CN112668629A - Intelligent warehousing method, system, equipment and storage medium based on picture identification - Google Patents
Intelligent warehousing method, system, equipment and storage medium based on picture identification Download PDFInfo
- Publication number
- CN112668629A CN112668629A CN202011550051.3A CN202011550051A CN112668629A CN 112668629 A CN112668629 A CN 112668629A CN 202011550051 A CN202011550051 A CN 202011550051A CN 112668629 A CN112668629 A CN 112668629A
- Authority
- CN
- China
- Prior art keywords
- threshold
- matching
- preset
- picture
- adjusting
- 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
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 18
- 238000012545 processing Methods 0.000 claims abstract description 15
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 238000001514 detection method Methods 0.000 claims abstract description 7
- 238000003708 edge detection Methods 0.000 claims abstract description 7
- 238000004590 computer program Methods 0.000 claims description 11
- 230000002159 abnormal effect Effects 0.000 claims description 9
- 230000008569 process Effects 0.000 description 15
- 238000013507 mapping Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 3
- 238000007726 management method Methods 0.000 description 3
- 238000004380 ashing Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000012790 confirmation Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000013500 data storage Methods 0.000 description 2
- 238000000151 deposition Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000009499 grossing Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 230000003068 static effect Effects 0.000 description 2
- 230000001131 transforming effect Effects 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000008094 contradictory effect Effects 0.000 description 1
- 230000010365 information processing Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the invention provides an intelligent warehousing method based on picture identification, which comprises the following steps: acquiring a picture of a part to be identified, preprocessing the picture of the part to be identified, and processing the picture by using an edge detection algorithm and a contour detection algorithm to obtain part characteristics of a frame of the part to be identified; respectively matching the part characteristics with a plurality of preset part characteristics in a pre-trained part recognition model to obtain a plurality of corresponding matching similarities; comparing the matching similarities with a preset first threshold respectively, when the matching similarities are lower than the first threshold, adjusting the threshold according to a preset threshold adjusting strategy, and determining a matching result according to the adjusted threshold; and when the matching result is successful, sending the part information corresponding to the matching similarity to a terminal where the user is located for the user to confirm. According to the invention, the recognition rate can be greatly improved and the user experience can be greatly improved by adjusting the threshold value after the matching fails.
Description
Technical Field
The embodiment of the invention relates to the technical field of artificial intelligence, in particular to an intelligent warehousing method, system, equipment and storage medium based on picture identification.
Background
The current automobile accessory storage management system identifies parts by acquiring a target picture containing the parts to be identified and then identifying the target picture by utilizing a pre-trained part identification model to obtain part information of the target picture. However, the existing automobile accessory integral parts are various in types, sizes and shapes, so that the identification difficulty is high, and the identification of automobile accessories is greatly influenced. The traditional accessory identification system affected by the method prompts abnormality after identification failure, and needs a user to upload pictures again, so that the user can perform secondary identification operation, and the user experience is greatly reduced. Meanwhile, the accessory storage position cannot be intelligently provided after the part information is identified, a user is required to input or select the storage position, the complexity of the system is increased, and the accessory storage position is not damaged easily.
Therefore, the invention aims to solve the technical problems of low recognition success rate caused by various types, sizes and shapes of parts, repeated uploading of pictures and low user experience during storage management.
Disclosure of Invention
In view of this, it is necessary to provide an intelligent warehousing method, an intelligent warehousing system, a computer device and a readable storage medium based on picture recognition, which aim to solve the technical problems of low recognition success rate caused by various types, sizes and shapes of parts, thereby causing repeated uploading of pictures and low user experience during warehousing management in the prior art.
In order to achieve the above object, an embodiment of the present invention provides an intelligent warehousing method based on picture identification, where the method includes:
acquiring a picture of a part to be identified, and preprocessing the picture of the part to be identified to obtain a target image;
processing the target image by using an edge detection algorithm and a contour detection algorithm in sequence to obtain a frame of the part to be identified and obtain part characteristics of the target image;
matching the part features with a plurality of preset part features in a pre-trained part recognition model respectively to obtain a plurality of corresponding matching similarities;
comparing the matching similarity with a preset first threshold respectively, when the matching similarity is lower than the first threshold, adjusting the threshold according to a preset threshold adjusting strategy, and determining a matching result according to the adjusted threshold;
and when one or more matching results are successful, sending the part information corresponding to the matching similarity to a terminal where a user is located for the user to confirm.
Optionally, the adjusting the threshold according to a preset threshold adjustment policy, and determining the matching result according to the adjusted threshold include:
adjusting the first threshold value to a preset second threshold value according to the preset threshold value adjusting strategy;
comparing a plurality of the matching similarity degrees with the second threshold value respectively;
when the matching similarity is larger than or equal to the second threshold, the matching is successful;
and when the matching similarity degrees are all smaller than the second threshold value, triggering the adjustment operation of the second threshold value until the matching similarity degrees are all smaller than a preset third threshold value, and quitting.
Optionally, the adjusting the threshold according to a preset threshold adjustment policy, and determining the matching result according to the adjusted threshold include:
obtaining identification data for a plurality of first live part features input to the part identification model, the identification data comprising: passing rate and false recognition rate;
adjusting the threshold value according to the passing rate and the false recognition rate of the first real-time part features to obtain an adjusted target threshold value;
comparing a plurality of the matching similarity degrees with the target threshold respectively;
when one or more matching similarity degrees are larger than or equal to the target threshold value, the matching is successful;
and when the matching similarities are smaller than the target threshold, triggering the operation of acquiring the identification data of the second real-time part features and adjusting the threshold according to the identification data of the second real-time part features until the matching similarities are smaller than a preset fourth threshold, and quitting.
