Intelligent cutting device and cutting method for luggage processing and production

Abstract

The invention discloses an intelligent cutting device and method for luggage processing and production. The intelligent cutting device comprises a cutting table, a rack, a three-axis motion module, a cutting actuator, a visual collection module and an intelligent control unit, and the visual collection module obtains a global image of leather on the cutting table; the intelligent control unit is used for processing the image data, automatically identifying leather contours and flaws, carrying out optimized layout and path planning for avoiding flaws on a single or multiple pieces of overlapped leather, and generating a cutting instruction; the method comprises the steps of visual collection and feature extraction, intelligent layout and path generation, dynamic collaborative cutting, completion processing, information filing and the like, the material utilization rate and the cutting precision are remarkably improved, technological parameters can be continuously optimized through data self-learning, and the cutting efficiency is improved. The technical problems that a traditional cutting machine cannot adapt to materials and process stacking errors and lacks process optimization are solved.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of luggage, leather products and flexible material processing equipment, and particularly relates to an intelligent cutting device and cutting method for luggage processing and production. BACKGROUND

[0002] In the production process of leather products such as luggage, leather shoes, and leather goods, cutting (or called blanking) is the first and crucial process. Traditional cutting mainly relies on manual operation, that is, workers draw lines on leather materials according to paper patterns, and then use handwork or simple machinery for stamping or cutting. This method has problems such as high labor intensity, low production efficiency, serious material waste (especially for expensive natural leather), high dependence on worker experience, and poor product consistency.

[0003] In order to improve efficiency, numerical control (NC) cutting has been gradually promoted in the industry. Such equipment usually presets a cutting path and drives a cutting knife or vibrating knife to perform cutting through a three-axis linkage system. However, most existing numerical control cutting still has significant limitations: Cannot adapt to materials: The cutting path is a fixed path programmed in advance, and cannot recognize the unique shape, size, flaws (such as scars, blood vessel lines), and actual placement position of each piece of leather. This leads to either the need for manual precise alignment, which is time-consuming and labor-intensive, or only regular materials can be cut, and irregular leather may either waste the corner material or may cut flaws into the finished product.

[0004] Cannot handle stacking errors: To improve efficiency, factories often use multi-layer stacking cutting. However, existing equipment cannot sense the misalignment between layers, and cutting according to the same path will cause size deviation between upper and lower layers, resulting in low yield.

[0005] Lack of process monitoring and optimization: The cutting process is mostly "open-loop" executed, and cannot detect the tool state, material movement or lifting in real time. When problems occur, it often causes the entire batch of materials to be scrapped. At the same time, the equipment does not have the ability to learn and optimize process parameters from historical data. SUMMARY

[0006] In view of the deficiencies in the prior art, the purpose of the present application is to provide an intelligent cutting device and cutting method for luggage processing and production, which can perceive materials, think about layout, accurately execute, and learn and optimize.

[0007] The technical solution adopted by the present application to solve its technical problems is: An intelligent cutting device for luggage processing and production, comprising: a cutting table for carrying the luggage leather to be processed; a rack arranged around the cutting table; A three-axis motion module is installed on the frame, including a transverse slide driven by a transverse moving motor, a longitudinal slide driven by a longitudinal moving motor, and a vertical mounting seat driven by a vertical moving motor; A cutting executor is installed on the vertical mounting seat for cutting leather; A visual acquisition module is fixedly installed at a high position of the frame, and its acquisition visual angle covers the entire working area of the cutting table; An intelligent control unit is electrically connected with each motor in the three-axis motion module and the cutting executor; The intelligent control unit is configured to: receive and process the leather image data acquired by the visual acquisition module; based on the processing result, automatically identify the effective area and defects of the leather, and plan the cutting path of the cutting executor for avoiding defects and optimizing the layout of single or multiple overlapped leathers; control the three-axis motion module and the cutting executor to cooperatively perform the cutting path.

[0008] As a preferred embodiment, the visual acquisition module includes at least one high-resolution industrial camera and a surrounding lighting source; and the surface of the cutting table is provided with a background reference mark or a calibration pattern in communication connection with the intelligent control unit.

[0009] As a preferred embodiment, the cutting executor is a fast-replaceable modular structure selected from one of a high-frequency vibration knife module, a laser cutting head, a circular knife cutting head, or a punching cutting head; and the vertical mounting seat is integrated with a pressure sensor and a displacement sensor for real-time monitoring of the cutting force and the pressing depth and forming a closed-loop feedback.

[0010] As a preferred embodiment, the cutting table is an independently controllable multi-zone vacuum adsorption table, and the intelligent control unit dynamically controls the adsorption start-stop and adsorption force of the corresponding zone according to the leather contour and placement position identified by the visual acquisition module.

