A data processing method, device, medium and product

Abstract

The application discloses a data processing method and device, medium and product. It relates to the sewing machine technical field. The method comprises the following steps: acquiring an operation image and / or Internet of Things data, wherein the operation image is collected by a camera, and the camera is arranged at a position opposite to a presser foot and an operation area; determining a general sewing time (GST) action sequence and a time measurement unit (TMU) value of each GST action according to the operation image and / or the Internet of Things data, wherein the GST actions in the GST action sequence belong to multiple processes; determining multiple processes corresponding to the GST action sequence according to the GST action sequence and a preset process template library; and determining standard working hours of each process according to the TMU value of the GST action in each process. Through the technical scheme, automatic calibration of accurate working hours of each process can be realized.

Description

Technical Field

[0001] The present invention relates to the field of sewing machine technology, and in particular to a data processing method, device, medium and product. Background Technology

[0002] In the process of upgrading the intelligent manufacturing industry of apparel, General Sewing Time (GST) serves as the core foundation of production management, directly determining the efficiency of production scheduling, the fairness of work price calculation, and the accuracy of capacity assessment.

[0003] Currently, most GST (Good Supply Practice) processes in the industry rely on manual analysis. The specific process involves: after capturing on-site video, skilled GST analysts analyze the video frame by frame to break down the GST actions, and then perform time analysis based on these breakdowns. This manual analysis method takes over 30 minutes per process, and labor costs are rising annually, failing to meet the flexible production needs of the garment industry for small-batch, multi-batch production. Summary of the Invention

[0004] This invention provides a data processing method, device, medium, and product to achieve automatic calibration of precise working hours for each process.

[0005] According to one aspect of the present invention, a data processing method is provided, comprising: Acquire work images and / or IoT data, wherein the work images are acquired by a camera, which is positioned directly facing the presser foot and the operating area; Based on the work images and / or the IoT data, determine the general sewing time GST action sequence and the time measurement unit (TMU) value of each GST action, wherein the GST actions in the GST action sequence belong to multiple processes; Based on the GST action sequence and the preset process template library, determine multiple processes corresponding to the GST action sequence; The standard working hours for each process are determined based on the TMU value of the GST action in each process.

[0006] According to another aspect of the present invention, an electronic device is provided, the electronic device comprising: At least one processor; and a memory communicatively connected to said at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the data processing method according to any embodiment of the present invention.

[0007] According to another aspect of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing computer instructions for causing a processor to execute and implement the data processing method described in any embodiment of the present invention.

[0008] According to another aspect of the present invention, a computer program product is provided, which, when executed by a processor, implements the data processing method as described in any of the embodiments of the present invention.

[0009] This invention acquires work images and / or IoT data, wherein the work images are captured by a camera positioned directly opposite the presser foot and the operating area; based on the work images and / or the IoT data, a general sewing time (GST) sequence and the Time Measurement Unit (TMU) value for each GST action are determined, wherein the GST actions in the GST sequence belong to multiple processes; based on the GST action sequence and a preset process template library, multiple processes corresponding to the GST action sequence are determined; based on the TMU value of the GST actions in each process, the standard working time of each process is determined; an integrated camera design is adopted in the sewing machine head to achieve natural millisecond-level synchronization between visual data and the device's native IoT data; by automatically recognizing GST actions, a continuous GST action sequence with precise timestamps is generated; based on the GST action sequence and the preset process template library, complex multi-process scenarios can be accurately identified, ultimately achieving automatic calibration of the precise working time of each process.

[0010] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of the present invention, nor is it intended to limit the scope of the invention. Other features of the invention will become readily apparent from the following description. Attached Figure Description

[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present invention and should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 This is a flowchart of a data processing method according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0013] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0014] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0015] It is understood that before using the technical solutions disclosed in the various embodiments of this disclosure, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in this disclosure in an appropriate manner in accordance with relevant laws and regulations, and user authorization should be obtained.

[0016] Example 1 Figure 1 This is a flowchart illustrating a data processing method provided in an embodiment of the present invention. This embodiment is applicable to data processing situations. The method can be executed by the data processing device in this embodiment of the present invention, which can be implemented in software and / or hardware, such as... Figure 1 As shown, the method specifically includes the following steps: S110, acquires images of the work and / or IoT data.

[0017] The work images are captured by a camera, which is positioned directly opposite the presser foot and the operating area.

[0018] In this embodiment, the camera can be a high-definition camera, directly integrated into the front of the sewing machine head, facing the presser foot and operating area without any obstruction of view. The camera acquires 1080P / 60fps video data, and shares a power supply and CAN bus interface with the sewing machine.

[0019] In this embodiment, the IoT data can be the native data directly read from the sewing machine's main controller. The acquisition frequency can be 100Hz. The IoT data may include at least one of the following: thread-cutting trigger timestamp (accurate to 1ms), cumulative stitch count within any time period, real-time stitch distance (unit: 0.1mm), motor start / stop timestamp and speed, presser foot lifting / lowering timestamp, and backstitch start / end timestamp.

[0020] It should be noted that the camera and the sewing machine main controller share the same 32-bit system clock, requiring no external Network Time Protocol (NTP) synchronization, and the time synchronization error is <1ms.

[0021] S120, based on the work image and / or the IoT data, determine the general sewing time GST action sequence and the time-measurement unit (TMU) value of each GST action.

[0022] The GST actions in the GST action sequence belong to multiple processes.

[0023] In this embodiment, the method for determining the General Sewing Time (GST) action sequence and the Time Measurement Unit (TMU) value of each GST action based on the work image can be as follows: If the work image is identified as a piece-picking action, then the piece-picking action is taken as a GST action, and the TMU value corresponding to the piece-picking type of the piece-picking action is taken as the TMU value of the piece-picking action; if the work image is identified as an alignment adjustment action, then the alignment adjustment action is taken as a GST action, and the TMU value corresponding to the alignment adjustment type of the alignment adjustment action is taken as the TMU value of the alignment adjustment action; if the work image is identified as a forming action, then the forming action is taken as a GST action, and the TMU value corresponding to the forming type of the forming action is taken as the TMU value of the forming action.

[0024] In this embodiment, the method for determining the General Sewing Time (GST) action sequence and the Time Measurement Unit (TMU) value of each GST action based on the work image and the IoT data can be as follows: The IoT data includes: device events and description information of the device events. If the device event is a thread-cutting triggered interruption event, the description information of the device event includes: a thread-cutting timestamp. Based on the thread-cutting timestamp, a first time range is determined; the work image within the first time range is identified to obtain the hand movement trajectory; the target action type is determined based on the hand movement trajectory; the thread-cutting action is taken as the GST action, and the TMU value corresponding to the target action type is taken as the TMU value of the thread-cutting action. In this embodiment, the method for determining the General Sewing Time (GST) action sequence and the Time Measurement Unit (TMU) value of each GST action based on the work image and the IoT data can also be as follows: The IoT data includes: equipment events and description information of the equipment events. If the equipment event is a presser foot lift event, the description information of the equipment event includes: a presser foot lift timestamp; if the equipment event is a presser foot descent event, the description information of the equipment event includes: a presser foot descent timestamp; based on the presser foot lift timestamp, a second time range is determined, and the work image within the second time range is identified to obtain the presser foot state; if the presser foot state is presser foot lift, then the presser foot lift action is taken as a GST action, and the TMU value of the presser foot lift action is obtained. Based on the presser foot descent timestamp, a third time range is determined, and the work image within the third time range is identified to obtain the presser foot state; if the presser foot state is presser foot descent, then the presser foot descent action is taken as a GST action, and the TMU value of the presser foot descent action is obtained. The method for determining the General Sewing Time (GST) action sequence and the Time Measurement Unit (TMU) value of each GST action based on the work image and the IoT data can also be as follows: The IoT data includes: equipment events and description information of the equipment events, the description information of the equipment events includes: sewing time period, cumulative stitch count within the sewing time period, and real-time stitch distance; determining the General Sewing Time (GST) action sequence and the TMU value of each GST action based on the work image and the IoT data includes: determining the sewing length based on the cumulative stitch count and real-time stitch distance within the sewing time period; determining a second time range based on the sewing time period; identifying the work image within the second time range to obtain the stitch type; determining the target weight based on the stitch type, wherein the stitch type includes: single-layer thread, double-layer dark straight line, double-layer bright straight line, and curved thread, the target weight of the single-layer thread is less than the target weight of the double-layer dark straight line, the target weight of the double-layer dark straight line is less than the target weight of the double-layer bright straight line, and the target weight of the double-layer bright straight line is less than the target weight of the curved thread; determining the TMU value of the sewing action based on the target weight, the sewing length, and the additional TMU value.

[0025] In this embodiment, the method for determining the General Sewing Time (GST) action sequence and the TMU value of each GST action based on IoT data can be as follows: The IoT data includes: device events. If the device event is a motor return needle trigger signal, then the return needle action is determined based on the motor return needle trigger signal. The return needle action includes: any one of computer manual return needle, automatic return needle, and flatbed press bar return needle. The return needle action is taken as the GST action, and the TMU value of the return needle action is taken as the TMU value of the GST action.

[0026] Optionally, the IoT data includes: device events and description information of the device events. If the device event is a wire-cutting triggered interruption event, the description information of the device event includes: wire-cutting timestamp.

[0027] In this embodiment, the wire-cutting trigger interruption event can be an interruption event triggered by cutting the wire with scissors.

[0028] Based on the work images and the IoT data, determine the general sewing time GST action sequence and the TMU value of each GST action, including: Determine the first time range based on the timestamp of the wire cutting.

[0029] In this embodiment, the wire-cutting timestamp can be the wire-cutting end timestamp.

[0030] In this embodiment, a first time range is determined based on the thread-cutting timestamp and a first time step. Specifically, the timestamp of the first time step before the thread-cutting timestamp is taken as the left endpoint of the first time range, and the timestamp of the first time step after the thread-cutting timestamp is taken as the right endpoint of the first time range. For example, the first time range can be [T-500ms, T+500ms], where T is the thread-cutting timestamp. The operation image within the first time range is recognized to obtain the hand movement trajectory.

[0031] In this embodiment, the work images within a first time range are processed in advance. If it is detected that the fabric leaves the presser foot area, the work images within the first time range are further processed to obtain the hand movement trajectory. If it is not detected that the fabric leaves the presser foot area, it is directly rejected and identified as a false trigger.

[0032] In this embodiment, the hand movement trajectory includes multiple trajectory points. For example, the hand movement trajectory can be: take scissors → cut the thread → put the scissors away.

[0033] The target action type is determined based on the hand movement trajectory.

[0034] In this embodiment, the target action type includes any one of the following: initial thread cutting, repeated thread cutting, thread cutting with a fixed blade, and cutting a series of cut pieces with scissors.

[0035] In this embodiment, if the hand motion trajectory includes the complete three-segment action of "grabbing scissors → cutting thread → putting down scissors", without a pre-positioned state of holding scissors, and the device outputs a thread-cutting interruption signal, then the target action type is determined to be the first thread cut, TMU=50. If the previous frame has already identified scissors being held, and the hand motion trajectory only includes the thread-cutting action without the action of picking up or putting down scissors, then the target action type is determined to be the repeated thread cut, TMU=25. If the hand motion trajectory does not include holding scissors, and only the fabric passes through the fixed blade area of ​​the device, coupled with a thread breakage signal, then the target action type is determined to be the fixed blade thread cut, TMU=30. If the hand motion trajectory includes the thread-cutting action, and the operation image within the first time range is identified, determining that multiple continuous pieces of fabric are cut simultaneously, and the number of fabric segments is ≥3, then the target action type is determined to be scissors cutting a series of cut pieces, TMU=49.

[0036] Treat the wire-cutting action as a GST action, and use the TMU value corresponding to the target action type as the TMU value of the wire-cutting action.

[0037] The wire-cutting action is the action corresponding to the wire-cutting trigger interrupt event.

[0038] In this embodiment, the IoT data includes: device events and descriptive information of the device events. If the device event is a non-scissor-induced thread breakage event, non-scissor-induced thread breakage events include: foot-operated thread breakage events and pneumatic thread breakage events. If the device event is a foot-operated thread breakage event, the descriptive information of the device event includes: a foot-operated thread breakage timestamp. The operation images within the time range corresponding to the foot-operated thread breakage timestamp are identified. If there are no scissors or awl tools in the operation image, and the fabric leaves the presser foot, then the action corresponding to the foot-operated thread breakage event is determined to be a foot-operated thread breakage action, and the foot-operated thread breakage action is taken as a GST action, and the TMU value of the foot-operated thread breakage action is obtained. If the device event is a pneumatic thread breakage event, the descriptive information of the device event includes: a pneumatic thread breakage timestamp. The operation images within the time range corresponding to the pneumatic thread breakage timestamp are identified. If there are no scissors or awl tools in the operation image, and the fabric leaves the presser foot, then the action corresponding to the pneumatic thread breakage event is determined to be a pneumatic thread breakage action, and the pneumatic thread breakage action is taken as a GST action, and the TMU value of the pneumatic thread breakage action is obtained.

[0039] In this embodiment, the method for determining the General Sewing Time (GST) action sequence and the TMU value of each GST action based on the work image and the IoT data can also be as follows: When the presser foot lifting signal is detected, the target time range is determined according to the timestamp corresponding to the time of lifting and lowering the presser foot, the work image within the target time range is obtained, the work image within the target time range is identified, if there are two layers of fabric in the image, and the two hands move in a wide range and correct the fabric edge by misalignment, and the fabric overlap changes from low to high, then the alignment and adjustment action is taken as the GST action, and the type of the alignment and adjustment action is determined to be adjusting the alignment of two pieces of fabric. The TMU value of adjusting the alignment of two pieces of fabric is obtained. For example, if the TMU value of adjusting the alignment of two pieces of fabric is 55, then the TMU value of the alignment and adjustment action is 55. For work images within the target time range, if only a single piece of fabric is being corrected, with one or both hands correcting one side of the position, and no double-layer fabric features are present, then the alignment and adjustment action is classified as a GST action. The type of alignment and adjustment action is determined to be single-piece alignment, and the TMU value for single-piece alignment is obtained. For example, if the TMU value for single-piece alignment is 40, then the TMU value for alignment and adjustment action is 40. For work images within the target time range, if the hand makes small reciprocating slides on the fabric surface with a movement amplitude <5cm and no large displacement, then the alignment and adjustment action is classified as a GST action, and the type of alignment and adjustment action is determined to be smoothing / fine-tuning the fabric piece. The TMU value for smoothing / fine-tuning the fabric piece is obtained. For example, if the TMU value for smoothing / fine-tuning the fabric piece is 22, then the TMU value for alignment and adjustment action is 22. For work images within the target time range, if the fabric is already held in the hand, with only slight twisting and straightening of the hand, and the fabric has not touched the presser foot, then the alignment and adjustment action is classified as a GST action. The type of alignment and adjustment action is determined to be adjusting the sewn material in hand, and the TMU value of adjusting the sewn material in hand is obtained. For example, if the TMU value of adjusting the sewn material in hand is 16, then the TMU value of the alignment and adjustment action is 16. For work images within the target time range, if the fabric is detached from the hand and falls, and the hand picks it up and re-holds it, and the action duration is short, then the alignment and adjustment action is classified as a GST action. The type of alignment and adjustment action is determined to be re-grabbing the fallen object, and the TMU value of re-grabbing the fallen object is obtained. For example, if the TMU value of re-grabbing the fallen object is 6, then the TMU value of the alignment and adjustment action is 6. The system identifies the work images within the target time range. If the fabric is slightly adjusted in position with one hand while the other hand remains still, the alignment and adjustment action is identified as a GST action. The type of alignment and adjustment action is determined to be single-handed control of the cut piece. The TMU value of the single-handed control of the cut piece is obtained. For example, if the TMU value of the single-handed control of the cut piece is 13, then the TMU value of the alignment and adjustment action is 13.For the operation images within the target time range, if the overall fabric translation distance is ≤30cm and there are no alignment or folding operations, the alignment and adjustment action is treated as a GST action, and the type of the alignment and adjustment action is determined to be short-distance material transfer. The TMU value of short-distance material transfer is obtained. For example, if the TMU value of short-distance material transfer is 10, then the TMU value of the alignment and adjustment action is 10. For the operation images within the target time range, if the fabric bends at a corner, and the timing signal of the presser foot falling is triggered synchronously, the alignment and adjustment action is treated as a GST action, and the type of the alignment and adjustment action is determined to be cornering and presser foot lowering. The TMU value of cornering and presser foot lowering is obtained. For example, if the TMU value of cornering and presser foot lowering is 33, then the TMU value of the alignment and adjustment action is 33. If the visual recognition identifies the outline of the zipper metal teeth and the hand pulls the zipper head back and forth, then the alignment and adjustment action is taken as the GST action, and the type of the alignment and adjustment action is determined to be zipper operation. The TMU value of the zipper operation is obtained. For example, if the TMU value of the zipper operation is 24, then the TMU value of the alignment and adjustment action is 24.

[0040] In this embodiment, when the presser foot lifting signal is detected, 300ms video frames before and after are captured simultaneously to determine whether it is accompanied by fabric cornering or zipper operation, so as to avoid simply identifying the presser foot action as a TZ type adjustment action.

[0041] Optionally, the general sewing time GST action sequence and the TMU value of each GST action are determined based on the work image, including: If the image of the operation is identified and a piece-picking action is obtained, then the piece-picking action is taken as a GST action, and the TMU value corresponding to the piece-picking type of the piece-picking action is taken as the TMU value of the piece-picking action.

[0042] In this embodiment, the action type is distinguished by four visual features: the number of key points of the hand, the timing of the left and right hand movements, the number of fabric instance segments, and the spatial coordinates of the presser foot area. The presser foot lifting signal is used for boundary verification. There is no hardware dependency throughout the process, and the action recognition is completed solely by the work image.

[0043] In this embodiment, the types of cutting pieces to be picked up include: picking up two pieces of cutting pieces at the same time and pairing them, picking up two pieces of cutting pieces one after the other and pairing them, moving the cutting pieces to the right position under the feet, picking up and pairing the accessories with one hand, and picking up and pairing the accessories with both hands.

[0044] In this embodiment, if the work image is recognized and the key points of the left and right hands are found to be simultaneously grasping downwards, the image is segmented into two independent pieces of fabric, and both hands move synchronously towards the presser foot area without any sequential grasping time difference, then the type of the cutting piece taking action is determined to be simultaneously taking two pieces of fabric and pairing them, with TMU=74. For example, it can be determined that there is no sequential grasping time difference based on the action sequence of the presser foot being in a raised state throughout the action, with no fabric entering the presser foot first and then the fabric being added. If the work image is recognized and there are two independent grasping trajectories, and the fabric segmentation image shows two pieces of fabric one after the other, with one hand first grasping the first piece of fabric and sending it into the presser foot area, then turning back to grasp the second piece of fabric and pairing it, with a significant time difference between the left and right hand actions, then the type of the cutting piece taking action is determined to be simultaneously taking two pieces of fabric and pairing them, with TMU=106. If the image of the operation is recognized and it shows that the presser foot is kept raised, there is no motor sewing start signal, the fabric has entered the presser foot projection area, the hand makes small horizontal movements and pushes and pulls the fabric to adjust its spatial position, and there is no long-distance material picking action, then the type of cutting piece picking action is determined to be moving the cutting piece under the presser foot, TMU=38. If the image of the operation is recognized and it shows that only one hand is used to grasp small accessories such as zippers, webbing, elastic, etc., and align them with the main cutting piece, while the other hand is kept fixed, then the type of cutting piece picking action is determined to be single-handed accessory picking and pairing, TMU=57; specifically, if the size of the grasped target is much smaller than the main cutting piece, it is determined to be an accessory. If the image of the operation is recognized and it shows that both hands work together to grasp long strip accessories (e.g., zippers, binding strips), and simultaneously align and pair them with the main cutting piece, then the type of cutting piece picking action is two-handed accessory picking and pairing, TMU=70.

[0045] If the work image is recognized and a piece placement action is obtained, then the piece placement action is taken as a GST action, and the TMU value corresponding to the piece placement type of the piece placement action is taken as the TMU value of the piece placement action.

[0046] In this embodiment, the fabric piece placement action refers to the placement of the fabric pieces after sewing is completed. The fabric piece placement types include: placing fabric pieces with one hand, placing fabric pieces with both hands, aligning a component with a single positioning point, and aligning a component with two positioning points.

[0047] In this embodiment, the coordinates of the final area of ​​the fabric movement are used to distinguish between simple placement and fixed-point alignment, and the key points of the left and right hands are used to distinguish between single-handed and double-handed material placement.

[0048] In this embodiment, if the image recognition shows that the fabric is moved to the side storage area of ​​the workbench with one hand and the fabric is removed from the presser foot area, then the fabric placement action is determined to be single-handed placement, TMU=23. If the image recognition shows that both hands work together to move the entire piece of sewn fabric and place it in the side area, then the fabric placement action is determined to be two-handed placement, TMU=42. If the image recognition shows that the fabric is precisely aligned with a positioning mark on the workbench during placement, with minor alignment adjustments, then the fabric placement action is determined to be component aligned with a single positioning point, TMU=27. If the image recognition shows that the fabric is simultaneously aligned with two positioning references, with a longer alignment adjustment time, then the fabric placement action is determined to be component aligned with two positioning points, TMU=48.

[0049] In this embodiment, the coordinates of the end point cloth determine whether it is a simple stacking type or a precise alignment type; the activation quantity of key points of the left and right hands distinguishes between single-handed and double-handed actions, and automatically matches the corresponding TMU value.

[0050] If the forming action is obtained by recognizing the work image, the forming action is taken as the GST action, and the TMU value corresponding to the forming type of the forming action is taken as the TMU value of the forming action.

[0051] In this embodiment, relying on the fabric contour deformation segmentation algorithm, the four major morphological changes of the fabric are identified: folding, indentation, turning, and ring shaping. It does not rely on device signals and completes the classification based on the pure visual geometric features of the fabric, distinguishing between ordinary folds and complex ring folds.

[0052] In this embodiment, the forming type includes any one of: folding fabric, pressing creases, unfolding folded pieces, and folding and smoothing the loops.

[0053] In this embodiment, if the image recognition shows that the fabric's planar outline is folded, multi-layered, and has regular creases at the boundaries, the forming type is determined to be folded fabric, TMU=40. If the image recognition shows that the fabric has been folded and the hand continues to press the folded edge without any further folding action, the forming type is determined to be crease pressing, TMU=30. If the image recognition shows that the stacked and folded fabric is unfolded in reverse, and the outline changes from multi-layered to single-layered, the forming type is determined to be unfolded folded fabric, TMU=23. If the image recognition shows that the fabric forms a ring / cylindrical structure, simultaneously completing the folding and smoothing actions, and the fabric deformation complexity is higher than ordinary folding, the forming type is determined to be ring folding and smoothing, TMU=82.

[0054] In this embodiment, simple folds and complex ring-shaped folds are distinguished by the change in the curvature of the fabric contour, and the corresponding TMU value is matched according to the duration of deformation, thus solving the pain point of manual identification of ordinary folds and ring-shaped folds.

[0055] In this embodiment, the method further includes: if the work image is identified as a corner-picking action, then the corner-picking action is taken as a GST action, and the TMU value corresponding to the corner-picking action type of the corner-picking action is taken as the TMU value of the piece-picking action. For example, if the work image is identified as a complete process including picking up the awl, picking the fabric corner, and putting the awl back, then the corner-picking action is taken as a GST action, and the corner-picking action type of the corner-picking action is determined to be the first corner-picking; the TMU value of the first corner-picking (e.g., TMU=55) is taken as the TMU value of the corner-picking action.

[0056] If the image of the operation is identified and it is found that the previous frame is holding the awl continuously and only performing the corner picking action, then the corner picking action of the awl is taken as the GST action, the corner picking action type of the awl corner picking action is determined to be repeated corner picking, and the repeated corner picking TMU value (e.g., TMU=30) is taken as the TMU value of the awl corner picking action.

[0057] Optionally, the IoT data includes: device events and descriptive information of the device events, wherein the descriptive information of the device events includes: sewing time period, cumulative number of stitches within the sewing time period, and real-time stitch distance.

[0058] Based on the work images and the IoT data, determine the general sewing time GST action sequence and the TMU value of each GST action, including: The sewing length is determined based on the cumulative number of stitches and the real-time stitch distance during the sewing period.

[0059] In this embodiment, the sewing length (cm) = the cumulative number of stitches during the sewing time period × the real-time stitch distance (mm) ÷ 10, with an accuracy of ±0.1mm.

[0060] A second time range is determined based on the sewing time period.

[0061] In this embodiment, the method for determining the second time range is similar to the method for determining the first time range, and will not be described again here.

[0062] The operation images within the second time range are identified to determine the line type.

[0063] The line types include: single-layer lines, double-layer dark straight lines, double-layer bright straight lines, and curved lines. The target weight of a single-layer line is less than the target weight of a double-layer dark straight line, the target weight of a double-layer dark straight line is less than the target weight of a double-layer bright straight line, and the target weight of a double-layer bright straight line is less than the target weight of a curved line.

[0064] The target weight is determined based on the described streak type.

[0065] In this embodiment, if the line type is a single-layer line, the target weight is determined to be 1.0; if the line type is a double-layer dark straight line, the target weight is determined to be 1.1; if the line type is a double-layer bright straight line, the target weight is determined to be 1.2; if the line type is a double-layer or single-layer curved line, the target weight is determined to be 1.4.

[0066] The TMU value of the sewing action is determined based on the target weight, the sewing length, and the additional TMU value.

[0067] In this embodiment, the method for determining the additional TMU value can be as follows: determine a third time range based on the device's return timestamp; identify the operation image within the third time range to obtain the return type; and determine the additional TMU value based on the return type.

[0068] In this embodiment, the TMU value of the sewing action can be determined based on the target weight, the sewing length, and the additional TMU value as follows: TMU value of sewing action = sewing length × 10 × difficulty coefficient + parking additional TMU.

[0069] Optionally, the description information of the device event may also include: device backspin timestamp.

[0070] Before determining the TMU value of the sewing action based on the target weight, the sewing length, and the additional TMU value, the method further includes: The third time range is determined based on the device's return needle timestamp.

[0071] The operation image within the third time range is identified to determine the return needle type.

[0072] In this embodiment, the needle return types include: approximately stop, exact stop, and precise stop.

[0073] Determine the additional TMU value based on the needle type.

[0074] In this embodiment, if the return needle type is approximately stop, that is, no return needle, the additional TMU value is determined to be 0; if the return needle type is exactly stop, that is, end return needle, the additional TMU value is determined to be 9; if the return needle type is precisely stop, that is, corner or open-line return needle, the additional TMU value is determined to be 20.

[0075] It should be noted that the machine model can identify whether a line is visible or hidden, and the machine image can be used to identify straight or curved lines.

[0076] S130, based on the GST action sequence and the preset process template library, determine multiple processes corresponding to the GST action sequence.

[0077] In this embodiment, all identified GST actions are arranged in chronological order to generate a continuous GST action sequence with precise timestamps.

[0078] Optionally, based on the GST action sequence and a preset process template library, multiple processes corresponding to the GST action sequence are determined, including: Based on the target sliding window and the GST action sequence, a sub-sequence is determined.

[0079] In this embodiment, let the complete continuous code sequence be S=[s1, s2, ..., s...]. n Let the process template library be T = {T1, T2, ..., T...} m The minimum length of the sliding window equals the shortest template length, and the maximum length of the sliding window equals the longest template length.

[0080] In this embodiment, the sliding window traversal is performed: starting from the beginning position of the sequence, the window is slidable with a step size of 1 to extract all possible subsequences S[i:i+l].

[0081] Obtain the weighted edit distance and target edit distance of each process template in the subsequence and the preset process template library.

[0082] In this embodiment, the target edit distance is the maximum possible edit distance.

[0083] In this embodiment, the user defines a process template library through an interactive interface. Each process template includes: process name and number, standard GST action sequence, optional code weight configuration, and similarity threshold (default 0.85). For example, a process template may include: Template ID: T001, Name: Zipper Installation, Code Sequence: PDXH, TZ2P, YZYJ, KZYJ, KZAN, F25MB, KZYJ, JX1C, TZFP, YZYJ, KZYJ, KZAN, F25MB, KZYJ, JX1C, BF1S.

[0084] The similarity between the subsequence and each process template in the preset process template library is determined based on the weighted edit distance and target edit distance of each process template in the subsequence and the preset process template library.

[0085] In this embodiment, the similarity between the subsequence and the process template is 1 - (weighted edit distance / target edit distance).

[0086] The process template corresponding to the subsequence is determined based on the similarity between the subsequence and each process template in the preset process template library.

[0087] In this embodiment, the process template that has a similarity greater than a similarity threshold with the subsequence can be determined as the process template corresponding to the subsequence.

[0088] The process template includes: template identifier, process name, and GST action sequence included in the process.

[0089] Optionally, obtain the weighted edit distance and target edit distance of each process template in the subsequence and the preset process template library, including: Obtain the weight of each GST action in the process template and the weight of each GST action in the subsequence.

[0090] In this embodiment, the weights of each GST action need to be set in advance. For example, the weight of the F-series sewing action is 3 (core action with the highest distinguishability), the weight of the JX-series thread cutting action is 2 (important feature, but cannot be used as a dividing point), and the weight of other basic actions is 1.

[0091] The sum of the weights of each GST action in the process template and the weights of each GST action in the subsequence is used as the target edit distance between the subsequence and the process template.

[0092] The weighted edit distance between the subsequence and the process template is determined based on the weights of the different GST actions in the subsequence and the process template.

[0093] In this embodiment, the weighted edit distance between the subsequence and the process template can be determined by: determining an edit type, wherein the edit type is used to adjust the GST actions in the subsequence to be the same as the GST actions in the process template; and determining the weighted edit distance based on the edit type and the weight of the GST action to be edited.

[0094] Optionally, the weighted edit distance between the subsequence and the process template is determined based on the weights of different GST actions in the subsequence and the process template, including: Determine the edit type, which is used to adjust the GST action in the sub-sequence to be the same as the GST action in the process template.

[0095] In this embodiment, the editing types include: deletion, replacement, and addition.

[0096] It should be noted that if multiple editing types exist to adjust the GST actions in the sub-sequence to be the same as the GST actions in the process template, the scheme with the smallest weighted editing distance is selected as the final editing type. For example, if the sub-sequence is GST action A, GST action B, GST action C, and GST action D, and the process template is GST action B, GST action C, GST action D, and GST action E, the editing types to adjust the sub-sequence to be the same as the GST actions in the process template include: Editing Type A (Replacement) and Editing Type B (Addition and Deletion). Specifically, the editing action corresponding to Editing Type A is to replace GST actions A, GST actions B, GST actions C, and GST actions D with GST actions B, GST actions C, GST actions D, and GST actions E. The editing action corresponding to Editing Type B is to delete GST action A in the sub-sequence and add GST action E after GST action D. The weighted edit distance for edit type A is the minimum of the sum of the first and second weights. The first weight is the sum of the weights of GST actions A, B, C, and D, and the second weight is the sum of the weights of GST actions B, C, D, and E. The weighted edit distance for edit type B is the sum of the weights of GST actions A and E. Since the weighted edit distance for edit type B is less than that for edit type A, edit type B is determined as the final edit type.

[0097] The weighted edit distance is determined based on the edit type and the weight of the GST action to be edited.

[0098] In this embodiment, if the edit type is deletion, the weight of the GST action deleted in the subsequence is used as the weighted edit distance; if the edit type is replacement, the minimum value of the weight of the GST action to be replaced in the subsequence and the weight of the GST action to be replaced in the process template is used as the weighted edit distance; if the edit type is addition, the weight of the GST action to be added in the subsequence is used as the weighted edit distance; if the edit type includes both deletion and addition, the sum of the weights of the GST actions deleted in the subsequence and the weights of the GST actions to be added is used as the weighted edit distance.

[0099] In this embodiment, if the subsequence consists of GST actions A, B, and C, and the process template consists of GST actions A, D, and C, then the edit type is to replace GST action B in the subsequence with GST action D in the process template, and the minimum value between the weights of GST action B and GST action D is used as the weighted edit distance. If the subsequence consists of GST actions A, B, and C, and the process template consists of GST actions A, C, and D, then the edit type is to replace GST actions B and C in the subsequence with GST actions C and D in the process template, and the minimum value between the sum of the weights of GST actions B and C and the sum of the weights of GST actions C and D is used as the weighted edit distance. If the subsequence consists of GST actions A, B, C, and D, and the process template consists of GST actions B, C, D, and E, then the edit type is to delete GST action A from the subsequence and add GST action E after GST action D. The sum of the weights of GST action A and GST action E is used as the weighted edit distance. If the subsequence consists of GST actions A, B, and C, and the process template consists of GST actions A, B, C, and D, then the edit type is to add GST action D to the subsequence and use the weight of GST action D as the weighted edit distance. If the subsequence consists of GST actions A, B, C, and D, and the process template consists of GST actions A, B, and C, then the edit type is to delete GST action D from the subsequence and use the weight of GST action D as the weighted edit distance.

[0100] S140, determine the standard working time for each process based on the TMU value of the GST action in each process.

[0101] In this embodiment, the sum of the TMU values ​​of the GST actions in each process is used as the standard working time for each process.

[0102] In a specific example, the template T is [F25MB, JX1C, TZ2P], with weights: F=3, JX=2, TZ=1, and the collected subsequence Ssub is [F25MB, BF1S, TZ2P]. The first F25MB is identical with no cost; the second position: the template JX1C (w=2) versus the collected BF1S (w=1) is a mismatch; the third TZ2P is identical with no cost. The optimal path is to replace the second code, with a cost of 2 (taking the larger weight / the minimum operation cost), so the weighted edit distance is 2. The sum of weights of T is F(3)+JX(2)+TZ(1)=6, the sum of weights of Ssub is F(3)+BF(1)+TZ(1)=5, the target edit distance is 6+5=11, and the similarity is 1-2 / 11≈0.82. When the similarity score is greater than or equal to the preset threshold, it is recorded as a valid match. Based on all valid matching results, a dynamic programming algorithm is used to solve the globally optimal splitting scheme: Define the state: dp[i] represents the optimal total similarity score for the first i codes. State transition equation: dp[i] = max{dp[j] + similarity score (j+1, i)} for all j < i where a match (j+1, i) exists. Initial condition: dp[0] = 0. Backtracking solution: Start backtracking from dp[n] to find the optimal splitting path, ensuring that the complete code sequence is covered, there is no overlap or omission between each process, and the total similarity score is the highest.

[0103] In another specific example, the complete on-site data acquisition code sequence S (total length n=9): Index: 123456789, S: PDXH, TZ2P, F25MB, KZYJ, JX1C, TZ2P, F25MB, KZYJ, JX1C, weight of each code: 1, 1, 3, 1, 2, 1, 3, 1, 2, process template library T (only 1 template, zipper T001), template T001 code sequence (length L=7): T: PDXH, TZ2P, F25MB, KZYJ, JX1C, TZ2P, F25MB, weight: 1, 1, 3, 1, 2, 1, 3, sliding window parameters: minimum window length = 7 (template length), maximum window length = 9, step size = 1. The sliding window iterates through all candidate subsequences: Window 1 (i=1~7): S1₋7=[PDXH, TZ2P, F25MB, KZYJ, JX1C, TZ2P, F25MB], completely identical to template T. Weighted edit distance = 0, similarity = 1.0 ≥ 0.85, valid match. Window 2 (i=2~8): S2₋8=[TZ2P, F25MB, KZYJ, JX1C, TZ2P, F25MB, KZYJ], weighted edit distance = 2, similarity ≈ 0.92 ≥ 0.85, valid match. Window 3 (i=3~9): S3₋9=[F25MB, KZYJ, JX1C, TZ2P, F25MB, KZYJ, JX1C], weighted edit distance = 5, similarity ≈ 0.79 < 0.85, invalid match. Summary of all valid matches: 2 valid matches, with a maximum similarity score of 1.0 for the interval [1, 7]. Based on the optimal splitting scheme, the time parameters for each sub-process are defined: Process start time = start timestamp of the first action in the process; Process end time = end timestamp of the last action in the process; Actual time consumed = end time - start time (accurate to 1ms); Standard time consumed = sum of TMU values ​​for all actions in the process; Efficiency value = Standard time consumed / Actual time consumed. A detailed GST analysis report is generated, including the complete action sequence, the start and end times of each action, and the TMU value. A multi-process time analysis report is generated, displaying the splitting results and time data for each process in a timeline format. Data can be exported in Excel and CSV formats and can be directly integrated with enterprise ERP and MES systems.

[0104] The technical solution provided in this embodiment can be pre-installed on sewing machines at the factory, offering plug-and-play functionality without the need for on-site camera installation and wiring, reducing deployment costs by over 90%. Furthermore, single-process analysis time is reduced from 30 minutes to less than 30 seconds, improving efficiency by over 98%, and one system can replace more than 20 professional GST analysts. It eliminates the subjectivity of manual analysis, ensuring consistent analysis results across different factories and batches, guaranteeing fairness in cost accounting. The technical solution provided in this embodiment also fully utilizes the sewing machine's native sensor data without incurring any additional hardware costs.

[0105] This embodiment employs an integrated camera design on the sewing machine head, achieving natural millisecond-level synchronization between visual data and the device's native IoT data. By automatically recognizing GST actions, it generates continuous GST action sequences with precise timestamps. It proposes a "sliding window multi-template matching + dynamic programming global optimal splitting" algorithm, which, without relying on thread cutting as a process boundary point, can accurately identify complex multi-process scenarios involving multiple thread cuts within a single process, ultimately achieving automatic calibration of precise time for each process. This embodiment improves GST analysis efficiency by over 98%, achieves a code recognition accuracy of 90%, and a complex multi-process splitting accuracy of 80%, completely solving the long-standing industry problem of multi-process splitting.

[0106] The technical solution of this embodiment acquires work images and / or IoT data, wherein the work images are captured by a camera positioned directly opposite the presser foot and the operating area; based on the work images and / or the IoT data, a general sewing time (GST) action sequence and the Time Measurement Unit (TMU) value of each GST action are determined, wherein the GST actions in the GST action sequence belong to multiple processes; based on the GST action sequence and a preset process template library, multiple processes corresponding to the GST action sequence are determined; based on the TMU value of the GST actions in each process, the standard working time of each process is determined; an integrated camera design integrated into the sewing machine head is adopted to achieve millisecond-level synchronization of work images and IoT data; by automatically recognizing GST actions, a continuous GST action sequence with precise timestamps is generated; based on the GST action sequence and the preset process template library, complex multi-process scenarios can be accurately identified, ultimately achieving automatic calibration of precise working times for each process.

[0107] Example 2 Figure 2A schematic diagram of an electronic device 10 that can be used to implement embodiments of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, wearable devices (e.g., helmets, glasses, watches, etc.), and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the invention described and / or claimed herein.

[0108] like Figure 2 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory 12 or a random access memory 13, communicatively connected to the at least one processor 11. The memory stores computer programs executable by the at least one processor. The processor 11 can perform various appropriate actions and processes based on the computer program stored in the read-only memory 12 or loaded from storage unit 18 into the random access memory 13. The random access memory 13 may also store various programs and data required for the operation of the electronic device 10. The processor 11, read-only memory 12, and random access memory 13 are interconnected via a bus 14. An input / output interface 15 is also connected to the bus 14.

[0109] Multiple components in electronic device 10 are connected to I / O interface 15, including: input unit 16, such as keyboard, mouse, etc.; output unit 17, such as various types of displays, speakers, etc.; storage unit 18, such as disk, optical disk, etc.; and communication unit 19, such as network card, modem, wireless transceiver, etc. Communication unit 19 allows electronic device 10 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks.

[0110] Processor 11 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of processor 11 include, but are not limited to, central processing units, graphics processing units, various special-purpose artificial intelligence computing chips, various processors running machine learning model algorithms, digital signal processors, and any suitable processor, controller, microcontroller, etc. Processor 11 performs the various methods and processes described above, such as data processing methods.

[0111] In some embodiments, the data processing method may be implemented as a computer program tangibly contained in a computer-readable storage medium, such as storage unit 18. In some embodiments, part or all of the computer program may be loaded and / or installed on electronic device 10 via read-only memory 12 and / or communication unit 19. When the computer program is loaded into random access memory 13 and executed by processor 11, one or more steps of the data processing method described above may be performed. Alternatively, in other embodiments, processor 11 may be configured to perform the data processing method by any other suitable means (e.g., by means of firmware).

[0112] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays, application-specific integrated circuits (ASICs), application-specific standard products (ASICs), systems-on-a-chip (SoCs), payload programmable logic devices, computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0113] Computer programs used to implement the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when executed by the processor, the computer programs cause the functions / operations specified in the flowcharts and / or block diagrams to be performed. The computer programs may be executed entirely on a machine, partially on a machine, or as a standalone software package, partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0114] In the context of this invention, a computer-readable storage medium can be a tangible medium that may contain or store a computer program for use by or in conjunction with an instruction execution system, apparatus, or device. A computer-readable storage medium may include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination thereof. Alternatively, a computer-readable storage medium may be a machine-readable signal medium. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory, read-only memory, erasable programmable read-only memory, optical fibers, portable compact disk read-only memory, optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0115] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a cathode ray tube or liquid crystal display) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user provides input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including sound input, voice input, or tactile input).

[0116] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as data servers), or computing systems that include middleware components (e.g., application servers), or computing systems that include frontend components (e.g., user computers with graphical user interfaces or web browsers through which users can interact with implementations of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., communication networks). Examples of communication networks include local area networks (LANs), wide area networks (WANs), blockchain networks, and the Internet.

[0117] A computing system can include clients and servers. Clients and servers are generally located far apart and typically interact through communication networks. The client-server relationship is created by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, also known as a cloud computing server or cloud host, which is a host product in the cloud computing service system to address the shortcomings of traditional physical hosts and virtual private servers, such as high management difficulty and weak business scalability.

[0118] It should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this invention can be executed in parallel, sequentially, or in different orders, as long as the desired result of the technical solution of this invention can be achieved, and this is not limited herein.

[0119] This invention also provides a computer program product, including a computer program that, when executed by a processor, implements the data processing method according to any embodiment of the invention.

[0120] In the implementation of a computer program product, computer program code for performing the operations of this invention can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0121] The specific embodiments described above do not constitute a limitation on the scope of protection of this invention. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this invention should be included within the scope of protection of this invention.

Claims

1. A data processing method, characterized in that, include: Acquire work images and / or IoT data, wherein the work images are acquired by a camera, which is positioned directly facing the presser foot and the operating area; Based on the work images and / or the IoT data, determine the general sewing time GST action sequence and the time measurement unit (TMU) value of each GST action, wherein the GST actions in the GST action sequence belong to multiple processes; Based on the GST action sequence and the preset process template library, determine multiple processes corresponding to the GST action sequence; The standard working hours for each process are determined based on the TMU value of the GST action in each process.

2. The method according to claim 1, characterized in that, The IoT data includes: device events and description information of the device events. If the device event is a wire-cutting triggered interruption event, the description information of the device event includes: wire-cutting timestamp. Based on the work images and the IoT data, determine the general sewing time GST action sequence and the TMU value of each GST action, including: Determine the first time range based on the timestamp of the wire cutting; The hand movement trajectory is obtained by recognizing the operation images within the first time range; The target action type is determined based on the described hand movement trajectory; The wire-cutting action is taken as a GST action, and the TMU value corresponding to the target action type is taken as the TMU value of the wire-cutting action, wherein the wire-cutting action is the action corresponding to the wire-cutting trigger interrupt event.

3. The method according to claim 1, characterized in that, Based on the work image, determine the general sewing time GST action sequence and the TMU value of each GST action, including: If the image of the operation is identified and a piece-picking action is obtained, then the piece-picking action is taken as a GST action, and the TMU value corresponding to the piece-picking type of the piece-picking action is taken as the TMU value of the piece-picking action. If the work image is identified and a piece placement action is obtained, then the piece placement action is taken as a GST action, and the TMU value corresponding to the piece placement type of the piece placement action is taken as the TMU value of the piece placement action. If the forming action is obtained by recognizing the work image, the forming action is taken as the GST action, and the TMU value corresponding to the forming type of the forming action is taken as the TMU value of the forming action.

4. The method according to claim 1, characterized in that, The IoT data includes: device events and descriptive information of the device events. The descriptive information of the device events includes: sewing time period, cumulative number of stitches within the sewing time period, and real-time stitch distance. Based on the work images and the IoT data, determine the general sewing time GST action sequence and the TMU value of each GST action, including: The sewing length is determined based on the cumulative number of stitches and the real-time stitch distance during the sewing period; Based on the sewing time period, a second time range is determined; The operation images within the second time range are identified to determine the line type; The target weight is determined according to the line type, wherein the line type includes: single-layer line, double-layer dark straight line, double-layer bright straight line and curved line, the target weight of the single-layer line is less than the target weight of the double-layer dark straight line, the target weight of the double-layer dark straight line is less than the target weight of the double-layer bright straight line, and the target weight of the double-layer bright straight line is less than the target weight of the curved line. The TMU value of the sewing action is determined based on the target weight, the sewing length, and the additional TMU value.

5. The method according to claim 4, characterized in that, The description information of the device event also includes: device backflip timestamp; Before determining the TMU value of the sewing action based on the target weight, the sewing length, and the additional TMU value, the method further includes: The third time range is determined based on the device's return needle timestamp; The operation image within the third time range is identified to determine the return needle type; Determine the additional TMU value based on the needle type.

6. The method according to claim 1, characterized in that, Based on the GST action sequence and the preset process template library, determine multiple processes corresponding to the GST action sequence, including: Based on the target sliding window and the GST action sequence, determine the sub-sequence; Obtain the weighted edit distance and target edit distance for each process template in the subsequence and the preset process template library; The similarity between the subsequence and each process template in the preset process template library is determined based on the weighted edit distance and target edit distance of each process template in the subsequence and the preset process template library. Based on the similarity between the subsequence and each process template in the preset process template library, the process template corresponding to the subsequence is determined, wherein the process template includes: template identifier, process name, and GST action sequence included in the process.

7. The method according to claim 6, characterized in that, Obtain the weighted edit distance and target edit distance for each process template in the subsequence and preset process template library, including: Obtain the weight of each GST action in the process template and the weight of each GST action in the subsequence; The sum of the weights of each GST action in the process template and the weights of each GST action in the subsequence is used as the target edit distance between the subsequence and the process template. The weighted edit distance between the subsequence and the process template is determined based on the weights of the different GST actions in the subsequence and the process template.

8. The method according to claim 7, characterized in that, Based on the weights of different GST actions in the subsequence and the process template, the weighted edit distance between the subsequence and the process template is determined, including: Determine the edit type, which is used to adjust the GST action in the sub-sequence to be the same as the GST action in the process template; The weighted edit distance is determined based on the edit type and the weight of the GST action to be edited.

9. An electronic device, characterized in that, The electronic device includes: At least one processor; and A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the data processing method according to any one of claims 1-8.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that cause a processor to execute the data processing method according to any one of claims 1-8.

11. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the data processing method according to any one of claims 1-8.

Images