Fabric separation method, device and system and storage medium

Abstract

The invention discloses a fabric separation method, device and system and a storage medium. The fabric separation method comprises the following steps: determining detection characteristics of a to-be-separated fabric based on a detection image of the to-be-cut fabric; the detection features at least comprise feature recognition lines of to-be-detected textures included in the to-be-separated fabric; obtaining a standard feature indicated by a standard image corresponding to the to-be-separated fabric; the standard feature at least comprises a feature standard line of a standard texture included in the standard image; determining whether the feature recognition line is within the deviation range of the feature standard line; and if yes, determining a target separation contour of the to-be-separated fabric by using the standard features and / or the detection features. According to the method, the feature point line related to the to-be-separated fabric is compared with the feature point line related to the standard fabric meeting the acceptance standard, whether deformation of the to-be-separated fabric is within the error range or not is determined, and it is determined that the whole separated product meets the acceptance requirement.

Description

Technical Field

[0001] This application relates to the field of industrial automation control and processing, and in particular to an automatic fabric separation method, apparatus, system and storage medium. Background Technology

[0002] Fabrics, such as those used in shoe uppers, are flexible and prone to deformation. Therefore, maintaining process consistency during automated fabric processing, such as cutting and slicing, to ensure the final product meets the cutting requirements is a problem that needs to be solved in this field. Summary of the Invention

[0003] To address the aforementioned problems, this application discloses a fabric separation method, apparatus, system, and storage medium. The method determines the final separation profile by comparing features after setting a range for error in feature point lines, rather than correcting the separation profile of the deformed fabric.

[0004] The first aspect of this application provides a fabric separation method, the method comprising: determining detection features of the fabric to be separated based on a detection image of the fabric to be separated; acquiring standard features indicated by a standard image corresponding to the fabric to be separated; determining whether the detection features are within the deviation range of the standard features; and if so, determining a target separation contour of the fabric to be separated using the standard features and / or the detection features.

[0005] By determining whether the detection features in the detection image of the fabric to be cut are within the deviation range of the standard features corresponding to the standard image that meets the acceptance criteria, it is determined whether the deformation of the shoe upper to be separated is within the error range. Based on the determination result, the target separation contour is determined with standard features, ensuring that the overall separation of the fabric meets the acceptance criteria.

[0006] According to some embodiments of this application, determining the detection features may include: processing the detection image using a pre-trained feature recognition model to determine the detection features; wherein, the training fabric samples of the feature recognition model are at least marked with feature marker lines of the sample textures included in the training fabric samples; the detection features at least include feature recognition lines of the textures to be detected included in the fabric to be cut, output by the feature recognition model.

[0007] By using a feature recognition model trained with a large number of fabric samples to confirm the features of the detected image, the recognition result can be obtained quickly and accurately.

[0008] According to some embodiments of this application, the standard feature includes feature standard lines of the standard texture included in the standard image; determining whether the detection feature is within the deviation range of the standard feature may include: forming a deviation region including the feature standard lines based on the feature standard lines; determining whether the detection feature is within the deviation range of the standard feature based on whether the feature recognition line is within the deviation region.

[0009] According to some embodiments of this application, forming a deviation region including the feature standard line may include: obtaining an allowable deviation distance; and expanding the feature standard line outward based on the allowable deviation distance to form the deviation region.

[0010] According to some embodiments of this application, the feature standard lines include one or more; the training fabric sample is further marked with marker points and / or marker lines for position alignment; the standard features include positioning points and / or positioning lines; the feature recognition model also outputs alignment points and / or alignment lines trained based on the marker points and / or the marker lines; determining whether the feature recognition line is within the deviation region may include: aligning the alignment points with the positioning points and / or the alignment lines with the positioning lines to determine one or more feature recognition lines corresponding to the one or more feature standard lines in position; determining whether the one or more feature recognition lines are all within the deviation region of the corresponding one or more feature standard lines; if so, determining that the detected feature is within the deviation range of the standard features.

[0011] By aligning the alignment point / location point and / or the alignment line / location line, the feature recognition line is aligned with the feature standard line in position, ensuring the accuracy of the comparison.

[0012] According to some embodiments of this application, the standard features include a preset cutting contour; determining the target separation contour of the fabric to be separated using the standard features may include: overlaying the preset separation contour onto the detection image of the fabric to be separated to generate the target separation contour.

[0013] According to some embodiments of this application, the generation of the preset separation contour based on the feature standard line may include: obtaining a positional relationship indicating the relative position between the feature standard line and the preset separation contour; and determining the preset separation contour based on the positional relationship using the feature standard line.

[0014] According to some embodiments of this application, the method may further include: updating the feature recognition model using the feature recognition line.

[0015] Continuously updating the feature recognition model can improve the model's recognition accuracy and efficiency.

[0016] According to some embodiments of this application, if the detected feature is outside the deviation range of the standard feature, the method is terminated.

[0017] Waste generation can be reduced by terminating the processing of fabrics that are outside the acceptable error range.

[0018] A second aspect of this application provides a fabric separation apparatus, the apparatus comprising: a determining module configured to determine detection features of the fabric to be separated based on a detection image of the fabric to be separated; an acquiring module configured to acquire standard features indicated by a standard image corresponding to the fabric to be separated; a comparison module configured to determine whether the detection features are within the deviation range of the standard features; and an execution module configured to determine a target separation contour of the fabric to be cut using the standard features and / or the detection features when it is determined that the detection features are within the deviation range of the standard features.

[0019] A third aspect of this application provides a processing system that may include a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, can implement the steps of the fabric separation method as described above.

[0020] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the fabric separation method described above.

[0021] The fifth aspect of this application provides a computer program product comprising a computer program that, when executed by a processor, implements the steps of the fabric separation method as described above.

[0022] A sixth aspect of this application provides a fabric separation system, the system comprising: an image capture component for acquiring a detection image of a fabric to be separated; a processing component for: determining detection features of the fabric to be separated based on the detection image; the detection features including at least feature recognition lines of the texture to be detected included in the fabric to be separated; acquiring standard features indicated by a standard image corresponding to the fabric to be cut; the standard features including at least feature standard lines of the standard texture included in the standard image; determining whether the feature recognition lines are within the deviation range of the feature standard lines; if so, determining a target separation contour of the fabric to be separated using the standard features and / or the standard features; and a separation component for separating the fabric to be separated based on the target separation contour.

[0023] The fabric separation method disclosed in this application determines whether the deformation of the shoe upper to be separated is within the error range by judging whether the detection features in the detection image of the fabric to be cut are within the deviation range of the standard features corresponding to the standard image that meets the acceptance criteria. Based on the judgment result, the target separation contour is determined by the standard features, ensuring that the overall separation of the fabric meets the acceptance criteria.

