Intelligent cutting piece sorting and sewing production line based on AI visual identification

Through AI visual recognition and blockchain technology, combined with robotic arms and sewing equipment, the accuracy, flexibility and coordination problems of traditional cutting and sewing production lines have been solved, achieving efficient automated production.

CN120666504APending Publication Date: 2025-09-19ZHEJIANG TEXTILE & FASHION COLLEGE
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202510782076.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-12
Publication Date
2025-09-19

AI Technical Summary

Technical Problem

Traditional cutting and sorting production lines rely on manual operation, which is inefficient and prone to errors. They have insufficient recognition accuracy, inefficient data management, weak flexible production capabilities, difficulty in adapting to complex scenes and lighting changes, and backward data storage and sharing methods.

Method used

It uses AI visual recognition modules and deep learning algorithms, combined with high-resolution industrial cameras and GPU servers, to achieve piece feature recognition and dynamic learning; uses a robotic arm sorting system and adaptive fixtures, and integrates blockchain storage technology to ensure secure data sharing; and configures multiple types of sewing equipment and quality detection sensors to achieve automated production.

Benefits of technology

It improves the accuracy of piece recognition and the flexibility of the production line, reduces human errors, ensures data security and synchronization, adapts to complex scenes and lighting changes, and improves production efficiency and quality controllability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure HDA0005446006270000011
    Figure HDA0005446006270000011
  • Figure HDA0005446006270000021
    Figure HDA0005446006270000021
  • Figure HDA0005446006270000031
    Figure HDA0005446006270000031
Patent Text Reader

Abstract

The invention discloses an intelligent cut-part sorting and sewing production line based on AI visual identification, and belongs to the field of visual identification, the intelligent cut-part sorting and sewing production line comprises a main conveying belt and a plurality of branch conveying belts, and an ironing module, an AI visual identification module and a sorting module are arranged above the main conveying belt; a sewing module is arranged above the shunting conveying belt; the ironing module comprises a transverse frame, a first motor is fixed to the right side of the transverse frame, the left end of an output shaft of the first motor penetrates into the transverse frame and is fixedly provided with a screw rod, the screw rod is rotationally connected with the interior of the transverse frame, and the outer surface of the screw rod is in threaded connection with two supports; through the dynamic learning ability of AI visual identification, the block chain data security mechanism and the self-adaptive design of the mechanical structure, the core pain points of a traditional production line in the aspects of precision, flexibility and collaboration are solved, and the method is particularly suitable for a multi-variety and small-batch intelligent garment production scene and has remarkable industrial application value.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of visual recognition technology, and in particular to an intelligent piece sorting and sewing production line based on AI visual recognition. Background Art

[0002] Pain points of traditional cut piece sorting and sewing production lines

[0003] High dependence on manual labor: The ironing, sorting and sewing processes of the cut pieces rely on manual operations, which is inefficient and prone to unstable quality due to human errors.

[0004] Insufficient recognition accuracy: Traditional visual inspection technology is unable to quickly and accurately identify the shape, color, pattern and other characteristics of complex pieces, especially poor adaptability to new fabrics or special cuts.

[0005] Inefficient data management: Standard data for cutting parts is stored in a scattered manner, making data synchronization between multiple production lines difficult. There is also a risk of data tampering, which affects production standardization.

[0006] Weak flexible production capabilities: The production line is unable to quickly adapt to process adjustments for new cut pieces, and frequent manual intervention in model training and equipment parameter settings is required.

[0007] Limitations of existing technologies

[0008] The lack of integration of AI visual recognition and deep learning algorithms results in delayed feature recognition of cut pieces;

[0009] Lack of dynamic adaptive mechanism, unable to cope with complex scenarios such as lighting changes and new types of cuts;

[0010] Data storage and sharing methods are backward and cannot meet the collaborative needs of large-scale intelligent manufacturing.

[0011] Therefore, an intelligent cutting piece sorting and sewing production line based on AI visual recognition is needed to solve the problems of traditional production lines in terms of precision, flexibility and coordination through the dynamic learning ability of AI visual recognition, blockchain data security mechanism and adaptive design of mechanical structure. Summary of the Invention

[0012] In response to the shortcomings of the existing technology, the present invention provides an intelligent cutting piece sorting and sewing production line based on AI visual recognition, which solves the problems raised in the above background technology.