[0011] As a preferred embodiment, a human-computer interaction interface and a data interface connected with the intelligent control unit are further included; the human-computer interaction interface is used for displaying the visual recognition result, the cutting path simulation, and the equipment state, and receiving parameter input and manual intervention instructions; and the data interface is used for receiving external CAD / CAM sheet data and uploading production data.

[0012] As a preferred embodiment, the path planning algorithm of the intelligent control unit includes: edge extraction and fitting based on the leather contour image; automatic layout of a plurality of preset cutting sheet patterns in the fitted effective contour, iterative calculation with the highest material utilization rate or the optimal cutting efficiency as the target; and setting a forbidden area for the identified defect area.

[0013] Another technical problem to be solved by the present application is to provide a luggage leather cutting method based on the intelligent cutting device of any one of the above, comprising the following steps: S1: visual acquisition and feature extraction: acquiring the global image of the leather placed on the cutting table through the visual acquisition module, identifying the effective contour, surface defects, pre-set positioning marks and stacking level information of the leather; S2: intelligent nesting and path generation: the intelligent control unit calls the pre-set cutting sheet pattern library, combines the feature information extracted in step S1, performs automatic nesting calculation, generates an optimized cutting path instruction set, and the path instruction set includes spatial coordinates, cutter start-stop and working parameters; S3: dynamic cooperative cutting: the intelligent control unit drives the three-axis motion module and the cutting executor to execute the cutting action along the optimized cutting path, and dynamically adjusts the motion parameters and cutting parameters according to the real-time feedback sensor data; S4: finishing processing and information archiving: after completing all path cutting, the equipment is automatically reset, and the intelligent control unit generates a production report containing information such as material utilization rate, defect position, actual cutting time, etc. and stores it.

[0014] As a preferred, in step S2, when multiple pieces of leather are identified to be stacked, the method further comprises: calculating the relative offset vector between each stack layer by image feature point matching algorithm, and performing translation and rotation compensation on the theoretical cutting path of each layer to generate the final execution path suitable for the current stack state.

[0015] As a preferred, in step S3, it further includes a real-time online monitoring step: through the visual acquisition module, the cut contour is locally imaged again during cutting intermittently, or the signals collected by the acoustic sensor and vibration sensor are compared with the pre-set threshold value, if the path deviation, material warping or tool abnormality is detected, it will be immediately suspended and alarmed, and the options of manual intervention or automatic retry are provided.

[0016] As a preferred, the method further comprises a self-learning optimization step: the intelligent control unit accumulates and analyzes historical cutting data and corresponding visual images, establishes a correlation model between leather type, texture direction, defect distribution and optimal nesting scheme, cutting speed, and pressing depth, which is used to optimize the cutting parameters of subsequent similar leathers.

[0017] The beneficial effects in the present application are: The visual acquisition module automatically identifies the leather outline and defects, while the intelligent control unit automatically completes the optimal layout and path planning, driving the equipment to execute the process. This completely replaces tedious steps such as manual marking and alignment, achieving full automation from "material loading" to "finished piece removal," resulting in an order-of-magnitude increase in production efficiency. Based on real-time visual information, the intelligent layout algorithm dynamically adjusts the pattern layout according to the actual shape and defect distribution of each piece of leather, maximizing the arrangement of cutting patterns within the effective area and automatically avoiding defective areas. For expensive natural leather, every percentage point increase in material utilization brings considerable economic benefits.

[0018] Visual positioning eliminates human placement errors; dynamic compensation for the offset of layered materials ensures the accuracy of each layer during multi-layer cutting; real-time online monitoring can promptly alarm or correct deviations. These measures collectively guarantee high precision and repeatability in the cutting process, thereby improving the consistency of the final product quality; by calculating the layer offset and performing path compensation through image feature matching algorithms, the equipment can stably and accurately complete the one-time cutting of multi-layer leather, solving the misalignment problem of traditional layered material cutting while ensuring efficiency, and demonstrating stronger process adaptability.

[0019] The equipment not only performs tasks but also monitors the process through sensors and vision, preventing quality incidents. More importantly, its self-learning function, which builds process models based on accumulated data, enables the equipment to continuously optimize cutting parameters for different material properties, achieving a virtuous cycle of becoming increasingly "intelligent" with use. This provides core equipment support for the digital and intelligent transformation of factories. Traditional cutting relies on the "eyes" and "hands" of experienced operators, while the intelligent system of this invention transforms experience into algorithms. Operators only need to load materials, call up solutions, and monitor processes, reducing reliance on specific skills and facilitating personnel training and large-scale production management. Attached Figure Description

[0020] Fig. 1 This is a schematic diagram of the overall structure of the intelligent cutting device for bag processing and production according to the present invention. Fig. 2 This is a flowchart of the intelligent cutting method for bag processing and production according to the present invention. Detailed Implementation

[0021] The present invention will be further described below with reference to the accompanying drawings and specific embodiments, so that those skilled in the art can better understand and implement the present invention. However, the embodiments are not intended to limit the present invention.