[0024] Additional aspects and advantages of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this application. Attached Figure Description

[0025] The disclosure of this application will become more readily understood with reference to the accompanying drawings. It will be readily understood by those skilled in the art that these drawings are for illustrative purposes only and are not intended to limit the scope of protection of this application. Furthermore, similar numbers in the drawings are used to denote similar components, wherein: Figure 1 These are exemplary schematic diagrams of fabric separation systems according to some embodiments of this application; Figure 2 This is an exemplary schematic diagram of a computing device for implementing a fabric separation method according to some embodiments of this application; Figure 3 This is an exemplary flowchart of a fabric separation method according to some embodiments of this application; Figure 4 These are exemplary schematic diagrams of training fabric samples according to some embodiments of this application; Figure 5 These are exemplary schematic diagrams illustrating fabric separation methods according to some embodiments of this application; Figure 6 These are exemplary schematic diagrams showing a feature comparison according to some embodiments of this application; Figure 7 This is an exemplary block diagram of a processing system for implementing a fabric separation method according to some embodiments of this application; Figure 8 This is an exemplary structural diagram of a fabric separation system according to some embodiments of this application. Detailed Implementation

[0026] To make the above-mentioned objectives, features, and advantages of this application more apparent and understandable, the specific embodiments of this application are described in detail below with reference to the accompanying drawings. Many specific details are set forth in the following description to provide a thorough understanding of this application. However, this application can be implemented in many other ways different from those described herein, and those skilled in the art can make similar modifications without departing from the spirit of this application. Therefore, this application is not limited to the specific embodiments disclosed below.

[0027] Unless otherwise defined, all technical and scientific terms used in this application have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains. The terminology used in this application and in its specification is for the purpose of describing particular embodiments only and is not intended to limit the application. The terms "first," "second," and similar terms used in this application do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed after the word and their equivalents, without excluding other elements or objects. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships, which may change accordingly when the absolute position of the described object changes. The terms "and / or" or "and / or" as used in this application include any and all combinations of one or more of the associated listed items.

[0028] Cutting is one means or method of achieving separation. The following examples will all use cutting. Currently, fabric cutting uses positioning laser cutting, which mainly involves using a camera to photograph the fabric surface and then locating the laser cutting position based on the outline and specific marker points. Additionally, fabric deformation can be corrected by adjusting the cutting outline.

[0029] However, the above methods have the following problems: First, the identified markers or contours need to be set individually for each different fabric, resulting in low efficiency. Second, alignment can only be based on specifically set markers, lacking the ability to determine whether there are positional deviations in other areas. When cutting based on marker positioning, other areas may not meet the cutting standards.

[0030] The fabric separation method disclosed in this application identifies the features of the fabric to be cut and compares them with the features carried in a standard image. Once determined to be within the error range, a suitable laser cutting profile is identified. This method can make an overall judgment on the fabric to be cut and determine the optimal laser cutting profile through calibration, rather than distorting and deforming the cutting profile according to the fabric in parallel. Ultimately, a shoe upper fabric that meets overall acceptance standards can be obtained.

[0031] The following description, with reference to the accompanying drawings, illustrates some preferred embodiments of the present application. It should be noted that the following description is for illustrative purposes only and is not intended to limit the scope of protection of the present application. The flowcharts used are for illustrating the operations performed by the system according to embodiments of the present application. It should be understood that the described operations are not necessarily performed precisely in sequence. Instead, various steps may be processed in reverse order or simultaneously. Furthermore, other operations may be added to these processes, or one or more steps may be removed from these processes.

[0032] Figure 1 This is an exemplary schematic diagram of a fabric separation system according to some embodiments of this application. In some embodiments, the fabric separation system 100 can be used for automatic identification and automatic cutting of fabric to be cut. Figure 1 As shown, the fabric separation system 100 may include a processing device 110, an execution platform 120, a terminal 130, a storage device 140, a network 150, and a data source 160.

[0033] The processing device 110 can be used to process information and / or data related to the fabric to be cut to perform one or more functions disclosed in this application. For example, the processing device 110 can receive a detection image of the fabric to be cut and determine the detection features of the fabric. The detection image can be captured by the execution platform 120, for example, the execution platform 120 is equipped with a camera device such as a camera or image sensor. As another example, the processing device 110 can acquire standard features indicated by a standard image corresponding to the fabric to be cut. This standard image can be pre-stored in the memory of the processing device 110 or in an external storage device, such as storage device 140. As yet another example, the processing device 110 can determine whether the detection feature is within the deviation range of the standard feature. Furthermore, the processing device 110 can use the standard feature to determine the target cutting profile of the fabric to be cut when it is determined that the detection feature is within the deviation range of the standard feature.

[0034] In some embodiments, the processing device 110 may be implemented by a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, the processing device 110 may be local or remote. For example, the processing device 110 may access information and / or data stored on storage device 140 via network 150, or receive information and / or data sent by data source 160. As another example, the processing device 110 may directly connect to storage device 140 to access stored information and / or data. In some embodiments, the processing device 110 may be implemented on a cloud platform. As merely an example, the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, multi-cloud, etc., or any combination of the above examples. In some embodiments, the processing device 110 may be implemented in conjunction with this application. Figure 2 This is implemented on the computing device shown. For example, the processing device 110 can be implemented on a computing device such as... Figure 2 Implemented on a computing device 200, as shown, including one or more components in the computing device 200.

[0035] In some embodiments, the processing device 110 may include one or more processing engines (e.g., a single-core processing engine or a multi-core processor). By way of example only, the processing device 110 may include one or more combinations of a central processing unit (CPU), an application-specific integrated circuit (ASIC), a special-purpose instruction set processor (ASIP), a graphics processing unit (GPU), a physical processing unit (PPU), a digital signal processor (DSP), a field-programmable gate array (FPGA), a programmable logic controller (PLC), a programmable logic device (PLD), a controller, a microcontroller unit (MCU), a reduced instruction set computer (RISC), a microprocessor, etc.

[0036] The execution platform 120 can be used to perform various operations on the fabric to be cut. For example, the execution platform 120 can be equipped with image capture components, such as a camera or image sensor, to take pictures of the fabric to be cut and obtain detection images. As another example, the execution platform 120 may include a three-axis stage (xyz), on which a laser cutter can be mounted to move in three dimensions under the control of a control signal to adjust the focusing position of the emitted laser beam, thereby cutting the fabric. In some embodiments, the execution platform 120 may have a guide belt, such as a conveyor belt, flow strip, or plate chain, for transporting or delivering the fabric to be cut. The three-axis stage is positioned above the guide belt and is equipped with the laser cutter. The three-axis stage can receive a cutting signal with the laser cutter's movement trajectory sent by the processing device 110, and move under the instruction of the signal, so that the laser beam emitted by the laser cutter cuts the fabric placed on the guide belt.

[0037] Terminal 130 can be the operating front end of processing device 110, and may include, but is not limited to, mobile device 130-1, tablet computer 130-2, laptop computer 130-3, desktop computer 130-4, or any combination thereof. The operator can input control commands corresponding to operations that the fabric separation system 100 can perform on terminal 130. For example, commands to call up standard images and the standard features carried by the standard images. In some embodiments, terminal 130 can be integrated with processing device 110. For example, the computing power (e.g., CPU, GPU, etc.) of terminal 130 can be used to implement the functions of processing device 110. The input ports of terminal 130 (e.g., the touch virtual keyboard of mobile device 130-1 and tablet computer 130-2, such as a smartphone, and the mouse and keyboard of laptop computer 130-3 and desktop computer 130-4) can be used for inputting operating commands.