[0013] Technical solution: To solve the above technical problems, according to one aspect of the present invention, more specifically, an intelligent piece sorting and sewing production line based on AI visual recognition, comprising a main conveyor belt and several branch conveyor belts. An ironing module, an AI visual recognition module, and a sorting module are provided above the main conveyor belt; a sewing module is provided above the branch conveyor belt;

[0014] The ironing module includes a horizontal frame, a motor 1 is fixed on the right side of the horizontal frame, the left end of the output shaft of the motor 1 passes through the interior of the horizontal frame and is fixed with a screw, the screw is rotatably connected to the interior of the horizontal frame, the outer surface of the screw is threadedly connected to two brackets, the bottom of the two brackets are jointly fixed with a fixed disk, the bottom of the fixed disk is rotatably connected to the disk frame, the upper surface of the disk frame is embedded with a motor 2, the top end of the output shaft of the motor 2 is fixed with a transmission wheel, the transmission wheel is in contact with the outer surface of the fixed disk, and a plurality of rotating racks are rotatably connected to the outer surface of the disk frame, the rotating rack is provided with movable slots through the front and back, a middle tube is fixed through the inside of the disk frame, the outer surface of the middle tube is rotatably connected to the fixed disk, and the outer surface of the middle tube A sleeve is slidably connected at the bottom, and a spring is fixed between the top of the sleeve and the lower surface of the disc frame. A water pan is integrally formed at the bottom of the sleeve, and a number of inclined frames are fixed on the upper surface of the water pan. The number of the inclined frames is the same as that of the rotating frame. A movable rod is movably connected inside the movable groove, and the top of the inclined frame is located inside the rotating frame and fixedly connected to the outer surface of the movable rod. A connecting pipe is fixed through the bottom of the rotating frame, and both ends of the connecting pipe are rotatably connected to ironing cylinders, and both ironing cylinders are rotatably connected to the outer surface of the rotating frame. A plurality of horizontal pipes are slidably connected inside the water pan, and the end of the horizontal pipe away from the water pan is located inside the rotating frame and communicated with the inside of the connecting pipe. Two hydraulic rods are fixed on the top of the horizontal frame.

[0015] Furthermore, the AI ​​visual recognition module uses a high-resolution industrial camera with lenses of different focal lengths to meet the shooting requirements of pieces of different sizes and adopts multi-angle image acquisition. It is equipped with a dedicated AI computing unit, such as a GPU server or edge computing device, to quickly process the collected image data and ensure the real-time performance of visual recognition.

[0016] The AI ​​visual recognition module uses deep learning algorithms, such as convolutional neural networks, to train and identify features such as the shape, color, pattern, and size of the cut pieces. An image preprocessing program is developed to perform denoising, enhancement, and edge detection on the collected images to improve recognition accuracy. At the same time, a cut piece database is established to store standard images and parameter information of various cut pieces to facilitate system comparison and analysis.

[0017] Furthermore, the sorting module adopts a robotic arm sorting system, and the end of the robotic arm is equipped with an adaptive clamp, which can be grasped and adjusted according to the shape and material of the cut piece; the sorting area is set between the main conveyor belt and multiple diversion conveyor belts, and the AI ​​visual recognition module is used to identify the cut piece and accurately push it to the corresponding diversion conveyor belt.

[0018] Furthermore, the sewing module integrates multiple types of sewing equipment, including but not limited to flat-bed sewing machines, overlock sewing machines, and special sewing machines. It accurately controls the conveying speed and position of the cut pieces through the diverter conveyor belt, and cooperates with the sewing equipment to complete complex sewing processes; it is equipped with quality detection sensors to perform real-time quality detection on the sewn products.

[0019] Furthermore, the top of the middle pipe is connected to an external hot water pipe.

[0020] Furthermore, the AI ​​visual recognition module further integrates a dynamic feature learning algorithm, which automatically updates the deep learning model parameters through the real-time collection of cutting piece image data to adapt to new cutting piece features or material changes;

[0021] The image preprocessing program embeds an adaptive threshold segmentation algorithm to automatically adjust the binarization threshold for images under different lighting conditions, thereby improving the edge detection accuracy in complex scenes;

[0022] The cutting piece database adopts blockchain distributed storage technology to store standard image hash values ​​and parameter information in blockchain nodes, ensuring that the data cannot be tampered with, and supports the secure sharing and synchronous update of cutting piece feature data among multiple production lines.

[0023] The beneficial effects of the intelligent piece sorting and sewing production line based on AI visual recognition of the present invention are:

[0024] (1) The ironing module of the present invention automatically stretches out wrinkles and evenly heats the pieces through mechanical linkage and hot water supply, thereby improving the flatness of the cut pieces, reducing subsequent errors, and achieving precise pre-processing.

[0025] (2) The present invention solves the core pain points of traditional production lines in terms of precision, flexibility and coordination through the dynamic learning ability of AI visual recognition, blockchain data security mechanism and adaptive design of mechanical structure. It is particularly suitable for intelligent clothing production scenarios with multiple varieties and small batches, and has significant industrial application value. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The present invention will be further described in detail below with reference to the accompanying drawings and specific implementation methods.

[0027] Figure 1 It is a structural schematic diagram of the present invention;

[0028] Figure 2 Schematic diagram of the structure of the ironing module of the present invention;

[0029] Figure 3 It is a structural schematic diagram of the water tray and tray frame in the present invention;

[0030] Figure 4Schematic diagram of the cross-sectional structure of the water tray and tray frame in the present invention;

[0031] Figure 5 Schematic diagram of the cross-section structure of the water tray and the ironing cylinder in the present invention;

[0032] Figure 6 It is a structural schematic diagram of the water tray and connecting pipe in the present invention.

[0033] In the figure: 1. Horizontal frame; 2. Motor 1; 3. Screw; 4. Bracket; 5. Fixed plate; 6. Plate frame; 7. Motor 2; 8. Drive wheel; 9. Rotating frame; 10. Movable groove; 11. Middle pipe; 12. Sleeve; 13. Spring; 14. Water tray; 15. Oblique frame; 16. Movable rod; 17. Connecting pipe; 18. Ironing cylinder; 19. Horizontal pipe; 20. Hydraulic rod. DETAILED DESCRIPTION

[0034] The present invention will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that, unless there is a conflict, the embodiments and features in the embodiments of the present application can be combined with each other.

[0035] In order to make the technical solution of the present invention clearer, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0036] Reference Figures 1-6 An intelligent piece sorting and sewing production line based on AI visual recognition includes a main conveyor belt and several branch conveyor belts. An ironing module, an AI visual recognition module, and a sorting module are installed above the main conveyor belt; a sewing module is installed above the branch conveyor belt. The operation steps are as follows:

[0037] S1, Pre-processing stage: The ironing module eliminates wrinkles on the cut pieces through mechanical movement and heat conduction, providing a flat detection surface for visual recognition;

[0038] S2, Identification and Sorting Stage: The AI ​​visual recognition module quickly analyzes the features of the cut pieces, and the sorting module implements “immediate identification and sorting” based on the results, avoiding efficiency losses caused by manual intervention;

[0039] S3, sewing and quality inspection stage: the diverter belt accurately controls the positioning of the cut pieces, and the sewing equipment is linked with the quality inspection sensor to ensure efficient production and controllable quality.

[0040] The ironing module includes a horizontal frame 1, a motor 2 is fixed on the right side of the horizontal frame 1, the left end of the output shaft of the motor 2 passes through the interior of the horizontal frame 1 and is fixed with a screw 3, the screw 3 is rotatably connected to the interior of the horizontal frame 1, the outer surface of the screw 3 is threadedly connected to two brackets 4, a fixed plate 5 is fixed to the bottom of the two brackets 4, the bottom of the fixed plate 5 is rotatably connected to a disc rack 6, a motor 2 7 is embedded on the upper surface of the disc rack 6, a transmission wheel 8 is fixed to the top of the output shaft of the motor 2 7, the transmission wheel 8 is in contact with the outer surface of the fixed disc 5, a plurality of rotating racks 9 are rotatably connected to the outer surface of the disc rack 6, the rotating rack 9 is provided with a movable slot 10 through the front and back, a middle tube 11 is fixed to the inside of the disc rack 6, the outer surface of the middle tube 11 is rotatably connected to the fixed disc 5, and a sleeve 12 is slidably connected to the lower outer surface of the middle tube 11, and the sleeve 1 A spring 13 is fixed between the top and the lower surface of the tray frame 6. A water pan 14 is integrally formed at the bottom of the sleeve 12. A plurality of inclined brackets 15 are fixed to the upper surface of the water pan 14. The number of inclined brackets 15 is the same as that of the rotating frame 9. A movable rod 16 is movably connected to the interior of the movable slot 10. The top of the inclined bracket 15 is located inside the rotating frame 9 and fixedly connected to the outer surface of the movable rod 16. A connecting pipe 17 is fixed through the bottom of the rotating frame 9. Both ends of the connecting pipe 17 are rotatably connected to ironing cylinders 18. Both ironing cylinders 18 are rotatably connected to the outer surface of the rotating frame 9. Multiple horizontal pipes 19 are slidably connected to the interior of the water pan 14. The ends of the horizontal pipes 19 away from the water pan 14 are located inside the rotating frame 9 and communicate with the interior of the connecting pipe 17. Two hydraulic rods 20 are fixed to the top of the horizontal frame 1. The top of the middle pipe 11 is connected to the external hot water pipe.

[0041] When the piece moves to the ironing module, the hydraulic rod 20 extends, causing the horizontal frame 1 and other structures to move downward, so that the ironing cylinder 18 and the water tray 14 are in contact with the piece. During the continuous downward pressure, the inclined frame 15 pushes the inner wall of the movable groove 10 through the movable rod 16, thereby causing the rotating frame 9 to rotate outward, so that the surface of the piece is stretched flat by the two ironing cylinders 18 at the bottom of the rotating frame 9, thereby avoiding errors in the finished product caused by wrinkles in the subsequent steps. At the same time, hot water enters the two ironing cylinders 18 through the middle pipe 11, the sleeve 12, the outward sliding horizontal pipe 19, and the connecting pipe 17, heating the ironing cylinders 18 and the water tray 14, thereby achieving an ironing effect; the disc frame 6 and other structures are rotated and adjusted by the motor 2 7 and the transmission wheel 8, and the disc frame 6 and other structures are displaced under the action of the motor 1 2 and the screw 3, so that the ironing operation can be carried out more comprehensively on pieces in different positions and at different positions of the pieces.

[0042] Preferably, the AI ​​visual recognition module uses a high-resolution industrial camera with lenses of different focal lengths to meet the shooting requirements of pieces of different sizes and adopts multi-angle image acquisition; a dedicated AI computing unit, such as a GPU server or edge computing device, is configured to quickly process the collected image data to ensure the real-time performance of visual recognition;

[0043] The AI ​​visual recognition module uses deep learning algorithms, such as convolutional neural networks (CNNs), to train and identify features such as the shape, color, pattern, and size of the pieces. An image preprocessing program is developed to perform denoising, enhancement, and edge detection on the collected images to improve recognition accuracy. At the same time, a piece database is established to store standard images and parameter information of various types of pieces to facilitate system comparison and analysis.

[0044] Preferably, the sorting module adopts a robotic arm sorting system, and the end of the robotic arm is equipped with an adaptive clamp, which can be grasped and adjusted according to the shape and material of the cut piece; the sorting area is set between the main conveyor belt and multiple diversion conveyor belts, and the AI ​​visual recognition module is used to identify the cut piece and accurately push it to the corresponding diversion conveyor belt.

[0045] Preferably, the sewing module integrates multiple types of sewing equipment, including but not limited to flat-bed sewing machines, overlock sewing machines, and special sewing machines. It accurately controls the conveying speed and position of the cut pieces through the diverter conveyor belt, and cooperates with the sewing equipment to complete complex sewing processes; and is equipped with quality detection sensors to perform real-time quality detection on the sewn products.

[0046] Preferably, the AI ​​visual recognition module further integrates a dynamic feature learning algorithm to automatically update the deep learning model parameters through the real-time collection of cutting piece image data to adapt to new cutting piece features or material changes;

[0047] The image preprocessing program embeds an adaptive threshold segmentation algorithm to automatically adjust the binarization threshold for images under different lighting conditions, thereby improving the edge detection accuracy in complex scenes;

[0048] The cutting piece database adopts blockchain distributed storage technology to store standard image hash values ​​and parameter information in blockchain nodes, ensuring that the data cannot be tampered with, and supports the secure sharing and synchronous update of cutting piece feature data among multiple production lines.

[0049] The above embodiments only express several implementation methods of the present invention, and their description is relatively specific and detailed, but they cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that for ordinary technicians in this field, without departing from the concept of the present invention,

[0050] Under the above conditions, several modifications and improvements can be made, which all fall within the scope of protection of the present invention.

[0051] Therefore, the scope of protection of the patent for this invention should be based on the attached claims.

Claims

1. An intelligent piece sorting and sewing production line based on AI visual recognition, comprising a main conveyor belt and several branch conveyor belts, characterized by: An ironing module, an AI visual recognition module, and a sorting module are provided above the main conveyor belt; a sewing module is provided above the diversion conveyor belt; The ironing module comprises a horizontal frame (1), a motor (2) is fixed on the right side of the horizontal frame (1), the left end of the output shaft of the motor (2) passes through the interior of the horizontal frame (1) and is fixed with a screw (3), the screw (3) is rotatably connected to the interior of the horizontal frame (1), the outer surface of the screw (3) is threadedly connected to two brackets (4), the bottom of the two brackets (4) are commonly fixed with a fixed plate (5), the bottom of the fixed plate (5) is rotatably connected to a plate rack (6), and the upper surface of the plate rack (6) is The surface is embedded with a second motor (7), a transmission wheel (8) is fixed on the top of the output shaft of the second motor (7), and the transmission wheel (8) is in contact with the outer surface of the fixed disk (5). The outer surface of the disk rack (6) is rotatably connected with a plurality of rotating racks (9), and the rotating racks (9) are provided with movable grooves (10) at the front and rear. A middle tube (11) is fixed inside the disk rack (6), and the outer surface of the middle tube (11) is rotatably connected with the fixed disk (5). A sleeve is slidably connected to the lower surface of the outer surface of the middle tube (11). (12), a spring (13) is fixed between the top of the sleeve (12) and the lower surface of the disc rack (6), a water tray (14) is integrally formed at the bottom of the sleeve (12), a plurality of inclined racks (15) are fixed on the upper surface of the water tray (14), the number of the inclined racks (15) is the same as that of the rotating rack (9), a movable rod (16) is movably connected inside the movable groove (10), the top of the inclined rack (15) is located inside the rotating rack (9) and is fixed to the outer surface of the movable rod (16) The bottom of the rotating frame (9) is fixed with a connecting pipe (17), both ends of the connecting pipe (17) are rotatably connected with ironing cylinders (18), and the two ironing cylinders (18) are rotatably connected to the outer surface of the rotating frame (9). The interior of the water tray (14) is slidably connected with a plurality of transverse pipes (19), and one end of the transverse pipe (19) away from the water tray (14) is located inside the rotating frame (9) and is connected to the interior of the connecting pipe (17). Two hydraulic rods (20) are fixed on the top of the transverse frame (1).

2. The intelligent piece sorting and sewing production line based on AI visual recognition according to claim 1, characterized in that: The AI ​​visual recognition module uses a high-resolution industrial camera with lenses of different focal lengths to capture images of pieces of varying sizes, and employs multi-angle image acquisition. It is equipped with a dedicated AI computing unit, such as a GPU server or edge computing device, to rapidly process the captured image data and ensure real-time visual recognition. The AI ​​visual recognition module uses deep learning algorithms, such as convolutional neural networks (CNNs), to train and identify features such as the shape, color, pattern, and size of the pieces. An image preprocessing program is developed to perform denoising, enhancement, and edge detection on the collected images to improve recognition accuracy. At the same time, a piece database is established to store standard images and parameter information of various types of pieces to facilitate system comparison and analysis.

3. The intelligent piece sorting and sewing production line based on AI visual recognition according to claim 1, characterized in that: The sorting module adopts a robotic arm sorting system, and the end of the robotic arm is equipped with an adaptive clamp, which can be grasped and adjusted according to the shape and material of the cut piece; the sorting area is set between the main conveyor belt and multiple diversion conveyor belts, and the AI ​​visual recognition module is used to identify the cut pieces and accurately push them to the corresponding diversion conveyor belt.

4. The intelligent piece sorting and sewing production line based on AI visual recognition according to claim 1, characterized in that: The sewing module integrates multiple types of sewing equipment, including but not limited to flat-bed sewing machines, overlock sewing machines, and special sewing machines. It accurately controls the conveying speed and position of the cut pieces through the diverter conveyor belt, and cooperates with the sewing equipment to complete complex sewing processes; it is equipped with quality detection sensors to perform real-time quality inspection on the sewn products.

5. The intelligent piece sorting and sewing production line based on AI visual recognition according to claim 1, characterized in that: The top of the middle pipe (11) is connected to an external hot water pipe.

6. The intelligent piece sorting and sewing production line based on AI visual recognition according to claim 2, characterized in that: The AI ​​visual recognition module further integrates a dynamic feature learning algorithm to automatically update deep learning model parameters through real-time collected piece image data to adapt to new piece features or material changes; The image preprocessing program embeds an adaptive threshold segmentation algorithm to automatically adjust the binarization threshold for images under different lighting conditions, thereby improving the edge detection accuracy in complex scenes; The cutting piece database adopts blockchain distributed storage technology to store standard image hash values ​​and parameter information in blockchain nodes, ensuring that the data cannot be tampered with, and supports the secure sharing and synchronous update of cutting piece feature data among multiple production lines.