[0022] Example See Figs. 1-2 As shown, this embodiment provides an intelligent cutting machine device for bag processing and production, comprising the following components: Cutting table 1: as a working platform, its table top adopts high-strength honeycomb structure, and the surface is paved with a cutting-resistant elastic pad. The inside of the cutting table 1 is divided into four independent vacuum suction partitions, each partition is connected to a high-power vacuum generator (not shown in the figure) through an independent electromagnetic valve and a control air pipe. The four corners of the table top are etched with a high-precision cross-shaped calibration pattern, which is used for the initial calibration of the visual system.

[0023] Frame 2: a gantry type frame structure welded by rectangular steel pipes, which has high rigidity and stability, and stands above and around the cutting table 1.

[0024] Three-axis motion module: X-axis (horizontal) moving device 3: including two precision linear guides 3a arranged in parallel on the top of the frame 2, a rigid slide 3b matched with the guide, and a set of ball screw transmission pair 3d driven by a servo motor (as a horizontal moving motor 3c). The servo motor 3c is directly connected with the screw 3d through a shaft coupling; Y-axis (vertical) moving device 4: the main body is a bridge frame vertically fixed with the X-axis slide 3b, and the bridge frame is also provided with linear guides and ball screws. The vertical moving motor 4a drives the screw, thereby driving a vertical guide mounting plate moving along the bridge frame; Z-axis (vertical) moving device 5: including two linear guides installed on the vertical guide mounting plate, a slide block seat (as a vertical mounting seat 5a), and a ball screw driven by a vertical moving motor 5b (preferably a servo motor) through a synchronous belt, for driving the slide block seat 5a to move up and down accurately. The slide block seat 5a is integrated with a multi-dimensional force sensor 5c and a linear encoder with a built-in grating ruler, for monitoring the cutting pressure and the depth of pressing in real time.

[0025] Cutting executor 6: in this embodiment, a high-frequency vibration knife module that can be quickly replaced is adopted. It is installed below the vertical mounting seat 5a through a standardized mechanical interface and an electrical and gas quick connector. The vibration knife module includes an eccentric cutter head driven by a high-speed motor, and the vibration frequency and amplitude can be adjusted through an intelligent control unit. A laser cutting head or a circular knife cutting head can also be replaced according to the material (such as superfine fiber, leather, PVC, etc.).

[0026] Visual acquisition module 7: including a 5 million pixel global shutter industrial camera 7a fixed on the central beam of the frame 2. A group of white light sources 7b with adjustable brightness is arranged in a ring shape directly below the lens of the camera 7a, to ensure uniform and shadow-free lighting on the leather surface, highlighting the texture and contour. The field of view of the camera 7a completely covers the effective working area of the cutting table 1.

[0027] Intelligent control unit: is the core of the device, which adopts an industrial computer and a programmable logic controller.

[0028] Industrial PC: responsible for running high-level tasks, including: receiving and processing images from camera 7a; running image processing and path planning software developed based on OpenCV and custom algorithm library; responsible for managing the human-machine interface; communicating with the factory MES system through Ethernet.

[0029] Programmable Logic Controller: responsible for low-level real-time control, including: receiving path point instructions generated by IPC, accurately controlling the interpolation motion of three servo motors 3c, 4a, 5b; controlling the electromagnetic valve switch of vacuum adsorption partition 1a; collecting data from force sensor 5c and encoder; controlling the start and stop of cutting executor 6 and power.

[0030] Industrial PC and Programmable Logic Controller exchange data through high-speed field bus (such as EtherCAT) to ensure command synchronization.

[0031] Data interface: the intelligent control unit provides standard Ethernet interface and USB interface for importing CAD sheet files in DXF, PLT, etc. format, and exporting CSV format reports containing material utilization, cutting time, device operation log, etc.

[0032] The device in this embodiment works according to the following steps: Step S1: visual acquisition and feature extraction, the operator lays one or more (usually 2-4 layers) of the box bag leather flat on the cutting table 1. Start the "scan" command through the human-machine interface. The intelligent control unit controls the lighting source 7b to light up at the best brightness, and triggers the industrial camera 7a to take a high-resolution color or grayscale image.

[0033] The image is transmitted to the IPC for the following processing; a) Image preprocessing and calibration: using the calibration pattern on the table, the image is corrected for distortion and the pixel coordinates are mapped to mechanical coordinates.

[0034] b) Leather contour and defect recognition: the edge detection algorithm (such as Canny operator) is used to extract the boundary between leather and background, and an accurate polygon contour is fitted. At the same time, threshold segmentation and texture analysis algorithms are used to identify scar, color difference, blood vessel lines and other defect areas on the surface of the leather, and mark them as "no entry zone".

[0035] c) Layer recognition and offset calculation: if it is multi-layer leather, the algorithm will find natural feature points (such as texture intersection points, small holes) or artificial pasted marker points on each layer of leather, and calculate the translation and rotation offset (ΔX, ΔY, Δθ) of the lower layer relative to the uppermost layer through feature matching algorithm (such as SIFT or ORB).

[0036] Step S2: Intelligent nesting and path generation, the operator selects several cutting patterns (such as backpack front and back, side pockets, etc.) from the pattern library that are needed for this production. The intelligent control unit calls its path planning algorithm; a) Automatic nesting: With the primary goal of maximizing material utilization, the selected patterns are automatically arranged as rigid bodies within the effective leather contour polygon (subtracting the forbidden area) obtained in step S1. The algorithm uses heuristic search (such as genetic algorithm or simulated annealing algorithm) for iterative optimization to find the optimal layout. For multi-layer cutting, the system will independently calculate an optimized layout for each layer based on the offset compensation.

[0037] b) Path generation and optimization: After determining the final layout, the system generates a continuous cutting contour path for each pattern. Then all paths are globally optimized, including: merging common edges to reduce idle travel; planning the best start and end points; setting appropriate cutting and exit angles for the vibrating knife to prevent material tearing. Finally, a complete numerical control (NC) program is generated, which includes three-axis coordinate sequences, cutting speed, knife height, and vibrating knife power parameters.

[0038] Step S3: Dynamic collaborative cutting; a) Preparation: The intelligent control unit automatically opens the relevant vacuum suction partition 1a according to the leather contour position, firmly fixing the leather.

[0039] b) Execution: PLC reads the NC program and accurately drives the three-axis motion module, making the cutting implement 6's knife tip move along the planned path. When cutting straight lines and curves, the system maintains a constant cutting depth through the Z-axis servo motor 5b according to the material thickness preset value. The force sensor 5c provides real-time feedback on cutting resistance, and if the resistance abnormally increases (such as encountering a particularly dense area), the system will adjust the feed speed or increase the vibration power to achieve adaptive cutting.

[0040] c) Online monitoring: After completing the cutting of a pattern, the cutting implement 6 will raise the knife to a safe height and pause briefly. At this time, the camera 7a can quickly take a photo of the local key contour of the pattern, and by comparing it with the theoretical contour, the cutting quality is verified. In addition, PLC continuously monitors the current signal of the vibrating motor, and if the current waveform shows abnormal fluctuations, it may indicate that the tool is worn or broken, and the system will immediately pause and alarm.

[0041] Step S4: Finishing processing and information archiving; After all the paths are executed, the Z-axis is lifted to the highest point, and the X and Y axes return to the original point. The vacuum suction is turned off. The operator removes the cut pieces and excess material. The intelligent control unit automatically generates a report for this operation, recording: the original leather image, the layout result image, the actual material utilization rate, the total cutting time, the identified defect information, and any alarm events. These data are stored in the local database and can be uploaded through the network.

[0042] Step S5: Self-learning optimization. The intelligent control unit runs a data learning module in the background. When the processing data of the same type of leather (such as cowhide of a specific origin, thickness, and tanning process) accumulates to a certain amount, the system analyzes the correlation between the "average defect rate" and "average cutting resistance distribution" of this type of leather and the "layout strategy", "cutting speed", "pressing depth" that have achieved the best results (high utilization rate, high quality, high efficiency) in history. When the same type of leather is processed again, the system will preferentially recommend these optimized process parameter combinations, thereby realizing continuous improvement of intelligence.

[0043] The above embodiments of the present application are not intended to limit the scope of protection of the present application, and the embodiments of the present application are not limited thereto. According to the above content of the present application, according to the ordinary technical knowledge and common methods in the art, other various forms of modifications, replacements or changes to the above structure of the present application, which do not deviate from the above basic technical idea of the present application, should fall within the scope of protection of the present application.

Claims

1. An intelligent cutting device for bag and luggage processing and production, characterized in that, include: A cutting table, used to hold the leather bags to be processed; A frame is arranged around the cutting table; A three-axis motion module, mounted on the frame, includes a transverse slide driven by a transverse movement motor, a longitudinal slide driven by a longitudinal movement motor, and a vertical mounting base driven by a vertical movement motor. A cutting actuator, mounted on the vertical mounting base, is used to cut leather; The visual acquisition module is fixedly installed at a high position on the frame, and its acquisition angle covers the entire working area of ​​the cutting table. The intelligent control unit is electrically connected to the vision acquisition module, each motor in the three-axis motion module, and the cutting actuator, respectively. The intelligent control unit is configured as follows: Receive and process the leather image data acquired by the visual acquisition module; Based on the processing results, the effective area and defects of the leather are automatically identified, and the cutting path of the cutting actuator for single or multiple overlapping leathers is planned to avoid defects and optimize the layout. The three-axis motion module is controlled to work in coordination with the cutting actuator to execute the cutting path.

2. The intelligent cutting device according to claim 1, characterized in that, The vision acquisition module includes at least one high-resolution industrial camera and a surrounding illumination source; the surface of the cutting table is provided with background reference marks or calibration patterns that are communicatively connected to the intelligent control unit.

3. The intelligent cutting device according to claim 1, characterized in that, The cutting actuator is a modular structure that can be quickly replaced, selected from one of a high-frequency vibrating knife module, a laser cutting head, a circular knife cutting head, or a stamping cutting head; the vertical mounting base integrates a pressure sensor and a displacement sensor to monitor the cutting force and pressing depth in real time and form a closed-loop feedback.

4. The intelligent cutting device according to claim 1, characterized in that, The cutting table is an independently controllable multi-zone vacuum adsorption table. The intelligent control unit dynamically controls the adsorption start / stop and adsorption force of the corresponding zones according to the leather outline and placement position identified by the vision acquisition module.

5. The intelligent cutting device according to claim 1, characterized in that, It also includes a human-machine interface and a data interface connected to the intelligent control unit; the human-machine interface is used to display visual recognition results, cutting path simulation, equipment status, and to receive parameter input and manual intervention commands; the data interface is used to receive external CAD / CAM sheet data and upload production data.

6. The intelligent cutting device according to claim 1, characterized in that, The path planning algorithm of the intelligent control unit includes: edge extraction and fitting based on the leather outline image; automatic sorting of multiple preset cutting pattern sheets within the fitted effective outline, and iterative calculation with the goal of maximizing material utilization or optimizing cutting efficiency; and setting a no-entry zone for the identified defective areas.

7. A method for cutting leather for bags based on the intelligent cutting device according to any one of claims 1-6, characterized in that, Includes the following steps: S1: Visual acquisition and feature extraction: The visual acquisition module acquires a global image of the leather placed on the cutting table and identifies the effective outline, surface defects, preset positioning marks and stacking layer information of the leather. S2: Intelligent nesting and path generation: The intelligent control unit calls the preset cutting pattern library, combines the feature information extracted in step S1, performs automatic nesting calculation, and generates an optimized cutting path instruction set. The path instruction set includes spatial coordinates, cutter start / stop and working parameters. S3: Dynamic Collaborative Cutting: The intelligent control unit drives the three-axis motion module and the cutting actuator to perform cutting actions along the optimized cutting path, and dynamically adjusts the motion parameters and cutting parameters based on real-time feedback sensor data; S4: Completion Processing and Information Archiving: After all paths have been cut, the equipment automatically resets, and the intelligent control unit generates and stores a production report containing information such as the material utilization rate, defect location, and actual cutting time.

8. The method for cutting leather for bags according to claim 7, characterized in that, In step S2, when multiple leather layers are identified to be stacked, the method further includes: calculating the relative offset vector between each stack using an image feature point matching algorithm, and performing translation and rotation compensation on the theoretical cutting path of each layer to generate a final execution path suitable for the current stack state.

9. The method for cutting leather for bags according to claim 7 or 8, characterized in that, Step S3 also includes a real-time online monitoring step: the visual acquisition module performs local secondary imaging of the cut contour during the cutting interval, or compares the signals collected by the acoustic sensor and vibration sensor with a preset threshold. If path deviation, material warping or tool abnormality is detected, the process is immediately paused and an alarm is triggered, while providing options for manual intervention or automatic retry.

10. The method for cutting leather for bags according to claim 7, characterized in that, The method also includes a self-learning optimization step: the intelligent control unit accumulates and analyzes historical cutting data and corresponding visual images to establish a correlation model between leather type, texture direction, defect distribution and optimal layout scheme, cutting speed and pressing depth, which is used to optimize the cutting parameters of subsequent similar leathers.

Images