Online image detection and fault prediction method and system for high-speed loom

Abstract

The invention relates to the technical field of woven fabric image detection, in particular to an online image detection and fault prediction method and system for a high-speed loom, and the method comprises the steps: shooting the high-speed loom through a high-definition camera, obtaining a detected woven fabric image set, sequentially extracting detected woven fabric images from the detected woven fabric image set, carrying out the sliding detection of the detected woven fabric images, and carrying out the fault prediction of the detected woven fabric images. The method comprises the steps of obtaining a complete sliding image set, calculating a local gray scale data set of each complete sliding image in the complete sliding image set to obtain a local gray scale data set, carrying out loom fault identification according to the local gray scale data set to obtain loom fault factors, summarizing the loom fault factors of each detected weaving image, and carrying out loom fault identification according to the summarized loom fault factors. And obtaining a loom fault factor set, and constructing a loom fault prediction model according to the loom fault factor set. According to the method, the accuracy and efficiency of fault prediction of the high-speed loom can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of weaving image detection, and in particular to an online image detection and fault prediction method and system for a high-speed loom. Background Art

[0002] In the modern textile industry, looms are key equipment for producing textiles. Their performance and efficiency directly affect product quality and production costs. With the advancement of technology, the speed and automation of looms are constantly improving, which puts higher requirements on fault prediction in the weaving process. Therefore, in order to ensure the normal operation of high-speed looms, it is necessary to perform high-precision fault prediction on the looms in operation.

[0003] At present, image prediction of high-speed looms is mainly achieved through manual inspection. This method is easily affected by human factors, thereby reducing the accuracy of high-speed loom fault prediction. At the same time, a large amount of manual participation often requires shutdown inspection, which will reduce the efficiency of high-speed loom fault prediction. Summary of the invention

[0004] The present invention provides a method and system for online image detection and fault prediction of a high-speed loom, the main purpose of which is to improve the accuracy and efficiency of fault prediction of the high-speed loom.

[0005] To achieve the above object, the present invention provides a high-speed loom online image detection and fault prediction method, comprising: Starting a pre-built high-speed loom, and at the same time starting the high-speed loom, starting a pre-built high-definition camera, and continuously photographing the high-speed loom with the high-definition camera to obtain a stored weaving image set; receiving an online image detection instruction, setting an online detection time based on the online image detection instruction, and recording an image captured by a high-definition camera at the online detection time as a current weaving image; According to the online detection time, a reference weaving image set is extracted from the stored weaving image set, and the current weaving image is merged with the reference weaving image set to obtain a detection weaving image set, wherein the detection weaving images in the detection weaving image set are arranged in order from earliest to latest according to the shooting time, and the current weaving image is arranged at the last position in the detection weaving image set; Extracting detection woven images in sequence from the detection woven image set, performing sliding detection on the detection woven images, and obtaining a complete sliding image set; Extracting complete sliding images in the complete sliding image set in sequence, and performing local grayscale deviation analysis on the complete sliding images based on the detected fabric images to obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value; Summarizing the local grayscale data groups of each complete sliding image to obtain a local grayscale data set, and performing loom fault identification according to the local grayscale data set to obtain a loom fault factor; Summarizing the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set; According to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model, and online image detection and fault prediction of the high-speed loom are realized based on the loom fault prediction model.

[0006] Optionally, the high-speed loom is continuously photographed by a high-definition camera to obtain a stored weaving image set, including: Based on the preset weaving start time, a high-definition camera is used to capture images of the high-speed loom to obtain a single-shot image; Supplementing the single shot image to the pre-constructed original weaving image set to obtain a target weaving image set, wherein the target weaving images in the target weaving image set are arranged in the order of shooting time, and the target weaving images with earlier shooting time are arranged in front; Acquire the number of target images in the target weaving image set, and determine whether the number of target images is greater than a preset maximum image capacity; If the number of target images is greater than the maximum image capacity, the first weaving image is extracted from the target weaving image set, and the first weaving image is removed from the target weaving image set to obtain a removed weaving image set; After waiting for the preset shooting time interval, record the current weaving moment; The original weaving image set and the weaving start time are updated respectively by using the discarded weaving image set and the current weaving time, and the step of using a high-definition camera to take an image of the high-speed loom based on the preset weaving start time is returned until the current weaving time is the preset instruction receiving time, and the discarded weaving image set at this time is recorded as the stored weaving image set; If the number of target images is not greater than the maximum image capacity, then after waiting for a preset shooting time interval, the current weaving moment is obtained, and the original weaving image set and the weaving start moment are updated respectively using the target weaving image set and the current weaving moment; Return to the step of taking images of the high-speed loom using a high-definition camera based on the preset weaving start time until the current weaving time is the instruction receiving time, and record the target weaving image set at this time as the stored weaving image set.

[0007] Optionally, performing sliding detection on the detected woven cloth image to obtain a complete sliding image set includes: Determining a detection sliding area according to the detected fabric image; Placing the detection sliding area at a preset original image position, setting a sliding speed and identifying a lateral sliding boundary of the detection fabric image, wherein the original image position is the upper left corner of the detection fabric image; According to the sliding speed and the lateral sliding boundary, the lateral sliding is performed in the detection fabric image by using the detection sliding area to obtain a unidirectional sliding image group; According to the preset detection accuracy, the longitudinal sliding interval is calculated, wherein the longitudinal sliding interval is expressed as: , in, Indicates the vertical sliding interval, Indicates the detection accuracy, Indicates the longitudinal dimension of the slide; Based on the longitudinal sliding interval, determining the longitudinal sliding position and identifying the longitudinal sliding boundary on the detected fabric image, wherein the longitudinal sliding position is the longitudinal sliding interval downward from the original image position in the longitudinal direction; The original image position is updated by using the longitudinal sliding position, and the step of sliding horizontally in the detection fabric image according to the sliding speed and the transverse sliding boundary and using the detection sliding area is returned until the detection sliding area contacts the transverse sliding boundary and the longitudinal sliding boundary at the same time; The one-way sliding image groups are aggregated to obtain a complete sliding image set.

[0008] Optionally, determining the detection sliding area according to the detected woven fabric image includes: Identify the horizontal pixel number and the vertical pixel number of the detected fabric image, and set the sliding area size according to the horizontal pixel number, the vertical pixel number and the detection accuracy, wherein the sliding area size includes: a sliding horizontal size and a sliding vertical size, and the sliding area size is expressed as: , in, Indicates the horizontal dimension of the slide. Indicates the number of horizontal pixels. Represents the pixel diameter of the detected pixel in the fabric image, Indicates the number of vertical pixels; A detection sliding area is set based on the sliding area size.

[0009] Optionally, the step of using the detection sliding area to perform lateral sliding in the detection fabric image to obtain a unidirectional sliding image group includes: Recording the detected sliding area at the position of the original image as the original sliding image; Based on the sliding speed, the detection sliding area is slid from the original image position from left to right, and after waiting for a preset sliding detection interval, the detection fabric image where the detection sliding area is located is recorded as the detection sliding image, and the position where the detection sliding area is located at this time is recorded as the detection image position; Supplementing the detected sliding image and the original sliding image to the pre-constructed original sliding image group to obtain a target sliding image group; The original image position and the original sliding image group are updated respectively by using the detected image position and the target sliding image group, and the step of sliding the detected sliding area from the original image position from left to right based on the sliding speed is returned until the detected sliding area contacts the lateral sliding boundary, and the target sliding image group at this time is recorded as a unidirectional sliding image group.

[0010] Optionally, based on detecting the weaving image, performing a local grayscale deviation analysis on the complete sliding image to obtain a local grayscale data group includes: Performing grayscale conversion on the detected fabric image to obtain a grayscale overall image, extracting an overall pixel point set from the grayscale overall image, identifying an overall grayscale value set of the overall pixel point set, and calculating an average overall grayscale value of the overall grayscale value set; The full sliding image is gray-scale converted to obtain a gray-scale local image, a local pixel point set is extracted from the gray-scale local image, and a local gray-scale value set of the local pixel point set is identified. According to the average overall gray-scale value and the local gray-scale value set, a local gray-scale deviation value is calculated, wherein the local gray-scale deviation value is expressed as: , in, Represents the local grayscale deviation value, Represents the number of local gray values ​​in the local gray value set, Represents the first local gray value concentration The local gray value, represents the average overall gray value; Performing damage clarity analysis on the complete sliding image to obtain a local image clarity value; The local grayscale deviation value and the local image clarity value are paired to obtain a local grayscale data set.

[0011] Optionally, performing damage clarity analysis on the complete sliding image to obtain a local image clarity value includes: The maximum local gray value in the local gray value set is identified, and the background gray threshold is calculated according to the maximum local gray value and a preset background gray ratio, wherein the background gray threshold is expressed as: , in, represents the background grayscale threshold, represents the maximum local gray value, Represents the background grayscale ratio; Based on the background grayscale threshold, determining a background grayscale value set in the local grayscale value set, wherein the background grayscale value in the background grayscale value set is not greater than the background grayscale threshold; The acceptable clarity ratio is set, and the local image clarity value is calculated using the following formula according to the background gray value set, the maximum local gray value and the acceptable clarity ratio: , in, Indicates the local image clarity value, Represents the number of background gray values ​​in the background gray value set, Represents the background gray value concentration Background gray value, Indicates acceptable clarity ratio.

[0012] Optionally, the performing loom fault identification according to the local grayscale data set to obtain the loom fault factor includes: The local grayscale data groups are sequentially extracted from the local grayscale data group set, and the fabric damage factor is calculated based on the local grayscale data group, wherein the fabric damage factor is expressed as: , in, represents the fabric damage factor; Determining whether the fabric damage factor is greater than a preset minimum damage factor; If the fabric damage factor is greater than the minimum damage factor, the complete sliding image is recorded as a damaged fabric image, and the damaged fabric images are summarized to obtain a damaged fabric image set; Counting the number of fabric damages of the damaged fabric images in the damaged fabric image set, and calculating an average damage factor of the damaged fabric image set, wherein the average damage factor is an average value of the fabric damage factors corresponding to all damaged fabric images in the damaged fabric image set; The loom failure factor is calculated according to the number of damaged fabrics and the average damage factor, wherein the loom failure factor is the product of the number of damaged fabrics and the average damage factor.

[0013] Optionally, performing fault prediction on the high-speed loom according to the loom fault factor set to obtain a loom fault prediction model includes: Determine the fault detection duration of each loom fault factor in the loom fault factor set to obtain a fault detection duration set, wherein the fault detection duration is the time interval between the shooting time of the detection weaving image corresponding to the loom fault factor and the online detection time; In the loom fault factor set and fault detection time set, respectively identify the current fault factor and current detection time corresponding to the current weaving image; According to the fault detection time set and the loom fault factor set, a fault factor change line segment is drawn, wherein the horizontal axis of the fault factor change line segment represents the fault detection time, and the vertical axis represents the loom fault factor; Setting a division precision value, dividing the fault factor change line segment based on the division precision value to obtain a division change line segment group, wherein the number of the division change line segments is the same as the division precision value; Identify the line segment slope of each partitioning change line segment in the partitioning change line segment group to obtain a line segment slope group, wherein the line segment slope is the slope value of the first and last coordinate points of the partitioning change line segment; The maximum line segment slope and the minimum line segment slope are identified in the line segment slope group, and the controllable line segment slope is set according to the maximum line segment slope and the minimum line segment slope. According to the controllable line segment slope, the current fault factor, the current detection time and the preset prediction time variable, a loom fault prediction model is constructed, wherein the controllable line segment slope and the loom fault prediction model are respectively expressed as: , in, represents the minimum line segment slope, represents the slope of the controllable line segment, represents the maximum line segment slope, represents the loom fault prediction model, represents the prediction duration variable, Indicates the current detection duration. Indicates the current fault factor.

[0014] To achieve the above object, the present invention also provides an online image detection and fault prediction system for a high-speed loom, comprising: A weaving image shooting module is used to start a pre-built high-speed loom, and at the same time as starting the high-speed loom, start a pre-built high-definition camera, and use the high-definition camera to continuously shoot the high-speed loom to obtain a stored weaving image set, receive an online image detection instruction, set an online detection time based on the online image detection instruction, and record the image captured by the high-definition camera at the online detection time as the current weaving image; A local image extraction module is used to extract a reference woven image set from a stored woven image set according to the online detection time, merge the current woven image with the reference woven image set to obtain a detection woven image set, wherein the detection woven images in the detection woven image set are arranged in the order of shooting time from earliest to latest, and the current woven image is ranked last in the detection woven image set, extract detection woven images in the detection woven image set in turn, perform sliding detection on the detection woven images, and obtain a complete sliding image set; A fault factor analysis module is used to sequentially extract complete sliding images from the complete sliding image set, perform local grayscale deviation analysis on the complete sliding images based on the detected weaving images, and obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value, summarize the local grayscale data group of each complete sliding image, obtain a local grayscale data group set, perform loom fault identification according to the local grayscale data group set, and obtain a loom fault factor; A prediction model building module is used to summarize the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set, and according to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model.

[0015] In order to solve the above problem, the present invention further provides an electronic device, the electronic device comprising: a memory storing at least one instruction; and The processor executes the instructions stored in the memory to implement the above-mentioned high-speed loom online image detection and fault prediction method.

[0016] In order to solve the above problems, the present invention also provides a computer-readable storage medium, in which at least one instruction is stored. The at least one instruction is executed by a processor in an electronic device to implement the above-mentioned high-speed loom online image detection and fault prediction method.

[0017] In order to solve the problems described in the background technology, the present invention first starts a high-definition camera and uses the high-definition camera to continuously shoot the high-speed loom to obtain a stored weaving image set. This step ensures that the operating status of the high-speed loom can be monitored in real time. The image set obtained by continuous shooting provides basic data for subsequent image detection and fault prediction. Then, the current weaving image is obtained based on the online image detection instruction. This step improves the flexibility and timeliness of image detection and improves the accuracy of subsequent high-speed loom fault prediction. By performing sliding detection on the detected weaving image, a complete sliding image set is obtained, which realizes continuous and comprehensive analysis of the weaving image, so that the obtained complete sliding image set can represent the trend of local image changes of the detected weaving image, which improves the coverage and continuity of image detection, and thus improves the accuracy of high-speed loom fault detection. Furthermore, based on the detected weaving image, the complete sliding image is analyzed for local grayscale deviation. Analysis is performed to obtain a local grayscale data set. In this step, by analyzing the local grayscale deviation and clarity of the complete sliding image, subtle changes in the detected weaving image can be identified, providing more accurate local features for fault identification. Then, according to the local grayscale data set, loom fault identification is performed to obtain a loom fault factor, wherein the loom fault factor provides an intuitive numerical representation for evaluating the fault condition of the high-speed loom. At the same time, a corresponding relationship between the loom fault factor and the specific image is established, making the fault analysis more specific. Finally, according to the loom fault factor set, a loom fault prediction model is constructed, which can realize the automatic prediction of the future state of the loom, and can discover potential faults in advance, thereby achieving the purpose of quickly predicting faults for high-speed looms. At the same time, the construction of the loom fault prediction model also greatly reduces the consumption of human resources, thereby improving the efficiency of fault prediction for high-speed looms. Therefore, the present invention can improve the accuracy and efficiency of high-speed loom fault prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic flow chart of a method for online image detection and fault prediction of a high-speed loom provided by an embodiment of the present invention; Figure 2 A functional module diagram of a high-speed loom online image detection and fault prediction system provided by an embodiment of the present invention; Figure 3 A schematic diagram of the structure of an electronic device for implementing the high-speed loom online image detection and fault prediction method provided by one embodiment of the present invention.

[0019] Description of reference numerals: 1. Electronic device; 10. Processor; 11. Memory; 12. Bus.

[0020] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0021] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0022] The embodiment of the present application provides a method for online image detection and fault prediction of a high-speed loom. The execution subject of the method for online image detection and fault prediction of a high-speed loom includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for online image detection and fault prediction of a high-speed loom can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0023] Reference Figure 1 FIG. 1 is a flow chart of a method for online image detection and fault prediction of a high-speed loom provided by an embodiment of the present invention. In this embodiment, the method for online image detection and fault prediction of a high-speed loom includes: S1. Start a pre-built high-speed loom, and at the same time start a pre-built high-definition camera, and use the high-definition camera to continuously shoot the high-speed loom to obtain a stored weaving image set.

[0024] It can be understood that the high-speed loom refers to a loom that needs to be predicted for faults, the high-definition camera refers to a device that can photograph the fabric in the high-speed loom, the fixed position of the high-definition camera can capture a panoramic image of the fabric being woven in the high-speed loom, and the stored weaving image set refers to a set of images of the fabric in the high-speed loom that has been captured by the high-definition camera before receiving the online image detection instruction.

[0025] It should be explained that in order to ensure that the weaving images obtained about the high-speed loom have a time-varying trend, it is necessary to obtain multiple images of the high-speed loom at different times. However, the storage space in the high-definition camera is limited. At the same time, in order to reduce the complexity of subsequent image processing, it is necessary to eliminate images that exceed the storage space of the high-definition camera. The elimination method is: first eliminate the images with an earlier shooting time.

[0026] In detail, the high-speed loom is continuously photographed by a high-definition camera to obtain a stored weaving image set, including: Based on the preset weaving start time, a high-definition camera is used to capture images of the high-speed loom to obtain a single-shot image; Supplementing the single shot image to the pre-constructed original weaving image set to obtain a target weaving image set, wherein the target weaving images in the target weaving image set are arranged in the order of shooting time, and the target weaving images with earlier shooting time are arranged in front; Acquire the number of target images in the target weaving image set, and determine whether the number of target images is greater than a preset maximum image capacity; If the number of target images is greater than the maximum image capacity, the first weaving image is extracted from the target weaving image set, and the first weaving image is removed from the target weaving image set to obtain a removed weaving image set; After waiting for the preset shooting time interval, record the current weaving moment; The original weaving image set and the weaving start time are updated respectively by using the discarded weaving image set and the current weaving time, and the step of using a high-definition camera to take an image of the high-speed loom based on the preset weaving start time is returned until the current weaving time is the preset instruction receiving time, and the discarded weaving image set at this time is recorded as the stored weaving image set; If the number of target images is not greater than the maximum image capacity, then after waiting for a preset shooting time interval, the current weaving moment is obtained, and the original weaving image set and the weaving start moment are updated respectively using the target weaving image set and the current weaving moment; Return to the step of taking images of the high-speed loom using a high-definition camera based on the preset weaving start time until the current weaving time is the instruction receiving time, and record the target weaving image set at this time as the stored weaving image set.

[0027] It should be explained that the weaving start time refers to the time when the high-speed loom is started, the single-shot image refers to the image of fabric being woven in the high-speed loom captured by a high-definition camera, the original weaving image set refers to the set of single-shot images captured before the current single-shot image is obtained. When this is the first time to capture the image of the high-speed loom, the original weaving image set is an empty set, the target weaving image set refers to the original weaving image set after being supplemented with a single-shot image, and the target image number refers to the number of target weaving images in the target weaving image set.

[0028] It can be understood that the maximum image capacity refers to an artificially set image number constant, which is the maximum number of images that a high-definition camera can store. The first weaving image is the target weaving image that is set in the target weaving image set and arranged in the first position. The eliminated weaving image refers to the target weaving image set after eliminating the first weaving image. The shooting time interval refers to a manually set time interval, which indicates the time interval between two image captures by the high-definition camera. The current weaving moment refers to the current moment. The instruction receiving moment refers to the moment when the online image detection instruction is received. The detailed definition of the online image detection instruction will be given later.

[0029] S2. Receive an online image detection instruction, set an online detection time based on the online image detection instruction, and record the image captured by the high-definition camera at the online detection time as the current weaving image.

[0030] It can be understood that the online image detection instruction refers to an instruction initiated manually to perform image detection on a specific high-speed loom, the online detection moment refers to the moment when the online image instruction is received, and the current weaving image refers to the image of the high-speed loom captured by the high-definition camera at the online detection moment.

[0031] For example, Xiao Wang is an operator in a weaving factory. One day, Xiao Wang receives an online image detection instruction and needs to predict faults for a specific model of high-speed loom. Xiao Wang then calls up the high-definition camera of the high-speed loom, records the moment when the online image detection instruction is received as the online detection moment, and records the other image sets stored in the high-definition camera as the stored weaving image sets.

[0032] S3. According to the online detection time, extract a reference weaving image set from the stored weaving image set, merge the current weaving image with the reference weaving image set to obtain a detection weaving image set, wherein the detection weaving images in the detection weaving image set are arranged in descending order according to the shooting time, and the current weaving image is ranked last in the detection weaving image set.

[0033] It should be explained that since there are many stored weaving image sets stored in the high-definition camera, in order to reduce the complexity of subsequent image processing and improve the efficiency of fault prediction, it is possible to retrieve images from the stored weaving image set with shooting times close to the online detection time as reference weaving images according to the online detection time, thereby obtaining a reference weaving image set.

[0034] For example, the time of an online detection is 20min20s. Xiao Wang retrieves the 10 images closest to 20min20s from the stored weaving image set to form a control weaving image set, and merges the current weaving image taken at the time of 20min20s with the 10 control weaving images to obtain the detection weaving image set.

[0035] S4. Extracting detection woven fabric images in sequence from the detection woven fabric image set, performing sliding detection on the detection woven fabric images, and obtaining a complete sliding image set.

[0036] It can be understood that the complete sliding image set refers to a set of images corresponding to all sliding areas recorded during the process of performing sliding detection on the woven fabric image.

[0037] In detail, the sliding detection is performed on the detected weaving image to obtain a complete sliding image set, including: Determining a detection sliding area according to the detected fabric image; Placing the detection sliding area at a preset original image position, setting a sliding speed and identifying a lateral sliding boundary of the detection fabric image, wherein the original image position is the upper left corner of the detection fabric image; According to the sliding speed and the lateral sliding boundary, the lateral sliding is performed in the detection fabric image by using the detection sliding area to obtain a unidirectional sliding image group; According to the preset detection accuracy, the longitudinal sliding interval is calculated, wherein the longitudinal sliding interval is expressed as: , in, Indicates the vertical sliding interval, Indicates the detection accuracy, Indicates the longitudinal dimension of the slide; Based on the longitudinal sliding interval, determining the longitudinal sliding position and identifying the longitudinal sliding boundary on the detected fabric image, wherein the longitudinal sliding position is the longitudinal sliding interval downward from the original image position in the longitudinal direction; The original image position is updated by using the longitudinal sliding position, and the step of sliding horizontally in the detection fabric image according to the sliding speed and the transverse sliding boundary and using the detection sliding area is returned until the detection sliding area contacts the transverse sliding boundary and the longitudinal sliding boundary at the same time; The one-way sliding image groups are aggregated to obtain a complete sliding image set.

[0038] It can be understood that the detection sliding area refers to a rectangular area whose size is the size of the sliding area, the original image position refers to the position in the detection fabric image that is located in the upper left corner and occupies the shape of the detection sliding area, and the sliding speed refers to the speed at which the artificially set detection sliding area slides in the detection fabric image, which can be expressed as: the number of pixels sliding per second.

[0039] Furthermore, the horizontal sliding boundary refers to the rightmost boundary of the detected fabric image, the vertical sliding interval refers to the distance between two adjacent detection sliding areas in the vertical direction of the detected fabric image, and its unit is the number of pixels, and the vertical sliding position refers to the distance below the original image position from the original image position. The longitudinal sliding boundary refers to the bottommost boundary of the detected fabric image.

[0040] In detail, the step of determining the detection sliding area according to the detected fabric image includes: Identify the horizontal pixel number and the vertical pixel number of the detected fabric image, and set the sliding area size according to the horizontal pixel number, the vertical pixel number and the detection accuracy, wherein the sliding area size includes: a sliding horizontal size and a sliding vertical size, and the sliding area size is expressed as: , in, Indicates the horizontal dimension of the slide. Indicates the number of horizontal pixels. Represents the pixel diameter of the pixel point in the detected fabric image, Indicates the number of vertical pixels; A detection sliding area is set based on the sliding area size.

[0041] It should be explained that the horizontal pixel count refers to the number of pixel points on the horizontal axis in the detected woven image, wherein the horizontal axis can be regarded as the shorter side of the detected woven image, and the vertical pixel count refers to the number of pixel points on the vertical axis in the detected woven image, wherein the vertical axis can be regarded as the longer side of the detected woven image.

[0042] In detail, the method of using the detection sliding area to perform lateral sliding in the detection fabric image to obtain a unidirectional sliding image group includes: Recording the detected sliding area at the position of the original image as the original sliding image; Based on the sliding speed, the detection sliding area is slid from the original image position from left to right, and after waiting for a preset sliding detection interval, the detection fabric image where the detection sliding area is located is recorded as the detection sliding image, and the position where the detection sliding area is located at this time is recorded as the detection image position; Supplementing the detected sliding image and the original sliding image to the pre-constructed original sliding image group to obtain a target sliding image group; The original image position and the original sliding image group are updated respectively by using the detected image position and the target sliding image group, and the step of sliding the detected sliding area from the original image position from left to right based on the sliding speed is returned until the detected sliding area contacts the lateral sliding boundary, and the target sliding image group at this time is recorded as a unidirectional sliding image group.

[0043] It should be explained that the original sliding image refers to the detection sliding area at the original image position, the detection sliding image refers to the detection woven image where the current detection sliding area is located, the detection image position refers to the position of the current detection sliding area in the detection woven image, the original sliding image group refers to the set of detection sliding images and original sliding images before the current detection sliding image and the original sliding image are obtained, and the target sliding image group is the original sliding image group after being supplemented by the detection sliding image and the original sliding image.

[0044] S5. Extract complete sliding images in the complete sliding image set in sequence, and perform local grayscale deviation analysis on the complete sliding images based on the detection of the weaving images to obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value.

[0045] It should be explained that the local grayscale deviation value refers to a numerical value used to represent the degree of difference in grayscale values ​​between the complete sliding image and the detected woven image, wherein, since the complete sliding image is an area in the detected woven image, the local grayscale deviation can be used to represent the possibility that the complete sliding image includes a damaged area in the detected woven image, and the larger the local grayscale deviation value, the greater the possibility that the corresponding complete sliding image contains a damaged area, wherein the damaged area is the location where stains or damage appear on the woven fabric, but the error of representing the damage in the complete sliding image only by the local grayscale deviation value is large, when the grayscale value difference between the damaged area and other areas in the complete sliding image is not large, it is impossible to determine whether the woven fabric damage information in the complete sliding image is correct information, for example: due to the blur caused by noise, the local grayscale deviation value of the complete sliding image will be large, so the local image clarity value is introduced to reflect the clarity of the damaged area in the complete sliding image, that is, the local image clarity value refers to a numerical value used to represent the clarity of the damaged area in the complete sliding image.

[0046] In detail, based on the detection of the weaving image, the local grayscale deviation analysis is performed on the complete sliding image to obtain a local grayscale data group, including: Performing grayscale conversion on the detected fabric image to obtain a grayscale overall image, extracting an overall pixel point set from the grayscale overall image, identifying an overall grayscale value set of the overall pixel point set, and calculating an average overall grayscale value of the overall grayscale value set; The full sliding image is gray-scale converted to obtain a gray-scale local image, a local pixel point set is extracted from the gray-scale local image, and a local gray-scale value set of the local pixel point set is identified. According to the average overall gray-scale value and the local gray-scale value set, a local gray-scale deviation value is calculated, wherein the local gray-scale deviation value is expressed as: , in, Represents the local grayscale deviation value, Represents the number of local gray values ​​in the local gray value set, Represents the first The local gray value, represents the average overall gray value; Performing damage clarity analysis on the complete sliding image to obtain a local image clarity value; The local grayscale deviation value and the local image clarity value are paired to obtain a local grayscale data set.

[0047] It can be understood that the grayscale overall image refers to the grayscale image of the detected woven image, wherein the color space of the detected woven image before grayscale conversion is the RGB color space, the overall pixel point set refers to the set of all pixel points in the grayscale overall image, the overall grayscale value set refers to the set of grayscale values ​​corresponding to all overall pixel points in the overall pixel point set, the average overall grayscale value refers to the average of all overall grayscale values ​​in the overall grayscale value set, the grayscale local image refers to the grayscale image of the complete sliding image, the local pixel points refer to the pixel points in the complete sliding image, and the local grayscale value refers to the grayscale value corresponding to the local pixel points.

[0048] It should be noted that the damage or stains on the fabric in the detected fabric image are caused by the failure of the high-speed loom. The larger the loom failure factor in the detected fabric image taken at a certain moment, the more serious the failure of the high-speed loom at this moment. Therefore, the failure factor of the high-speed loom can be predicted to obtain the predicted failure factor. When the predicted failure factor reaches a maximum failure threshold set by an artificial person, it means that the high-speed loom is about to have an uncontrollable failure, which will cause greater damage to the fabric being woven at this moment.

[0049] In detail, the damage clarity analysis of the complete sliding image to obtain the local image clarity value includes: The maximum local gray value in the local gray value set is identified, and the background gray threshold is calculated according to the maximum local gray value and a preset background gray ratio, wherein the background gray threshold is expressed as: , in, represents the background grayscale threshold, represents the maximum local gray value, Represents the background grayscale ratio; Based on the background grayscale threshold, determining a background grayscale value set in the local grayscale value set, wherein the background grayscale value in the background grayscale value set is not greater than the background grayscale threshold; The acceptable clarity ratio is set, and the local image clarity value is calculated using the following formula according to the background gray value set, the maximum local gray value and the acceptable clarity ratio: , in, Indicates the local image clarity value, Represents the number of background gray values ​​in the background gray value set, Represents the background gray value concentration Background gray value, Indicates acceptable clarity ratio.

[0050] It can be understood that the maximum local grayscale value refers to the maximum value in the local grayscale value set, and the background grayscale ratio refers to the minimum proportion of the normal woven fabric area in the complete sliding image that is artificially set, wherein the normal woven fabric area refers to the woven fabric area without damage and stains, and the background grayscale threshold refers to the maximum grayscale value used to represent the normal woven fabric area in the complete sliding image. When the grayscale value of a certain position in the complete sliding image is greater than the background grayscale threshold, it can be considered that there is woven damage or stains at that position.

[0051] S6. Summarize the local grayscale data groups of each complete sliding image to obtain a local grayscale data set, and perform loom fault identification based on the local grayscale data set to obtain a loom fault factor.

[0052] It can be understood that the loom fault factor refers to a numerical value used to indicate the severity of a fault currently occurring in the high-speed loom.

[0053] In detail, the loom fault identification is performed based on the local grayscale data set to obtain the loom fault factor, including: The local grayscale data groups are sequentially extracted from the local grayscale data group set, and the fabric damage factor is calculated based on the local grayscale data group, wherein the fabric damage factor is expressed as: , in, represents the fabric damage factor; Determining whether the fabric damage factor is greater than a preset minimum damage factor; If the fabric damage factor is greater than the minimum damage factor, the complete sliding image is recorded as a damaged fabric image, and the damaged fabric images are summarized to obtain a damaged fabric image set; Counting the number of fabric damages of the damaged fabric images in the damaged fabric image set, and calculating an average damage factor of the damaged fabric image set, wherein the average damage factor is an average value of the fabric damage factors corresponding to all damaged fabric images in the damaged fabric image set; The loom failure factor is calculated according to the number of damaged fabrics and the average damage factor, wherein the loom failure factor is the product of the number of damaged fabrics and the average damage factor.

[0054] It can be understood that the fabric damage factor refers to a value used to represent the degree of fabric damage in the complete sliding image. The fabric damage factor is a value obtained by integrating the local image clarity value and the local grayscale deviation value. The fabric damage number is the number of damaged fabric images in the damaged fabric image set. The mathematical form of the loom failure factor is expressed as: ,in, represents the loom failure factor, Indicates the amount of fabric damage. Represents the average breakage factor.

[0055] S7. Summarize the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set.

[0056] It should be explained that the loom fault factor set is a set of loom fault factors of each detected loom image in the detected loom image set.

[0057] S8. According to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model, and online image detection and fault prediction of the high-speed loom is implemented based on the loom fault prediction model.

[0058] It can be understood that the loom fault prediction model refers to a model used to predict faults of high-speed looms. The correlation coefficient of the loom prediction model is calculated through the loom fault factor set, and the required prediction time is input into the loom prediction model, so that the loom failure factor of the high-speed loom after a prediction time can be obtained. Relevant personnel can set the input prediction time as needed, and perform fault assessment on the high-speed loom through the output loom failure factor.

[0059] For example, Xiao Zhang is an operator in a weaving factory. Once, Xiao Zhang obtained a loom fault prediction model for a high-speed loom through a loom fault factor set. Then Xiao Zhang needed to predict the fault condition of the high-speed loom after 10 minutes. So Xiao Zhang input 10 minutes as the prediction duration variable into the loom fault prediction model, and the output obtained was a predicted fault factor. Xiao Zhang found that the predicted fault factor had exceeded the pre-set controllable fault factor, so Xiao Zhang reported the situation to his superiors.

[0060] In detail, the fault prediction of the high-speed loom is performed according to the loom fault factor set to obtain a loom fault prediction model, including: Determine the fault detection duration of each loom fault factor in the loom fault factor set to obtain a fault detection duration set, wherein the fault detection duration is the time interval between the shooting time of the detection weaving image corresponding to the loom fault factor and the online detection time; In the loom fault factor set and fault detection time set, respectively identify the current fault factor and current detection time corresponding to the current weaving image; According to the fault detection time set and the loom fault factor set, a fault factor change line segment is drawn, wherein the horizontal axis of the fault factor change line segment represents the fault detection time, and the vertical axis represents the loom fault factor; Setting a division precision value, dividing the fault factor change line segment based on the division precision value to obtain a division change line segment group, wherein the number of the division change line segments is the same as the division precision value; Identify the line segment slope of each partitioning change line segment in the partitioning change line segment group to obtain a line segment slope group, wherein the line segment slope is the slope value of the first and last coordinate points of the partitioning change line segment; The maximum line segment slope and the minimum line segment slope are identified in the line segment slope group, and the controllable line segment slope is set according to the maximum line segment slope and the minimum line segment slope. According to the controllable line segment slope, the current fault factor, the current detection time and the preset prediction time variable, a loom fault prediction model is constructed, wherein the controllable line segment slope and the loom fault prediction model are respectively expressed as: , in, represents the minimum line segment slope, represents the slope of the controllable line segment, represents the maximum line segment slope, represents the loom fault prediction model, represents the prediction duration variable, Indicates the current detection duration. Indicates the current fault factor.

[0061] It can be understood that the fault detection duration refers to the time interval between the shooting moment of the detection weaving image corresponding to the fault factor of the loom and the moment of starting the high-speed loom.

[0062] For example, the start-up time of a high-speed loom is 20 minutes and 20 seconds. After a period of time, the high-definition camera captures a detection weaving image of the high-speed loom. At this time, the time of capturing the detection weaving image is 30 minutes and 20 seconds. After analyzing this detection weaving image, a loom fault factor is obtained. The fault detection time corresponding to the loom fault factor is: 30 minutes and 20 seconds - 20 minutes and 20 seconds = 10 minutes.

[0063] Furthermore, the current fault factor and the current detection time refer to the loom fault factor and the fault detection time corresponding to the current weaving image respectively, the fault factor change line segment refers to the line segment used to represent the changing relationship between the fault detection time and the loom fault factor, the division accuracy value refers to an artificially set constant, and the division accuracy value represents the number of divisions when the fault factor change line segment is divided, the divided change line segment refers to the fault factor change line segment after division, wherein the detailed steps of dividing the fault factor change line segment are: dividing the fault factor change line segment into u equal parts, wherein u represents the division accuracy value, the line segment slope refers to the slope of the divided change line segment, and its numerical value is expressed as: the slope between the first and last coordinate points of the divided change line segment, and the technology of calculating the slope using two coordinate points is the prior art and will not be repeated here, and the predicted time variable refers to the independent variable in the loom fault prediction model.

[0064] In order to solve the problems described in the background technology, the present invention first starts a high-definition camera and uses the high-definition camera to continuously shoot the high-speed loom to obtain a stored weaving image set. This step ensures that the operating status of the high-speed loom can be monitored in real time. The image set obtained by continuous shooting provides basic data for subsequent image detection and fault prediction. Then, the current weaving image is obtained based on the online image detection instruction. This step improves the flexibility and timeliness of image detection and improves the accuracy of subsequent high-speed loom fault prediction. By performing sliding detection on the detected weaving image, a complete sliding image set is obtained, which realizes continuous and comprehensive analysis of the weaving image, so that the obtained complete sliding image set can represent the trend of local image changes of the detected weaving image, which improves the coverage and continuity of image detection, and thus improves the accuracy of high-speed loom fault detection. Furthermore, based on the detected weaving image, the complete sliding image is analyzed for local grayscale deviation. Analysis is performed to obtain a local grayscale data set. In this step, by analyzing the local grayscale deviation and clarity of the complete sliding image, subtle changes in the detected weaving image can be identified, providing more accurate local features for fault identification. Then, according to the local grayscale data set, loom fault identification is performed to obtain a loom fault factor, wherein the loom fault factor provides an intuitive numerical representation for evaluating the fault condition of the high-speed loom. At the same time, a corresponding relationship between the loom fault factor and the specific image is established, making the fault analysis more specific. Finally, according to the loom fault factor set, a loom fault prediction model is constructed, which can realize the automatic prediction of the future state of the loom, and can discover potential faults in advance, thereby achieving the purpose of quickly predicting faults for high-speed looms. At the same time, the construction of the loom fault prediction model also greatly reduces the consumption of human resources, thereby improving the efficiency of fault prediction for high-speed looms. Therefore, the present invention can improve the accuracy and efficiency of high-speed loom fault prediction.

[0065] like Figure 2 , which is a functional module diagram of a high-speed loom online image detection and fault prediction system provided by an embodiment of the present invention.

[0066] The high-speed loom online image detection and fault prediction system 100 of the present invention can be installed in an electronic device. According to the functions to be implemented, the high-speed loom online image detection and fault prediction system 100 can include a weaving image shooting module 101, a local image extraction module 102, a fault factor analysis module 103 and a prediction model construction module 104. The module of the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0067] The weaving image shooting module 101 is used to start a pre-built high-speed loom, and at the same time as starting the high-speed loom, start a pre-built high-definition camera, and use the high-definition camera to continuously shoot the high-speed loom to obtain a stored weaving image set, receive an online image detection instruction, set an online detection time based on the online image detection instruction, and record the image captured by the high-definition camera at the online detection time as the current weaving image; The local image extraction module 102 is used to extract a reference weaving image set from the stored weaving image set according to the online detection time, merge the current weaving image with the reference weaving image set to obtain a detection weaving image set, wherein the detection weaving images in the detection weaving image set are arranged in the order of shooting time from the earliest to the latest, and the current weaving image is ranked last in the detection weaving image set, extract detection weaving images in the detection weaving image set in turn, perform sliding detection on the detection weaving images, and obtain a complete sliding image set; The fault factor analysis module 103 is used to sequentially extract complete sliding images from the complete sliding image set, perform local grayscale deviation analysis on the complete sliding images based on the detected weaving images, and obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value, summarize the local grayscale data group of each complete sliding image, obtain a local grayscale data group set, perform loom fault identification according to the local grayscale data group set, and obtain a loom fault factor; The prediction model building module 104 is used to summarize the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set, and according to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model.

[0068] In detail, the modules in the high-speed loom online image detection and fault prediction system 100 in the embodiment of the present invention are used in the same manner as described above. Figure 1 The high-speed loom online image detection and fault prediction method described in the text are the same technical means and can produce the same technical effects, so they will not be repeated here.

[0069] like Figure 3 FIG. 1 is a schematic diagram of the structure of an electronic device for implementing an online image detection and fault prediction method for a high-speed loom provided by an embodiment of the present invention.

[0070] The electronic device 1 may include a processor 10, a memory 11 and a bus 12, and may also include a computer program stored in the memory 11 and executable on the processor 10, such as a high-speed loom online image detection and fault prediction method program.

[0071] The memory 11 includes at least one type of readable storage medium, and the readable storage medium includes a flash memory, a mobile hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a magnetic memory, a disk, an optical disk, etc. In some embodiments, the memory 11 may be an internal storage unit of the electronic device 1, such as a mobile hard disk of the electronic device 1. In other embodiments, the memory 11 may also be an external storage device of the electronic device 1, such as a plug-in mobile hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (SecureDigital, SD) card, a flash card (Flash Card), etc. equipped on the electronic device 1. Further, the memory 11 also includes an internal storage unit of the electronic device 1 and an external storage device. The memory 11 can not only be used to store application software and various types of data installed in the electronic device 1, such as the code of the high-speed loom online image detection and fault prediction method program, but also can be used to temporarily store data that has been output or is to be output.

[0072] The processor 10 may be composed of an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or a plurality of packaged integrated circuits with the same or different functions, including one or more central processing units (CPUs), microprocessors, digital processing chips, graphics processors, and combinations of various control chips, etc. The processor 10 is the control core (Control Unit) of the electronic device, and uses various interfaces and lines to connect various components of the entire electronic device, and executes or executes programs or modules (such as high-speed loom online image detection and fault prediction method programs, etc.) stored in the memory 11, and calls data stored in the memory 11 to execute various functions of the electronic device 1 and process data.

[0073] The bus 12 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The bus 12 may be divided into an address bus, a data bus, a control bus, etc. The bus 12 is configured to realize connection and communication between the memory 11 and at least one processor 10, etc.

[0074] Figure 3 Only an electronic device with components is shown, and those skilled in the art will understand that Figure 3The structure shown does not constitute a limitation on the electronic device 1, and may include fewer or more components than shown in the figure, or combine certain components, or arrange the components differently.

[0075] For example, although not shown, the electronic device 1 may also include a power source (such as a battery) for supplying power to various components. Preferably, the power source may be logically connected to the at least one processor 10 through a power management system, so that the power management system can realize functions such as charging management, discharging management, and power consumption management. The power source may also include any components such as one or more DC or AC power sources, recharging systems, power failure detection circuits, power converters or inverters, and power status indicators. The electronic device 1 may also include a variety of sensors, Bluetooth modules, Wi-Fi modules, etc., which will not be repeated here.

[0076] Furthermore, the electronic device 1 may also include a network interface. Optionally, the network interface may include a wired interface and / or a wireless interface (such as a WI-FI interface, a Bluetooth interface, etc.), which is generally used to establish a communication connection between the electronic device 1 and other electronic devices.

[0077] Optionally, the electronic device 1 may further include a user interface, which may be a display, an input unit (such as a keyboard), or a standard wired interface or a wireless interface. Optionally, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, and an OLED (Organic Light-Emitting Diode) touch device. The display may also be appropriately referred to as a display screen or a display unit, which is used to display information processed in the electronic device 1 and to display a visual user interface.