[0038] Storage device 140 can store data and / or instructions. In some embodiments, storage device 140 can store data obtained after processing an inspection image of the fabric to be cut by processing device 110. For example, feature identifiers obtained after identifying texture features of the fabric to be cut in the inspection image. In some embodiments, storage device 140 can store data and / or instructions for execution or use by processing device 110, which can implement the exemplary methods in this application by executing or using the data and / or instructions. In some embodiments, storage device 140 can be part of processing device 110. In some embodiments, storage device 140 can include mass storage, removable storage, volatile read-write storage (RAM), read-only storage (ROM), etc., or any combination thereof. Exemplary mass storage can include disks, optical disks, solid-state drives, etc. Exemplary removable storage can include flash drives, floppy disks, optical disks, memory cards, compact disks, magnetic tapes, etc. Exemplary RAM can include dynamic RAM (DRAM), double-rate synchronous dynamic RAM (DDR SDRAM), static RAM (SRAM), thyristor RAM (T-RAM), and zero-capacitance RAM (Z-RAM), etc. Exemplary ROMs may include mask ROMs (MROMs), programmable ROMs (PROMs), erasable programmable ROMs (PEROMs), electronically erasable programmable ROMs (EEPROMs), optical disc ROMs (CD-ROMs), and digital universal disk ROMs, etc. In some embodiments, storage device 140 may be a distributed storage system. In some embodiments, storage device 140 may be implemented on a cloud platform. By way of example only, the cloud platform may include private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-layer cloud, etc., or any combination thereof. For example, some algorithms or data in this application may be stored on a cloud platform and updated regularly. Processing device 110 accesses these algorithms or data through a network to achieve the unification and interaction of algorithms or data across the entire platform.

[0039] Network 150 can facilitate the exchange of information and / or data. In some embodiments, one or more components of the fabric separation system 100 (e.g., processing device 110, execution platform 120, terminal 130, storage device 140, and data source 160) can transmit information to other components of the fabric separation system 100 via network 150. For example, processing device 110 can acquire a detection image of the fabric to be cut from execution platform 120 via network 150. In some embodiments, network 150 can be any form of wired or wireless network, or any combination thereof. As an example only, network 150 can be a wired network, fiber optic network, telecommunications network, intranet, Internet, local area network (LAN), wide area network (WAN), wireless local area network (WLAN), metropolitan area network (MAN), public switched telephone network (PSTN), Bluetooth network, bee network, near field communication (NFC) network, Global System for Mobile Communications (GSM) network, code division multiple access (CDMA) network, time division multiple access (TDMA) network, general packet radio service (GPRS) network, enhanced data rate GSM evolution (GSM Evolution) network, etc. The network 150 may include one or more combinations of the following: EDGE network, Wideband Code Division Multiple Access (WCDMA) network, High-Speed ​​Downlink Packet Access (HSDPA) network, Long Term Evolution (LTE) network, User Datagram Protocol (UDP) network, Transmission Control Protocol / Internet Protocol (TCP / IP) network, Short Message Service (SMS) network, Wireless Application Protocol (WAP) network, Ultra Wideband (UWB) network, Mobile Communication (1G, 2G, 3G, 4G, 5G) network, Wi-Fi, Li-Fi, Narrowband Internet of Things (NB-IoT), and infrared communication. In some embodiments, network 150 may include one or more network access points. For example, network 150 may include wired or wireless network access points such as base stations and / or Internet switching points. Through these network access points, one or more components of the fabric separation system 100 may connect to network 150 to exchange information and / or data.

[0040] Data source 160 can provide various types of fabric data, such as images and / or videos of various fabrics. This fabric data can be used to generate standard images or to train the models involved in the methods disclosed in this application. For example, this fabric data can be sent to terminal 130 via network 150 for preliminary labeling, and then, after positive and negative sample classification, it can be transmitted via network 150 to processing device 110 for further model training. In some embodiments, data source 160 may be integrated with storage device 140, and data source 160 may be part of storage device 140.

[0041] Figure 2This is a block diagram illustrating an exemplary processing device according to some embodiments of this application. The computing device 200 may include any components used to implement the system described in the embodiments of this application. For example, the computing device 200 may be implemented using hardware, software programs, firmware, or a combination thereof. For convenience, only one processing device is shown in the figure; however, the computing functions related to the fabric separation system 100 described in the embodiments of this application may be implemented in a distributed manner by a set of similar platforms to distribute the system's processing load.

[0042] In some embodiments, computing device 200 may include processor 210, memory 220, input / output 230, and communication port 240. In some embodiments, the processor (e.g., CPU) 210 may execute program instructions as one or more processors. In some embodiments, the memory 220 may include different forms of program memory and data memory, such as hard disk, read-only memory (ROM), random access memory (RAM), etc., for storing various data files processed and / or transmitted by the computer. In some embodiments, the input / output 230 may be used to support input / output between computing device 200 and other components. In some embodiments, the communication port 240 may be connected to a network for data communication. Exemplary processing devices may include program instructions executed by processor 210 stored in read-only memory (ROM), random access memory (RAM), and / or other types of non-transitory storage media. The methods and / or processes of the embodiments of this application may be implemented as program instructions. Computing device 200 may also receive programs and data disclosed in this application via network communication.

[0043] For ease of understanding, Figure 2 Only one processor is illustrated in the illustration. However, it should be noted that the computing device 200 in this embodiment may include multiple processors. Therefore, the operations and / or methods implemented by one processor as described in this embodiment may also be implemented jointly or independently by multiple processors. For example, if, in this application, the processor of computing device 200 executes operations A and B, it should be understood that operations A and B may also be executed jointly or independently by two different processors of computing device 200 (e.g., the first processor executes operation A, the second processor executes operation B, or the first and second processors jointly execute operations A and B).

[0044] Figure 3This is an exemplary flowchart of a fabric separation method according to some embodiments of this application. In some embodiments, process 300 can be executed by processing system 700. For example, process 300 can be stored in a storage device (such as the built-in storage unit of processing system 700 or an external storage device) in the form of a program or instructions, which, when executed, can implement process 300. Figure 3 As shown, process 300 may include the following operations.

[0045] Step 310: Determine the detection features of the fabric to be cut based on the detection image of the fabric to be cut. This step can be performed by the determination module 710.

[0046] In some embodiments, the fabric to be cut can be a textile manufactured by various weaving methods (e.g., knitting or weaving). This textile can be used to form the upper. It is understood that the upper formed by this method can have specific weaving textures, such as raised points, recessed points, holes, etc., arranged in a certain pattern, or repeating weaving patterns. These weaving textures on the upper serve aesthetic purposes and enhance comfort, durability, and breathability. Generally, weaving produces large rolls of fabric, including multiple textile uppers of the same or different sizes or types, which need to be cut from the large roll. Based on this, the fabric to be cut in this application can be this large roll of woven fabric. Alternatively, the large roll of woven fabric may undergo preliminary cutting to form multiple pieces. Each piece includes a textile upper. This piece can also be considered the fabric to be cut.

