Workpiece quality AI detection method, device and system

Abstract

The invention provides a workpiece quality AI detection method, device and system. The workpiece quality AI detection method comprises the steps of detecting a workpiece according to a preset actual detection item and a deviation value, and judging whether the workpiece is qualified or not according to whether a detection result is within an allowable deviation value range or not. According to the invention, each marking item in the mechanical engineering design drawing of the to-be-detected workpiece can be identified, and each identified marking item is used as an actual detection item of the workpiece. Thus, it is avoided that in the workpiece quality detection process, annotation items in the mechanical engineering design drawing are interpreted and recognized on site, and the detection efficiency is improved. Meanwhile, by automatically identifying the annotation items, updating changes of the drawing can be adapted in real time, it is ensured that the detection standard and the drawing requirement are kept consistent, and the problem caused by the fact that the drawing is not updated in time is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of workpiece detection, and particularly to an AI detection method, device and system for workpiece quality. Background Art

[0002] In industrial production, problems of workpiece quality detection are often faced, including dimension detection, appearance detection, etc. In order to achieve workpiece quality detection, in the prior art, it is first necessary to observe and / or measure the workpiece with the aid of detection tools, and then compare each detection item marked in the mechanical engineering design drawing of the workpiece with the observation and / or measurement results to determine whether the quality of the workpiece meets the relevant detection requirements.

[0003] However, in the above detection method, it relies on manual interpretation and lacks professional technical analysis ability. At the same time, there may be problems such as untimely update and unclear marking in the mechanical engineering design drawings, which may lead to misunderstandings and misjudgments in the detection process. In addition, the storage, management and retrieval of drawings may also become a major challenge. Therefore, in view of the above problems, it is necessary to propose a further solution. Summary of the Invention

[0004] The present invention aims to provide an AI detection method, device and system for workpiece quality to overcome the deficiencies in the prior art.

[0005] The object of the present application is achieved by the following technical solutions:

[0006] In a first aspect, the present application provides an AI detection method for workpiece quality, which includes:

[0007] Detect the workpiece according to the preset actual detection items and deviation values, and judge whether the workpiece is qualified according to whether the detection result is within the allowable deviation value range;

[0008] Among them, the preset actual detection items and deviation values are obtained through the following steps:

[0009] S1. Upload the mechanical engineering design drawing corresponding to the workpiece to be detected;

[0010] S2. Identify each marked item of the workpiece in the mechanical engineering design drawing according to the image features in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece;

[0011] S3. Calculate the allowable deviation values of each marked item according to the detection standard and detection level.

[0012] As an improvement of the AI detection method for workpiece quality of the present invention, the step S1 includes: storing the mechanical engineering design drawing information corresponding to the workpiece to be detected in a server or cloud storage.

[0013] As an improvement to the AI inspection method for workpiece quality of the present invention, step S1 includes: storing the mechanical engineering design drawing information corresponding to the workpiece to be inspected in a database, generating the ID information of the mechanical engineering design drawing information in the database, and uploading the mechanical engineering design drawing corresponding to the workpiece to be inspected by accessing the ID information.

[0014] As an improvement to the AI inspection method for workpiece quality of the present invention, the ID information is a two-dimensional code set on the mechanical engineering design drawing.

[0015] As an improvement to the AI inspection method for workpiece quality of the present invention, "identifying each marked item of the workpiece in the mechanical engineering design drawing according to the image features in the mechanical engineering design drawing" includes:

[0016] Reading at least one of the line, shape, and text information in the mechanical engineering design drawing, and identifying each marked item of the workpiece in the mechanical engineering design drawing by at least one of edge detection, shape matching, and text recognition methods according to the reading result.

[0017] As an improvement to the AI inspection method for workpiece quality of the present invention, "summarizing the identified marked items as the actual inspection items of the workpiece" includes:

[0018] After identifying each marked item of the workpiece in the mechanical engineering design drawing, numbering each marked item in sequence, and summarizing the numbered marked items in a table form, thereby serving as the actual inspection items of the workpiece.

[0019] As an improvement to the AI inspection method for workpiece quality of the present invention, in step S3, the allowable deviation value of each marked item is calculated through the following algorithm in combination with the inspection standard and inspection level:

[0020] Allowable deviation value Δ = Δ 理论 (1 + k(1 - C pk ))), Δ 理论 is the theoretical deviation value, C pk is the process capability index, and k is the inspection level adjustment coefficient.

[0021] As an improvement to the AI inspection method for workpiece quality of the present invention, "inspecting the workpiece according to the preset actual inspection items and deviation values, and judging whether the workpiece is qualified according to whether the inspection result is within the allowable deviation value range" includes:

[0022] Measuring each actual inspection item of the workpiece through a measuring mechanism;

[0023] Uploading the measurement result to compare with the deviation value corresponding to each actual inspection item, and displaying the comparison result;

[0024] When the comparison result shows that the measurement result is within the allowable deviation range, it is determined that the workpiece meets the detection requirements; otherwise, it is determined that the workpiece does not meet the detection requirements.

[0025] In a second aspect, the present application provides a workpiece size detection device, which includes:

[0026] A detection end, which can detect the workpiece according to the preset actual detection items and deviation values, and determine whether the workpiece is qualified according to whether the detection result is within the allowable deviation range;

[0027] Among them, the preset actual detection items and deviation values are obtained through the following steps:

[0028] S1. Upload the mechanical engineering design drawing corresponding to the workpiece to be detected;

[0029] S2. According to the image features in the mechanical engineering design drawing, identify each marked item of the workpiece in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece;

[0030] S3. Calculate the allowable deviation values of each marked item according to the detection standard and detection level.

[0031] In a third aspect, the present application provides a workpiece size detection system, which includes:

[0032] An upload module, which is used to upload the mechanical engineering design drawing corresponding to the workpiece to be detected;

[0033] An AI recognition module, which is used to identify each marked item of the workpiece in the mechanical engineering design drawing according to the image features in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece;

[0034] A calculation module, which is used to calculate the allowable deviation values of each marked item according to the detection standard and detection level;

[0035] A detection module, which is used to detect the workpiece according to the preset actual detection items and deviation values, and determine whether the workpiece is qualified according to whether the detection result is within the allowable deviation range.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] The present invention can identify each marked item in the mechanical engineering design drawing of the workpiece to be detected, and use the identified marked items as the actual detection items of the workpiece. In this way, when detecting the quality of the workpiece, it avoids the on-site interpretation and identification of the marked items in the mechanical engineering design drawing, improving the detection efficiency. At the same time, by automatically identifying the marked items, it can also adapt to the update and change of the drawing in real time, ensuring that the detection standard is consistent with the drawing requirements and avoiding the problems caused by untimely drawing updates.

[0038] Furthermore, the present invention can also calculate the allowable deviation values of each marked item based on a specific algorithm. In this way, it ensures the reliability of the deviation values used in actual detection, which is conducive to improving the accuracy and precision of the workpiece detection results. That is, by precisely defining the allowable deviation range, it can more effectively distinguish normal changes from abnormal states and reduce false alarms or missed detections caused by improper deviation settings. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0040] Figure 1 It is a flowchart of the workpiece quality AI detection method in Embodiment 1 of the present invention;

[0041] Figure 2 It is a schematic diagram of the modules of the workpiece size detection device in Embodiment 2 of the present invention;

[0042] Figure 3 It is a schematic diagram of the modules of the workpiece size detection system in Embodiment 3 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0043] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0044] The present invention relates to the technical field of workpiece quality detection.

[0045] Among them, the "workpiece" mainly refers to mechanical parts, such as shaft parts, sleeve parts, disc parts, etc. The quality detection of the "workpiece" mainly includes dimension detection, appearance detection, etc.

[0046] Specifically, dimensional inspection is an important part of the inspection of machined parts. It determines whether the product meets the design requirements by measuring the dimensions of the product in various directions. Appearance inspection is a preliminary inspection of machined parts to determine whether the appearance of the product meets the design requirements and satisfies the product's service performance. For example, the requirements for surface roughness, flatness, planeness, perpendicularity, etc. of machined parts are inspected.

[0047] In view of the current situation in the existing workpiece inspection technology where the drawing updates are not timely and the markings are not clear, which may lead to misunderstandings and misjudgments during the inspection process, the present invention proposes a further solution.

[0048] Embodiment 1

[0049] This embodiment provides an AI inspection method for workpiece quality. This AI inspection method for workpiece quality can identify each marked item in the mechanical engineering design drawing of the workpiece to be inspected, and use the identified marked items as the actual inspection items for the workpiece. In this way, when performing workpiece quality inspection, the need to interpret and identify the marked items in the mechanical engineering design drawing on-site is avoided, improving the inspection efficiency. At the same time, by automatically identifying the marked items, it can also adapt to the updates and changes of the drawing in real time, ensuring that the inspection standards are consistent with the drawing requirements and avoiding problems caused by untimely drawing updates.

[0050] As Figure 1 shown, the AI inspection method for workpiece quality in this embodiment includes:

[0051] S1. Upload the mechanical engineering design drawing corresponding to the workpiece to be inspected.

[0052] The mechanical engineering design drawing contains the structural information and dimensional information of the workpiece, which is used to guide the production and inspection of the workpiece. This step is used to upload the mechanical engineering design drawing corresponding to the workpiece to be inspected to the system, so as to facilitate the identification of the structural information and dimensional information in the mechanical engineering design drawing.

[0053] According to different upload methods, in one implementation, the mechanical engineering design drawing information corresponding to the workpiece to be inspected can be stored in the server or cloud storage, that is, directly upload the drawing information to the server or cloud storage of the system for storage, so as to achieve high availability and reliability.

[0054] In another alternative implementation, the drawing information can be uploaded to a material database in advance, and then the data can be read by accessing, so as to improve the upload efficiency.

[0055] Specifically, store the mechanical engineering design drawing information corresponding to the workpiece to be detected in the material database, generate the ID information of the mechanical engineering design drawing information in the material database, and this ID information forms the access entrance to the drawing information. This ID information can be a QR code set on the mechanical engineering design drawing. At this time, the drawing information stored in the material database can be read by scanning this QR code to upload the mechanical engineering design drawing corresponding to the workpiece to be detected.

[0056] S2. According to the image features in the mechanical engineering design drawing, identify each marked item of the workpiece in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece.

[0057] This step S2 specifically includes:

[0058] S21. Read at least one of the line, shape, and text information in the mechanical engineering design drawing. According to the reading result, identify each marked item of the workpiece in the mechanical engineering design drawing by at least one of edge detection, shape matching, and text recognition methods.

[0059] The line, shape, and text information in the mechanical engineering design drawing are the image features in the mechanical drawing. At this time, the reading result can be identified by at least one of edge detection, shape matching, and text recognition methods to facilitate clarifying its specific marked type.

[0060] Specifically, the outline of the object, including straight lines, curves, dotted lines, solid lines, etc., can be identified through edge detection. Shape matching can further compare the detected outline with a preset AI image recognition model to quickly identify the type, features, etc. of the workpiece. Text recognition can extract information such as name, material, dimension marking, technical requirements, etc. through OCR technology to assist in the judgment of the marked type.

[0061] S22. After identifying each marked item of the workpiece in the mechanical engineering design drawing, number each marked item in sequence, summarize the numbered marked items in tabular form, and then use them as the actual detection items of the workpiece.

[0062] Among them, numbering and summarizing the identified marked items is conducive to subsequent centralized search for each actual detection item of this workpiece. Specifically, when numbering each marked item, the marked items shown in step S21 can be used as the objects, and numbers can be generated in sequence along the edge of the workpiece in a clockwise or counterclockwise direction. Then, a summary table is newly created in a preset format to reflect information such as the marked item name and number in the table. In one implementation, the newly created summary table is as shown in Table 1 below:

[0063] Serial number Actual test item Reference value Upper deviation limit Lower deviation limit 1 07 Reference value 1 Upper limit 1-1 Upper limit 1-2 2 REVSON5 Reference value 2 Upper limit 2-1 Upper limit 2-2 3 EC0 Reference value 3 Upper limit 3-1 Upper limit 3-2 4 REV Reference value 4 Upper limit 4-1 Upper limit 4-2

[0064] Table 1

[0065] S3. Calculate the allowable deviation values for each marked item according to the detection standard and detection level.

[0066] Among them, the detection standard refers to the relevant markings based on which the actual detection items are carried out, such as national standards or industry standards, etc. The detection level refers to the precision level of detection, that is, the allowable deviation size during detection. In this embodiment, various detection standards can be collected through the capabilities of the large model and the corresponding detection levels can be matched, and it can be inferred whether the workpiece is qualified according to the detection results. In step S3, the allowable deviation values for each marked item are calculated through the following algorithm in combination with the detection standard and detection level:

[0067] The allowable deviation value Δ = Δ 理论 (1 + k(1 - C pk ))), Δ 理论 is the theoretical deviation value, C pk is the process capability index, and k is the detection level adjustment coefficient. Among them, the process capability index is specifically an ability index that measures the ability of the process to stably produce products that meet the quality requirements according to the detection standard. The detection level adjustment coefficient is determined according to the skills of the detection personnel and the detection environment.

[0068] Therefore, step S3 calculates the allowable deviation values for each marked item based on a specific algorithm. In this way, the reliability of the deviation values based on which the actual detection is carried out is ensured, which is conducive to improving the accuracy and precision of the workpiece detection results. That is, by precisely defining the allowable deviation range, normal changes and abnormal states can be more effectively distinguished, and false alarms or missed detections caused by improper deviation settings can be reduced.

[0069] S4. Detect the workpiece according to the preset actual detection items and deviation values, and judge whether the workpiece is qualified according to whether the detection results are within the allowable deviation values.

[0070] This step S4 specifically includes:

[0071] S41. Measure each actual detection item of the workpiece through a measuring mechanism.

[0072] Among them, the measuring mechanism can be a tool that can measure items such as the dimensions of the workpiece. For example, length measuring tools, angle measuring tools, surface roughness measuring tools, etc.

[0073] In addition, for the convenience of uploading the measurement results subsequently. The above-mentioned measuring tools can be selected as tools with wireless transmission functions, such as rulers with built-in Bluetooth modules. In this way, after the measurement is completed, the measurement results can be automatically uploaded through the built-in Bluetooth module, thereby improving the detection efficiency.

[0074] Specifically, one measurement action of the measurement tool is a data upload cycle. During this cycle, the Bluetooth module automatically reads the temporarily stored measurement result data and uploads it.

[0075] S42. Upload the measurement result to compare it with the deviation value corresponding to each actual detection item, and display the comparison result.

[0076] Among them, when uploading the measurement result to compare it with the deviation value corresponding to each actual detection item, the summary table with information such as the marked item name, label, deviation value, etc. can be accessed, and then the measurement result of the actual detection item is compared with the corresponding deviation value in the summary table.

[0077] S43. When the comparison result shows that the measurement result is within the allowable deviation value range, it is determined that the workpiece meets the detection requirements; otherwise, it is determined that the workpiece does not meet the detection requirements.

[0078] Embodiment 2

[0079] Based on different application scenarios, this embodiment provides a workpiece size detection device. The application scenario corresponding to this embodiment is a handheld detection device, and the detection personnel can upload the actual measurement result to the handheld detection device, so that the detection device can automatically detect whether the workpiece is qualified.

[0080] As Figure 2 shown, the workpiece size detection device 100 of this embodiment includes: a detection end 10. The detection end 10 can specifically be a tablet computer, an industrial handheld terminal, etc.

[0081] In this embodiment, the detection end 10 includes: a memory 11 and a processor 12. Among them, the memory 11 can receive the measurement result of the workpiece, and the processor 12 can detect the workpiece according to the preset actual detection items and deviation values, and determine whether the workpiece is qualified according to whether the detection result is within the allowable deviation value range.

[0082] Among them, the preset actual detection items and deviation values are obtained through the following steps:

[0083] S1. Upload the mechanical engineering design drawing corresponding to the workpiece to be detected;

[0084] S2. According to the image features in the mechanical engineering design drawing, identify each marked item of the workpiece in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece;

[0085] S3. Calculate the allowable deviation value of each marked item according to the detection standard and detection level.

[0086] The technical details involved in this embodiment can be referred to those described in Embodiment 1, and will not be repeated here.

[0087] Embodiment III

[0088] Based on different application scenarios, this embodiment provides a workpiece size detection system. The application scenario corresponding to this embodiment is a detection system including a front-end detection module and a background server. Among them, the background server integrates an uploading module, an AI recognition module, and a calculation module. At this time, the detection module receives the actual detection results from the detector and realizes the detection of the workpiece by accessing the background server.

[0089] As Figure 3 shown, the workpiece size detection system 200 of this embodiment includes:

[0090] An uploading module 201, which is used to upload the mechanical engineering design drawings corresponding to the workpiece to be detected;

[0091] An AI recognition module 202, which is used to identify each marked item of the workpiece in the mechanical engineering design drawings according to the image features in the mechanical engineering design drawings, and summarize the identified marked items as the actual detection items of the workpiece;

[0092] A calculation module 203, which is used to calculate the allowable deviation values of each marked item according to the detection standard and the detection level;

[0093] A detection module 204, which is used to detect the workpiece according to the preset actual detection items and deviation values, and judge whether the workpiece is qualified according to whether the detection result is within the allowable deviation value range.

[0094] The technical details involved in this embodiment can be referred to those described in Embodiment I, and will not be repeated here.

[0095] In summary, the present invention can identify each marked item in the mechanical engineering design drawings of the workpiece to be detected, and use the identified marked items as the actual detection items of the workpiece. In this way, when performing workpiece quality detection, the on-site interpretation and recognition of the marked items in the mechanical engineering design drawings are avoided, and the detection efficiency is improved. At the same time, by automatically identifying the marked items, the drawings can be updated and changed in real time, ensuring that the detection standard is consistent with the drawing requirements, and avoiding the problems caused by untimely drawing updates.

[0096] Furthermore, the present invention can also calculate the allowable deviation values of each marked item based on a specific algorithm. In this way, the reliability of the deviation values used in actual detection is ensured, which is beneficial to improving the accuracy and precision of the workpiece detection results. That is, by accurately defining the allowable deviation range, normal changes and abnormal states can be more effectively distinguished, and false alarms or missed alarms caused by improper deviation settings can be reduced.

[0097] For those skilled in the art, it is obvious that the present invention is not limited to the details of the above-described exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention. Therefore, in any aspect, the embodiments should be regarded as exemplary and non-limiting. The scope of the present invention is defined by the appended claims rather than the above description. Therefore, all changes falling within the meaning and scope of the equivalent elements of the claims are intended to be embraced within the present invention. Any reference signs in the claims should not be construed as limiting the claims involved.

[0098] In addition, it should be understood that although this specification is described according to embodiments, not every embodiment only contains an independent technical solution. This narrative way of the specification is only for clarity. Those skilled in the art should regard the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An AI detection method for workpiece quality, characterized in that, The AI detection method for workpiece quality includes: Detect the workpiece according to the preset actual detection items and deviation values, and judge whether the workpiece is qualified based on whether the detection results are within the allowable deviation values; Among them, the preset actual detection items and deviation values are obtained through the following steps: S1. Upload the mechanical engineering design drawing corresponding to the workpiece to be detected; S2. According to the image features in the mechanical engineering design drawing, identify each marked item of the workpiece in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece; S3. Calculate the allowable deviation values for each marked item according to the detection standard and detection level.

2. The workpiece quality AI detection method according to claim 1, characterized in that The step S1 includes: storing the mechanical engineering design drawing information corresponding to the workpiece to be detected in the server or cloud storage.

3. The workpiece quality AI detection method according to claim 1, wherein, The step S1 includes: storing the mechanical engineering design drawing information corresponding to the workpiece to be detected in the database, generating the ID information of the mechanical engineering design drawing information in the database, and uploading the mechanical engineering design drawing corresponding to the workpiece to be detected in the form of accessing the ID information.

4. The workpiece quality AI detection method according to claim 3, characterized in that, The ID information is a QR code set on the mechanical engineering design drawing.

5. The AI detection method for workpiece quality according to claim 1, wherein, The "identifying each marked item of the workpiece in the mechanical engineering design drawing according to the image features in the mechanical engineering design drawing" includes: Reading at least one of the line, shape, and text information in the mechanical engineering design drawing, and identifying each marked item of the workpiece in the mechanical engineering design drawing by at least one of edge detection, shape matching, and text recognition methods according to the reading result.

6. The workpiece quality AI detection method according to claim 5, wherein, The "summarizing the identified marked items as the actual detection items of the workpiece" includes: After identifying each marked item of the workpiece in the mechanical engineering design drawing, number each marked item in sequence, and summarize the numbered marked items in a table form, and then use them as the actual detection items of the workpiece.

7. The workpiece quality AI detection method according to claim 1, characterized in that, In step S3, the allowable deviation values for each marked item are calculated through the following algorithm in combination with the detection standard and detection level: Allowable deviation value Δ = Δ 理论 (1 + k(1 - C pk )),Δ 理论 is the theoretical deviation value, C pk is the process capability index, and k is the detection level adjustment coefficient.

8. The workpiece quality AI detection method according to claim 1, characterized in that, The "detecting the workpiece according to the preset actual detection items and deviation values, and judging whether the workpiece is qualified based on whether the detection results are within the allowable deviation values" includes: Measuring each actual detection item of the workpiece through a measuring mechanism; Uploading the measurement result to compare with the deviation value corresponding to each actual detection item, and displaying the comparison result; When the comparison result shows that the measurement result is within the allowable deviation value range, it is judged that the workpiece meets the detection requirements; otherwise, it is judged that the workpiece does not meet the detection requirements.

9. A workpiece size detection device, characterized in that, The workpiece size detection device includes: A detection end, which can detect the workpiece according to the preset actual detection items and deviation values, and judge whether the workpiece is qualified based on whether the detection results are within the allowable deviation values; Among them, the preset actual detection items and deviation values are obtained through the following steps: S1. Upload the mechanical engineering design drawing corresponding to the workpiece to be detected; S2. According to the image features in the mechanical engineering design drawing, identify each marked item of the workpiece in the mechanical engineering design drawing, and summarize the identified marked items as the actual detection items of the workpiece; S3. Calculate the allowable deviation values for each marked item according to the detection standard and detection level.

10. A workpiece size detection system, characterized in that, The workpiece dimension detection system includes: An upload module, which is used to upload the mechanical engineering design drawings corresponding to the workpiece to be detected; An AI recognition module, which is used to recognize each marked item of the workpiece in the mechanical engineering design drawings according to the image features in the mechanical engineering design drawings, and summarize the recognized marked items as the actual detection items of the workpiece; A calculation module, which is used to calculate the allowable deviation values for each marked item according to the detection standard and detection level; A detection module, which is used to detect the workpiece according to the preset actual detection items and deviation values, and judge whether the workpiece is qualified according to whether the detection result is within the allowable deviation value range.

Images