculation model to obtain the image similarity, compare the image similarity with the dynamically updated second threshold, and mark the identified image as the difference image when the image similarity is greater than the dynamically updated second threshold.
8. The express item identification search system of claim 4, wherein: if the standard express delivery information includes a plurality of standard express delivery data, the express delivery identification unit (34) includes:
a shooting subunit (341) configured to perform short-distance shooting on the express labels on the respective express items corresponding to the difference images to obtain corresponding express label images;
the information identification subunit (342) is connected with the shooting subunit (341) and is used for identifying the text content on each express label image to obtain the express label information, and the express label information contains a plurality of express data;
the second calculating subunit (343), which is connected to the information identifying subunit (342), is configured to calculate, according to a preset text similarity algorithm, a text similarity between each piece of express delivery data and the corresponding standard express delivery data, and further calculate, according to each piece of text similarity, the information similarity score;
a second comparing subunit (344), connected to the second calculating subunit (343), configured to compare the information similarity score with the third threshold, mark the difference image as the final image output when the information similarity score is greater than the third threshold, and generate a monitoring call instruction when the information similarity score is not greater than the third threshold;
a monitoring analysis subunit (345), connected to the second comparison subunit (344), configured to call, according to the monitoring call instruction, a first monitoring video that is over against the shelf layer in the process of transferring the express mail to the shelf layer, and mark a placement area of the express mail on the shelf layer in a plurality of consecutive image frames of the first monitoring video;
and the marking subunit (346) is connected with the monitoring and analyzing subunit (345) and is used for introducing an initial marking model, taking a plurality of continuous image frames as input, taking the placement area in each continuous image frame as output, retraining the initial marking model to obtain an area marking model, and further inputting the overall image into the area marking model to obtain a predicted placement area.
9. The express item identification search system of claim 8, wherein: the express item identification unit (34) further comprises a ground monitoring subunit (347) which is respectively connected with the second comparison subunit (344) and used for calling a second monitoring video which is over against the ground in the process of transporting the express item to the shelf layer and a third monitoring video which faces the top of each shelf layer according to the monitoring calling instruction, respectively subtracting the first appearance moment of the express item on the ground in a data frame of the second monitoring video and the falling moment of the express item in the data frame of the third monitoring video to obtain a falling time difference, and associating the position area of the express item on the ground with the predicted falling area on the corresponding shelf layer when the falling time difference is smaller than a preset time threshold.
10. An express mail identification and search method applied to the express mail identification and search system of any one of claims 1 to 9, comprising:
the method comprises the following steps that S1, a scanning and labeling module (1) scans all the express items before the express items enter an express post station to obtain characteristic images of the express items, and characteristic labels are pasted on the express items according to the characteristic images;
s2, distributing corresponding shelf information for each express item by the shelving module (2) and transferring each express item to a shelf layer of a shelf, wherein the shelf information is associated with the corresponding characteristic image and the standard express information;
s3, an image acquisition unit (31) acquires the overall image of each express item placed on the corresponding shelf layer according to the shelf information;
s4, a first processing unit (32) analyzes the overall image to obtain an identification image of each express and a label image of the corresponding characteristic label, processes the identification image according to the label image to obtain a color characteristic index, and outputs the identification image as a final image when the color characteristic index is larger than a first threshold value;