[0078] The high-speed loom online image detection and fault prediction method program stored in the memory 11 of the electronic device 1 is a combination of multiple instructions. When running in the processor 10, it can achieve: Starting a pre-built high-speed loom, and at the same time starting the high-speed loom, starting a pre-built high-definition camera, and continuously photographing the high-speed loom with the high-definition camera to obtain a stored weaving image set; receiving an online image detection instruction, setting an online detection time based on the online image detection instruction, and recording an image captured by a high-definition camera at the online detection time as a current weaving image; According to the online detection time, a reference weaving image set is extracted from the stored weaving image set, and the current weaving image is merged with the reference weaving image set to obtain a detection weaving image set, wherein the detection weaving images in the detection weaving image set are arranged in order from earliest to latest according to the shooting time, and the current weaving image is arranged at the last position in the detection weaving image set; Extracting detection woven images in sequence from the detection woven image set, performing sliding detection on the detection woven images, and obtaining a complete sliding image set; Extracting complete sliding images in the complete sliding image set in sequence, and performing local grayscale deviation analysis on the complete sliding images based on the detected fabric images to obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value; Summarizing the local grayscale data groups of each complete sliding image to obtain a local grayscale data set, and performing loom fault identification according to the local grayscale data set to obtain a loom fault factor; Summarizing the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set; According to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model, and online image detection and fault prediction of the high-speed loom are realized based on the loom fault prediction model.

[0079] Specifically, the specific implementation method of the processor 10 for the above instructions can refer to Figures 1 to 3 The description of the relevant steps in the corresponding embodiments will not be repeated here.

[0080] Furthermore, if the module / unit integrated in the electronic device 1 is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. The computer-readable storage medium can be volatile or non-volatile. For example, the computer-readable medium can include: any entity or system that can carry the computer program code, a recording medium, a USB flash drive, a mobile hard disk, a magnetic disk, an optical disk, a computer memory, and a read-only memory (ROM).

[0081] The present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor of an electronic device, the computer program can implement: Starting a pre-built high-speed loom, and at the same time starting the high-speed loom, starting a pre-built high-definition camera, and continuously photographing the high-speed loom with the high-definition camera to obtain a stored weaving image set; receiving an online image detection instruction, setting an online detection time based on the online image detection instruction, and recording an image captured by a high-definition camera at the online detection time as a current weaving image; According to the online detection time, a reference weaving image set is extracted from the stored weaving image set, and the current weaving image is merged with the reference weaving image set to obtain a detection weaving image set, wherein the detection weaving images in the detection weaving image set are arranged in order from earliest to latest according to the shooting time, and the current weaving image is arranged at the last position in the detection weaving image set; Extracting detection woven images in sequence from the detection woven image set, performing sliding detection on the detection woven images, and obtaining a complete sliding image set; Extracting complete sliding images in the complete sliding image set in sequence, and performing local grayscale deviation analysis on the complete sliding images based on the detected fabric images to obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value; Summarizing the local grayscale data groups of each complete sliding image to obtain a local grayscale data set, and performing loom fault identification according to the local grayscale data set to obtain a loom fault factor; Summarizing the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set; According to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model, and online image detection and fault prediction of the high-speed loom are realized based on the loom fault prediction model.

[0082] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only illustrative, and actual implementation may have other division methods.

[0083] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0084] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.

[0085] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0086] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A high-speed loom online image detection and fault prediction method, characterized in that: The method comprises: Starting a pre-built high-speed loom, and at the same time starting the high-speed loom, starting a pre-built high-definition camera, and continuously photographing the high-speed loom with the high-definition camera to obtain a stored weaving image set; receiving an online image detection instruction, setting an online detection time based on the online image detection instruction, and recording an image captured by a high-definition camera at the online detection time as a current weaving image; According to the online detection time, a reference weaving image set is extracted from the stored weaving image set, and the current weaving image is merged with the reference weaving image set to obtain a detection weaving image set, wherein the detection weaving images in the detection weaving image set are arranged in order from earliest to latest according to the shooting time, and the current weaving image is arranged at the last position in the detection weaving image set; Extracting detection woven images in sequence from the detection woven image set, performing sliding detection on the detection woven images, and obtaining a complete sliding image set; Extracting complete sliding images in the complete sliding image set in sequence, and performing local grayscale deviation analysis on the complete sliding images based on the detected fabric images to obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value; Summarizing the local grayscale data groups of each complete sliding image to obtain a local grayscale data set, and performing loom fault identification according to the local grayscale data set to obtain a loom fault factor; Summarizing the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set; According to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model, and online image detection and fault prediction of the high-speed loom are realized based on the loom fault prediction model.

2. The high-speed loom online image detection and fault prediction method according to claim 1, characterized in that: The high-speed loom is continuously photographed by a high-definition camera to obtain a stored weaving image set, including: Based on the preset weaving start time, a high-definition camera is used to capture images of the high-speed loom to obtain a single-shot image; Supplementing the single shot image to the pre-constructed original weaving image set to obtain a target weaving image set, wherein the target weaving images in the target weaving image set are arranged in the order of shooting time, and the target weaving images with earlier shooting time are arranged in front; Acquire the number of target images in the target weaving image set, and determine whether the number of target images is greater than a preset maximum image capacity; If the number of target images is greater than the maximum image capacity, the first weaving image is extracted from the target weaving image set, and the first weaving image is removed from the target weaving image set to obtain a removed weaving image set; After waiting for the preset shooting time interval, record the current weaving moment; The original weaving image set and the weaving start time are updated respectively by using the discarded weaving image set and the current weaving time, and the step of using a high-definition camera to take an image of the high-speed loom based on the preset weaving start time is returned until the current weaving time is the preset instruction receiving time, and the discarded weaving image set at this time is recorded as the stored weaving image set; If the number of target images is not greater than the maximum image capacity, then after waiting for a preset shooting time interval, the current weaving moment is obtained, and the original weaving image set and the weaving start moment are updated respectively using the target weaving image set and the current weaving moment; Return to the step of taking images of the high-speed loom using a high-definition camera based on the preset weaving start time until the current weaving time is the instruction receiving time, and record the target weaving image set at this time as the stored weaving image set.

3. The high-speed loom online image detection and fault prediction method according to claim 2, characterized in that: The sliding detection is performed on the detected fabric image to obtain a complete sliding image set, including: Determining a detection sliding area according to the detected fabric image; Placing the detection sliding area at a preset original image position, setting a sliding speed and identifying a lateral sliding boundary of the detection fabric image, wherein the original image position is the upper left corner of the detection fabric image; According to the sliding speed and the lateral sliding boundary, the lateral sliding is performed in the detection fabric image by using the detection sliding area to obtain a unidirectional sliding image group; According to the preset detection accuracy, the longitudinal sliding interval is calculated, wherein the longitudinal sliding interval is expressed as: ， in, Indicates the vertical sliding interval, Indicates the detection accuracy, Indicates the longitudinal dimension of the slide; Based on the longitudinal sliding interval, determining the longitudinal sliding position and identifying the longitudinal sliding boundary on the detected fabric image, wherein the longitudinal sliding position is the longitudinal sliding interval downward from the original image position in the longitudinal direction; The original image position is updated by using the longitudinal sliding position, and the step of sliding horizontally in the detection fabric image according to the sliding speed and the transverse sliding boundary and using the detection sliding area is returned until the detection sliding area contacts the transverse sliding boundary and the longitudinal sliding boundary at the same time; The one-way sliding image groups are aggregated to obtain a complete sliding image set.

4. The high-speed loom online image detection and fault prediction method according to claim 3, characterized in that: The step of determining the detection sliding area according to the detected fabric image comprises: Identify the horizontal pixel number and the vertical pixel number of the detected fabric image, and set the sliding area size according to the horizontal pixel number, the vertical pixel number and the detection accuracy, wherein the sliding area size includes: a sliding horizontal size and a sliding vertical size, and the sliding area size is expressed as: ， in, Indicates the horizontal dimension of the slide. Indicates the number of horizontal pixels. Represents the pixel diameter of the detected pixel in the fabric image, Indicates the number of vertical pixels; A detection sliding area is set based on the sliding area size.

5. The high-speed loom online image detection and fault prediction method according to claim 4, characterized in that: The method of using the detection sliding area to perform lateral sliding in the detection fabric image to obtain a unidirectional sliding image group includes: Recording the detected sliding area at the position of the original image as the original sliding image; Based on the sliding speed, the detection sliding area is slid from the original image position from left to right, and after waiting for a preset sliding detection interval, the detection fabric image where the detection sliding area is located is recorded as the detection sliding image, and the position where the detection sliding area is located at this time is recorded as the detection image position; Supplementing the detected sliding image and the original sliding image to the pre-constructed original sliding image group to obtain a target sliding image group; The original image position and the original sliding image group are updated respectively by using the detected image position and the target sliding image group, and the step of sliding the detected sliding area from the original image position from left to right based on the sliding speed is returned until the detected sliding area contacts the lateral sliding boundary, and the target sliding image group at this time is recorded as a unidirectional sliding image group.

6. The high-speed loom online image detection and fault prediction method according to claim 5, characterized in that: Based on the detection of the weaving image, the local grayscale deviation analysis is performed on the complete sliding image to obtain a local grayscale data group, including: Performing grayscale conversion on the detected fabric image to obtain a grayscale overall image, extracting an overall pixel point set from the grayscale overall image, identifying an overall grayscale value set of the overall pixel point set, and calculating an average overall grayscale value of the overall grayscale value set; The full sliding image is gray-scale converted to obtain a gray-scale local image, a local pixel point set is extracted from the gray-scale local image, and a local gray-scale value set of the local pixel point set is identified. According to the average overall gray-scale value and the local gray-scale value set, a local gray-scale deviation value is calculated, wherein the local gray-scale deviation value is expressed as: ， in, Represents the local grayscale deviation value, Represents the number of local gray values ​​in the local gray value set, Represents the first The local gray value, represents the average overall gray value; Performing damage clarity analysis on the complete sliding image to obtain a local image clarity value; The local grayscale deviation value and the local image clarity value are paired to obtain a local grayscale data set.

7. The high-speed loom online image detection and fault prediction method according to claim 6, characterized in that: The performing damage clarity analysis on the complete sliding image to obtain a local image clarity value includes: The maximum local gray value in the local gray value set is identified, and the background gray threshold is calculated according to the maximum local gray value and a preset background gray ratio, wherein the background gray threshold is expressed as: ， in, represents the background grayscale threshold, represents the maximum local gray value, Represents the background grayscale ratio; Based on the background grayscale threshold, determining a background grayscale value set in the local grayscale value set, wherein the background grayscale value in the background grayscale value set is not greater than the background grayscale threshold; The acceptable clarity ratio is set, and the local image clarity value is calculated using the following formula according to the background gray value set, the maximum local gray value and the acceptable clarity ratio: ， in, Indicates the local image clarity value, Represents the number of background gray values ​​in the background gray value set, Represents the background gray value concentration Background gray value, Indicates acceptable clarity ratio.

8. The high-speed loom online image detection and fault prediction method according to claim 7, characterized in that: The method of performing loom fault identification based on the local grayscale data set to obtain the loom fault factor includes: The local grayscale data groups are sequentially extracted from the local grayscale data group set, and the fabric damage factor is calculated based on the local grayscale data group, wherein the fabric damage factor is expressed as: ， in, represents the fabric damage factor; Determining whether the fabric damage factor is greater than a preset minimum damage factor; If the fabric damage factor is greater than the minimum damage factor, the complete sliding image is recorded as a damaged fabric image, and the damaged fabric images are summarized to obtain a damaged fabric image set; Counting the number of fabric damages of the damaged fabric images in the damaged fabric image set, and calculating an average damage factor of the damaged fabric image set, wherein the average damage factor is an average value of the fabric damage factors corresponding to all damaged fabric images in the damaged fabric image set; The loom failure factor is calculated according to the number of damaged fabrics and the average damage factor, wherein the loom failure factor is the product of the number of damaged fabrics and the average damage factor.

9. The high-speed loom online image detection and fault prediction method according to claim 8, characterized in that: The method of performing fault prediction on the high-speed loom according to the loom fault factor set to obtain a loom fault prediction model comprises: Determine the fault detection duration of each loom fault factor in the loom fault factor set to obtain a fault detection duration set, wherein the fault detection duration is the time interval between the shooting time of the detection weaving image corresponding to the loom fault factor and the online detection time; In the loom fault factor set and fault detection time set, respectively identify the current fault factor and current detection time corresponding to the current weaving image; According to the fault detection time set and the loom fault factor set, a fault factor change line segment is drawn, wherein the horizontal axis of the fault factor change line segment represents the fault detection time, and the vertical axis represents the loom fault factor; Setting a division precision value, dividing the fault factor change line segment based on the division precision value to obtain a division change line segment group, wherein the number of the division change line segments is the same as the division precision value; Identify the line segment slope of each partitioning change line segment in the partitioning change line segment group to obtain a line segment slope group, wherein the line segment slope is the slope value of the first and last coordinate points of the partitioning change line segment; The maximum line segment slope and the minimum line segment slope are identified in the line segment slope group, and the controllable line segment slope is set according to the maximum line segment slope and the minimum line segment slope. According to the controllable line segment slope, the current fault factor, the current detection time and the preset prediction time variable, a loom fault prediction model is constructed, wherein the controllable line segment slope and the loom fault prediction model are respectively expressed as: ， in, represents the minimum line segment slope, represents the slope of the controllable line segment, represents the maximum line segment slope, represents the loom fault prediction model, represents the prediction duration variable, Indicates the current detection duration. Indicates the current fault factor.

10. A high-speed loom online image detection and fault prediction system, characterized in that: The system comprises: A weaving image shooting module is used to start a pre-built high-speed loom, and at the same time as starting the high-speed loom, start a pre-built high-definition camera, and use the high-definition camera to continuously shoot the high-speed loom to obtain a stored weaving image set, receive an online image detection instruction, set an online detection time based on the online image detection instruction, and record the image captured by the high-definition camera at the online detection time as the current weaving image; A local image extraction module is used to extract a reference woven image set from a stored woven image set according to the online detection time, merge the current woven image with the reference woven image set to obtain a detection woven image set, wherein the detection woven images in the detection woven image set are arranged in the order of shooting time from earliest to latest, and the current woven image is ranked last in the detection woven image set, extract detection woven images in the detection woven image set in turn, perform sliding detection on the detection woven images, and obtain a complete sliding image set; A fault factor analysis module is used to sequentially extract complete sliding images from the complete sliding image set, perform local grayscale deviation analysis on the complete sliding images based on the detected weaving images, and obtain a local grayscale data group, wherein the local grayscale data group includes: a local grayscale deviation value and a local image clarity value, summarize the local grayscale data group of each complete sliding image, obtain a local grayscale data group set, perform loom fault identification according to the local grayscale data group set, and obtain a loom fault factor; A prediction model building module is used to summarize the loom fault factors of each detected loom image to obtain a loom fault factor set, wherein the loom fault factors in the loom fault factor set correspond one-to-one to the detected loom images in the detected loom image set, and according to the loom fault factor set, fault prediction is performed on the high-speed loom to obtain a loom fault prediction model.

Images