CN110405540A - A kind of artificial intelligence breaking detection system and method - Google Patents
A kind of artificial intelligence breaking detection system and method Download PDFInfo
- Publication number
- CN110405540A CN110405540A CN201910616439.XA CN201910616439A CN110405540A CN 110405540 A CN110405540 A CN 110405540A CN 201910616439 A CN201910616439 A CN 201910616439A CN 110405540 A CN110405540 A CN 110405540A
- Authority
- CN
- China
- Prior art keywords
- tool
- image
- breaking
- cutter
- breaking detection
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/09—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool
- B23Q17/0904—Arrangements for observing, indicating or measuring on machine tools for indicating or measuring cutting pressure or for determining cutting-tool condition, e.g. cutting ability, load on tool before or after machining
- B23Q17/0909—Detection of broken tools
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
- B23Q17/00—Arrangements for observing, indicating or measuring on machine tools
- B23Q17/24—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves
- B23Q17/2452—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces
- B23Q17/2457—Arrangements for observing, indicating or measuring on machine tools using optics or electromagnetic waves for measuring features or for detecting a condition of machine parts, tools or workpieces of tools
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of artificial intelligence breaking detection systems, including image capture module, tool magazine, Cloud Server, image processing module, and wherein Cloud Server is used to obtain breaking detection model based on the training of cutter intersection point feature;Image processing module loads the breaking detection model that training obtains in Cloud Server, is then based on the breaking detection model and calculates Tool monitoring result for pre-processing to tool image.The invention discloses a kind of artificial intelligence knife-breaking detecting methods, by shooting tool image after processing is completed, and to carrying out auxiliary wire tag, then the intersection point feature training breaking detection model between tool nose and auxiliary line is extracted, and incremental training is carried out to the tool image of judgement error, constantly breaking detection model is updated, it may be implemented to detect a variety of different types of cutters, the influence for reducing external environment, substantially increases the generalization ability of breaking detection model and the accuracy rate of detection.
Description
Technical field
The invention belongs to intelligent numerical control machine tool fields, more particularly, to a kind of artificial intelligence breaking detection system and side
Method.
Background technique
In order to make numerically-controlled machine tool it is more intelligent, automation, guarantee numerical control machine high accurate, high speed, it is efficient from
Dynamicization operation becomes one of the main direction of studying of numerically-controlled machine tool.In process, numerically-controlled machine tool process tool can be with adding
Different degrees of damage occurs for work process, influences processing efficiency.Therefore, cutter breakdown situation is detected in time, can be reduced subsequent
The scrappage of part reduces the loss of lathe, guarantees the processing efficiency of lathe.
The knife-breaking detecting method of existing view-based access control model is broadly divided into detection indirectly and directly two kinds of detection, wherein inspection indirectly
Survey detects the part after processing, in patent CN109540919A, carries out by using level-one lifting device to part
Then detection realizes the inspection to the left and right sides of part by the electrooptical device of the right and left and fixed device detection device
It surveys, then a photoelectric detection system is set below workbench, feature bottom is detected, to judge whether to break
Knife, this is a kind of indirect mode, and the cost of this method not only whole equipment is relatively high, but also needs for different types of
Processing part does different judgment method processing, and the usable range of whole equipment is narrow.The side that another kind directly detects
Method directly shoots cutter, judges whether cutter wears, logical first in the patent that number is CN109500657A
The threshold value for crossing setting obtains the image of corresponding process tool, by the way that image is carried out gray processing, obtains corresponding bianry image,
After carrying out Morphological scale-space, the profile of bianry image is extracted, calculates the area ratio of the profile after extracting and the normal cutter of calibration,
Judge its relationship between preset threshold, realizes the detection to cutter.This method needs to demarcate the area of normal cutter in advance
As reference value, when demarcating to the entire cutter area on picture, the influence of artificial calibrated error is bigger, in addition,
When extracting profile, it is illuminated by the light bigger with background influence, generalization ability is weaker, and when tool category is more, serious forgiveness is smaller,
Accuracy is lower.
It is therefore proposed that the problem of a kind of generalization ability is strong, accuracy is high breaking detection system and method are urgent need to resolve.
Summary of the invention
In view of the drawbacks of the prior art, it is an object of the invention to propose a kind of artificial intelligence breaking detection system and side
Method, it is intended to solve the prior art due to using the area for the normal cutter demarcated in advance as reference value carry out breaking judgement when by people
The lower problem of accuracy caused by work calibration and illumination and Pekinese are affected.
To achieve the above object, one aspect of the present invention provides a kind of artificial intelligence breaking detection system, comprising:
Image capture module, tool magazine, Cloud Server, image processing module;
Wherein, the output end of image capture module is connected with the input terminal of image processing module, image capture module and knife
It is spaced a distance between library, is communicated between image processing module and Cloud Server by Ethernet;
The cutter to be processed that image capture module is used to shoot in tool magazine obtains the calibration information of position of tool tip, and claps
The tool image that cutter after processing is completed obtains blurred background and cutter object highlights is taken the photograph, and is transferred to image processing module
In;
Tool magazine is used to store cutter to be processed and cutter to be detected after processing is completed;
Cloud Server is used to obtain breaking detection model based on the training of cutter intersection point feature;
Image processing module loads the breaking detection that training obtains in Cloud Server for pre-processing to tool image
Model is then based on the breaking detection model and calculates Tool monitoring result.
It is further preferred that image capture module includes endoscope, using the cutter in endoscope shooting tool magazine, cutter is not
When being processed, tool image is shot, and demarcate the location information where tool nose, after the completion of tool sharpening again by cutter
Tool magazine is gained, using the cutter to be detected of endoscope shooting after processing is completed, the tool image of blurred background is obtained, to highlight
Cutter object.
It is further preferred that image processing module can be the IPC equipped with AI chip.
It is further preferred that image capture module, tool magazine, image processing module are embeddable into lathe;
It is further preferred that another aspect of the present invention provides a kind of artificial intelligence knife-breaking detecting method, including following step
It is rapid:
S1, image processing module connect Cloud Server, and breaking detection model is downloaded from Cloud Server;
S2, it is taken pictures to all cutters to be processed in tool magazine and demarcates its position of tool tip;
S3, cutter to be processed gain the corresponding position of tool magazine after processing is completed, and take pictures to it, obtain to be checked
Survey tool image;
S4, tool image to be detected is pre-processed based on the position of tool tip information demarcated, is obtained with auxiliary mark
The tool image of note;
S5, intersection point feature between the tool nose and auxiliary line of tool image to be detected is extracted;
S6, intersection point feature is input in breaking detection model trained in advance, obtains breaking testing result;
S7, the tool image sample that will test mistake are transmitted on Cloud Server, obtain error sample collection.
It is further preferred that pretreated method is carried out to tool image in above-mentioned steps the following steps are included:
S41, on the basis of the position of tool tip information demarcated, tool image is cut out, the knife of fixed size is obtained
Has image;
S42, gray proces are carried out to the image after cutting, obtains gray level image;
S43, the fixed position in gained gray level image draw auxiliary line, obtain the cutter figure with aid mark
Picture.
It is further preferred that the auxiliary line position of all pretreatment images is all identical.
It is further preferred that the item number of auxiliary line is more than or equal to 1, and there are certain tilt angles and parallel to each other;
It is further preferred that the item number of auxiliary line is 3.
It is further preferred that the intersection point between point of a knife and auxiliary line is more, a possibility that being breaking, is smaller.
It is further preferred that the friendship between the tool nose and auxiliary line of tool image can be extracted using GoogleNet
Point feature.