step S5, when the color characteristic index is not larger than a first threshold value, a second processing unit (33) inputs each identification image and the characteristic image into a preset similarity calculation model to obtain image similarity, and when the image similarity is larger than a second threshold value, the identification images are marked as difference images;
and S6, performing close-range shooting analysis on the express corresponding to each difference image by an express identification unit (34) to obtain express label information, processing the express label information and standard express information to obtain an information similarity score, and marking the difference image as a final image to be output when the information similarity score is larger than a third threshold value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310220757.0A CN115965900B (en) | 2023-03-09 | 2023-03-09 | Express item identification searching system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310220757.0A CN115965900B (en) | 2023-03-09 | 2023-03-09 | Express item identification searching system and method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN115965900A true CN115965900A (en) | 2023-04-14 |
| CN115965900B CN115965900B (en) | 2023-06-09 |
Family
ID=85888657
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310220757.0A Active CN115965900B (en) | 2023-03-09 | 2023-03-09 | Express item identification searching system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115965900B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116486346A (en) * | 2023-06-16 | 2023-07-25 | 济南雷圣自动化技术有限公司 | Carton conveying method and device |
| CN117854091A (en) * | 2024-01-15 | 2024-04-09 | 金锋馥(滁州)科技股份有限公司 | Method for extracting information of multi-surface dense labels of packages based on image feature detection |
| CN117854091B (en) * | 2024-01-15 | 2024-06-07 | 金锋馥(滁州)科技股份有限公司 | Method for extracting information of multi-surface dense labels of packages based on image feature detection |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102161040A (en) * | 2011-01-14 | 2011-08-24 | 北京交通大学 | Logistic sorting system based on color sensor |
| CN108090728A (en) * | 2017-12-28 | 2018-05-29 | 上海传英信息技术有限公司 | A kind of express delivery information input method and input system based on intelligent terminal |
| CN110674713A (en) * | 2019-08-23 | 2020-01-10 | 北京市商汤科技开发有限公司 | Express query method, express delivery warehouse-out method, express delivery pickup method and related device |
| CN111368730A (en) * | 2020-03-04 | 2020-07-03 | 上海东普信息科技有限公司 | Method, device and equipment for identifying abnormal actions of express sorting and storage medium |
| CN211294479U (en) * | 2020-03-07 | 2020-08-18 | 林昌全 | Back adhesive label paper |
| CN111709283A (en) * | 2020-05-07 | 2020-09-25 | 顺丰科技有限公司 | Method and device for detecting state of logistics piece |
| CN111767973A (en) * | 2020-07-02 | 2020-10-13 | 林昌全 | Design method of logistics management software function based on associated label |
| WO2021196836A1 (en) * | 2020-03-30 | 2021-10-07 | 北京沃东天骏信息技术有限公司 | Method and apparatus for positioning express parcel |
| CN113592376A (en) * | 2021-06-11 | 2021-11-02 | 中国农业大学 | Intelligent access information management system applied to express post house |
| CN114047750A (en) * | 2021-10-26 | 2022-02-15 | 南京理工大学 | Express delivery warehousing method based on mobile robot |
| CN114169301A (en) * | 2021-11-30 | 2022-03-11 | 上海东普信息科技有限公司 | Electronic surface list convergence number-taking method, device, equipment and storage medium |
| CN114186654A (en) * | 2022-02-17 | 2022-03-15 | 节时科技(深圳)有限公司 | Post house system, method, equipment and storage medium for automatic sorting and recovery |
| CN114842456A (en) * | 2022-06-29 | 2022-08-02 | 北京科技大学 | Logistics distribution method based on unmanned express delivery vehicle |
| CN115311626A (en) * | 2022-08-30 | 2022-11-08 | 金锋馥(滁州)科技股份有限公司 | Express package detection and identification algorithm based on deep learning |
| CN115303598A (en) * | 2022-09-13 | 2022-11-08 | 深圳市利和兴股份有限公司 | Intelligent labeling system and working method thereof |
-
2023
- 2023-03-09 CN CN202310220757.0A patent/CN115965900B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102161040A (en) * | 2011-01-14 | 2011-08-24 | 北京交通大学 | Logistic sorting system based on color sensor |
| CN108090728A (en) * | 2017-12-28 | 2018-05-29 | 上海传英信息技术有限公司 | A kind of express delivery information input method and input system based on intelligent terminal |
| CN110674713A (en) * | 2019-08-23 | 2020-01-10 | 北京市商汤科技开发有限公司 | Express query method, express delivery warehouse-out method, express delivery pickup method and related device |
| CN111368730A (en) * | 2020-03-04 | 2020-07-03 | 上海东普信息科技有限公司 | Method, device and equipment for identifying abnormal actions of express sorting and storage medium |
| CN211294479U (en) * | 2020-03-07 | 2020-08-18 | 林昌全 | Back adhesive label paper |