Optionally, the method further comprises:
and when the matching similarity is higher than the first threshold, sending identification library information corresponding to the part characteristics in the part identification model and corresponding warehousing position information to the terminal so that the terminal can display the identification library information and the warehousing position information.
Optionally, the method further comprises:
and when the matching result is matching failure, sending abnormal information to the terminal.
Optionally, the method further comprises:
acquiring target part information confirmed by the user;
and adding target part characteristic information corresponding to the target part information into the part identification model, and sending identification library information corresponding to the target part characteristic information and corresponding storage position information to the terminal so that the terminal can display the identification library information and the storage position information.
Optionally, the method further comprises:
and uploading the target part information to a block chain.
In order to achieve the above object, an embodiment of the present invention provides an intelligent warehousing system based on picture identification, where the method includes:
the device comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring a picture of a part to be identified and preprocessing the picture of the part to be identified to obtain a target image;
the processing module is used for processing the target image by sequentially utilizing an edge detection algorithm and a contour detection algorithm to obtain a frame of the part to be identified and obtain the part characteristics of the target image;
the matching module is used for matching the part characteristics with a plurality of preset part characteristics in a pre-trained part recognition model respectively to obtain a plurality of corresponding matching similarities;
the threshold adjusting module is used for comparing the matching similarity degrees with a preset first threshold respectively, adjusting the threshold according to a preset threshold adjusting strategy when the matching similarity degrees are lower than the first threshold, and determining a matching result according to the adjusted threshold;
and the sending module is used for sending the part information corresponding to the matching similarity to a terminal where a user is located when one or more matching results are successful in matching so as to be confirmed by the user.
In order to achieve the above object, an embodiment of the present invention further provides a computer device, a memory of the computer device, a processor, and a computer program stored on the memory and executable on the processor, where the computer program, when executed by the processor, implements the steps of the smart warehousing method based on picture recognition as described above.
In order to achieve the above object, an embodiment of the present invention further provides a computer-readable storage medium, in which a computer program is stored, where the computer program is executable by at least one processor, so as to cause the at least one processor to execute the steps of the smart warehousing method based on picture recognition as described above.
According to the intelligent warehousing method based on picture recognition, the processed part features are matched with the part recognition models trained in advance, after matching fails, threshold adjustment is carried out through a threshold adjustment strategy, then matching is carried out again until matching is successful or when the matching is adjusted to the lowest recognition model matching threshold, matching still fails, abnormal information is returned, and then manual intervention is carried out. According to the invention, the recognition rate can be greatly improved by adjusting the threshold value after the matching fails, the situation that a user takes a picture again or submits a picture operation after the recognition fails once is avoided, and the user experience is greatly improved.
Drawings
FIG. 1 is a flowchart illustrating exemplary steps of a smart warehousing method based on picture recognition according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating an exemplary step of adjusting the threshold according to the preset threshold adjustment policy and determining the matching result according to the adjusted threshold in step S400 in fig. 1;
fig. 3 is a flowchart illustrating another exemplary step of adjusting the threshold according to the preset threshold adjustment policy and determining the matching result according to the adjusted threshold in step S400 in fig. 1;
FIG. 4 is a schematic diagram of a program module of an intelligent warehousing system based on image recognition according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating a hardware architecture of a computer device according to an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.
It should be noted that the descriptions relating to "first", "second", etc. in the present invention are for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In addition, technical solutions between the various embodiments may be combined with each other, but must be realized by a person skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not be within the protection scope required by the present invention.
Example one
Referring to fig. 2, a flowchart illustrating steps of a smart warehousing method based on picture recognition according to an embodiment of the present invention is shown. It is to be understood that the flow charts in the embodiments of the present method are not used to limit the order in which the steps are performed. It should be noted that, in the present embodiment, the server is taken as an execution subject to perform an exemplary description, and specifically includes the following steps:
step S100: the method comprises the steps of obtaining a picture of a part to be identified, and preprocessing the picture of the part to be identified to obtain a target image.
The target picture can be acquired through uploading by a user or through a camera shooting mode. The part to be identified comprises an automobile part.
In an exemplary embodiment, the preprocessing the target picture may specifically include: performing ashing treatment on the target picture to obtain a first intermediate picture; performing black and white processing on the first intermediate picture by using a binarization algorithm to obtain a second intermediate picture; and denoising, smoothing and transforming the second intermediate picture to obtain the target image. By preprocessing the picture of the part to be identified, the important characteristics of the picture can be enhanced.
Step S200: and processing the target image by using an edge detection algorithm and a contour detection algorithm in sequence to obtain a frame of the part to be identified and obtain the part characteristics of the target image.
And obtaining the specific shape, size and type information of the part to be identified by obtaining the frame of the part to be identified so as to carry out subsequent feature matching.
Step S300: and respectively matching the part characteristics with a plurality of preset part characteristics in a pre-trained part recognition model to obtain a plurality of corresponding matching similarities.
Specifically, the matching the part features with a plurality of preset part features in the part recognition models trained in advance respectively to obtain matching similarity specifically includes: and performing similarity calculation on the part features and each preset part feature in the part recognition model to obtain the matching similarity of the part features and each preset part feature in the part recognition model.
Step S400: and comparing the matching similarity with a preset first threshold respectively, when the matching similarity is lower than the first threshold, adjusting the threshold according to a preset threshold adjusting strategy, and determining a matching result according to the adjusted threshold.