[0047] In some embodiments, the fabric to be cut can be placed on a cutting component for transport and cutting. For example... Figure 1 The execution platform 120 shown can be used to perform this operation. For example, the fabric to be cut can be placed on the guide belt of the execution platform 120 for transport. Simultaneously, an image capture unit mounted on the execution platform 120 can acquire an image of the fabric to be cut to obtain the detection image. Any component that generates images / videos based on optical or electronic imaging principles, such as a multi-camera system, a high-speed camera, an industrial camera (e.g., based on a CCD / CMOS sensor), a line scan camera, or a 3D camera, can serve as the aforementioned image capture unit. The determination module 710 can communicate with the image capture unit to obtain the detection image, for example, via a network 150.

[0048] After acquiring the detection image of the fabric to be cut, the determination module 710 can process the detection image to obtain the detection features. In conjunction with the foregoing description, the detection features can be the woven texture on the fabric to be cut. Alternatively, the detection features can include texture features. For example, they can be presented as dots, lines, etc., identifying continuous woven textures such as sequentially arranged holes or patterns, and distinguishing different woven texture areas, for example, using curved feature lines as separators. Additionally, the intersections between areas of continuous or repetitive woven textures, or the turning points, midpoints, etc., of these areas can also be used as detection features, such as feature points. In some embodiments, the detection features can at least include feature recognition lines of the texture to be detected included in the fabric to be cut.

[0049] For example, methods such as Gray-Level Co-occurrence Matrix (GLCM), Gray-Level Run-Length Matrix (GLRLM), Gray-Level Size Zone Matrix (GLSZM), and Local Binary Pattern (LBP) based on the statistical characteristics or local structure of the woven texture; frequency domain and multi-scale analysis methods such as Fourier Transform (FT), Gabor Filter, and Wavelet Transform (WT); or mathematical morphology algorithms and texture primitive models based on the basic units and repetition rules of the woven texture; and deep learning methods such as Convolutional Neural Networks (CNN), attention mechanisms and Transformers, multi-task and self-supervised learning models can all be used in this application to determine the detection features of the detected image. The determination module 710 can call these methods and / or models to process the detected image to determine the detection features.

[0050] In some embodiments, a pre-trained feature recognition model can be used to process the detected image. The feature recognition model can be a deep learning model. As an exemplary but not limiting illustration, the feature recognition model can include, but is not limited to, VGG (VGG16 / VGG19), ResNet, DenseNet, EfficientNet, TextCNN, DeepTexture, MCNN, Gabor-CNN, Vision Transformer (ViT), Swin Transformer, TextureTransformer, ConViT, MoCo, SimCLR, MAE, StyleGAN, Diffusion Models, etc., or any combination thereof, for example, a multi-model combination such as CNN+Transformer. After training, the feature recognition model can be stored, for example, in the built-in storage unit of the processing system 700 or in an external storage device, such as storage device 140. The determination module 710 can communicate with the built-in storage unit or external storage device via network 150 to obtain the feature recognition model for processing the detected image.

[0051] Training the feature recognition model can be performed using a large number of training fabric samples. For example, these training fabric samples could be images of woven shoe uppers with various weave textures. Before training, the training fabric samples can be annotated with feature marker lines. These feature marker lines can also be called training ground truth. This can be done by manually drawing the feature point lines of the weave texture on the training fabric samples. (See reference...) Figure 4 The illustrated diagram shows an example of a training fabric sample, where the labeled training ground truth values ​​include the boundaries that distinguish different weave texture regions (which can be referred to as sample texture regions in the context of the training fabric sample). Figure 4 As shown in Figure a), for continuous sample texture regions such as patterns or mesh areas, the region outline is formed based on the boundary line (e.g. Figure 4 As shown in b), and the reference points for intersections, turning points, midpoints, etc., in regular / repetitive sample texture regions or between sample texture regions. After labeling, the training fabric samples can be input into the initial model for training. Based on the difference between the model's output and the training ground truth, such as the difference in position and direction between the separator lines of the sample texture in the model output and the separator lines previously labeled on the training fabric sample, the model parameters (e.g., learning rate, weights, biases, etc.) can be adjusted via backpropagation. Training can be stopped when preset conditions are met (e.g., the number of training rounds reaches a predetermined number, or the accuracy of the recognition result reaches a preset threshold, such as 99.5%). The final model obtained is the feature recognition model.

[0052] The determining module 710 can input the detected image into the feature recognition model to obtain the feature recognition line output by the feature recognition model regarding the weave texture (also referred to as the texture to be detected in this application) in the fabric to be cut, as the detection feature. Corresponding to the label represented on the training fabric sample used to train the feature recognition model, the feature recognition line may also include the boundary line distinguishing different textures to be detected on the fabric to be cut, the outline of the area where the continuous texture to be detected is located, and / or the intersection, turning point, midpoint, etc. related to the regular / repetitive textures to be detected.

[0053] Step 320: Obtain the standard features indicated by the standard image corresponding to the fabric to be cut. This step can be performed by the acquisition module 720.

[0054] In some embodiments, the standard image may be an image of a standard fabric with the same weave texture as the fabric to be cut. This same weave texture can be understood as a weave pattern, including the size, position, shape, and type of the weave texture area. The standard fabric may be a shoe upper fabric that meets the generation acceptance criteria. The standard image may be obtained in advance by photographing the standard fabric using an image capture device that is the same as or similar to the one described above. The standard features may include standard feature lines for the weave texture (which may also be referred to as standard texture in this application) on the standard fabric. Similarly, these standard feature lines may also include texture dividing lines, texture area outlines, texture area intersections, turning points, midpoints, etc. The determination of these points and lines can be done manually in advance. For example, similar to the feature marker lines in the aforementioned training fabric samples, these are the annotations for training ground values. In some embodiments, the standard features may also include a preset cutting contour. Cutting the standard fabric according to the preset cutting contour yields a shoe upper fabric that generally meets production / use standards for subsequent shoe upper manufacturing. The preset cutting contour may be determined based on empirical values. It is understandable that the shoe upper manufacturing process involves a large amount of shoe upper fabric. These fabrics encounter various situations during the process from weaving to cutting to assembly into shoes. For example, uneven placement during cutting can damage the fabric, and fabric deformation can cause the product to deviate from standards when cut according to the original contour. These experiences can be used to define the preset cutting contour. For instance, based on the weaving texture of different shoe upper fabrics, and considering various situations including shoe upper deformation, the relative position (including distance and direction) between the cutting line and the standard feature lines of these texture areas can be obtained through statistical analysis to arrive at the preset cutting contour. Similarly, the standard image can be pre-stored in a built-in storage unit such as the processing system 700 or an external storage device such as storage device 140. The acquisition module 720 can communicate with the built-in storage unit or external storage device via network 150 to obtain the standard image.

[0055] In some embodiments, the number of feature reference lines included in the standard feature may be different from the number of feature marker lines in the training fabric sample. For example, it may be fewer than the number of feature marker lines. The feature reference lines may be feature point lines associated with key texture regions. These key texture regions may be feature point lines associated with texture regions located in the center, symmetrically located on both sides, or near the edge. As another example, the feature reference lines of the standard feature may be feature point lines of a woven texture used to determine the preset cutting contour. For example, feature point lines of a woven texture located on the periphery of a standard image.

[0056] In some embodiments, the number of standard images can be one or more. When multiple standard images exist, they can correspond to different weave textures, such as different weave materials, different sizes, different patterns, etc., and thus different fabrics to be cut. In this case, to determine the standard image corresponding to the fabric to be cut currently being processed, the acquisition module 720 can use a matching algorithm / model to determine the required standard image from the multiple standard images. Existing template matching algorithms can be applied to this process. Another possible implementation is that the acquisition module 720 can receive an externally input signal, such as an image selection signal input by an operator from the terminal 130, and transmit it to the acquisition module 720 via the network 150. This image selection signal can directly indicate which standard image is being used. Based on this, the acquisition module 720 can directly obtain the required image from the multiple standard images.

[0057] Step 330: Determine whether the detected feature is within the deviation range of the standard feature. This step can be performed by the comparison module 730.

[0058] Based on the foregoing explanation, due to the consistency of the weave texture between the fabric to be cut and the standard fabric, it can be known that the feature identification lines obtained from the detection image are theoretically consistent with the feature standard lines of the standard image. For example, the feature identification lines and the feature standard lines coincide. Alternatively, the linear features included in the feature identification lines and the linear features included in the feature standard lines have consistent routing, with the distance within the allowable error range. In other words, if, in practice, the feature identification lines and the feature standard lines meet the above conditions, then cutting the fabric to be cut based on the preset cutting contour can yield a shoe upper fabric that meets production standards. Even if the fabric to be cut is deformed, this is because the preset cutting contour has already taken into account the corresponding situation.

[0059] Therefore, the comparison module 730 can form a deviation region containing the feature standard line based on the feature standard line, and determine whether the detected feature is within the deviation range of the standard feature based on whether the feature recognition line is within the deviation region. The deviation region can be the same as or similar to the aforementioned allowable error range. To determine the deviation region, the comparison module 730 can obtain the allowable deviation distance and expand the feature standard line outward based on the allowable deviation distance to form the deviation region. The allowable deviation distance can be determined based on empirical values. The allowable deviation distance can be different for feature point lines corresponding to different weave textures. For example, for a weave texture in the middle position, the allowable deviation distance corresponding to its boundary line or area contour line can be larger, such as ±3mm. For a weave texture in the edge position, such as a weave texture near the cutting line, the allowable deviation distance corresponding to its boundary line or area contour line can be smaller, such as ±0.5mm. After obtaining the allowable deviation distance, the feature standard line can be expanded outward. The expansion distance can be the allowable deviation distance. For example, if the allowable deviation distance is ±3mm, then for each point on the characteristic standard line, an area can be expanded outwards by 3mm to form an expanded region. When all the expanded regions overlap, the final deviation region including the characteristic standard line is formed.

[0060] In some embodiments, to ensure accurate correspondence between the feature identification lines and the feature standard lines—that is, to ensure that the feature point lines corresponding to the woven areas at the same position correspond to each other—the comparison module 730 can perform alignment based on point-line alignment. For example, during the training process of the aforementioned feature recognition model, the labels on the training fabric samples may be marked with alignment points and / or lines. For instance, the training fabric sample may be an image of existing shoe upper fabric, which is generally symmetrical. Therefore, the equidistant points or midpoints on the axis of symmetry can be marked as the alignment points, and part or all of the axis of symmetry can be marked as the alignment lines. Thus, after the feature recognition model is trained, processing the detection image of the fabric to be cut will also output corresponding alignment points and / or alignment lines. Similarly, the standard image may also be marked with positioning points and / or positioning lines to constitute the standard features, or it may be the equidistant points on the axis of symmetry of the woven texture in the standard image. Based on this, the comparison module 730 can first align the alignment point and the positioning point, and / or the alignment line and the positioning line, thereby determining one or more feature marker lines corresponding to the one or more feature standard lines in terms of position. After alignment, the comparison module 730 can determine whether all one or more feature marker lines are within the deviation area of ​​the corresponding one or more feature standard lines. If so, the comparison module 730 can determine that the detected feature is within the deviation range of the standard feature. If one or more feature marker lines are not within the deviation area of ​​the corresponding feature standard line, the comparison module 730 can determine that the detected feature is outside the deviation range of the standard feature. When determining whether a feature marker line is within the deviation area of ​​the feature standard line, the comparison module 730 only compares whether the feature marker lines corresponding to the feature standard line meet the requirements, while other feature marker lines without corresponding feature standard lines do not need to be considered. This can save judgment time and improve overall efficiency to a certain extent.

[0061] After determining the feature identification line corresponding to the corresponding feature standard line, if the detected feature is within the deviation range of the standard feature, process 300 can proceed to step 340.

[0062] Step 340: Determine the target cutting profile of the fabric to be cut using the standard features and / or the detection features. This step can be performed by the execution module 740.

[0063] It is understood that the detection feature being within the deviation range of the standard feature can include the feature identification line completely coinciding with the feature standard line, and the feature identification line being within the outer range of the feature standard line. Complete coincidence indicates that the detection image of the fabric to be cut is completely consistent with the standard image of the standard fabric. In this case, the preset cutting contour marked on the standard image can be directly used to cut the fabric to be cut. When the feature identification line is within the outer range of the feature standard line, it can be considered that the fabric to be cut has been deformed due to placement or other reasons, causing the feature points of the weave texture to shift. In conjunction with the foregoing description, even if the fabric to be cut is deformed, the deformation is still within the allowable range. Cutting the fabric to be cut using the preset cutting contour can still yield a product that meets the standard. At this time, the execution module 740 can overlay the preset cutting contour onto the detection image to generate a target cutting contour for cutting the fabric to be cut. For example, the set of pixel coordinates of the target cutting contour on the detection image is consistent with the set of pixel coordinates of the preset cutting contour. First, a coordinate transformation from the image coordinate system to the camera coordinate system can be performed, for example, by using the inverse operation of the camera's intrinsic parameter matrix through projection transformation. The laser cutter needs to move in the world coordinate system to cut the fabric to be cut. This can be achieved by transforming the coordinates from the camera coordinate system to the world coordinate system, for example, through the camera's extrinsic parameter matrix. This ultimately yields the motion trajectory for controlling the laser cutter, for example, the laser cutter mounted on the execution platform 120. Based on this motion trajectory, the laser cutter's motion is controlled, and its emitted laser beam is controlled to cut the fabric to be cut, ultimately obtaining a standard-compliant shoe upper fabric.

[0064] In some embodiments, the final cutting contour can also be determined based on the detection features. The execution module 740 can acquire the positional relationship indicating the relative position between the feature marker line and the target cutting contour, and determine the target cutting contour based on the positional relationship using the feature marker line. This positional relationship can be the positional relationship used above to describe the relative position between the preset cutting contour and the feature standard line. Specific generation steps can also refer to the aforementioned related content. Generating the target cutting contour based on the feature marker line can accurately outline the portion of the fabric to be cut, whether the fabric to be cut is undeformed or the degree of deformation is within an acceptable range (i.e., within the error range). Compared to directly using the preset cutting contour, the target cutting contour generated based on the feature marker line can yield a relatively superior upper fabric when the fabric to be cut is deformed.

[0065] The fabric separation method disclosed in this application compares the feature point lines of the fabric to be cut with those of a standard fabric to determine whether they are within the error range. Based on this comparison, the cutting contour is determined. This fabric separation method is applicable to fabric deformation and does not require deformation correction or deformation of the outer frame contour to obtain the final cutting contour.

[0066] In some embodiments, process 300 may further include: updating the feature recognition model using the feature recognition lines. The feature recognition lines are detection features for the detection image obtained based on the feature recognition model. After the feature recognition lines are output, they can be verified, for example, manually. If verified as correct, the detection image can be used as a training fabric sample, and the corresponding feature recognition lines can be used as feature marker lines to further train the feature recognition model. In some implementations, the detection images can be accumulated, for example, to 1000 images, before uniformly retraining the feature recognition model to achieve the update purpose. When the feature recognition lines are confirmed to be incorrect, i.e., in the case of recognition error, the detection image and the corresponding feature recognition lines can also be used as training fabric samples to update the feature recognition model. For example, the detection image can be used as a negative sample, and the feature recognition lines as negative markers. Correspondingly, the detection image corresponding to the feature recognition lines that are confirmed as correct is a positive sample. In this way, the recognition accuracy and efficiency of the feature recognition model can be continuously improved.

[0067] In some embodiments, if it is determined in step 330 that the detected feature is outside the deviation range of the standard feature, process 300 can be terminated directly. This indicates that the deformation of the fabric to be cut is large and exceeds the error range. In this case, no further processing is necessary. This can improve cutting efficiency and reduce waste.

[0068] It should be noted that the above-mentioned Figure 3 The descriptions of the various steps in this application are merely for illustrative purposes and do not limit the scope of this application. Those skilled in the art can learn from the guidance of this application. Figure 3 Various modifications and changes have been made to the various steps in the process. However, these modifications and changes are still within the scope of this application.

[0069] Figure 5 A flowchart illustrating the fabric separation method provided in this application is shown as an example. Figure 5 As shown, Figure 5 (a) is a detection image of the fabric to be cut, which includes multiple woven texture areas. Figure 5Image (b) shows the use of a feature identification model to process the detected image for texture recognition, resulting in multiple dividing lines and region contours used to identify textured areas. This feature recognition model can be trained using a large number of images of woven shoe upper fabrics, on which the boundaries and contours of the woven textures are marked. Figure 5 (c) represents the standard features of the standard image, which are multiple feature points and lines. Figure 5 In the middle (d), the lines of these feature points are expanded, which means the lines are thickened based on the allowable error range. Figure 5 In section (e), error comparison is performed. First, feature alignment is conducted using the "dry" type alignment points / alignment lines in (b) and the "dry" type positioning points / positioning lines in (d). Then, it is compared whether the feature lines in (b) are within the range of the thickened feature lines in (d). If so, further steps can be taken. Figure 5 Laser cutting in (f). The final cutting profile is generated according to the preset cutting profile shown in (c) and the fabric to be cut is then cut.

[0070] Figure 6 An exemplary schematic diagram of feature comparison is shown. For example... Figure 6 As shown in (a), it is a standard image marked with standard features, including curve BF1 on the left, loop line BF2 on the right, and positioning point BP and positioning line BL. Figure 6 Figure (b) shows the expansion of the feature line. For example, the curve BF1 on the left is expanded by ±1 mm to obtain the expanded line BR1, and the loop line BF2 on the right is expanded by ±3 mm to obtain the expanded loop line BR2. Figure 6 Image (c) shows the detected image of the fabric to be cut, marked with detection features, including curve TF1 and a loop curve on the left, loop curve TF2 on the right, two feature points, alignment point FP, and positioning line TL. During feature comparison, first, positioning point BP and alignment point FP are aligned, and positioning line BL and alignment line FL are aligned. Then, the aligned feature lines / points are compared. For example, it is determined whether curve TF1 is within the range of the outer expansion line BR1, and whether loop curve TF2 is within the range of the outer expansion loop line BR2. The loop curve on the left and the two feature points on the right, as marked in the detection image, are not compared.

[0071] Figure 7 This is an exemplary block diagram of a processing system implementing the above-described fabric separation method according to some embodiments of this application. This processing system compares the texture feature lines of the fabric to be cut with the texture feature lines of a standard fabric to determine whether they are within the error range and, based on this, determines the final cutting contour of the fabric to be cut. Figure 7As shown, the processing system 700 may include a determining module 710, an acquiring module 720, a comparing module 730, and an executing module 740.

[0072] The determination module 710 can be configured to determine detection features of the fabric to be cut based on a detection image of the fabric to be cut. The determination module 710 can acquire a detection image of the fabric to be cut by communicating with an image capture unit that captures an image of the fabric. In some embodiments, the determination module 710 can process the detection image using a pre-trained feature recognition model to determine the detection features. The feature recognition model is determined after training with a large number of training fabric samples. Each training fabric sample may be marked with feature marker lines representing the sample texture included in the training fabric sample. By processing the detection image using the feature marker model, the determination module 710 can obtain the feature recognition lines representing the detectable texture included in the fabric to be cut.

[0073] The acquisition module 720 can be configured to acquire standard features indicated by a standard image corresponding to the fabric to be cut. The standard image can be an image of a standard fabric with the same weave texture as the fabric to be cut. The standard fabric can be a shoe upper fabric that meets the generation acceptance criteria. The standard image can be obtained in advance by photographing the standard fabric using an image capture device that is the same as or similar to the one described above. The standard features can include standard feature lines for the weave texture on the standard fabric.

[0074] The comparison module 730 can be configured to determine whether the detected feature is within the deviation range of a standard feature. The comparison module 730 can form a deviation region encompassing the feature standard line based on the feature standard line, and determine whether the detected feature is within the deviation region based on whether the feature identification line is within the deviation region. To form the deviation region, the comparison module 730 can obtain an allowable deviation distance and expand the feature standard line outward based on the allowable deviation distance to form the deviation region. The allowable deviation distance can be determined based on empirical values, and different allowable deviation distances can exist for different feature standard lines. During comparison, the comparison module 730 can align the alignment points and / or alignment lines output for the detected image with the positioning points and / or positioning lines identified on the standard image, thereby determining one or more feature identification lines corresponding to the one or more feature standard lines in position. After alignment, the comparison module 730 can determine whether the one or more feature identification lines are all within the deviation region of the corresponding one or more feature standard lines. If so, the comparison module 730 can determine that the detected feature is within the deviation range of the standard feature. If one or more feature marker lines are not within the deviation range of the corresponding feature standard line, the comparison module 730 can determine that the detected feature is outside the deviation range of the standard feature.

[0075] The execution module 740 can be configured to determine a target cutting profile for the fabric to be cut using the standard feature and / or the detection feature after determining that the detected feature is within the deviation range of the standard feature. The execution module 740 can overlay the preset cutting profile onto the detection image to generate a target cutting profile for cutting the fabric. For example, the set of pixel coordinates of the target cutting profile on the detection image is consistent with the set of pixel coordinates of the preset cutting profile. After transformation from image coordinates to camera coordinates and then to world coordinates, a motion trajectory for controlling a laser cutter, such as one mounted on the execution platform 120, is obtained.

[0076] In some embodiments, the processing system 700 may further include a training module (not shown in the figure). This training module can be used for training and updating the feature recognition model described above. For example, the training module can utilize the detected image and the feature recognition lines output for the detected image to further train the feature recognition model for updating purposes.

[0077] Further descriptions of the aforementioned components can be found in this application. Figures 1-6 part.

[0078] It should be understood that Figure 7The systems and modules shown can be implemented in various ways. For example, in some embodiments, the systems and modules can be implemented by hardware, software, or a combination of both. The hardware portion can be implemented using dedicated logic; the software portion can be stored in memory and executed by an appropriate instruction execution system, such as a microprocessor or dedicated-design hardware. Those skilled in the art will understand that the methods and systems described above can be implemented using computer-executable instructions and / or included in processor control code, for example, on a carrier medium such as a disk, CD, or DVD-ROM, a programmable memory such as read-only memory (firmware), or a data carrier such as an optical or electronic signal carrier. The systems and modules of this application can be implemented not only by hardware circuits such as very large-scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field-programmable gate arrays, programmable logic devices, etc., but also by software executed by various types of processors, or by a combination of the aforementioned hardware circuits and software (e.g., firmware).

[0079] It should be noted that the above description of the modules is for ease of description only and should not be construed as limiting this application to the scope of the embodiments described. It is understood that those skilled in the art, after understanding the principle of the system, may arbitrarily combine the modules or construct subsystems connected to other modules without departing from this principle. For example, the modules may share a single storage module, or each module may have its own separate storage module. Such modifications are all within the scope of protection of this application.

[0080] Figure 8 These are exemplary structural diagrams of a fabric separation system according to some embodiments of this application. Figure 8 As shown, the fabric separation system 800 may include a conveying component 810, an image capture component 820, a processing component 830, and a laser cutting component 840. The various components in the fabric separation system 800 can be interconnected in various ways. For example, they can be connected via wired connections or wireless communication.

[0081] The conveying assembly 810 can be used to carry and convey the fabric to be cut. For example, the conveying assembly 810 can be a conveyor belt, a flow strip, a chain conveyor, or other structure used for transporting items. The fabric to be cut can be placed on it for conveying. In some implementations, the conveying assembly 810 can also have an adsorption structure. For example, a vacuum adsorption generator can be used to draw air into the space between the fabric to be cut and the bearing surface of the conveying assembly 810 to create negative pressure, so that the fabric to be cut adheres to the bearing surface and does not shift.

[0082] The image capture component 820 can be used to acquire a detection image of the fabric to be cut. For example, the image capture component 820 can be mounted above the conveying component 810 to capture images of the fabric to be cut placed thereon. For example, the image capture component 820 can be any imaging device that generates images / videos based on optical or electronic imaging principles, such as a multi-camera system, a high-speed camera, an industrial camera (e.g., based on a CCD / CMOS sensor), a line scan camera, or a 3D camera. The image capture component 820 can have a built-in supplementary lighting element to provide illumination during image capture of the fabric to be cut, ensuring clear imaging. Alternatively, the environment in which the fabric separation system 100 is located can provide a light source to prevent the environment from becoming dark and to guarantee image quality.

[0083] The processing component 830 can perform the above-described fabric separation method, including but not limited to determining the detection features of the fabric to be cut based on the detection image of the fabric to be cut; the detection features include at least the feature recognition lines of the texture to be detected included in the fabric to be cut; acquiring the standard features indicated by the standard image corresponding to the fabric to be cut; the standard features include at least the feature standard lines of the standard texture included in the standard image; determining whether the feature recognition lines are within the deviation range of the feature standard lines; if so, determining the target cutting contour of the fabric to be cut using the standard features and / or the standard features.

[0084] In some embodiments, the processing component 830 can be any device with computing power, including but not limited to industrial computers, computers, main control chips (e.g., ARM, DSC, DSP, etc.), programmable logic controllers (PLCs), programmable logic devices (PLDs), microcontrollers (MCUs), etc. For example, the processing component 830 can be described in this application. Figure 1 Implemented on the processing device 110 described herein, or in this application Figure 2 Implemented on the computing device shown.

[0085] The cutting assembly 840 can be used to cut the fabric to be cut based on the target cutting profile. The cutting assembly 840 may include a laser cutter, and the target cutting profile may carry a motion trajectory indicating the movement of the laser cutter. The laser cutter can move based on the motion trajectory, while emitting a laser beam to cut the fabric to be cut, ultimately obtaining a shoe upper fabric that meets acceptance standards.

[0086] Data / information transmission between the various components of the fabric separation system 800 can be achieved through various information transmission types, including but not limited to fieldbus (e.g., PROFIBUS, MODBUS, DeviceNet, CANopen, etc.), Ethernet (e.g., EtherNet / IP, PROFINET, EtherCAT, Modbus TCP, etc.), wireless communication (e.g., Wi-Fi™, Bluetooth™, ZigBee™, LoRa™, etc.), serial communication (e.g., RS-232, RS-485, etc.), parallel communication, fiber optic communication, OPC (OLE for Process Control), Internet of Things (IIoT) protocols (e.g., MQTT, AMQP, CoAP, etc.), Time-Sensitive Networking (TSN), 5G networks, etc. Other methods capable of information transmission can also be applied to this application.

[0087] Furthermore, the various components of the fabric separation system 800 can be combined. For example, the conveying assembly 810, the image capture assembly 820, and the laser cutting assembly 840 can be integrated into a laser cutting platform with a camera. Other adjustments and variations are also within the scope of this application.

[0088] This application has described the basic concepts. Obviously, for those skilled in the art, the above detailed disclosure is merely illustrative and does not constitute a limitation of this application. Although not explicitly stated herein, those skilled in the art may make various modifications, improvements, and corrections to this application. Such modifications, improvements, and corrections are suggested in this application, and therefore such modifications, improvements, and corrections still fall within the spirit and scope of the exemplary embodiments of this application.

[0089] Furthermore, this application uses specific terms to describe its embodiments. For example, "an embodiment," "one embodiment," and / or "some embodiments" refer to a particular feature, structure, or characteristic related to at least one embodiment of this application. Therefore, it should be emphasized and noted that "an embodiment," "one embodiment," or "an alternative embodiment" mentioned twice or more in different locations in this application do not necessarily refer to the same embodiment. In addition, certain features, structures, or characteristics in one or more embodiments of this application can be appropriately combined.