It is further preferred that the method for breaking detection model is obtained in step S6 the following steps are included:
S61, cloud server end is judged with the presence or absence of trained breaking detection model, if it does not exist, acquisition is broken knife
The image data set of tool and normal cutter goes to step S62 as training set;If it exists, then step S63 is gone to;
S62, breaking detection model is trained on Cloud Server based on gained training set, goes to S63;
S63, when on Cloud Server error sample concentrate number of samples be greater than can train threshold value C when, in current breaking
Incremental training is carried out based on error sample collection on the basis of detection model, updates breaking detection model.
It is further preferred that in step S62 training breaking detection model method the following steps are included:
S621, the tool image in training set is pre-processed, building has the training set of aid mark;
Friendship between S622, extraction tool nose and auxiliary line with every width tool image in the training set of aid mark
Point feature;
S623, using the intersection point feature of cutter normal in training set as positive sample, it is special with the intersection point of breakage tool in training set
Sign is negative sample, is trained to classifier, obtains breaking detection model.
It is further preferred that SENet can be used as classifier.
Contemplated above technical scheme through the invention can achieve the following beneficial effects compared with prior art:
1, the present invention provides a kind of artificial intelligence knife-breaking detecting methods, by extracting between tool nose and auxiliary line
Intersection point feature trains breaking detection model, and carries out incremental training to the tool image of judgement error and constantly detect mould to breaking
Type is updated, and can be detected to a variety of different types of cutters, be substantially increased the generalization ability of breaking detection model
And the accuracy rate of detection.
2, the present invention provides a kind of breaking detection systems, shoot tool image using endoscope, can make tool image
Background parts it is fuzzy, to highlight cutter object, endoscope is at low cost, and simple to install is easy to maintain.In addition in acquisition
Tool image is the image after the completion of tool sharpening, reduces the interference of environmental factor, and it is right during tool sharpening to avoid
There are coolant liquid and workpiece occlusion issue when it is shot, the effective influence for reducing light in whole process is adding
It also can achieve relatively good detection effect in the case that work environment is relatively severe.
3, the present invention provides a kind of artificial intelligence knife-breaking detecting methods demarcates to be processed in advance before detection starting
The position of tool tip information of cutter determines the range of entire cutter by position of tool tip information, and this processing method is relative to existing
The tool-information that the method that some determines cutter range using threshold value obtains is more accurate, is influenced more by shooting background
It is small.
4, the present invention provides a kind of artificial intelligence breaking detection system, the training breaking detection model on Cloud Server,
It can be timely synchronized on different lathes after model modification, more flexible, the data that whole system may be implemented carry out
Summarize, is conducive to carry out processing analysis to the data of entire lathe, while also having relatively good maintainability.
Detailed description of the invention
Fig. 1 is a kind of artificial intelligence breaking detection system provided by the embodiment of the present invention;
Fig. 2 is a kind of artificial intelligence knife-breaking detecting method provided by the embodiment of the present invention;
Fig. 3 is a kind of training method of breaking detection model provided by the embodiment of the present invention;
Fig. 4 is the normal tool image with aid mark after pretreatment provided by the embodiment of the present invention;
Fig. 5 is the breakage tool image with aid mark after pretreatment provided by the embodiment of the present invention.
Specific embodiment
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
The invention proposes a kind of artificial intelligence breaking detection systems, as shown in Figure 1, including image capture module 1, tool magazine
2, Cloud Server 3, image processing module 4;
Wherein, the output end of image capture module 1 is connected with the input terminal of image processing module 4, image capture module 1 with
It is spaced a distance between tool magazine 2, is communicated between image processing module 4 and Cloud Server 3 by Ethernet;
The cutter to be processed and calibration that image capture module 1 is used to shoot in tool magazine 2 obtain the calibration letter of position of tool tip
Breath, and the tool image that the cutter of shooting after processing is completed obtains blurred background and cutter object highlights, and it is transferred to figure
As in processing module 4;
Tool magazine 2 is used to store cutter to be processed and cutter to be detected after processing is completed;
Cloud Server 3 is used to obtain breaking detection model based on the training of cutter intersection point feature;
Image processing module 4 loads the breaking inspection that training obtains in Cloud Server for pre-processing to tool image
Model is surveyed, the breaking detection model is then based on and calculates Tool monitoring result.
Specifically, image capture module 1 includes endoscope, headlamp, USB data line, endoscope is fixed on bracket
In numerically-controlled machine tool, the position of cutter 10cm in tool magazine, with USB data line by host computer phase in endoscope and numerically-controlled machine tool
Even.Headlamp, due to tool magazine position dark, can make the cutter in photo more using headlamp for illuminating tool magazine area
Clearly.Using the cutter in endoscope shooting tool magazine, when cutter is not processed, tool image is shot, and demarcate tool nose
The location information at place, cutter gains tool magazine after the completion of tool sharpening, using endoscope shooting after processing is completed to be detected
Cutter obtains the tool image of blurred background to highlight cutter object, and image data is transferred to image by USB data line
In processing module.There are coolant liquid and workpiece occlusion issue when being shot during tool sharpening to it, need cutter
It is hyperphoric to tool magazine from processing district, avoid occlusion issue.
Specifically, training obtains breaking detection model and is loaded into image processing module on Cloud Server, by image
The tool image taken in acquisition module is transmitted in image processing module, after the completion of pretreatment, using image processing module
In breaking detection model breaking detection is carried out to treated image.
Specifically, image processing module can be the IPC equipped with AI chip.
Specifically, another aspect of the present invention provides a kind of artificial intelligence knife-breaking detecting method, as shown in Fig. 2, include with
Lower step:
S1, image processing module connect Cloud Server, judge that the breaking detection model in image processing module whether there is,
If it does not exist, breaking detection model is downloaded from Cloud Server, error message is returned to if model failed download, algorithm terminates;
If it exists, step 2 is gone to;
S2, it is taken pictures using endoscope to all cutters to be processed in tool magazine and demarcates its position of tool tip;
S3, in machine tooling, the tool changing signal of poll lathe, after capturing tool changing signal, after processing is completed to be detected
Cutter returns to the corresponding position of tool magazine, is taken pictures using endoscope to cutter, obtains tool image to be detected;
S4, tool image to be detected is pre-processed based on the position of tool tip information demarcated, is obtained with auxiliary mark
The tool image of note;
S5, intersection point feature between the tool nose and auxiliary line of tool image to be detected is extracted using GoogleNet;Tool
Body, intersected by auxiliary line with tool nose, enhances the change of gradient of near intersections pixel, so that when extracting characteristics of image
It is easier to extract intersection point, number of intersections and position can reflect cutter length again, thus accomplish this qualitative features conversion of length
For intersection position and this quantitative characteristic of quantity, to keep training faster more acurrate.
S6, intersection point feature is input in breaking detection model trained in advance, breaking testing result is obtained, if being predicted as
Breaking then sends alarm signal to lathe, otherwise continues poll tool changing information.
S7, judge whether the breaking result of detection is correct, the tool image sample that will test mistake is transmitted to Cloud Server
On, error sample collection is obtained, algorithm terminates.
Specifically, in the method for cloud server end training breaking detection model, as shown in Figure 3, comprising the following steps:
S61, cloud server end is judged with the presence or absence of trained breaking detection model, if it does not exist, acquisition is broken knife
The image data set of tool and normal cutter goes to step S62 as training set;If it exists, then step S63 is gone to;Specifically, instruction
Practicing the image concentrated is the tool image comprising point of a knife;
S62, breaking detection model is trained on Cloud Server based on gained training set, goes to S63;
S63, judge whether number of samples that error sample on Cloud Server is concentrated is greater than and can train threshold value C, if more than
Threshold value C can be trained, then incremental training is carried out based on error sample collection on the basis of current breaking detection model, updates breaking inspection
Model is surveyed, algorithm terminates;Otherwise, current breaking detection model is gained, and algorithm terminates.Specifically, C value is 4000
, specifically, normal cutter sample size is 2000, breakage tool sample size is 2000, by examining in current breaking
It surveys on the basis of the existing parameter of model and incremental training is carried out to model, the serious forgiveness and accuracy rate of model can be improved.
Specifically, including: the step of training breaking detection model in step S62
S621, the tool image in training set is pre-processed, building has the training set of aid mark;
S622, the cutter knife using GoogleNet model extraction with every width tool image in the training set of aid mark
Intersection point feature between point and auxiliary line;
S623, using the intersection point feature of cutter normal in training set as positive sample, it is special with the intersection point of breakage tool in training set
Sign is negative sample, is trained to SENet classifier, obtains breaking detection model.
Specifically, in above-mentioned steps pretreated method is carried out to tool image the following steps are included:
S41, with the position (x, y) where point of a knife in the tool image demarcated before tool sharpening on the basis of, take x or so two
100 pixels of each 50 pixels in side and y or more are cut out the tool image after processing, obtain 100 × 100
Tool image;
S42, gray proces are carried out to the image after cutting, obtains gray level image;
S43,3 parallel auxiliary for having certain tilt angle are drawn in gained gray level image lower half portion fixed position
Line obtains the tool image with aid mark.Specifically, the auxiliary line position of all pretreatment images is all identical.Specifically,
It is illustrated in figure 4 the pretreated normal tool image with aid mark, is illustrated in figure 5 pretreated band aid mark
Breakage tool image.It can be seen from the figure that being intersected by auxiliary line with tool nose, the quantity of the longer intersection point of cutter is more
More, the intersection point of normal cutter and auxiliary line is significantly more than breaking cutter, and in addition there are certain tilt angles to increase for auxiliary line
Add the number of intersection point make to detect it is more accurate.
Provided artificial intelligence breaking detection system and method through the invention, by extracting tool nose and auxiliary line
Between intersection point feature training breaking detection model, and the tool image for collecting judgement error carries out incremental training, continuous right
Breaking detection model is updated, and allows to detect a variety of different types of cutters, substantially increases breaking detection
The generalization ability of model and the accuracy rate of detection.100 × 100 tool image addition auxiliary line feature comprising point of a knife is come
Training pattern makes the accuracy rate of model improve 3%.The size of training set used herein is 4000, wherein breaking and normal knife
Have all kinds of 2000, the test set data used are 4000, wherein breaking and normal cutter all kinds of 2000, and model is on test set
Final accuracy rate is 98.6%, and accuracy rate is higher.
As it will be easily appreciated by one skilled in the art that the foregoing is merely illustrative of the preferred embodiments of the present invention, not to
The limitation present invention, any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should all include
Within protection scope of the present invention.
Claims (8)
1. a kind of artificial intelligence breaking detection system, which is characterized in that including image capture module, tool magazine, Cloud Server, image
Handle model;
The output end of described image acquisition module is connected with the input terminal of described image processing module, described image acquisition module with
It is spaced a distance between the tool magazine, is led between described image processing module and the Cloud Server by Ethernet
News;
The cutter to be processed that described image acquisition module is used to shoot in tool magazine obtains the calibration information of position of tool tip, and claps
The tool image that cutter after processing is completed obtains blurred background and cutter object highlights is taken the photograph, and is transferred to image processing module
In;
The tool magazine is used to store cutter to be processed and cutter to be detected after processing is completed;
The Cloud Server is used to obtain breaking detection model based on the training of cutter intersection point feature;
Described image processing module loads the breaking detection that training obtains in Cloud Server for pre-processing to tool image
Model is then based on the breaking detection model and calculates Tool monitoring result.
2. breaking detection system according to claim 1, which is characterized in that described image acquisition module includes endoscope,
Using the cutter in endoscope shooting tool magazine, when cutter is not processed, tool image is shot, and demarcate where tool nose
Cutter is rechanged after the completion of tool sharpening and returns to tool magazine by location information, using endoscope shooting after processing is completed to be detected
Cutter obtains the tool image of blurred background, to highlight cutter object.
3. breaking detection system according to claim 1, which is characterized in that described image processing module is equipped with AI chip
IPC.
4. a kind of artificial intelligence knife-breaking detecting method, which comprises the following steps:
S1, image processing module connect Cloud Server, and breaking detection model is downloaded from Cloud Server;
S2, it is taken pictures to all cutters to be processed in tool magazine and demarcates its position of tool tip;
S3, cutter to be processed gain the corresponding position of tool magazine after processing is completed, and take pictures to it, obtain knife to be detected
Has image;
S4, the tool image to be detected is pre-processed based on the position of tool tip information demarcated, is obtained with auxiliary mark
The tool image of note;
S5, intersection point feature between the tool nose and auxiliary line of the tool image to be detected is extracted;
S6, the intersection point feature is input in breaking detection model trained in advance, obtains breaking testing result;
S7, the tool image sample that will test mistake are transmitted on Cloud Server, obtain error sample collection.
5. knife-breaking detecting method according to claim 4, which is characterized in that carry out pretreated method packet to tool image
Include following steps:
S41, on the basis of the position of tool tip information demarcated, tool image is cut out, the cutter figure of fixed size is obtained
Picture;
S42, gray proces are carried out to the image after cutting, obtains gray level image;
S43, the fixed position in the gray level image draw auxiliary line, obtain the tool image with aid mark.
6. knife-breaking detecting method according to claim 4 or 5, which is characterized in that the item number of the auxiliary line is more than or equal to
1, and there are certain tilt angles and parallel to each other.
7. knife-breaking detecting method according to claim 4, which is characterized in that obtain the breaking detection mould trained in advance
The method of type the following steps are included:
S61, judge cloud server end with the presence or absence of trained breaking detection model, if it does not exist, acquisition breakage tool and
The image data set of normal cutter goes to step S62 as training set;If it exists, then step S63 is gone to;
S62, breaking detection model is trained on Cloud Server based on the training set, goes to S63;
S63, when on Cloud Server error sample concentrate number of samples be greater than can train threshold value C when, current breaking detect
Incremental training is carried out based on error sample collection on the basis of model, updates breaking detection model.
8. knife-breaking detecting method according to claim 7, which is characterized in that the method packet of the training breaking detection model
Include following steps:
S621, the tool image in training set is pre-processed, building has the training set of aid mark;
S622, the intersection point extracted in the training set with aid mark between the tool nose and auxiliary line of every width tool image are special
Sign;
S623, using the intersection point feature of cutter normal in training set as positive sample, the intersection point feature with breakage tool in training set is
Negative sample is trained classifier, obtains breaking detection model.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910616439.XA CN110405540B (en) | 2019-07-09 | 2019-07-09 | Artificial intelligence broken cutter detection system and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910616439.XA CN110405540B (en) | 2019-07-09 | 2019-07-09 | Artificial intelligence broken cutter detection system and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110405540A true CN110405540A (en) | 2019-11-05 |
CN110405540B CN110405540B (en) | 2020-07-10 |
Family
ID=68360863
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910616439.XA Active CN110405540B (en) | 2019-07-09 | 2019-07-09 | Artificial intelligence broken cutter detection system and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110405540B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112247675A (en) * | 2020-11-04 | 2021-01-22 | 苏州众盈恒信息科技有限公司 | System and method for detecting service life of cutter based on big data autonomous learning |
CN113126964A (en) * | 2021-03-31 | 2021-07-16 | 成都飞机工业(集团)有限责任公司 | CATIA-based efficient cutter checking programming method |
CN115365889A (en) * | 2022-09-17 | 2022-11-22 | 杭州鹏润电子有限公司 | Method, system and storage medium for detecting knife breaking |
CN115741230A (en) * | 2022-09-30 | 2023-03-07 | 成都飞机工业(集团)有限责任公司 | Online broken cutter detection system and method |
CN115847187A (en) * | 2023-02-27 | 2023-03-28 | 成都大金航太科技股份有限公司 | Real-time monitoring system for deep and narrow groove turning |
Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6201567B1 (en) * | 1995-12-20 | 2001-03-13 | Komatsu Ltd. | Turn broach abnormality sensing apparatus |
US20100242693A1 (en) * | 2009-03-31 | 2010-09-30 | Toshiba Kikai Kabushiki Kaisha | Cutting-edge position detecting method and cutting-edge position detecting apparatus |
CN102528562A (en) * | 2012-02-28 | 2012-07-04 | 上海大学 | On-line automatic tool setting and breakage detection device for minitype milling tool |
CN104117876A (en) * | 2013-04-28 | 2014-10-29 | 郑州大学 | Method and device for checking whether cutting tool is broken or abraded or not |
CN106584209A (en) * | 2016-11-10 | 2017-04-26 | 哈尔滨理工大学 | Real-time online monitoring method for tool wear based on cloud manufacturing |
CN206855141U (en) * | 2017-04-01 | 2018-01-09 | 深圳市蓝海永兴实业有限公司 | A kind of milling cutter wears on-line measuring device |
KR20180027078A (en) * | 2016-09-06 | 2018-03-14 | 한국생산기술연구원 | Machining equipment and method for inspection of tool wear and defects |
CN108692709A (en) * | 2018-04-26 | 2018-10-23 | 济南浪潮高新科技投资发展有限公司 | A kind of farmland the condition of a disaster detection method, system, unmanned plane and cloud server |
CN109191367A (en) * | 2018-08-02 | 2019-01-11 | 哈尔滨理工大学 | The joining method of tool wear image and the life-span prediction method of cutter |
CN109500657A (en) * | 2018-11-14 | 2019-03-22 | 华中科技大学 | A kind of knife-breaking detecting method and system of view-based access control model |
CN109822398A (en) * | 2019-03-25 | 2019-05-31 | 华中科技大学 | A kind of numerically-controlled machine tool breaking detection system and method based on deep learning |
CN109919941A (en) * | 2019-03-29 | 2019-06-21 | 深圳市奥特立德自动化技术有限公司 | Internal screw thread defect inspection method, device, system, equipment and medium |
-
2019
- 2019-07-09 CN CN201910616439.XA patent/CN110405540B/en active Active
Patent Citations (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US6201567B1 (en) * | 1995-12-20 | 2001-03-13 | Komatsu Ltd. | Turn broach abnormality sensing apparatus |
US20100242693A1 (en) * | 2009-03-31 | 2010-09-30 | Toshiba Kikai Kabushiki Kaisha | Cutting-edge position detecting method and cutting-edge position detecting apparatus |
CN102528562A (en) * | 2012-02-28 | 2012-07-04 | 上海大学 | On-line automatic tool setting and breakage detection device for minitype milling tool |
CN104117876A (en) * | 2013-04-28 | 2014-10-29 | 郑州大学 | Method and device for checking whether cutting tool is broken or abraded or not |
KR20180027078A (en) * | 2016-09-06 | 2018-03-14 | 한국생산기술연구원 | Machining equipment and method for inspection of tool wear and defects |
CN106584209A (en) * | 2016-11-10 | 2017-04-26 | 哈尔滨理工大学 | Real-time online monitoring method for tool wear based on cloud manufacturing |
CN206855141U (en) * | 2017-04-01 | 2018-01-09 | 深圳市蓝海永兴实业有限公司 | A kind of milling cutter wears on-line measuring device |
CN108692709A (en) * | 2018-04-26 | 2018-10-23 | 济南浪潮高新科技投资发展有限公司 | A kind of farmland the condition of a disaster detection method, system, unmanned plane and cloud server |
CN109191367A (en) * | 2018-08-02 | 2019-01-11 | 哈尔滨理工大学 | The joining method of tool wear image and the life-span prediction method of cutter |
CN109500657A (en) * | 2018-11-14 | 2019-03-22 | 华中科技大学 | A kind of knife-breaking detecting method and system of view-based access control model |
CN109822398A (en) * | 2019-03-25 | 2019-05-31 | 华中科技大学 | A kind of numerically-controlled machine tool breaking detection system and method based on deep learning |
CN109919941A (en) * | 2019-03-29 | 2019-06-21 | 深圳市奥特立德自动化技术有限公司 | Internal screw thread defect inspection method, device, system, equipment and medium |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112247675A (en) * | 2020-11-04 | 2021-01-22 | 苏州众盈恒信息科技有限公司 | System and method for detecting service life of cutter based on big data autonomous learning |
CN113126964A (en) * | 2021-03-31 | 2021-07-16 | 成都飞机工业(集团)有限责任公司 | CATIA-based efficient cutter checking programming method |
CN113126964B (en) * | 2021-03-31 | 2022-01-25 | 成都飞机工业(集团)有限责任公司 | CATIA-based efficient cutter checking programming method |
CN115365889A (en) * | 2022-09-17 | 2022-11-22 | 杭州鹏润电子有限公司 | Method, system and storage medium for detecting knife breaking |
CN115741230A (en) * | 2022-09-30 | 2023-03-07 | 成都飞机工业(集团)有限责任公司 | Online broken cutter detection system and method |
CN115847187A (en) * | 2023-02-27 | 2023-03-28 | 成都大金航太科技股份有限公司 | Real-time monitoring system for deep and narrow groove turning |
CN115847187B (en) * | 2023-02-27 | 2023-05-05 | 成都大金航太科技股份有限公司 | Real-time monitoring system for deep and narrow groove turning |
Also Published As
Publication number | Publication date |
---|---|
CN110405540B (en) | 2020-07-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110405540A (en) | A kind of artificial intelligence breaking detection system and method | |
CN109822398B (en) | Numerical control machine tool broken cutter detection system and method based on deep learning | |
CN107948465B (en) | A kind of method and apparatus that detection camera is disturbed | |
CN111680632A (en) | Smoke and fire detection method and system based on deep learning convolutional neural network | |
CN102446266A (en) | Device, system and method for automatically identifying industrial number | |
CN110458093B (en) | Safety belt detection method based on driver monitoring system and corresponding equipment | |
CN112949578B (en) | Vehicle lamp state identification method, device, equipment and storage medium | |
CN111461260A (en) | Target detection method, device and equipment based on feature fusion and storage medium | |
CN105957300B (en) | A kind of wisdom gold eyeball identification is suspicious to put up masking alarm method and device | |
CN111950523A (en) | Ship detection optimization method and device based on aerial photography, electronic equipment and medium | |
CN112247675A (en) | System and method for detecting service life of cutter based on big data autonomous learning | |
MX2023011465A (en) | Object imaging and detection systems and methods. | |
CN111339927A (en) | Intelligent work state identification system for personnel in electric power business hall | |
CN114419493A (en) | Image annotation method and device, electronic equipment and storage medium | |
CN104715250A (en) | Cross laser detection method and device | |
CN110817674A (en) | Method, device and equipment for detecting step defect of escalator and storage medium | |
EP1565873B1 (en) | Particle extraction for automatic flow microscope | |
CN113205133B (en) | Tunnel water stain intelligent identification method based on multitask learning | |
CN112016514B (en) | Traffic sign recognition method, device, equipment and storage medium | |
CN113781407A (en) | Screw detection method, device and computer-readable storage medium | |
CN114965484A (en) | Defect detection system | |
CN115512283A (en) | Parcel image processing method and device, computer equipment and storage medium | |
CN112052824A (en) | Gas pipeline specific object target detection alarm method, device and system based on YOLOv3 algorithm and storage medium | |
CN112446884A (en) | Method and device for positioning power transmission line in laser point cloud and terminal equipment | |
CN117557570B (en) | Rail vehicle abnormality detection method and system |
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 |