Specifically, when the matching similarity between the part features and each preset part feature in the part recognition model is lower than the first threshold, it indicates that the part features are not successfully matched for the first time, and there may be a situation that a picture is not clear. For example: if the matching similarity between the part features and each preset part feature in the part identification model is smaller than a first threshold value 88, and at this time, it indicates that no part meeting the threshold value is matched with the part to be identified, the first threshold value 88 is adjusted for one or more times according to a preset threshold value adjustment strategy, and a matching result is determined according to the adjusted threshold value.
In an exemplary embodiment, as shown in fig. 2, the adjusting of the threshold value according to the preset threshold value adjusting policy in step S400, and determining the matching result according to the adjusted threshold value may include steps S201 to S205.
Step S201: adjusting the first threshold value to a preset second threshold value according to the preset threshold value adjusting strategy;
step S202: comparing the matching similarity with the second threshold respectively, if the matching similarity is greater than or equal to the second threshold, executing step S203, otherwise executing step S204;
step S203: matching is successful;
step S204: and triggering the adjustment operation of the second threshold until the matching similarity is smaller than a preset third threshold, and quitting the matching process, wherein the matching process fails.
In an exemplary embodiment, a threshold and passing rate and false recognition rate mapping relation table is preset according to a threshold and passing rate and false recognition rate mapping relation, and in the mapping relation table, the smaller the threshold is, the higher the passing rate is, the higher the false recognition rate is, and the more inaccurate the recognition result is. For example: when the threshold value is 68, the passing rate is 98 percent, and the false recognition rate is one thousandth; when the threshold value is 78, the passing rate is 92%, and the false recognition rate is one ten thousandth; when the threshold value is 88, the passing rate is 85%, and the false recognition rate is one hundred thousand. It should be noted that, in order to ensure the passing rate and the false recognition rate, the threshold 68 may be set as a minimum threshold. In practical applications, the third threshold may be the minimum threshold 68 in the mapping table. In the matching process, the recognition is not accurate due to the unclear picture of the part to be recognized or the complex and various sizes or shapes of the same part, and at the moment, the intelligent error correction of the part matching can be realized by reducing the threshold value. The preset threshold adjustment strategy may be: the adjustment is carried out according to the sequence of the threshold value from large to small, the adjustment is carried out according to the sequence of the passing rate from large to small and/or the adjustment is carried out according to the sequence of the error recognition rate from small to large. Of course, the preset threshold adjustment strategy may also be to select a threshold according to a manually set threshold, a passing rate and/or a false recognition rate and according to a mapping relationship among the threshold, the passing rate and the false recognition rate, so as to adjust the threshold.
Specifically, in an exemplary embodiment, when the part features are matched with a pre-trained part recognition model, a maximum threshold is preferentially set as a first threshold, and when matching similarities are lower than the maximum threshold, the threshold is adjusted according to the preset threshold adjustment strategy, so that the false recognition rate is reduced.
It should be noted that, when the preset threshold adjustment strategy is to perform adjustment according to the passing rate and/or the false recognition rate, the threshold to be adjusted is determined according to the passing rate and/or the false recognition rate by searching the mapping relationship table of the threshold and the passing rate and the false recognition rate. In the embodiment of the present invention, the presence of matching similarity greater than or equal to the second threshold indicates that one or more matching similarities greater than or equal to the second threshold exist.
For example, after the threshold is adjusted from the first threshold 88 to the second threshold 78, the matching similarity is all smaller than the second threshold, the threshold is continuously adjusted to the minimum threshold 68, and if the matching similarity is still all smaller than the minimum threshold 68, it is determined that the matching fails. Of course, in practical applications, when the threshold is adjusted from the first threshold 88 to the second threshold 78, one or more matching similarities higher than the second threshold are all determined to be successful.
According to the embodiment of the invention, the efficiency and the accuracy of subsequent part identification are greatly improved by adjusting the threshold value.
In an exemplary embodiment, as shown in fig. 3, the adjusting of the threshold value according to the preset threshold value adjusting policy in step S400, and determining the matching result according to the adjusted threshold value may further include step S301 to step S305.
Step S301: obtaining identification data for a plurality of first real-time part features input to the part identification model, the identification data comprising: passing rate and false recognition rate;
step S302: adjusting the threshold value according to the passing rate and the false recognition rate of the first real-time parts to obtain an adjusted target threshold value;
step S303: comparing the matching similarity degrees with the target threshold respectively, if one or more matching similarity degrees are greater than or equal to the target threshold, executing step S304, otherwise executing step S305;
step S304: judging that the matching is successful;
step S305: and triggering the acquisition operation of the identification data of the second real-time part characteristics, and adjusting the threshold value according to the identification data of the second real-time part characteristics until the matching similarity is smaller than a preset fourth threshold value, and quitting the matching failure.
According to the embodiment of the invention, the threshold value is automatically adjusted according to the matching result of the real-time part data without manually adjusting the threshold value, so that the efficiency and the accuracy of subsequent identification can be greatly improved. And when the matching similarity is smaller than the target threshold, the matching result of the time is also included in the next target threshold adjusting process, so that the adjustment of the real-time threshold is realized, and the accuracy of subsequent identification is improved.
In an exemplary embodiment, the method further comprises: and when the matching similarity is higher than the first threshold value, transmitting the identification library information corresponding to the part characteristics and the corresponding warehousing position information to a terminal where a user is located, so that the terminal can display the identification library information and the warehousing position information.
It should be noted that the existence of matching similarity higher than the first threshold in the embodiment of the present invention means that when at least one of the matching degrees is higher than the first threshold, that is, one or more of the matching degrees are higher than the first threshold.
Specifically, when the matching similarity is higher than the first threshold, the part feature is successfully matched for the first time. For example: and when the part features are matched with a pre-trained part recognition model, setting the first threshold value to be 88, and if the matching similarity of the calculated part features and a preset part feature in the part recognition model is greater than or equal to 88, successfully matching. The storage position information corresponding to the part characteristics is sent to the terminal, so that the terminal can display the storage position information, and the part storage is safer and more convenient. For example: if treat the discernment part is fragile article, through with fragile article storage position send to the terminal, then the terminal shows storage position information has greatly improved treat the security level of discernment part in depositing the handling, brings the facility for the user simultaneously.
Step S500: and when one or more matching results are successful, sending the part information corresponding to the matching similarity to a terminal where a user is located for the user to confirm.
Specifically, after the threshold is adjusted, there may be a possibility that all the part features of the plurality of pieces of part information are successfully matched with the part features, and at this time, manual confirmation is required to ensure that the matching result is only one of the plurality of pieces of part information.
In an exemplary embodiment, the method further comprises: and when the matching result is matching failure, sending abnormal information to the terminal.
Specifically, after the matching is determined to fail, it indicates that the part to be identified does not exist in the smart warehousing system or the part to be identified is not identified due to the reason of the smart warehousing system, and at this time, the abnormal information is sent to the user terminal so as to facilitate manual intervention, and subsequent operations are executed according to a manual intervention result.
In an exemplary embodiment, the method further comprises: acquiring target part information confirmed by the user; and adding target part characteristic information corresponding to the target part information into the part identification model, and sending identification library information corresponding to the target part characteristic information and corresponding storage position information to the terminal so that the terminal can display the identification library information and the storage position information. The identification library information refers to characteristic information corresponding to the target part characteristic information in the part identification model, and the storage position information refers to a storage position where the target part information is stored. The matched part feature information is added into the part identification model, so that the part feature types in the part identification model are enriched, and the subsequent part feature matching efficiency is improved.
In an exemplary embodiment, the method further comprises: and uploading the target part information to a block chain. The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism and an encryption algorithm. A block chain (Blockchain), which is essentially a decentralized database, is a series of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, so as to verify the validity (anti-counterfeiting) of the information and generate a next block. The blockchain may include a blockchain underlying platform, a platform product services layer, and an application services layer.
According to the intelligent warehousing method based on picture recognition, the processed part characteristics are matched with the part recognition models trained in advance, after matching fails, threshold value adjustment is carried out through a threshold value adjustment strategy, then matching is carried out again until matching is successful or when the matching is adjusted to the lowest recognition model matching threshold value, matching is still failed, abnormal information is returned, and manual intervention is carried out. According to the invention, the recognition rate can be greatly improved by adjusting the threshold value after the matching fails, the situation that a user takes a picture again or submits a picture operation after the primary recognition fails is avoided, and the user experience is greatly improved.
Example two
Referring to fig. 4, a schematic diagram of a program module of an intelligent warehousing system based on picture recognition according to an embodiment of the present invention is shown. The intelligent warehousing system based on the picture recognition can be applied to a server. In this embodiment, the smart warehousing system 20 based on picture recognition may include or be divided into one or more program modules, and the one or more program modules are stored in a storage medium and executed by one or more processors to implement the present invention and the smart warehousing method based on picture recognition described above. The program module referred to in the embodiments of the present invention refers to a series of computer program instruction segments capable of performing specific functions, and is more suitable for describing the execution process of the smart warehousing system 20 based on picture recognition in the storage medium than the program itself. The following description will specifically describe the functions of the program modules in this embodiment:
the obtaining module 401 is configured to obtain a picture of a part to be identified, and preprocess the picture of the part to be identified to obtain a target image.
The obtaining module 401 may obtain the target picture through uploading by a user or through taking a picture by a camera. The part to be identified comprises an automobile part.
In an exemplary embodiment, the preprocessing the target picture may specifically include: performing ashing treatment on the target picture to obtain a first intermediate picture; performing black and white processing on the first intermediate picture by using a binarization algorithm to obtain a second intermediate picture; and denoising, smoothing and transforming the second intermediate picture to obtain the target image. By preprocessing the picture of the part to be identified, the important characteristics of the picture can be enhanced.
And the processing module 402 is configured to process the target image by using an edge detection algorithm and a contour detection algorithm in sequence to obtain a frame of the part to be identified, so as to obtain a part feature of the target image.
And obtaining the specific shape, size and type information of the part to be identified by obtaining the frame of the part to be identified so as to carry out subsequent feature matching.
A matching module 403, configured to match the part features with a plurality of preset part features in a pre-trained part recognition model, respectively, to obtain a plurality of corresponding matching similarities.
Specifically, the matching module 403 may be specifically configured to: and performing similarity calculation on the part features and each preset part feature in the part recognition model to obtain the matching similarity of the part features and each preset part feature in the part recognition model.
A threshold adjusting module 404, configured to compare the multiple matching similarities with a preset first threshold, and when the multiple matching similarities are lower than the first threshold, adjust the threshold according to a preset threshold adjusting policy, and determine a matching result according to the adjusted threshold.
Specifically, when the matching similarity between the part feature and each preset part feature in the part recognition model is lower than the first threshold, it indicates that the part feature is not successfully matched for the first time, and there may be a situation that a picture is unclear, and at this time, the threshold adjustment module 404 adjusts the first threshold, and then determines a matching result according to the adjusted threshold, so as to improve the recognition rate, and meanwhile, avoid repeated photographing or picture submitting operations by a user, and provide convenience for the user. For example: if the matching similarity between the part feature and each preset part feature in the part recognition model is smaller than a first threshold 88, which indicates that no part meeting the threshold is matched with the part to be recognized, the threshold adjustment module 404 adjusts the first threshold 88 for one or more times according to a preset threshold adjustment strategy, and determines a matching result according to the adjusted threshold.
In an exemplary embodiment, the threshold adjusting module 404 may specifically include: a threshold adjusting unit and a comparing unit.
The threshold adjusting unit is configured to: adjusting the first threshold value to a preset second threshold value according to the preset threshold value adjusting strategy;
the comparison unit is used for: comparing the matching similarity with the second threshold respectively, and if the matching similarity is greater than or equal to the second threshold, the matching is successful; otherwise, triggering the adjustment operation of the second threshold value until the matching similarity is smaller than a preset third threshold value, and quitting the matching process, wherein the matching process fails.
In an exemplary embodiment, a threshold and passing rate and false recognition rate mapping relation table is preset according to a threshold and passing rate and false recognition rate mapping relation, and in the mapping relation table, the smaller the threshold is, the higher the passing rate is, the higher the false recognition rate is, and the more inaccurate the recognition result is. For example: when the threshold value is 68, the passing rate is 98 percent, and the false recognition rate is one thousandth; when the threshold value is 78, the passing rate is 92%, and the false recognition rate is one ten thousandth; when the threshold value is 88, the passing rate is 85%, and the false recognition rate is one hundred thousand. It should be noted that, in order to ensure the passing rate and the false recognition rate, the threshold 68 may be set as a minimum threshold. In practical applications, the third threshold is the minimum threshold 68 in the mapping table. In the matching process, the recognition is not accurate due to the unclear picture of the part to be recognized or the complex and various sizes or shapes of the same part, and at the moment, the intelligent error correction of the part matching can be realized by reducing the threshold value. The preset threshold adjustment strategy may be: and adjusting according to the sequence of the threshold value from large to small, adjusting according to the sequence of the passing rate from large to small and/or adjusting according to the sequence of the error recognition rate from small to large. Of course, the preset threshold adjustment strategy may also be to select a threshold according to a manually set threshold, a passing rate and/or a false recognition rate and according to a mapping relationship among the threshold, the passing rate and the false recognition rate, so as to adjust the threshold.
Specifically, in an exemplary embodiment, when the part features are matched with a pre-trained part recognition model, a maximum threshold is preferentially set as a first threshold, and when matching similarities are lower than the maximum threshold, the threshold is adjusted according to the preset threshold adjustment strategy, so that the false recognition rate is reduced.
It should be noted that, when the preset threshold adjustment strategy is to perform adjustment according to the passing rate and/or the false recognition rate, the threshold to be adjusted is determined according to the passing rate and/or the false recognition rate by searching the mapping relationship table of the threshold and the passing rate and the false recognition rate. In the embodiment of the present invention, the presence of matching similarity greater than or equal to the second threshold indicates that one or more matching similarities greater than or equal to the second threshold exist.
For example, after the threshold is adjusted from the first threshold 88 to the second threshold 78, the matching similarity is all smaller than the second threshold, the threshold is continuously adjusted to the minimum threshold 68, and if the matching similarity is still all smaller than the minimum threshold 68, it is determined that the matching fails. Of course, in practical applications, when the threshold is adjusted from the first threshold 88 to the second threshold 78, one or more matching similarities higher than the second threshold are all determined to be successful.
According to the embodiment of the invention, the efficiency and the accuracy of subsequent part identification are greatly improved by adjusting the threshold value.
In an exemplary embodiment, the threshold adjustment module 404 may further include an obtaining unit.
The acquisition unit is configured to: obtaining identification data for a plurality of first live part features input to the part identification model, the identification data comprising: pass rate and false positive rate.
The threshold adjusting unit is further configured to: adjusting the threshold value according to the passing rate and the false recognition rate of the first real-time parts to obtain an adjusted target threshold value;
the comparison unit is further configured to: comparing the matching similarity with the target threshold respectively, and if one or more matching similarity is greater than or equal to the target threshold, the matching is successful; and otherwise, triggering the acquisition operation of the identification data of the plurality of second real-time part features, and adjusting the threshold value according to the identification data of the plurality of second real-time part features until the matching similarity is smaller than a preset fourth threshold value, and quitting the matching process, wherein the matching process fails.
According to the embodiment of the invention, the threshold value is automatically adjusted according to the matching result of the real-time part data without manually adjusting the threshold value, so that the efficiency and the accuracy of subsequent identification can be greatly improved. And when the matching similarity is smaller than the target threshold, the matching result of the time is also included in the next target threshold adjusting process, so that the adjustment of the real-time threshold is realized, and the accuracy of subsequent identification is improved.
In an exemplary embodiment, the smart warehousing system 20 based on picture recognition may further include a sending unit for: and when the matching similarity is higher than the first threshold value, transmitting the identification library information corresponding to the part characteristics and the corresponding warehousing position information to a terminal where a user is located, so that the terminal can display the identification library information and the warehousing position information.
It should be noted that the existence of matching similarity higher than the first threshold in the embodiment of the present invention means that when at least one of the matching degrees is higher than the first threshold, that is, one or more of the matching degrees are higher than the first threshold.
Specifically, when the matching similarity is higher than the first threshold, the part feature is successfully matched for the first time. For example: and when the part features are matched with a pre-trained part recognition model, setting the first threshold value to be 88, and if the matching similarity of the calculated part features and a preset part feature in the part recognition model is greater than or equal to 88, successfully matching. The warehousing position information corresponding to the part characteristics is sent to the terminal, and then the terminal displays the warehousing position information, so that the part warehousing is safer and more convenient. For example: if treat the discernment part is fragile article, through with fragile article storage position send to the terminal, then the terminal shows storage position information has greatly improved treat the security level of discernment part in depositing the handling, brings the facility for the user simultaneously.
A sending module 405, configured to send, when one or more matching results are matching success, the part information corresponding to the matching similarity to a terminal where a user is located, so that the user can confirm the part information.
Specifically, after the threshold is adjusted, there may be a possibility that all the part features of the plurality of pieces of part information are successfully matched with the part features, and at this time, manual confirmation is required to ensure that the matching result is only one of the plurality of pieces of part information.
In an exemplary embodiment, the sending module 405 may further be configured to: and when the matching result is matching failure, sending abnormal information to the terminal.
Specifically, after the matching is determined to fail, it indicates that the part to be identified does not exist in the smart warehousing system 20 or the part to be identified is not identified due to the smart warehousing system 20, and at this time, the sending module 405 sends the abnormal information to the user so as to facilitate manual intervention and execute subsequent operations according to a result of the manual intervention.
In an exemplary embodiment, the obtaining module 401 is further configured to: and acquiring the target part information confirmed by the user.
In an exemplary embodiment, the smart warehousing system 20 based on picture recognition further includes an adding unit. The adding unit is used for adding target part characteristic information corresponding to the target part information into the part identification model; the sending unit is used for sending the identification library information corresponding to the target part characteristic information and the corresponding warehousing position information to the terminal so that the terminal can display the identification library information and the warehousing position information. The identification library information refers to the characteristic information of the target part corresponding to the characteristic information in the part identification model, and the storage position information refers to the storage position of the target part information in storage. The matched part feature information is added into the part identification model, so that the part feature types in the part identification model are enriched, and the follow-up part feature matching efficiency is improved.
In an exemplary embodiment, the smart warehousing system 20 based on picture recognition further includes an uploading unit for: and uploading the target part information to a block chain. The block chain is a novel application mode of computer technologies such as distributed data storage, point-to-point transmission, a consensus mechanism and an encryption algorithm. A block chain (Blockchain), which is essentially a decentralized database, is a string of data blocks associated by using a cryptographic method, and each data block contains information of a batch of network transactions, which is used for verifying the validity (anti-counterfeiting) of the information and generating a next block. The blockchain may include a blockchain underlying platform, a platform product services layer, and an application services layer.
According to the intelligent warehousing method based on picture recognition, the processed part characteristics are matched with the part recognition models trained in advance, after matching fails, threshold value adjustment is carried out through a threshold value adjustment strategy, then matching is carried out again until matching is successful or when the matching is adjusted to the lowest recognition model matching threshold value, matching is still failed, abnormal information is returned, and manual intervention is carried out. According to the invention, the recognition rate can be greatly improved by adjusting the threshold value after the matching fails, the situation that a user takes a picture again or submits a picture operation after the primary recognition fails is avoided, and the user experience is greatly improved.
EXAMPLE III
Referring to fig. 5, a hardware architecture diagram of a computer device according to a third embodiment of the present invention is shown. While the computer device 2 includes, but is not limited to, a memory 21, a process 22, and a network interface 23 communicatively coupled to each other via a system bus, FIG. 5 illustrates only the computer device 2 having components 21-23, it is to be understood that not all of the illustrated components are required and that more or fewer components may alternatively be implemented. In the present embodiment, the computer device 2 is a device capable of automatically performing numerical calculation and/or information processing in accordance with a preset or stored instruction. For example, a smart phone, a tablet computer, a notebook computer, a desktop computer, a rack server, a blade server, a tower server, or a rack server (including an independent server or a server cluster composed of a plurality of servers) that can execute programs, and the like.
The memory 21 includes at least one type of readable storage medium including a flash memory, a hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, etc. In some embodiments, the memory 21 may be an internal storage unit of the computer device 2, such as a hard disk or a memory of the computer device 2. In other embodiments, the memory may also be an external storage device of the computer device 2, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like, which are provided on the computer device 2. Of course, the memory 21 may also comprise both an internal storage unit of the computer device 2 and an external storage device thereof. In this embodiment, the memory 21 is generally used for storing an operating system installed on the computer device 2 and various application software, such as program codes of the smart warehousing system 20 based on picture recognition. Further, the memory 21 may also be used to temporarily store various types of data that have been output or are to be output.
The processor 22 may be a Central Processing Unit (CPU), controller, microcontroller, microprocessor, or other data Processing chip in some embodiments. The processor 22 is typically used to control the overall operation of the computer device 2. In this embodiment, the processor 22 is configured to run the program codes stored in the memory 21 or process data, such as running the smart warehousing system 20 based on picture recognition.
The network interface 23 may comprise a wireless network interface or a wired network interface, and the network interface 23 is generally used for establishing communication connection between the computer device 2 and other electronic devices. For example, the network interface 23 is used to connect the computer device 2 to an external terminal through a network, establish a data transmission channel and a communication connection between the computer device 2 and the external terminal, and the like. The network may be a wireless or wired network such as an Intranet (Intranet), the Internet (Internet), a Global System for Mobile communications (GSM), Wideband Code Division Multiple Access (WCDMA), a 4G network, a 5G network, Bluetooth (Bluetooth), or Wi-Fi.
It is noted that fig. 5 only shows the computer device 2 with components 21-23, but it is to be understood that not all shown components are required to be implemented, and that more or less components may be implemented instead.
Example four
The present embodiment also provides a computer-readable storage medium, on which a computer program is stored, and the computer program, when executed by a processor, implements the steps of the smart warehousing method based on picture recognition in the embodiments.
In this embodiment, the computer-readable storage medium includes a Flash memory, a hard disk, a multimedia Card, a Card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a Read Only Memory (ROM), an Electrically Erasable Programmable Read Only Memory (EEPROM), a Programmable Read Only Memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application store, etc., and in other embodiments, the computer-readable storage medium may also be an external storage device of the computer apparatus, such as a plug-in hard disk equipped on the computer apparatus, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash Card, etc. Of course, the computer-readable storage medium may also include both internal and external storage devices of the computer device. In this embodiment, the computer-readable storage medium is generally used to store an operating system and various application software installed in the computer device, for example, the program code of the smart warehousing method based on picture recognition in the embodiment, and the like. Further, the computer-readable storage medium may also be used to temporarily store various types of data that have been output or are to be output.
It will be apparent to those skilled in the art that the modules or steps of the embodiments of the invention described above may be implemented by a general purpose computing device, they may be centralized on a single computing device or distributed across a network of multiple computing devices, and alternatively, they may be implemented by program code executable by a computing device, such that they may be stored in a storage device and executed by a computing device, and in some cases, the steps shown or described may be performed in an order different than that described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple ones of them may be fabricated into a single integrated circuit module. Thus, embodiments of the invention are not limited to any specific combination of hardware and software.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. An intelligent warehousing method based on picture identification is characterized by comprising the following steps:
acquiring a picture of a part to be identified, and preprocessing the picture of the part to be identified to obtain a target image;
processing the target image by using an edge detection algorithm and a contour detection algorithm in sequence to obtain a frame of the part to be identified and obtain part characteristics of the target image;
matching the part features with a plurality of preset part features in a pre-trained part recognition model respectively to obtain a plurality of corresponding matching similarities;
comparing the matching similarity with a preset first threshold respectively, when the matching similarity is lower than the first threshold, adjusting the threshold according to a preset threshold adjusting strategy, and determining a matching result according to the adjusted threshold;
and when one or more matching results are successful, sending the part information corresponding to the matching similarity to a terminal where a user is located for the user to confirm.
2. The smart warehousing method as claimed in claim 1, wherein the adjusting the threshold according to a preset threshold adjusting strategy and determining the matching result according to the adjusted threshold comprises:
adjusting the first threshold value to a preset second threshold value according to the preset threshold value adjusting strategy;
comparing a plurality of the matching similarity degrees with the second threshold value respectively;
when the matching similarity is larger than or equal to the second threshold, the matching is successful;
and when the matching similarity degrees are all smaller than the second threshold value, triggering the adjustment operation of the second threshold value until the matching similarity degrees are all smaller than a preset third threshold value, and quitting.
3. The smart warehousing method as claimed in claim 1, wherein the adjusting the threshold according to a preset threshold adjusting strategy and determining the matching result according to the adjusted threshold comprises:
obtaining identification data for a plurality of first live part features input to the part identification model, the identification data comprising: passing rate and false recognition rate;
adjusting the threshold value according to the passing rate and the false recognition rate of the first real-time part features to obtain an adjusted target threshold value;
comparing a plurality of the matching similarity degrees with the target threshold respectively;
when one or more matching similarity degrees are larger than or equal to the target threshold value, the matching is successful;
and when the matching similarities are smaller than the target threshold, triggering the operation of acquiring the identification data of the second real-time part features and adjusting the threshold according to the identification data of the second real-time part features until the matching similarities are smaller than a preset fourth threshold.
4. The smart warehousing method of claim 1, wherein the method further comprises:
and when the matching similarity is higher than the first threshold, sending identification library information corresponding to the part characteristics in the part identification model and corresponding warehousing position information to the terminal so that the terminal can display the identification library information and the warehousing position information.
5. The smart warehousing method of claim 1, wherein the method further comprises:
and when the matching result is matching failure, sending abnormal information to the terminal.
6. The smart warehousing method of claim 1, wherein the method further comprises:
acquiring target part information confirmed by the user;
and adding target part characteristic information corresponding to the target part information into the part identification model, and sending identification library information corresponding to the target part characteristic information and corresponding storage position information to the terminal so that the terminal can display the identification library information and the storage position information.
7. The smart warehousing method of claim 6, wherein the method further comprises:
and uploading the target part information to a block chain.
8. An intelligent warehousing system based on picture recognition is characterized by comprising:
the device comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring a picture of a part to be identified and preprocessing the picture of the part to be identified to obtain a target image;
the processing module is used for processing the target image by sequentially utilizing an edge detection algorithm and a contour detection algorithm so as to obtain a frame of the part to be identified and obtain the part characteristics of the target image;
the matching module is used for matching the part characteristics with a plurality of preset part characteristics in a pre-trained part recognition model respectively to obtain a plurality of corresponding matching similarities;
the threshold adjusting module is used for comparing the matching similarities with a preset first threshold respectively, adjusting the threshold according to a preset threshold adjusting strategy when the matching similarities are lower than the first threshold, and determining a matching result according to the adjusted threshold;
and the sending module is used for sending the part information corresponding to the matching similarity to a terminal where a user is located when one or more matching results are successful in matching so as to be confirmed by the user.