| WO2021196836A1 (en) * | 2020-03-30 | 2021-10-07 | 北京沃东天骏信息技术有限公司 | Method and apparatus for positioning express parcel |
| CN111709283A (en) * | 2020-05-07 | 2020-09-25 | 顺丰科技有限公司 | Method and device for detecting state of logistics piece |
| CN111767973A (en) * | 2020-07-02 | 2020-10-13 | 林昌全 | Design method of logistics management software function based on associated label |
| CN113592376A (en) * | 2021-06-11 | 2021-11-02 | 中国农业大学 | Intelligent access information management system applied to express post house |
| CN114047750A (en) * | 2021-10-26 | 2022-02-15 | 南京理工大学 | Express delivery warehousing method based on mobile robot |
| CN114169301A (en) * | 2021-11-30 | 2022-03-11 | 上海东普信息科技有限公司 | Electronic surface list convergence number-taking method, device, equipment and storage medium |
| CN114186654A (en) * | 2022-02-17 | 2022-03-15 | 节时科技(深圳)有限公司 | Post house system, method, equipment and storage medium for automatic sorting and recovery |
| CN114842456A (en) * | 2022-06-29 | 2022-08-02 | 北京科技大学 | Logistics distribution method based on unmanned express delivery vehicle |
| CN115311626A (en) * | 2022-08-30 | 2022-11-08 | 金锋馥(滁州)科技股份有限公司 | Express package detection and identification algorithm based on deep learning |
| CN115303598A (en) * | 2022-09-13 | 2022-11-08 | 深圳市利和兴股份有限公司 | Intelligent labeling system and working method thereof |
Non-Patent Citations (4)
| Title |
|---|
| MIWA TOKUTAKE 等: "A method for revising package image colors to express brand perceptions better", 《COLOR》, vol. 44, no. 5, pages 798 - 810, XP071625787, DOI: 10.1002/col.22400 * |
| 吴蓬勃 等: "基于多模态的无序堆叠快递包裹机械臂视觉抓取系统", 《包装工程》, vol. 43, no. 15, pages 68 - 76 * |
| 张瑜: "快递智能自动分拣系统设计研究", 微型电脑应用, vol. 36, no. 4, pages 118 - 120 * |
| 李恩宇: "基于单目视觉的六自由度机械臂分拣系统设计", 《中国优秀硕士学位论文全文数据库 信息科技辑》, vol. 2020, no. 1, pages 140 - 793 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116486346A (en) * | 2023-06-16 | 2023-07-25 | 济南雷圣自动化技术有限公司 | Carton conveying method and device |
| CN116486346B (en) * | 2023-06-16 | 2023-09-05 | 济南雷圣自动化技术有限公司 | Carton conveying method and device |
| CN117854091A (en) * | 2024-01-15 | 2024-04-09 | 金锋馥(滁州)科技股份有限公司 | Method for extracting information of multi-surface dense labels of packages based on image feature detection |
| CN117854091B (en) * | 2024-01-15 | 2024-06-07 | 金锋馥(滁州)科技股份有限公司 | Method for extracting information of multi-surface dense labels of packages based on image feature detection |
Also Published As
| Publication number | Publication date |
|---|---|
| CN115965900B (en) | 2023-06-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CA2841721C (en) | Package vision sort system and method | |
| CN109409175B (en) | Settlement method, device and system | |
| US9578107B2 (en) | Data capture and identification system and process | |
| CN115965900A (en) | Express item identification and search system and method | |
| CN106710083A (en) | Intelligent express method based on data analysis and terminal system | |
| CN109092696B (en) | Sorting system and sorting method | |
| CN107656960B (en) | Automatic matching system for managing lost articles of subway | |
| CN111401813A (en) | Logistics sorting system based on image recognition | |
| CN110570600A (en) | Express sending and taking method and system based on certificate card and biological characteristic information | |
| CN108701239B (en) | Article positioning method and system | |
| CN1129082C (en) | Image processing apparatus | |
| CN110163584A (en) | A kind of data processing method based on electronics red packet, device and relevant device | |
| CN112381293A (en) | Intelligent express delivery distribution system and method based on big data | |
| CN112445321A (en) | Article processing method and device and electronic equipment | |
| CN114186654B (en) | Post house system, method, equipment and storage medium for automatic sorting and recovery | |
| US20140195451A1 (en) | Unmanned system for accepting mail and method thereof | |
| CN106682861A (en) | Logistics parcel sorting and identification method | |
| JP6875760B1 (en) | Filtering processing system | |
| US7480394B2 (en) | Method and arrangement for recognizing objects in mail item images, their position and reading their postal information | |
| CN104677499B (en) | The system and method for fruit is selected using electronic equipment | |
| CN111695621B (en) | Method for detecting matching of customized article and order based on deep learning | |
| CN115860605A (en) | Warehouse logistics detection system based on visual identification | |
| CN109063855A (en) | Aid in treatment method and device for the identification of vehicle vehicle condition | |
| CN110555219A (en) | Three-dimensional CAD model similarity retrieval system and method based on image recognition | |
| CN113978987A (en) | Pallet object packaging and picking method, device, equipment and medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |