CN117437224B - Gas cylinder defect image data processing system and method based on artificial intelligence - Google Patents
Gas cylinder defect image data processing system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN117437224B CN117437224B CN202311753250.8A CN202311753250A CN117437224B CN 117437224 B CN117437224 B CN 117437224B CN 202311753250 A CN202311753250 A CN 202311753250A CN 117437224 B CN117437224 B CN 117437224B
- Authority
- CN
- China
- Prior art keywords
- gas cylinder
- detection
- data
- marked
- feature
- 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.)
- Active
Links
- 230000007547 defect Effects 0.000 title claims abstract description 37
- 238000012545 processing Methods 0.000 title claims abstract description 35
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 17
- 238000000034 method Methods 0.000 title claims abstract description 13
- 238000001514 detection method Methods 0.000 claims abstract description 212
- 230000002159 abnormal effect Effects 0.000 claims abstract description 104
- 238000004364 calculation method Methods 0.000 claims abstract description 20
- 238000011897 real-time detection Methods 0.000 claims abstract description 14
- 238000013500 data storage Methods 0.000 claims description 18
- 238000007689 inspection Methods 0.000 claims description 11
- 238000006243 chemical reaction Methods 0.000 claims description 9
- 238000012821 model calculation Methods 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 4
- 239000000654 additive Substances 0.000 claims description 3
- 230000008859 change Effects 0.000 claims description 3
- 239000011521 glass Substances 0.000 claims description 3
- 238000010079 rubber tapping Methods 0.000 claims description 3
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 238000012549 training Methods 0.000 claims description 3
- 239000002023 wood Substances 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 abstract description 2
- 238000012360 testing method Methods 0.000 description 12
- 238000005516 engineering process Methods 0.000 description 7
- 230000009471 action Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 230000001953 sensory effect Effects 0.000 description 3
- 230000007797 corrosion Effects 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Health & Medical Sciences (AREA)
- Economics (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Marketing (AREA)
- Medical Informatics (AREA)
- Evolutionary Computation (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Tourism & Hospitality (AREA)
- Artificial Intelligence (AREA)
- Development Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Educational Administration (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Primary Health Care (AREA)
- Filling Or Discharging Of Gas Storage Vessels (AREA)
Abstract
The invention discloses a gas cylinder defect image data processing system and method based on artificial intelligence, and belongs to the technical field of gas cylinder defect image data processing. The invention comprises the following steps: the system comprises a real-time detection image database module, a gas cylinder history detection big database module, an AI calculation module and a warning module; constructing a detection image database by using a real-time detection image database module, storing the detection image database when the system transmits the feature detection data of each detection of the real-time gas cylinder, and constructing a gas cylinder defect image processing port to call a marked abnormal object; constructing a non-safety gas cylinder data judging model, substituting the called marked abnormal object into the non-safety gas cylinder data judging model to form judging data; adding a warning sign to the gas cylinder with judging data meeting the judgment of the unsafe gas cylinder to remind an administrator to patrol; the invention can realize the intelligent processing of the defect image of the gas cylinder, provide more accurate depth on the defect evaluation of the gas cylinder and further improve the safety coefficient.
Description
Technical Field
The invention relates to the technical field of gas cylinder defect image data processing, in particular to a gas cylinder defect image data processing system and method based on artificial intelligence.
Background
The periodic inspection of the gas cylinder refers to the compliance verification activity of a special equipment inspection mechanism on the safety condition of the gas cylinder according to a certain time period and the related safety technical specification and the related standard. The regular test items of the gas cylinder comprise appearance test, sound test, internal test, bottle mouth thread test, weight and volume measurement, water pressure test, bottle valve test and air tightness test, and in the current technical means, the appearance test, sound test, internal test, bottle mouth thread test and the like all need manual judgment to carry out sensory analysis.
However, the current sensory analysis technology has a great limitation, and mainly relies on a few experiences of so-called "pedestrians", and the evaluation result lacks scientificity, objectivity and comparability. With the development of intelligent technology, the image comparison technology is gradually applied, the current image comparison technology is only aimed at single item inspection and single item comparison, and most of the time, single item inspection results are judged only through a calibration threshold value, and the correlation between single item inspection and single item inspection is ignored, for example, the corrosion phenomenon in the threshold value appears in the interior and the exterior of a gas cylinder at the same time, in the common inspection, both inspection can pass, but in the actual use, the corrosion speed of the gas cylinder can be rapidly increased, so that accidents are caused, and no related technology is available in the detection correlation aiming at the gas cylinder at the present.
Disclosure of Invention
The invention aims to provide a gas cylinder defect image data processing system and method based on artificial intelligence, so as to solve the problems in the background technology.