9. A computer device, characterized by a computer device memory, a processor and a computer program stored on the memory and executable on the processor, which computer program, when executed by the processor, performs the steps of the smart warehousing method based on picture recognition as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, in which a computer program is stored, which computer program is executable by at least one processor to cause the at least one processor to perform the steps of the smart warehousing method based on picture identification as claimed in any one of claims 1 to 7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011550051.3A CN112668629A (en) | 2020-12-24 | 2020-12-24 | Intelligent warehousing method, system, equipment and storage medium based on picture identification |
| PCT/CN2021/126013 WO2022134828A1 (en) | 2020-12-24 | 2021-10-25 | Intelligent warehousing method and system based on picture recognition, and device and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011550051.3A CN112668629A (en) | 2020-12-24 | 2020-12-24 | Intelligent warehousing method, system, equipment and storage medium based on picture identification |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112668629A true CN112668629A (en) | 2021-04-16 |
Family
ID=75408336
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011550051.3A Pending CN112668629A (en) | 2020-12-24 | 2020-12-24 | Intelligent warehousing method, system, equipment and storage medium based on picture identification |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN112668629A (en) |
| WO (1) | WO2022134828A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469015A (en) * | 2021-06-29 | 2021-10-01 | 浙江大华技术股份有限公司 | Face recognition method and device, electronic equipment and computer storage medium |
| CN113537237A (en) * | 2021-06-25 | 2021-10-22 | 西安交通大学 | Intelligent sensing method, system and device for multi-feature part quality information |
| WO2022134828A1 (en) * | 2020-12-24 | 2022-06-30 | 深圳壹账通智能科技有限公司 | Intelligent warehousing method and system based on picture recognition, and device and storage medium |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116071532B (en) * | 2023-03-16 | 2023-07-18 | 无锡广盈集团有限公司 | Intelligent door control method and system for outdoor cabinet |
| CN116747525A (en) * | 2023-08-21 | 2023-09-15 | 成都初心互动科技有限公司 | Automatic studio script detection method, device, equipment and medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106909893B (en) * | 2017-02-09 | 2020-12-29 | 北京小米移动软件有限公司 | Fingerprint identification method and device |
| CN109117822A (en) * | 2018-08-31 | 2019-01-01 | 贵州大学 | A kind of part case segmentation recognition method based on deep learning |
| CN111104542B (en) * | 2020-01-07 | 2023-04-18 | 成都睿琪科技有限责任公司 | Part identification management method and device |
| CN111738321B (en) * | 2020-06-12 | 2023-08-08 | 腾讯音乐娱乐科技（深圳）有限公司 | Data processing method, device, terminal equipment and storage medium |
| CN112668629A (en) * | 2020-12-24 | 2021-04-16 | 深圳壹账通智能科技有限公司 | Intelligent warehousing method, system, equipment and storage medium based on picture identification |
-
2020
- 2020-12-24 CN CN202011550051.3A patent/CN112668629A/en active Pending
-
2021
- 2021-10-25 WO PCT/CN2021/126013 patent/WO2022134828A1/en unknown
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2022134828A1 (en) * | 2020-12-24 | 2022-06-30 | 深圳壹账通智能科技有限公司 | Intelligent warehousing method and system based on picture recognition, and device and storage medium |
| CN113537237A (en) * | 2021-06-25 | 2021-10-22 | 西安交通大学 | Intelligent sensing method, system and device for multi-feature part quality information |
| CN113537237B (en) * | 2021-06-25 | 2024-01-16 | 西安交通大学 | Multi-feature part quality information intelligent sensing method, system and device |
| CN113469015A (en) * | 2021-06-29 | 2021-10-01 | 浙江大华技术股份有限公司 | Face recognition method and device, electronic equipment and computer storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2022134828A1 (en) | 2022-06-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112668629A (en) | Intelligent warehousing method, system, equipment and storage medium based on picture identification | |
| CN110543815A (en) | Training method of face recognition model, face recognition method, device, equipment and storage medium | |
| CN110751041A (en) | Certificate authenticity verification method, system, computer equipment and readable storage medium | |
| CN110825857B (en) | Multi-round question and answer identification method and device, computer equipment and storage medium | |
| CN112560964A (en) | Method and system for training Chinese herbal medicine pest and disease identification model based on semi-supervised learning | |
| CN111339979B (en) | Image recognition method and image recognition device based on feature extraction | |
| CN110378249B (en) | Text image inclination angle recognition method, device and equipment | |
| CN111355941A (en) | Image color real-time correction method, device and system | |
| CN111444775A (en) | Face key point correction method and device and computer equipment | |
| CN110458184B (en) | Optical character recognition assistance method, device, computer equipment and storage medium | |
| CN112434556A (en) | Pet nose print recognition method and device, computer equipment and storage medium | |
| CN111767965A (en) | Image matching method and device, electronic equipment and storage medium | |
| CN111310710A (en) | Face detection method and system | |
| WO2019095587A1 (en) | Face recognition method, application server, and computer-readable storage medium | |
| CN111680181A (en) | Abnormal object identification method and terminal equipment | |
| CN112699871B (en) | Method, system, device and computer readable storage medium for identifying field content | |
| CN110610117A (en) | Face recognition method, face recognition device and storage medium | |
| CN110751105B (en) | Finger image acquisition method and device and storage medium | |
| CN111428708A (en) | License plate checking method and device, computer equipment and storage medium | |
| CN112036362A (en) | Image processing method, image processing device, computer equipment and readable storage medium | |
| CN111950562A (en) | Picture processing method and device, terminal equipment and storage medium | |
| CN111382687A (en) | Face detection method and system | |
| CN107872451B (en) | User identity authentication method and identity authentication device | |
| CN111199182A (en) | Lost object method, system and storage medium based on intelligent community | |
| CN112650741B (en) | Method, system, equipment and readable storage medium for identifying and correcting abnormal data |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| REG | Reference to a national code |
Ref country code: HK Ref legal event code: DE Ref document number: 40049897 Country of ref document: HK |
|
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |