Device for detecting solid geometrical shape of flat steel wire
By using deep neural network combination image processing technology, intelligently analyzes the length value of the physical geometric shape of flat steel wires, solve the lack of the length of the flat steel wire in the existing technology, and achieve an effective guarantee of product quality.
Patent Information
- Application Number
- CN202411581501.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-06-04
- Filing Date
- 2024-11-07
- Publication Date
- 2025-01-28
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The prior art lacks mature and effective solutions to detect length values in the solid geometry of flat steel wires, resulting in the inability to effectively guarantee the quality of the manufactured product.
Adopt the deep neural network after multiple training that completes the total amount, combined with the imaging focal length of the surveillance of the video equipment, the coordinates and depth of field value of the pixels of the flat steel in the image segmentation The length value of the physical geometric shape of the flat steel wire, and send the length deviation or effective command through the parameter appraisal device.
Intelligent authentication of the length of the flat steel wire is achieved, ensuring the quality of the product's quality, running stable and designing intelligence.
Smart Images
Figure CN119374485A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the field of stainless steel bodies, and in particular to a device for detecting the solid geometric shape of a flat steel wire. Background Art
[0002] Flat steel wire is made of high-quality wire flattening machine and rolled into flat steel wire. The round cross-section wire rod drawn to a certain size is rolled with flat rollers to obtain high-precision flat steel wire with large width-to-thickness ratio. Flat steel wire can be used to manufacture various corresponding products. For example, flat wire mesh belt is a mesh belt woven from flat steel wire. Its characteristics are smooth transmission and no slipping. The main materials are carbon steel wire and stainless steel wire. It is mainly used in food production lines, with specifications of herringbone mesh belt and diamond mesh belt. It is mainly used in the biscuit industry. It is favored by many biscuit manufacturers.
[0003] In order to ensure that the flat steel wire used to manufacture various subsequent products is within the standard value range in size, for example, within the standard length value range, thereby improving the quality of various subsequent products manufactured using the flat steel wire, it is necessary to perform customized inspection and analysis on the length value in the solid geometry of each piece of flat steel wire manufactured. However, there is a lack of mature and effective solutions in the prior art. Summary of the invention
[0004] In order to solve the technical problems in the prior art, the present invention provides a detection device for the solid geometric shape of a flat steel wire, which intelligently analyzes the length value in the solid geometric shape of the flat steel wire according to the imaging focal length of the monitoring video equipment, the coordinate values corresponding to the pixels of the image block where the flat steel wire is located, the depth of field values corresponding to the pixels of the image block where the flat steel wire is located, and the total number of pixels of the multi-layer optimized image by using a deep neural network that has completed multiple trainings of a preset total amount, and when the error between the length value in the solid geometric shape of the flat steel wire intelligently analyzed and the design length value of the flat steel wire is greater than or equal to the set error limit, a length deviation command is issued, and when the error between the length value in the solid geometric shape of the flat steel wire intelligently analyzed and the design length value of the flat steel wire is less than the set error limit, a length validity command is issued, thereby completing the intelligent certification of whether the length of the latest flat steel wire conveyed by wire drawing meets the standard.
[0005] According to the present invention, a device for detecting the physical geometric shape of a flat steel wire is provided, the device comprising: A wire drawing device is used to perform wire drawing on a stainless steel body to obtain individual flat steel wires, and to deliver the individual flat steel wires one by one to the bottom of the surveillance video device; An action cooperation device is connected to the wire drawing processing device and the monitoring and recording device respectively, and is used to perform synchronous control of a single wire conveying action of the wire drawing processing device and a single frame imaging action of the monitoring and recording device, so as to trigger the monitoring and recording device to perform imaging of a single flat steel wire that is recently conveyed to the bottom of the monitoring and recording device after the wire drawing processing device completes the conveying action of each flat steel wire; A monitoring video device, used for performing imaging of a single flat steel wire that is recently transported to the bottom of the monitoring video device, so as to obtain a corresponding transport scene image; A multi-layer optimization device, connected to the monitoring video device, is used to sequentially perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing, and Gaussian white noise removal processing on the received transport scene image to obtain and output a corresponding multi-layer optimized image; A network application device is connected to the multi-layer optimization device, and is used to receive the multi-layer optimization image, extract the coordinate values corresponding to the pixel points of the image block where the flat steel wire is located in the multi-layer optimization image, and extract the depth of field values corresponding to the pixel points of the image block where the flat steel wire is located, and use a deep neural network that has completed multiple trainings of a preset total amount to intelligently analyze the length value of the solid geometric shape of the flat steel wire according to the imaging focal length of the monitoring video device, the coordinate values corresponding to the pixel points of the image block where the flat steel wire is located, the depth of field values corresponding to the pixel points of the image block where the flat steel wire is located, and the total number of pixels of the multi-layer optimization image, wherein the value of the preset total amount is proportional to the maximum noise amplitude in the multi-layer optimization image; A parameter identification device is connected to the network application device, and is used to issue a length deviation command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by intelligent analysis and the design length value of the flat steel wire is greater than or equal to the set error limit, and is also used to issue a length validity command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by intelligent analysis and the design length value of the flat steel wire is less than the set error limit.
[0006] It can be seen that the present invention mainly has the following significant technical effects: First: extract the coordinate values corresponding to each pixel point of the image block where the flat steel wire is located in the multi-layer optimized image, and extract the depth of field values corresponding to each pixel point of the image block where the flat steel wire is located, so as to provide key basic data for the subsequent intelligent analysis of the length value of the solid geometric shape of the flat steel wire; Second, a deep neural network that has completed multiple trainings for a preset total amount is used to intelligently analyze the length value of the solid geometric shape of the flat steel wire according to the imaging focal length of the surveillance video device, the coordinate values corresponding to the pixels of the image block where the flat steel wire is located, the depth of field values corresponding to the pixels of the image block where the flat steel wire is located, and the total number of pixels of the multi-layer optimized image. The value of the preset total amount is proportional to the maximum noise amplitude in the multi-layer optimized image. Third: a parameter identification device is used to issue a length deviation command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by intelligent analysis and the design length value of the flat steel wire is greater than or equal to the set error limit. It is also used to issue a length validity command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by intelligent analysis and the design length value of the flat steel wire is less than the set error limit, thereby completing the intelligent authentication of whether the length of the flat steel wire delivered by the latest wire drawing meets the standard.
[0007] The flat steel wire solid geometric shape detection device of the present invention has stable operation and intelligent design. Since the solid geometric shape of the current flat steel wire can be intelligently analyzed by using a deep neural network that has completed multiple trainings of a preset total amount, the quality of the corresponding subsequent products of the flat steel wire is guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] The embodiments of the present invention will be described below with reference to the accompanying drawings, wherein:
[0009] Figure 1 It is a schematic diagram of the internal structure of a device for detecting the solid geometric shape of a flat steel wire according to the first embodiment of the present invention.
[0010] Figure 2 Schematic diagram of the internal structure of a device for detecting the solid geometric shape of a flat steel wire according to a second embodiment of the present invention.
[0011] Figure 3 Schematic diagram of the internal structure of a device for detecting the solid geometric shape of a flat steel wire according to a third embodiment of the present invention. DETAILED DESCRIPTION
[0012] The embodiments of the device for detecting the solid geometric shape of a flat steel wire according to the present invention will be described in detail below with reference to the accompanying drawings.
[0013] Figure 1 This is a schematic diagram of the internal structure of a device for detecting the solid geometric shape of a flat steel wire according to a first embodiment of the present invention, wherein the device comprises: A wire drawing device is used to perform wire drawing on a stainless steel body to obtain individual flat steel wires, and to deliver the individual flat steel wires one by one to the bottom of the surveillance video device; An action cooperation device is connected to the wire drawing processing device and the monitoring and recording device respectively, and is used to perform synchronous control of a single wire conveying action of the wire drawing processing device and a single frame imaging action of the monitoring and recording device, so as to trigger the monitoring and recording device to perform imaging of a single flat steel wire that is recently conveyed to the bottom of the monitoring and recording device after the wire drawing processing device completes the conveying action of each flat steel wire; For example, the action cooperation device is connected to the wire drawing processing device and the monitoring video device respectively, and is used to perform the synchronous control of the single conveying action of the wire drawing processing device and the single frame imaging action of the monitoring video device, so as to trigger the monitoring video device to perform imaging of the single flat steel wire that is most recently conveyed to the bottom of the monitoring video device after the wire drawing processing device completes the conveying action of one flat steel wire each time. The method includes: using a PLC device to implement the action cooperation device, which is connected to the wire drawing processing device and the monitoring video device respectively, and is used to perform the synchronous control of the single conveying action of the wire drawing processing device and the single frame imaging action of the monitoring video device, so as to trigger the monitoring video device to perform imaging of the single flat steel wire that is most recently conveyed to the bottom of the monitoring video device after the wire drawing processing device completes the conveying action of one flat steel wire each time; A monitoring video device, used for performing imaging of a single flat steel wire that is recently transported to the bottom of the monitoring video device, so as to obtain a corresponding transport scene image; A multi-layer optimization device, connected to the monitoring video device, is used to sequentially perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing, and Gaussian white noise removal processing on the received transport scene image to obtain and output a corresponding multi-layer optimized image; A network application device is connected to the multi-layer optimization device, and is used to receive the multi-layer optimization image, extract the coordinate values corresponding to the pixel points of the image block where the flat steel wire is located in the multi-layer optimization image, and extract the depth of field values corresponding to the pixel points of the image block where the flat steel wire is located, and use a deep neural network that has completed multiple trainings of a preset total amount to intelligently analyze the length value of the solid geometric shape of the flat steel wire according to the imaging focal length of the monitoring video device, the coordinate values corresponding to the pixel points of the image block where the flat steel wire is located, the depth of field values corresponding to the pixel points of the image block where the flat steel wire is located, and the total number of pixels of the multi-layer optimization image, wherein the value of the preset total amount is proportional to the maximum noise amplitude in the multi-layer optimization image; A parameter identification device is connected to the network application device, and is used to issue a length deviation command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by the intelligent analysis and the design length value of the flat steel wire is greater than or equal to the set error limit, and is also used to issue a length validity command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by the intelligent analysis and the design length value of the flat steel wire is less than the set error limit; Among them, the multi-layer optimization device is connected to the monitoring video equipment, and is used to sequentially perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing and Gaussian white noise removal processing on the received transmission scene image to obtain and output the corresponding multi-layer optimized image, including: using different image processing components to respectively perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing and Gaussian white noise removal processing.
[0014] Figure 2 Schematic diagram of the internal structure of a device for detecting the solid geometric shape of a flat steel wire according to a second embodiment of the present invention.
[0015] Compared to Figure 1 According to the second embodiment of the present invention, the detection device for the physical geometric shape of the flat steel wire may further include: A signal strengthening component is connected to the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, respectively, and is used to strengthen the signals of the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, so as to obtain signal-enhanced images corresponding to the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, respectively; Wherein, respectively performing signal enhancement on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device comprises: the front-end signal enhancement performs Wiener filtering processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device; Among them, respectively performing signal enhancement on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device also includes: a first mid-end signal enhancement performs wavelet filtering processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device; Among them, respectively performing signal enhancement on the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, further comprising: the second mid-end signal enhancement respectively performs salt and pepper noise filtering processing on the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device; Among them, signal enhancement is performed on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and also includes: terminal signal enhancement performs distortion correction processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device.
[0016] Figure 3 Schematic diagram of the internal structure of a device for detecting the solid geometric shape of a flat steel wire according to a third embodiment of the present invention.
[0017] Compared to Figure 1 According to the third embodiment of the present invention, the detection device for the physical geometric shape of the flat steel wire may further include: A sequence generation device, connected to the multi-layer optimization device, the network application device, the parameter identification device and the action coordination device, respectively, for respectively performing on-site supply of respective working clock sequences of the multi-layer optimization device, the network application device, the parameter identification device and the action coordination device; Wherein, the sequence generation device is connected to the multi-layer optimization device, the network application device, the parameter identification device and the action coordination device respectively, and is used to respectively perform on-site supply of respective working clock sequences of the multi-layer optimization device, the network application device, the parameter identification device and the action coordination device, including: the sequence generation device is a quartz oscillator device; And wherein, the sequence generating device is respectively connected to the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and is used to respectively execute the on-site supply of the respective working clock sequences of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, including: using a CPLD device designed using the VHDL language to implement the sequence generating device.
[0018] In addition, in the detection device for the solid geometric shape of the flat steel wire, different image processing components are used to respectively perform sharpening processing based on the Kirsch operator, image spatial domain enhancement processing and Gaussian white noise removal processing, including: using a sharpening processing component to perform sharpening processing based on the Kirsch operator on the received image data, using an enhancement processing component to perform image spatial domain enhancement processing on the received image data, and using a filtering processing component to perform Gaussian white noise removal processing on the received image data.
[0019] Furthermore, the present invention is not limited to the above-mentioned embodiments themselves, and the constituent elements can be deformed and embodied in the implementation stage without departing from the scope of its purpose. In addition, various inventions can be formed by appropriately combining multiple constituent elements disclosed in the above-mentioned embodiments. For example, several constituent elements can be deleted from all the constituent elements displayed in this embodiment. Furthermore, constituent elements of different embodiments can also be appropriately combined.
Claims
1. A device for detecting the physical geometric shape of a flat steel wire, characterized in that: The device comprises: A wire drawing device is used to perform wire drawing on a stainless steel body to obtain individual flat steel wires, and to deliver the individual flat steel wires one by one to the bottom of the surveillance video device; An action cooperation device is connected to the wire drawing processing device and the monitoring and recording device respectively, and is used to perform synchronous control of a single wire conveying action of the wire drawing processing device and a single frame imaging action of the monitoring and recording device, so as to trigger the monitoring and recording device to perform imaging of a single flat steel wire that is recently conveyed to the bottom of the monitoring and recording device after the wire drawing processing device completes the conveying action of each flat steel wire; A monitoring video device, used for performing imaging of a single flat steel wire that is recently transported to the bottom of the monitoring video device, so as to obtain a corresponding transport scene image; A multi-layer optimization device, connected to the monitoring video device, is used to sequentially perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing, and Gaussian white noise removal processing on the received transport scene image to obtain and output a corresponding multi-layer optimized image; A network application device is connected to the multi-layer optimization device, and is used to receive the multi-layer optimization image, extract the coordinate values corresponding to the pixel points of the image block where the flat steel wire is located in the multi-layer optimization image, and extract the depth of field values corresponding to the pixel points of the image block where the flat steel wire is located, and use a deep neural network that has completed multiple trainings of a preset total amount to intelligently analyze the length value of the solid geometric shape of the flat steel wire according to the imaging focal length of the monitoring video device, the coordinate values corresponding to the pixel points of the image block where the flat steel wire is located, the depth of field values corresponding to the pixel points of the image block where the flat steel wire is located, and the total number of pixels of the multi-layer optimization image, wherein the value of the preset total amount is proportional to the maximum noise amplitude in the multi-layer optimization image; A parameter identification device is connected to the network application device, and is used to issue a length deviation command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by intelligent analysis and the design length value of the flat steel wire is greater than or equal to the set error limit, and is also used to issue a length validity command when the error between the length value in the solid geometric shape of the flat steel wire analyzed by intelligent analysis and the design length value of the flat steel wire is less than the set error limit.
2. The device for detecting the solid geometric shape of a flat steel wire according to claim 1, characterized in that: A multi-layer optimization device is connected to the monitoring video recording device and is used to sequentially perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing and Gaussian white noise removal processing on the received transmission scene image to obtain and output a corresponding multi-layer optimized image, including: using different image processing components to respectively perform Kirsch operator-based sharpening processing, image spatial domain enhancement processing and Gaussian white noise removal processing.
3. The device for detecting the solid geometric shape of a flat steel wire according to claim 2, characterized in that: The device also includes: A signal strengthening component is connected to the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, respectively, and is used to strengthen the signals of the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, so as to obtain signal-enhanced images corresponding to the output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, respectively; Among them, signal enhancement is performed on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, including: the front-end signal enhancement performs Wiener filtering processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device.
4. The device for detecting the solid geometric shape of a flat steel wire according to claim 3, characterized in that: Signal enhancement is performed on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and also includes: a first mid-end signal enhancement performs wavelet filtering processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device.
5. The device for detecting the solid geometric shape of a flat steel wire according to claim 4, characterized in that: The signal enhancement is performed on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and also includes: the second mid-end signal enhancement performs salt and pepper noise filtering processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device.
6. The device for detecting the solid geometric shape of a flat steel wire according to claim 5, characterized in that: The signal enhancement is performed on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device to obtain respective signal enhanced images corresponding to the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device. The terminal signal enhancement also performs distortion correction processing on the respective output image signals of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device.
7. The device for detecting the solid geometric shape of a flat steel wire according to any one of claims 3 to 6, characterized in that: The device also includes: A sequence generating device is respectively connected to the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and is used to respectively perform on-site supply of respective working clock sequences of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device.
8. The device for detecting the solid geometric shape of a flat steel wire according to claim 7, characterized in that: A sequence generating device is respectively connected to the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and is used to respectively execute on-site supply of respective working clock sequences of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, including: the sequence generating device is a quartz oscillator device.
9. The device for detecting the solid geometric shape of a flat steel wire according to claim 7, characterized in that: A sequence generating device is respectively connected to the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, and is used to respectively execute on-site supply of respective working clock sequences of the multi-layer optimization device, the network application device, the parameter identification device and the action collaboration device, including: using a CPLD device designed using VHDL language to implement the sequence generating device.