In order to solve the technical problems, the invention provides the following technical scheme: the gas cylinder defect image data processing method based on artificial intelligence comprises the following steps:
s1, constructing a gas cylinder defect image processing port, wherein the gas cylinder defect image processing port calls a marked abnormal object, and the marked abnormal object is stored in a detection image database;
s2, the detection image database stores feature detection data of each detection of the gas cylinder in real time, and the marked abnormal objects are converted in real time according to the marked model, so that the marked abnormal objects are ensured to be in the detection image database and only one of the marked abnormal objects is ensured to be in the detection image database;
s3, calling a gas cylinder history detection big database, wherein the gas cylinder history detection big database comprises two types of gas cylinders which are respectively marked as a safe gas cylinder and a non-safe gas cylinder, and acquiring feature detection data of each detection of the non-safe gas cylinder;
s4, constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data, and if the judgment data meets the unsafe gas cylinder judgment, adding a warning sign to the gas cylinder to remind an administrator to patrol.
According to the technical scheme, each detection of the gas cylinder comprises the following steps: appearance detection, sound detection, internal detection and bottle thread detection; the appearance detection is used for detecting the appearance of the gas cylinder to form a plurality of groups of cylinder appearance images; the sound detection means that under the condition that no additives or other obstacles interfere the vibration of the bottle body, a wood hammer is used for tapping the wall of the bottle, the vibrating sound passes through a microphone, then is amplified by an electron tube and a transformer, a waveform image is displayed from the vertical axis of the oscilloscope, a loop is formed by a thyratron, a relay and a rectifier, the waveform is transversely unfolded from one end to the other end along the transverse axis of the oscilloscope, and the waveform image is recorded; the internal detection is extended into the bottle mouth by means of an endoscope, so that a plurality of groups of image data in the bottle body are formed; the bottleneck thread detection forms an amplified image by means of a magnifying glass, and thread condition image data are generated;
the marking model includes:
acquiring feature detection data under each item of detection, calling bottle standard images of the feature detection data set by a system, and comparing similarity between the images:
；
wherein,similarity between the image representing the feature detection data and the bottle standard image of the feature detection data set by the system; />Any data feature under the image representing the feature detection data; />Correspondence of bottle standard image representing characteristic detection data set by system +.>Is a data feature of the same data; />Representing a serial number; t represents the number of features selected on the image;
the feature detection data such as the appearance detection includes: each item is provided with a standard image, such as a crack, a bottle body image with or without the crack is built in the system, the system calculates the crack similarity between the crack detected in real time and the standard image, and the more similar represents the closer to the standard image, namely the smaller represents the crack; the less similar, the larger the crack;
based on the similarity between the image of each feature detection data and the bottle standard image of the feature detection data set by the system, calculating an average value to form a marked abnormal value of the feature detection data, sequencing the marked abnormal values in order from large to small to form an abnormal sequence, and marking the feature detection data corresponding to the maximum value as a marked abnormal object;
after the marked abnormal object is substituted into the unsafe gas cylinder data judging model, deleting the marked abnormal object in the abnormal sequence, and simultaneously continuously selecting the characteristic detection data corresponding to the maximum value in the sequence to be marked as the marked abnormal object.
According to the above technical scheme, the safety gas cylinder and the non-safety gas cylinder comprise:
when all single feature detection data of any gas cylinder meet a set threshold value of a system, recording that the gas cylinder is in a safe mode, marking a safe service life according to a factory nameplate of the gas cylinder, and if the gas cylinder with any marked safe service life does not reach the safe service life, carrying out feature detection data early warning, wherein the marked gas cylinder is a non-safe gas cylinder;
the safety gas cylinder refers to a gas cylinder other than the non-safety gas cylinder.
According to the above technical scheme, the non-safety gas cylinder data judgment model comprises:
acquiring feature detection data of each detection of the non-safety gas cylinder, selecting feature detection data of each detection of the previous time of the non-safety gas cylinder, wherein in all the single feature detection data, a mark which is completely similar to a bottle standard image of the feature detection data set by a system is used as an index feature, a mark which meets a threshold set by the system but is not completely similar is used as a non-index feature, and establishing a non-index feature set for each non-safety gas cylinder;
acquiring the called marked abnormal object, and selecting a non-index feature set with the marked abnormal object to participate in model training:
removing marked abnormal objects from each non-index feature set, sorting feature detection data according to occurrence frequency to form a first feature sequence, setting a frequency threshold value, not counting the first feature sequence, selecting the feature detection data detected by each item of real-time gas cylinder according to the first feature sequence, if a certain feature detection data is marked as an index feature in the feature detection data detected by each item of real-time gas cylinder, not selecting, forming the selected feature detection data into a second feature sequence, recording the total number n of features of the second feature sequence, and intelligently generating the number P of combination modes in the second feature sequence:
；
wherein,is the number of features within the combination;
calculating the probability duty ratio of each combination in the non-index feature set with the marked abnormal object, and generating the non-safety gas cylinder judgment probability of the marked abnormal object:
；
wherein H represents the judging probability of the unsafe gas cylinder marking the abnormal object; j represents any combination mode;the probability duty ratio of the combination mode j in the non-index feature set is indicated; k represents a constant of a measurement coefficient, is a constant term smaller than 1 and larger than 0, and is generally infinitely close to 0 and is far smaller than 1 under the condition of value;
continuously calling the marked abnormal object, calculating the judging probability of the unsafe gas cylinder of the marked abnormal object, setting a calibration value by the system, if the judging probability of the unsafe gas cylinder of the marked abnormal object exceeds the calibration value by V, representing that judging data meet the judging of the unsafe gas cylinder, adding a warning mark to the gas cylinder, and reminding an administrator to patrol; if the number of the unsafe gas cylinders of the marked abnormal objects exceeding the standard value does not reach V after the marked abnormal objects are called, the marked abnormal objects are marked as safe gas cylinders, and V is manually set by a person. By means of the method, the processing progress of the image data of the gas cylinder defect can be accelerated, the abnormal object is marked generally, probability influence which can be brought by the abnormal object is largest, the abnormal object is selected continuously according to the sequence, if the larger probability influence successfully reaches V, calculation is not needed to be carried out later, the working frequency is further reduced, the processing speed is improved, and the V is generally set to be three.
An artificial intelligence based gas cylinder defect image data processing system, the system comprising: the system comprises a real-time detection image database module, a gas cylinder history detection big database module, an AI calculation module and a warning module;
the real-time detection image database module is used for constructing a detection image database, storing the detection image database when the system transmits the feature detection data of each detection of the real-time gas cylinder, and constructing a gas cylinder defect image processing port to call a marked abnormal object; the gas cylinder history detection big database module is used for storing characteristic detection data of various historic detections of different types of gas cylinders; the AI calculation module is used for constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data; the warning module is used for judging whether the judging data meet the judgment of the unsafe gas cylinder, adding a warning sign to the gas cylinder and reminding an administrator to patrol;
the output ends of the real-time detection image database module and the gas cylinder history detection large database module are connected with the input end of the AI calculation module; the output end of the AI calculation module is connected with the input end of the warning module.
According to the technical scheme, the real-time detection image database module comprises a detection image database unit and a mark conversion unit;
the detection image database unit is used for constructing a detection image database and storing the detection image database when the system transmits the characteristic detection data of each detection of the real-time gas cylinder; the marking conversion unit is used for constructing a gas cylinder defect image processing port, converting marked abnormal objects in real time according to a marking model, ensuring that only one marked abnormal object exists in the detection image database, and calling the marked abnormal object at the same time;
the output end of the detection image database unit is connected with the input end of the mark conversion unit.
According to the technical scheme, the gas cylinder history detection big database module comprises a safety gas cylinder data storage unit and a non-safety gas cylinder data storage unit;
the safety gas cylinder data storage unit is used for storing characteristic detection data of various historical detections of the safety gas cylinder; the non-safety gas cylinder data storage unit is used for storing characteristic detection data of various historical detections of the non-safety gas cylinder;
when all single feature detection data of any gas cylinder meet a set threshold value of a system, recording that the gas cylinder is in a safe mode, marking a safe service life according to a factory nameplate of the gas cylinder, and if the gas cylinder with any marked safe service life does not reach the safe service life, carrying out feature detection data early warning, wherein the marked gas cylinder is a non-safe gas cylinder;
the safety gas cylinder refers to a gas cylinder except for a non-safety gas cylinder;
and the output ends of the safety gas cylinder data storage unit and the non-safety gas cylinder data storage unit are connected to the AI calculation module.
According to the technical scheme, the AI calculation module comprises a model calculation unit and a judgment unit;
the model calculation unit is used for constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data; the judging unit is used for calculating the judging probability of the unsafe gas cylinder for marking the abnormal object, setting a calibration value by the system, and recording the quantity that the judging probability of the unsafe gas cylinder for marking the abnormal object exceeds the calibration value;
the output end of the model calculation unit is connected with the input end of the judging unit.
According to the above technical scheme, the warning module includes:
when judging that the judging data meets the judgment of the unsafe gas cylinder, adding a warning sign to the gas cylinder by using a red mark, outputting the warning sign to the inside of the system, and reminding an administrator to change the inspection period of the gas cylinder.
Compared with the prior art, the invention has the following beneficial effects: the invention can replace the current sensory analysis technology in an image processing mode, and improves the scientificity, objectivity and comparability of detection. Meanwhile, a series of brand new judging modes are provided for the image processing process, so that the defects caused by single detection in the current environment can be overcome, the safety degree is further improved, and the accident risk is reduced.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a flow chart of the gas cylinder defect image data processing method based on artificial intelligence.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, in a first embodiment, a method for processing image data of a gas cylinder defect based on artificial intelligence is provided:
firstly, constructing a gas cylinder defect image processing port, wherein the gas cylinder defect image processing port is a computer processing platform, and the main function is used for calling a marked abnormal object which is stored in a detection image database and is formed based on feature detection data of various detections of the gas cylinder stored in the detection image database in real time;
the marking modes comprise: acquiring feature detection data under each item of detection, calling bottle standard images of the feature detection data set by a system, and comparing similarity between the images:
；
wherein,similarity between the image representing the feature detection data and the bottle standard image of the feature detection data set by the system; />Any data feature under the image representing the feature detection data; />Correspondence of bottle standard image representing characteristic detection data set by system +.>Is a data feature of the same data; />Representing a serial number; m represents the number of features selected on the image;
based on the similarity between the image of each feature detection data and the bottle standard image of the feature detection data set by the system, calculating an average value to form a marked abnormal value of the feature detection data, sequencing the marked abnormal values in order from large to small to form an abnormal sequence, and marking the feature detection data corresponding to the maximum value as a marked abnormal object;
after the marked abnormal object is substituted into the non-safety gas cylinder data judging model, deleting the marked abnormal object in the abnormal sequence, and simultaneously continuously selecting the feature detection data corresponding to the maximum value in the sequence to be marked as the marked abnormal object, so that the marked abnormal object is ensured to exist in the detection image database and only exist in the detection image database.
Secondly, calling a marked abnormal object by using a gas cylinder defect image processing port, forming a non-safety gas cylinder data judging model based on a gas cylinder history detection large database, wherein the gas cylinder history detection large database comprises two types of gas cylinders which are respectively marked as a safety gas cylinder and a non-safety gas cylinder,
the safety gas cylinder and the non-safety gas cylinder comprise:
when all single feature detection data of any gas cylinder meet a set threshold value of a system, recording that the gas cylinder is in a safe mode, marking a safe service life according to a factory nameplate of the gas cylinder, and if the gas cylinder with any marked safe service life does not reach the safe service life, carrying out feature detection data early warning, wherein the marked gas cylinder is a non-safe gas cylinder;
the safety gas cylinder refers to a gas cylinder other than the non-safety gas cylinder.
Each item of detection of the gas cylinder comprises: appearance detection, sound detection, internal detection and bottle thread detection; the appearance detection is used for detecting the appearance of the gas cylinder to form a plurality of groups of cylinder appearance images; the sound detection means that under the condition that no additives or other obstacles interfere the vibration of the bottle body, a wood hammer is used for tapping the wall of the bottle, the vibrating sound passes through a microphone, then is amplified by an electron tube and a transformer, a waveform image is displayed from the vertical axis of the oscilloscope, a loop is formed by a thyratron, a relay and a rectifier, the waveform is transversely unfolded from one end to the other end along the transverse axis of the oscilloscope, and the waveform image is recorded; the internal detection is extended into the bottle mouth by means of an endoscope, so that a plurality of groups of image data in the bottle body are formed; the bottleneck thread detection forms an amplified image by means of a magnifying glass, and thread condition image data are generated;
acquiring feature detection data of each detection of the non-safety gas cylinder, selecting feature detection data of each detection of the previous time of the non-safety gas cylinder, wherein in all the single feature detection data, a mark which is completely similar to a bottle standard image of the feature detection data set by a system is used as an index feature, a mark which meets a threshold set by the system but is not completely similar is used as a non-index feature, and establishing a non-index feature set for each non-safety gas cylinder;
finally, acquiring the called marked abnormal object, and selecting a non-index feature set with the marked abnormal object to participate in model training:
removing marked abnormal objects from each non-index feature set, sorting feature detection data according to occurrence frequency to form a first feature sequence, setting a frequency threshold value, not counting the first feature sequence, selecting the feature detection data detected by each item of real-time gas cylinder according to the first feature sequence, if a certain feature detection data is marked as an index feature in the feature detection data detected by each item of real-time gas cylinder, not selecting, forming the selected feature detection data into a second feature sequence, recording the total number n of features of the second feature sequence, and intelligently generating the number P of combination modes in the second feature sequence:
；
wherein,is the number of features within the combination;
calculating the probability duty ratio of each combination in the non-index feature set with the marked abnormal object, and generating the non-safety gas cylinder judgment probability of the marked abnormal object:
；
wherein H represents the judging probability of the unsafe gas cylinder marking the abnormal object; j represents any combination mode;the probability duty ratio of the combination mode j in the non-index feature set is indicated; k represents a constant of a measurement coefficient, which is a constant term smaller than 1 and larger than 0;
continuously calling the marked abnormal object, calculating the judging probability of the unsafe gas cylinder of the marked abnormal object, setting a calibration value by the system, if the judging probability of the unsafe gas cylinder of the marked abnormal object exceeds the calibration value by V, representing that judging data meet the judging of the unsafe gas cylinder, adding a warning mark to the gas cylinder, and reminding an administrator to patrol; if the number of the unsafe gas cylinders of the marked abnormal objects exceeding the standard value does not reach V after the marked abnormal objects are called, the marked abnormal objects are marked as safe gas cylinders, and V is manually set by a person.
In a second embodiment, an artificial intelligence-based gas cylinder defect image data processing system is provided, the system including: the system comprises a real-time detection image database module, a gas cylinder history detection big database module, an AI calculation module and a warning module;
the real-time detection image database module is used for constructing a detection image database, storing the detection image database when the system transmits the feature detection data of each detection of the real-time gas cylinder, and constructing a gas cylinder defect image processing port to call a marked abnormal object; the gas cylinder history detection big database module is used for storing characteristic detection data of various historic detections of different types of gas cylinders; the AI calculation module is used for constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data; the warning module is used for judging whether the judging data meet the judgment of the unsafe gas cylinder, adding a warning sign to the gas cylinder and reminding an administrator to patrol;
the output ends of the real-time detection image database module and the gas cylinder history detection large database module are connected with the input end of the AI calculation module; the output end of the AI calculation module is connected with the input end of the warning module.
The real-time detection image database module comprises a detection image database unit and a mark conversion unit;
the detection image database unit is used for constructing a detection image database and storing the detection image database when the system transmits the characteristic detection data of each detection of the real-time gas cylinder; the marking conversion unit is used for constructing a gas cylinder defect image processing port, converting marked abnormal objects in real time according to a marking model, ensuring that only one marked abnormal object exists in the detection image database, and calling the marked abnormal object at the same time;
the output end of the detection image database unit is connected with the input end of the mark conversion unit.
The gas cylinder history detection big database module comprises a safety gas cylinder data storage unit and a non-safety gas cylinder data storage unit;
the safety gas cylinder data storage unit is used for storing characteristic detection data of various historical detections of the safety gas cylinder; the non-safety gas cylinder data storage unit is used for storing characteristic detection data of various historical detections of the non-safety gas cylinder;
when all single feature detection data of any gas cylinder meet a set threshold value of a system, recording that the gas cylinder is in a safe mode, marking a safe service life according to a factory nameplate of the gas cylinder, and if the gas cylinder with any marked safe service life does not reach the safe service life, carrying out feature detection data early warning, wherein the marked gas cylinder is a non-safe gas cylinder;
the safety gas cylinder refers to a gas cylinder except for a non-safety gas cylinder;
and the output ends of the safety gas cylinder data storage unit and the non-safety gas cylinder data storage unit are connected to the AI calculation module.
The AI calculation module comprises a model calculation unit and a judgment unit;
the model calculation unit is used for constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data; the judging unit is used for calculating the judging probability of the unsafe gas cylinder for marking the abnormal object, setting a calibration value by the system, and recording the quantity that the judging probability of the unsafe gas cylinder for marking the abnormal object exceeds the calibration value;
the output end of the model calculation unit is connected with the input end of the judging unit.
The warning module comprises:
when judging that the judging data meets the judgment of the unsafe gas cylinder, adding a warning sign to the gas cylinder by using a red mark, outputting the warning sign to the inside of the system, and reminding an administrator to change the inspection period of the gas cylinder.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (6)
1. The gas cylinder defect image data processing method based on artificial intelligence is characterized by comprising the following steps of: the method comprises the following steps:
s1, constructing a gas cylinder defect image processing port, wherein the gas cylinder defect image processing port calls a marked abnormal object, and the marked abnormal object is stored in a detection image database;
s2, the detection image database stores feature detection data of each detection of the gas cylinder in real time, and the marked abnormal objects are converted in real time according to the marked model, so that the marked abnormal objects are ensured to be in the detection image database and only one of the marked abnormal objects is ensured to be in the detection image database;
s3, calling a gas cylinder history detection big database, wherein the gas cylinder history detection big database comprises two types of gas cylinders which are respectively marked as a safe gas cylinder and a non-safe gas cylinder, and acquiring feature detection data of each detection of the non-safe gas cylinder;
s4, constructing a non-safety gas cylinder data judgment model, substituting the non-safety gas cylinder data judgment model based on the called marked abnormal object to form judgment data, and if the judgment data meets the non-safety gas cylinder judgment, adding a warning sign to the gas cylinder to remind an administrator to patrol;
each item of detection of the gas cylinder comprises: appearance detection, sound detection, internal detection and bottle thread detection; the appearance detection is used for detecting the appearance of the gas cylinder to form a plurality of groups of cylinder appearance images; the sound detection means that under the condition that no additives or other obstacles interfere the vibration of the bottle body, a wood hammer is used for tapping the wall of the bottle, the vibrating sound passes through a microphone, then is amplified by an electron tube and a transformer, a waveform image is displayed from the vertical axis of the oscilloscope, a loop is formed by a thyratron, a relay and a rectifier, the waveform is transversely unfolded from one end to the other end along the transverse axis of the oscilloscope, and the waveform image is recorded; the internal detection is extended into the bottle mouth by means of an endoscope, so that a plurality of groups of image data in the bottle body are formed; the bottleneck thread detection forms an amplified image by means of a magnifying glass, and thread condition image data are generated;
the marking model includes:
acquiring feature detection data under each item of detection, calling bottle standard images of the feature detection data set by a system, and comparing similarity between the images:
；
wherein,similarity between the image representing the feature detection data and the bottle standard image of the feature detection data set by the system; />Any one of the data features of the image representing the feature detection data; />Correspondence of bottle standard image representing characteristic detection data set by system +.>Is a data feature of the same data; />Representing a serial number; t represents the number of features selected on the image;
based on the similarity between the image of each feature detection data and the bottle standard image of the feature detection data set by the system, calculating an average value to form a marked abnormal value of the feature detection data, sequencing the marked abnormal values in order from large to small to form an abnormal sequence, and marking the feature detection data corresponding to the maximum value as a marked abnormal object;
after the marked abnormal object is substituted into the non-safety gas cylinder data judging model, deleting the marked abnormal object in the abnormal sequence, and simultaneously continuously selecting the characteristic detection data corresponding to the maximum value in the sequence to be marked as the marked abnormal object;
the safety gas cylinder and the non-safety gas cylinder comprise:
when all single feature detection data of any gas cylinder meet a set threshold value of a system, recording that the gas cylinder is in a safe mode, marking a safe service life according to a factory nameplate of the gas cylinder, and if the gas cylinder with any marked safe service life does not reach the safe service life, carrying out feature detection data early warning, wherein the marked gas cylinder is a non-safe gas cylinder;
the safety gas cylinder refers to a gas cylinder except for a non-safety gas cylinder;
the unsafe gas cylinder data judging model comprises:
acquiring feature detection data of each detection of the non-safety gas cylinder, selecting feature detection data of each detection of the previous time of the non-safety gas cylinder, wherein in all the single feature detection data, a mark which is completely similar to a bottle standard image of the feature detection data set by a system is used as an index feature, a mark which meets a threshold set by the system but is not completely similar is used as a non-index feature, and establishing a non-index feature set for each non-safety gas cylinder;
acquiring the called marked abnormal object, and selecting a non-index feature set with the marked abnormal object to participate in model training:
removing marked abnormal objects from each non-index feature set, sorting feature detection data according to occurrence frequency to form a first feature sequence, setting a frequency threshold value, not counting the first feature sequence, selecting the feature detection data detected by each item of real-time gas cylinder according to the first feature sequence, if a certain feature detection data is marked as an index feature in the feature detection data detected by each item of real-time gas cylinder, not selecting, forming the selected feature detection data into a second feature sequence, recording the total number n of features of the second feature sequence, and intelligently generating the number P of combination modes in the second feature sequence:
；
wherein,is the number of features within the combination;
calculating the probability duty ratio of each combination in the non-index feature set with the marked abnormal object, and generating the non-safety gas cylinder judgment probability of the marked abnormal object:
；
wherein H represents the judging probability of the unsafe gas cylinder marking the abnormal object; j represents any combination mode;the probability duty ratio of the finger combination mode j in the non-index feature set; k represents a constant of a measurement coefficient, which is a constant term smaller than 1 and larger than 0;
continuously calling the marked abnormal object, calculating the judging probability of the unsafe gas cylinder of the marked abnormal object, setting a calibration value by the system, if the judging probability of the unsafe gas cylinder of the marked abnormal object exceeds the calibration value by V, representing that judging data meet the judging of the unsafe gas cylinder, adding a warning mark to the gas cylinder, and reminding an administrator to patrol; if the number of the unsafe gas cylinders of the marked abnormal objects exceeding the standard value does not reach V after the marked abnormal objects are called, the marked abnormal objects are marked as safe gas cylinders, and V is manually set by a person.
2. An artificial intelligence based gas cylinder defect image data processing system, which uses the artificial intelligence based gas cylinder defect image data processing method as set forth in claim 1, wherein: the system comprises: the system comprises a real-time detection image database module, a gas cylinder history detection big database module, an AI calculation module and a warning module;
the real-time detection image database module is used for constructing a detection image database, storing the detection image database when the system transmits the feature detection data of each detection of the real-time gas cylinder, and constructing a gas cylinder defect image processing port to call a marked abnormal object; the gas cylinder history detection big database module is used for storing characteristic detection data of various historic detections of different types of gas cylinders; the AI calculation module is used for constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data; the warning module is used for judging whether the judging data meet the judgment of the unsafe gas cylinder, adding a warning sign to the gas cylinder and reminding an administrator to patrol;
the output ends of the real-time detection image database module and the gas cylinder history detection large database module are connected with the input end of the AI calculation module; the output end of the AI calculation module is connected with the input end of the warning module.
3. The artificial intelligence based cylinder defect image data processing system of claim 2, wherein: the real-time detection image database module comprises a detection image database unit and a mark conversion unit;
the detection image database unit is used for constructing a detection image database and storing the detection image database when the system transmits the characteristic detection data of each detection of the real-time gas cylinder; the marking conversion unit is used for constructing a gas cylinder defect image processing port, converting marked abnormal objects in real time according to a marking model, ensuring that only one marked abnormal object exists in the detection image database, and calling the marked abnormal object at the same time;
the output end of the detection image database unit is connected with the input end of the mark conversion unit.
4. The artificial intelligence based cylinder defect image data processing system of claim 2, wherein: the gas cylinder history detection big database module comprises a safety gas cylinder data storage unit and a non-safety gas cylinder data storage unit;
the safety gas cylinder data storage unit is used for storing characteristic detection data of various historical detections of the safety gas cylinder; the non-safety gas cylinder data storage unit is used for storing characteristic detection data of various historical detections of the non-safety gas cylinder;
when all single feature detection data of any gas cylinder meet a set threshold value of a system, recording that the gas cylinder is in a safe mode, marking a safe service life according to a factory nameplate of the gas cylinder, and if the gas cylinder with any marked safe service life does not reach the safe service life, carrying out feature detection data early warning, wherein the marked gas cylinder is a non-safe gas cylinder;
the safety gas cylinder refers to a gas cylinder except for a non-safety gas cylinder;
and the output ends of the safety gas cylinder data storage unit and the non-safety gas cylinder data storage unit are connected to the AI calculation module.
5. The artificial intelligence based cylinder defect image data processing system of claim 2, wherein: the AI calculation module comprises a model calculation unit and a judgment unit;
the model calculation unit is used for constructing an unsafe gas cylinder data judgment model, substituting the unsafe gas cylinder data judgment model based on the called marked abnormal object to form judgment data; the judging unit is used for calculating the judging probability of the unsafe gas cylinder for marking the abnormal object, setting a calibration value by the system, and recording the quantity that the judging probability of the unsafe gas cylinder for marking the abnormal object exceeds the calibration value;
the output end of the model calculation unit is connected with the input end of the judging unit.
6. The artificial intelligence based cylinder defect image data processing system of claim 2, wherein: the warning module comprises:
when judging that the judging data meets the judgment of the unsafe gas cylinder, adding a warning sign to the gas cylinder by using a red mark, outputting the warning sign to the inside of the system, and reminding an administrator to change the inspection period of the gas cylinder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311753250.8A CN117437224B (en) | 2023-12-20 | 2023-12-20 | Gas cylinder defect image data processing system and method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311753250.8A CN117437224B (en) | 2023-12-20 | 2023-12-20 | Gas cylinder defect image data processing system and method based on artificial intelligence |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117437224A CN117437224A (en) | 2024-01-23 |
| CN117437224B true CN117437224B (en) | 2024-03-29 |
Family
ID=89558568
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311753250.8A Active CN117437224B (en) | 2023-12-20 | 2023-12-20 | Gas cylinder defect image data processing system and method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117437224B (en) |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110260152A (en) * | 2019-06-12 | 2019-09-20 | 无锡泓瑞航天科技有限公司 | A kind of liquid petroleum gas vehicle vehicle gas cylinder wisdom monitoring system |
| CN110298839A (en) * | 2019-07-10 | 2019-10-01 | 北京滴普科技有限公司 | A kind of open defect intelligent distinguishing system based on data-driven |
| CN112033457A (en) * | 2020-08-11 | 2020-12-04 | 湖北特种设备检验检测研究院 | Gas cylinder safety state on-line monitoring device |
| CN112270262A (en) * | 2020-10-28 | 2021-01-26 | 福州市君则信息技术有限公司 | Intelligent filling gas cylinder detection system and method |
| US11145202B1 (en) * | 2020-05-01 | 2021-10-12 | United Services Automobile Association (Usaa) | Systems and methods for providing warnings of imminent hazards |
| CN113533497A (en) * | 2021-07-21 | 2021-10-22 | 北京航空航天大学 | Detection system and defect diagnosis method for gas cylinder acoustic emission signals in extreme environment |
| CN114062828A (en) * | 2021-12-04 | 2022-02-18 | 国网山西省电力公司电力科学研究院 | Transformer fault online monitoring and sensing device and method based on gas collecting box |
| CN115330802A (en) * | 2022-10-17 | 2022-11-11 | 山东大学 | Carbon fiber composite material gas cylinder X-ray image debonding defect extraction method |
| CN115330702A (en) * | 2022-08-01 | 2022-11-11 | 无锡雪浪数制科技有限公司 | Beverage bottle filling defect identification method based on deep vision |
| CN115494077A (en) * | 2022-10-18 | 2022-12-20 | 浙江蓝能燃气设备有限公司 | Gas cylinder detection device based on single sample learning AI algorithm and application method |
| CN115995119A (en) * | 2023-03-23 | 2023-04-21 | 山东特联信息科技有限公司 | Gas cylinder filling link illegal behavior identification method and system based on Internet of things |
| CN116597388A (en) * | 2023-07-18 | 2023-08-15 | 山东特联信息科技有限公司 | Method and system for monitoring hidden danger in filling process of vehicle gas cylinder based on image recognition |
| CN117252809A (en) * | 2023-08-01 | 2023-12-19 | 南京苏胜天信息科技有限公司 | Identification system and method based on AI industrial vision |
-
2023
- 2023-12-20 CN CN202311753250.8A patent/CN117437224B/en active Active
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110260152A (en) * | 2019-06-12 | 2019-09-20 | 无锡泓瑞航天科技有限公司 | A kind of liquid petroleum gas vehicle vehicle gas cylinder wisdom monitoring system |
| CN110298839A (en) * | 2019-07-10 | 2019-10-01 | 北京滴普科技有限公司 | A kind of open defect intelligent distinguishing system based on data-driven |
| US11145202B1 (en) * | 2020-05-01 | 2021-10-12 | United Services Automobile Association (Usaa) | Systems and methods for providing warnings of imminent hazards |
| CN112033457A (en) * | 2020-08-11 | 2020-12-04 | 湖北特种设备检验检测研究院 | Gas cylinder safety state on-line monitoring device |
| CN112270262A (en) * | 2020-10-28 | 2021-01-26 | 福州市君则信息技术有限公司 | Intelligent filling gas cylinder detection system and method |
| CN113533497A (en) * | 2021-07-21 | 2021-10-22 | 北京航空航天大学 | Detection system and defect diagnosis method for gas cylinder acoustic emission signals in extreme environment |
| CN114062828A (en) * | 2021-12-04 | 2022-02-18 | 国网山西省电力公司电力科学研究院 | Transformer fault online monitoring and sensing device and method based on gas collecting box |
| CN115330702A (en) * | 2022-08-01 | 2022-11-11 | 无锡雪浪数制科技有限公司 | Beverage bottle filling defect identification method based on deep vision |
| CN115330802A (en) * | 2022-10-17 | 2022-11-11 | 山东大学 | Carbon fiber composite material gas cylinder X-ray image debonding defect extraction method |
| CN115494077A (en) * | 2022-10-18 | 2022-12-20 | 浙江蓝能燃气设备有限公司 | Gas cylinder detection device based on single sample learning AI algorithm and application method |
| CN115995119A (en) * | 2023-03-23 | 2023-04-21 | 山东特联信息科技有限公司 | Gas cylinder filling link illegal behavior identification method and system based on Internet of things |
| CN116597388A (en) * | 2023-07-18 | 2023-08-15 | 山东特联信息科技有限公司 | Method and system for monitoring hidden danger in filling process of vehicle gas cylinder based on image recognition |
| CN117252809A (en) * | 2023-08-01 | 2023-12-19 | 南京苏胜天信息科技有限公司 | Identification system and method based on AI industrial vision |
Non-Patent Citations (4)
| Title |
|---|
| Defect Detection in the Winding Layer of CNG Composite‑Wound Cylindersvirtue of minimal mechanism is that it allows;Hui Lv 等;《International Journal of Precision Engineering and Manufacturing》;20211231;第22卷;1937–1946 * |
| 化工压力容器焊接质量缺陷及控制措施分析;扈然;《中国高新技术企业》;20161231(第25期);85-86 * |
| 基于机器视觉的气瓶内壁缺陷在线检测系统的研究;王英;《中国优秀硕士学位论文全文数据库 工程科技Ⅱ辑（月刊）》;20181215;第第2018年卷卷(第第12期期);C029-223 * |
| 车用气瓶缺陷在线检测技术分析;王东奎 等;《安防科技》;20201231(第28期);15 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117437224A (en) | 2024-01-23 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US7265662B2 (en) | Apparatus and method for inspecting containers | |
| CN106841403A (en) | A kind of acoustics glass defect detection method based on neutral net | |
| CN110490125A (en) | A kind of fueling area service quality detection system detected automatically based on gesture | |
| CN107240216A (en) | Based on 3DGIS+BIM technologies and artificial intelligence O＆M emergent alarm and fast response method | |
| CN113075065B (en) | Deep sea pipeline crack propagation monitoring and reliability evaluation system based on image recognition | |
| CN115077627B (en) | Multi-fusion environmental data supervision method and supervision system | |
| CN110674790A (en) | Abnormal scene processing method and system in video monitoring | |
| CN112819068A (en) | Deep learning-based real-time detection method for ship operation violation behaviors | |
| CN114842560B (en) | Computer vision-based construction site personnel dangerous behavior identification method | |
| CN112163760A (en) | Student learning concentration degree detection method and system | |
| CN117437224B (en) | Gas cylinder defect image data processing system and method based on artificial intelligence | |
| CN117576632B (en) | Multi-mode AI large model-based power grid monitoring fire early warning system and method | |
| CN113255550B (en) | Pedestrian garbage throwing frequency counting method based on video | |
| CN111191855A (en) | Water quality abnormal event identification and early warning method based on pipe network multi-element water quality time sequence data | |
| CN112215264B (en) | Steel rail abrasion detection method based on steel rail light band image | |
| CN117437740A (en) | Industrial fire risk early warning and analyzing method based on big data technology | |
| CN109902752A (en) | Bridge regular inspection classification method, device and equipment based on defect information big data | |
| CN116910619A (en) | Ship welding defect recognition system based on artificial intelligence | |
| CN115791803A (en) | Deep-buried tunnel surrounding rock blasting damage test system and test method | |
| CN115063427A (en) | Pollutant discharge monitoring image processing method for novel ship | |
| CN115311601A (en) | Fire detection analysis method based on video analysis technology | |
| CN113159230A (en) | Metallographic grade classification method based on deep learning | |
| CN113256560A (en) | Heading machine nose area intrusion detection method based on YOLOv5 | |
| CN111812116A (en) | On-line monitoring method and monitoring system for oil smoke pipeline | |
| CN116434508B (en) | Anti-collision monitoring big data early warning system and method based on Internet of things |
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 |