[0090] Furthermore, those skilled in the art will understand that aspects of this application can be described and illustrated through several patentable types or situations, including any new and useful combination of processes, machines, products, or substances, or any new and useful improvements thereof. Accordingly, aspects of this application can be implemented entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. All of the above hardware or software may be referred to as a “data block,” “module,” “engine,” “unit,” “component,” or “system.” Furthermore, aspects of this application may manifest as a computer product located on one or more computer-readable media, the product including computer-readable program code.

[0091] Computer storage media may contain a propagated data signal containing computer program code, for example, on baseband or as part of a carrier wave. This propagated signal may take various forms, including electromagnetic, optical, or suitable combinations thereof. Computer storage media can be any computer-readable medium other than a computer-readable storage medium, which can be used to communicate, propagate, or transmit a program for use by being connected to an instruction control system, apparatus, or device. The program code located on the computer storage medium can be propagated through any suitable medium, including radio, cable, fiber optic cable, RF, or similar media, or any combination of the above media.

[0092] The computer program code required for the operation of each part of this application can be written in any one or more programming languages, including object-oriented programming languages ​​such as Java, Scala, Smalltalk, Eiffel, JADE, Emerald, C++, C#, VB.NET, Python, etc., conventional procedural programming languages ​​such as C, Visual Basic, Fortran 2003, Perl, COBOL 2002, PHP, ABAP, dynamic programming languages ​​such as Python, Ruby, and Groovy, or other programming languages. This program code can run entirely on the user's computer, or as a standalone software package on the user's computer, or partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the latter case, the remote computer can be connected to the user's computer through any network, such as a local area network (LAN) or wide area network (WAN), or connected to an external computer (e.g., via the Internet), or in a cloud computing environment, or used as a service such as Software as a Service (SaaS).

[0093] Furthermore, unless expressly stated in the claims, the order of processing elements and sequences, the use of numbers and letters, or other names described in this application are not intended to limit the order of the processes and methods of this application. Although the foregoing disclosure has discussed some currently considered useful embodiments of the invention through various examples, it should be understood that such details are for illustrative purposes only, and the appended claims are not limited to the disclosed embodiments; rather, the claims are intended to cover all modifications and equivalent combinations that conform to the substance and scope of the embodiments of this application. For example, while the system components described above can be implemented using hardware devices, they can also be implemented solely through software solutions, such as installing the described system on existing servers or mobile devices.

[0094] Similarly, it should be noted that, in order to simplify the description of the present application and thus aid in the understanding of one or more embodiments of the invention, the foregoing description of the embodiments of the present application sometimes combines multiple features into a single embodiment, drawing, or description thereof. However, this disclosure method does not imply that the subject matter of the application requires more features than those mentioned in the claims. In fact, the embodiments contain fewer features than all the features of the single embodiments disclosed above.

[0095] Finally, it should be understood that the embodiments described in this application are merely illustrative of the principles of the embodiments of this application. Other modifications may also fall within the scope of this application. Therefore, alternative configurations of the embodiments of this application are considered as examples and not limitations, and are regarded as consistent with the teachings of this application. Accordingly, the embodiments of this application are not limited to the embodiments explicitly described and illustrated in this application.

Claims

1. A fabric separation method, characterized in that, The method includes: The detection features of the fabric to be separated are determined based on the detection image of the fabric to be separated; the detection features include at least the feature recognition lines of the texture to be detected included in the fabric to be separated; Obtain the standard features indicated by the standard image corresponding to the fabric to be separated; the standard features include at least the feature standard lines of the standard texture included in the standard image; Determine whether the feature recognition line is within the deviation range of the feature standard line; If so, the target separation profile of the fabric to be separated is determined using the standard features and / or the detection features.

2. The method according to claim 1, characterized in that, Determining the detection features includes: The detection image is processed using a pre-trained feature recognition model to determine the detection features; wherein the training fabric samples of the feature recognition model are identified with feature marker lines of the sample texture included in the training fabric samples.

3. The method according to claim 2, characterized in that, Determining whether the detected feature is within the deviation range of the standard feature includes: Based on the characteristic standard line, a deviation region containing the characteristic standard line is formed; Based on whether the feature recognition line is within the deviation region, it is determined whether the detected feature is within the deviation range of the standard feature.

4. The method according to claim 3, characterized in that, The formation of the deviation region containing the characteristic standard line includes: Obtain the allowable deviation distance; Based on the allowable deviation distance, the deviation region is formed by expanding the characteristic standard line outward.

5. The method according to claim 3, characterized in that, The feature standard lines include one or more; the training fabric samples are further marked with marker points and / or marker lines for position alignment; the standard features also include positioning points and / or positioning lines; the feature recognition model further outputs alignment points and / or alignment lines trained based on the marker points and / or the marker lines; determining whether the feature recognition line is within the deviation region includes: Align the alignment point and the positioning point and / or the alignment line and the positioning line to determine one or more feature recognition lines corresponding to the one or more feature standard lines in terms of position; Determine whether the one or more feature recognition lines are all within the deviation region of the corresponding one or more feature standard lines; If so, determine that the detected feature is within the deviation range of the standard feature. If the detected feature is outside the deviation range of the standard feature, the method is terminated.

6. The method according to claim 3, characterized in that, The standard features also include a preset separation profile; determining the target separation profile of the fabric to be separated using the standard features includes: The preset separation contour is overlaid onto the detection image of the fabric to be separated to generate the target separation contour.

7. The method according to claim 6, characterized in that, Determining the target separation contour of the fabric to be separated using the detection features includes: Obtain the positional relationship between the feature identification line and the target separation contour, indicating their relative positions; Based on the positional relationship, the target separation contour is determined using the feature identification lines.

8. The method according to claim 2, characterized in that, The method further includes: The feature recognition model is updated using the feature recognition line.

9. A fabric separating device, characterized in that, The device includes: The determination module is configured to determine the detection features of the fabric to be separated based on the detection image of the fabric to be separated; The acquisition module is configured to acquire standard features indicated by a standard image corresponding to the fabric to be separated; The comparison module is configured to determine whether the detected feature is within the deviation range of the standard feature; The execution module is configured to determine the target separation profile of the fabric to be separated using the standard feature and / or the detection feature when it is determined that the detection feature is within the deviation range of the standard feature.

10. A fabric separation system, characterized in that, The system includes: A transport component used to carry and transport the fabric to be separated; An image capture component is used to acquire a detection image of the fabric to be separated; Processing components, used for: The detection features of the fabric to be separated are determined based on the detection image of the fabric to be separated; the detection features include at least the feature recognition lines of the texture to be detected included in the fabric to be separated; Obtain the standard features indicated by the standard image corresponding to the fabric to be separated; the standard features include at least the feature standard lines of the standard texture included in the standard image; Determine whether the feature recognition line is within the deviation range of the feature standard line; If so, the target separation profile of the separating fabric is determined using the standard features and / or the detection features; and A separation component for separating the fabric to be separated based on the target separation profile.

Images