CN114004787A

CN114004787A - Steel wire rope damage detection method and device, terminal and storage medium

Info

Publication number: CN114004787A
Application number: CN202111109446.4A
Authority: CN
Inventors: 王浩; 孙晓腾; 李翀; 孙业栋; 王瑞明; 李兵; 王毅; 白晨; 杨媛媛; 郭荣坤
Original assignee: State Grid Corp of China SGCC; Marketing Service Center of State Grid Hebei Electric Power Co Ltd
Current assignee: State Grid Corp of China SGCC; Marketing Service Center of State Grid Hebei Electric Power Co Ltd
Priority date: 2021-09-22
Filing date: 2021-09-22
Publication date: 2022-02-01

Abstract

The invention relates to the technical field of nondestructive testing, in particular to a method, a device, a terminal and a storage medium for detecting damage of a steel wire rope, wherein the method comprises the following steps: acquiring a steel wire rope image; performing sharpening processing on the image to generate a preprocessed image; and determining whether the steel wire rope is damaged or not according to the preprocessed image. According to the embodiment of the method, the image is subjected to the sharpening processing, the definition of the image acquired under weak light and mechanical vibration can be improved, the damage identification is carried out through the sharpened image, the identification accuracy is improved, the possibility of false alarm is reduced, and the detection efficiency is improved.

Description

Steel wire rope damage detection method and device, terminal and storage medium

Technical Field

The invention relates to the technical field of nondestructive testing, in particular to a method, a device, a terminal and a storage medium for detecting damage of a steel wire rope.

Background

When the steel wire rope is used, under the action of alternating stress, the steel wires slide and wear frequently, so that the diameter of the steel wire rope is reduced, and microcracks are generated at the worn part of the traction steel wires under the action of bending stress to generate broken wires, so that the production safety problem is caused.

In the prior art, a method for preventing the production safety problem caused by the steel wire rope is to periodically check the quality of the steel wire rope, such as determining whether the steel wire rope has fuzzing and broken filaments; regularly checking the operation environment of the steel wire rope, such as determining whether the oil level of lubricating oil is within a standard range; through the operation, the steel wire rope with the hidden danger can be prevented from being further used to a certain extent and seriously abraded and broken, and finally a production accident is caused. For some application scenes requiring high reliability, the steel wire rope needs to be replaced regularly besides a method of manually observing the steel wire rope.

As known, the manual visual method has poor reliability and low efficiency; the regular replacement causes huge waste of the steel wire rope. As a statistic shows, more than 70% of the replaced cords have no strength loss.

Therefore, it is urgent to develop a nondestructive inspection technique for a steel wire rope.

Disclosure of Invention

The embodiment of the invention provides a steel wire rope damage detection method, a steel wire rope damage detection device, a terminal and a storage medium, which are used for solving the problem that the steel wire rope damage detection method in the prior art is low in efficiency.

In a first aspect, an embodiment of the present invention provides a method for detecting damage to a steel wire rope, including:

acquiring a steel wire rope image;

performing sharpening processing on the image to generate a preprocessed image;

and determining whether the steel wire rope is damaged or not according to the preprocessed image.

In one possible implementation manner, the performing sharpness processing on the image to generate a preprocessed image includes:

acquiring a trained DeblurgAN network;

and inputting the image into the DeblurgAN network to generate the preprocessed image.

In one possible implementation manner, the obtaining the trained DeblurGAN network includes:

constructing a generator and a discriminator;

acquiring a blurred image and a clear image, wherein the blurred image corresponds to the clear image;

inputting the blurred image into the generator to generate a pre-processed blurred image;

a judging step: inputting the preprocessed blurred image and the sharp image into the discriminator to generate a discrimination result;

and according to the judgment result, the generator generates a preprocessed blurred image again, and jumps to the judgment step until the preset times are reached.

In one possible implementation, the acquiring the blurred image and the sharp image includes:

aligning the first image acquisition device and the second image acquisition device to the same part of the steel wire rope;

starting the steel wire rope to enable the steel wire rope to move according to a working state;

acquiring a first image from the first image acquisition device and a second image from the second image acquisition device, wherein the first image and the second image are acquired at the same time, and the shutter speed of the first image acquisition device is lower than that of the second image acquisition device;

the first image is a blurred image, and the second image is a sharp image.

In one possible implementation manner, the determining whether the steel wire rope generates the damage according to the preprocessed image includes:

acquiring a trained YOLOV3 network;

carrying out capacity expansion on the preprocessed image to form a capacity expansion image set;

and inputting the expansion image set into the YOLOV3 network to determine whether the steel wire rope is damaged.

In one possible implementation, the obtaining a trained YOLOV3 network includes:

acquiring a steel wire rope damage image set;

expanding the damage image set in a turning and/or rotating mode to obtain an expanded damage image set;

inputting the volume-expanded damage image set into the Yolov3 network for training to form the trained Yolov3 network.

In one possible implementation manner, the expanding the preprocessed image to form an expanded image set includes:

and expanding the capacity of the preprocessed image in a turning and/or rotating mode to form the capacity-expanded image set.

In a second aspect, an embodiment of the present invention provides a device for detecting damage to a steel wire rope, including:

the image acquisition module is used for acquiring a steel wire rope image;

the image processing module is used for carrying out sharpening processing on the image to generate a preprocessed image; and the number of the first and second groups,

and the damage determining module is used for determining whether the steel wire rope generates damage according to the preprocessed image.

In a third aspect, an embodiment of the present invention provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, where the processor implements the steps of the method according to the first aspect or any one of the possible implementation manners of the first aspect when executing the computer program.

In a fourth aspect, the present invention provides a computer-readable storage medium, which stores a computer program that, when executed by a processor, implements the steps of the method as described in the first aspect or any one of the possible implementations of the first aspect.

Compared with the prior art, the implementation mode of the invention has the following beneficial effects:

according to the steel wire rope damage detection method disclosed by the embodiment of the invention, the image is subjected to the sharpening treatment, so that the definition of the image acquired under weak light and mechanical vibration can be improved, the damage identification is carried out through the sharpened image, the identification accuracy is improved, the possibility of false alarm is reduced, and the detection efficiency is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the embodiments or the prior art description will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art that other drawings can be obtained according to these drawings without inventive labor.

Fig. 1 is a flowchart of a method for detecting damage to a steel wire rope according to an embodiment of the present invention;

FIG. 2 is a diagram of a DeblurgAN network architecture according to an embodiment of the present invention;

FIG. 3 is a block diagram of a YOLOV3 network according to an embodiment of the present invention;

fig. 4 is a functional block diagram of a steel wire rope damage detection device according to an embodiment of the present invention;

fig. 5 is a functional block diagram of a terminal according to an embodiment of the present invention.

Detailed Description

In the following description, for purposes of explanation and not limitation, specific details are set forth, such as particular system structures, techniques, etc. in order to provide a thorough understanding of the embodiments of the invention. It will be apparent, however, to one skilled in the art that the present invention may be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known systems, devices, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

In the prior art, the steel wire rope nondestructive detection method is mainly divided into a detection method based on magnetic flux leakage, a detection method based on acoustic emission, a detection method based on computer vision and the like. The detection method based on magnetic flux leakage mainly judges local damage such as broken wires, pitting corrosion and the like through a magnetic flux leakage technology. The detection method based on acoustic emission mainly detects the wire breakage and elastic deformation of the steel rope by measuring elastic waves caused by the structural change of the steel wire rope. The detection method based on computer vision mainly utilizes a CCD (Charge Coupled Device, CCD for short) camera to shoot an optical image of the surface of the steel wire rope, and extracts the wire breaking characteristics through the computer vision method to detect the wire breaking.

The detection device based on computer vision is simple in structure, the same set of device can adapt to steel wire ropes of different sizes to compare with magnetic flux leakage detection, and detection results are basically not affected by a lift-off value. Compared with acoustic emission detection, the device is easier to design and manufacture.

However, computer vision detection also has inherent defects, such as low detection accuracy, and the false alarm often brings trouble to normal production.

One of the creative efforts of the applicant of the present invention is to find that the image collected in real time is blurred and the quality cannot be guaranteed due to the slight vibration of the camera caused by the working of the large hoisting equipment and the uncertainty of the movement of the steel rope due to the particularity of the working environment condition of the steel rope.

The second creative work of the applicant of the present invention is that one point closely related to the detection accuracy is as follows: if the image is subjected to the sharpening processing, the possibility of false alarm can be greatly reduced on the premise that the image quality is guaranteed.

The third creative work of the applicant of the present invention is to adopt which processing mode to preprocess the acquired target image, to keep the clearer texture and abundant details in the image and to be closer to the real image, and to reduce the influence of the motion on the image acquisition to the maximum extent, thereby improving the accuracy rate of steel rope defect identification and the robustness of the judgment system.

In order to make the objects, technical solutions and advantages of the present invention more apparent, the following description is made with reference to the accompanying drawings.

The following is a detailed description of the embodiments of the present invention, which is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.

Fig. 1 is a flowchart of a method for detecting damage to a steel wire rope according to an embodiment of the present invention.

As shown in fig. 1, which shows an implementation flowchart of a method for detecting damage to a steel wire rope according to an embodiment of the present invention, the method is detailed as follows:

in step 101, a wire rope image is acquired.

In step 102, the image is sharpened to generate a preprocessed image.

In some embodiments, step 102 comprises: and acquiring the trained DeblurgAN network.

Wherein the obtaining of the trained DeblurGAN network comprises:

a generator and an arbiter are constructed.

And acquiring a blurred image and a clear image, wherein the blurred image corresponds to the clear image.

And inputting the blurred image into the generator to generate a pre-processed blurred image.

A judging step: and inputting the preprocessed blurred image and the clear image into the discriminator to generate a discrimination result.

Wherein the acquiring of the blurred image and the sharp image includes:

and aligning the first image acquisition device and the second image acquisition device to the same part of the steel wire rope.

And starting the steel wire rope to enable the steel wire rope to move according to the working state.

And acquiring a first image from the first image acquisition device and a second image from the second image acquisition device, wherein the first image and the second image are acquired at the same time, and the shutter speed of the first image acquisition device is lower than that of the second image acquisition device.

The first image is a blurred image, and the second image is a sharp image.

Illustratively, the theory of generating a countermeasure network (GAN) is a zero-sum (zero-sum) game, and from the viewpoint of the network itself, two parties of the game are a Generator (Generator, abbreviated as G) and a Discriminator (Discriminator, abbreviated as D), and the game is continuously played by the two parties of the game G and D, so that the network can generate data with false or false.

The DeblurGAN is a method for blind motion blur removal based on the GAN method, which regards Image blur removal as a special Image2Image task, and performs training and learning based on wGAN and content loss, thereby achieving excellent performance in terms of SSIM and visual effect.

The basic principle is as follows: the sharp image and the blurred image are a pair of data in a training data set, when the blurred image is sent into the generator, the generator can reconstruct a generated image, each training can calculate the countermeasure loss and the content loss between the sharp image and the image generated by the generator in the data set, and the countermeasure loss is used for restricting the generated image to have high definition and good authenticity; content loss is used to restrict the generated image from being consistent with the sharp image in the data set in content. The classifier outputs a reconstructed clear image as a probability value of a real image, the generator or the classifier is determined to continue training according to the Wassertein distance, the Wassertein distance is the distance between two probability distributions, namely the minimum consumption of the two probability distributions under an optimal path in a visual way, the smaller the value of the Wassertein distance is, the better the generalization performance of the network is represented, the quality of the image output by the generator is more real, and after training is completed, the network G can be generated to be a vivid image.

As shown in fig. 2, the specific composition of the DeblurGAN, wherein the generator main body adopts a structure of a residual network, the inside of the network is composed of a convolution module of a basic unit of a convolution neural network, a normalization module and an activation function module, and the discriminator is composed of a convolution module of a basic unit of a convolution neural network, a normalization module and an activation function module.

Specifically, the generator network structure comprises a convolution module, an up-sampling and down-sampling module, a residual error module and an output module; after 1-layer convolution (7 × 7Conv), an example normalization layer (InstanceNorm) and an activation layer (ReLU), the input blurred picture Im keeps the size of input data unchanged; then, 2 times of Upsampling (Upsampling) is carried out, and the number of image feature layers is increased to 256; then, 9 layers of residual convolutions (each residual convolution comprises a 3 × 3 convolution layer 3 × 3conv, an example normalization layer InstanceNorm and a ReLU activation layer) are performed; then, 2 times of Downsampling (Downsampling), 1 layer of convolution (7 × 7) and an activation function (tanh) are carried out and output; the discriminator network structure comprises a convolution module, a down-sampling module and an output module; consists of 1 convolution (4 × 4Conv), 4 down-samples (Upsampling), 1 convolution (1 × 1Conv) and an activation function (sigmoid); the inputs to the discriminator network D are: a clear picture Iq corresponding to the fuzzy picture Im and the fuzzy picture Im pass through a generator G to obtain a steel rope generated picture Om; and the discriminator network D judges the similarity of the two input pictures.

For the present patent application, in an implementable technical solution, the generator and the discriminator are trained respectively by using the blurred image and the sharp image until the loss convergence of the final discriminator output tends to 0. The parameters of the fixed generator and the discriminator are the trained DeblurgAN network.

A method for acquiring a blurred image and a sharp image is to align two image acquisition devices, such as cameras or cameras, to the same position of a steel wire rope, so that the image areas captured by the two image acquisition devices are completely consistent.

The device is then started so that the wire rope moves according to its operating state. Then, the two image acquisition devices are respectively adjusted to enable the shutter speed of one image acquisition device to be lower than that of the other image acquisition device, the image acquired by the image acquisition device with the low shutter speed is a visual blurred image, and the image acquired by the image acquisition device with the high shutter speed is a visual clear image.

And finally, at the same moment, respectively acquiring two images from the two image acquisition devices, wherein the image acquired by the image acquisition device with the slow shutter speed is a blurred image, and the image acquired by the image acquisition device with the fast shutter speed is a clear image. And two images acquired at the same time are corresponding images.

And repeating the steps to obtain a plurality of groups of images to form two data sets, wherein one data set is a fuzzy image data set, and the other data set is a clear image data set.

A blurred image and a sharp image corresponding thereto are inputted to the generator and the discriminator, respectively. The generator processes the blurred image according to the initial parameters to generate a preprocessed blurred image, the preprocessed blurred image is compared with the clear image through the discriminator to output a discrimination result, the discrimination result is used as a parameter basis of the generator, the generator generates the preprocessed blurred image again, and the preprocessed blurred image is input into the discriminator again to be compared with the clear image. And reciprocating in such a way until a preset reciprocating number is reached.

One possible way of determining the predetermined number of reciprocations is to determine by observing the loss value of the discriminator. For example, there is a possibility that the loss value decreases rapidly in the initial period and gradually becomes stable in the later period, and when the loss converges to 0, the total is repeated N times, and the predetermined number of reciprocating times is determined as N.

In step 103, it is determined whether the wire rope is damaged according to the preprocessed image.

In some embodiments, step 103 comprises:

a trained YOLOV3 network is obtained.

And expanding the capacity of the preprocessed image to form a capacity-expanded image set.

Wherein the obtaining a trained YOLOV3 network comprises:

acquiring a steel wire rope damage image set;

Wherein, the expanding the preprocessed image to form an expanded image set includes:

Illustratively, the YOLOV3 model is proposed based on YOLOV1 and YOLOV2, and as shown in fig. 3, the model increases a feature extraction network from a Darknet-19 layer to a Darknet-53 layer, and on the basis, introduces the idea of multi-scale detection, and increases a multi-feature Fusion Pyramid (FPN) structure in the network, thereby improving the detection level of small targets.

In a YOLOV3 network model, a Darknet-53 feature extraction network utilizes a depth residual error network to add network depth to a 53-layer, so that the problem of gradient explosion caused by network deepening is avoided while effective expression of features is ensured.

The Darknet-53 feature extraction network is shown as a dashed box on the left of FIG. 3, and comprises 5 different residual blocks, wherein a convolution operation is arranged in the middle of each residual block, and the convolution in the graph represents convolution; the multiple under the residual block in the figure indicates that this residual block has been repeated several times, e.g. (X4) indicates that this block has been repeated 4 times; after the Darknet-53 network performs convolution operation with a convolution kernel size of 3x3x32 on an input picture, convolution calculation with a convolution kernel size of 3x3x64 and a moving stride of 2 is performed, and then the input picture is subjected to convolution operation for 53 times respectively through 1 residual block, 2 residual blocks, 8 residual blocks and 4 residual blocks, which is also the source of the Darknet-53 network name.

The network structure of the multi-feature pyramid performs information fusion on the feature map of the lowest layer after up-sampling the feature map of the lowest layer and the feature map of the high layer, and predicts the target to be measured by using the feature maps of different levels. The information fusion is generally to perform up-sampling operation on a feature map of a low-level feature map by using a bilinear interpolation algorithm, and then to fuse the up-sampled feature with a previous-level feature.

For the present patent application, first, a damage image of the steel wire rope is acquired, the damage image including broken wire, worn and deformed steel wire rope, the images constituting a damage image set. And the images are rotated and turned, so that an image set of multiple times of the damaged image set is further obtained, and the expansion of the damaged image is realized. Then, labeling is respectively carried out on each image, and the emphasis is used for labeling the coordinates of the defects and the types of the defects in the images. And finally, inputting the marked image into a built YOLOV3 network, training the network, and selecting proper model training parameters before the training step so as to achieve the best convergence effect as possible with the least training time.

After the YOLOV3 network training is completed, the effect after the training can be verified by obtaining a verification set, for example, in an implementable mode, M images are collected in the verification set, wherein N images are damaged steel wire rope images, the trained YOLOV3 network is used for identification, if the accuracy reaches a target value, such as 98%, the training is determined to be successful, otherwise, the training is performed again.

When the trained YOLOV3 network is put into use, if the acquired images generate a plurality of images in a turning and/or rotating mode, the acquired images are expanded, and then the expanded images are input into the trained YOLOV3 network, so that the identification accuracy can be improved.

According to the embodiment of the steel wire rope damage identification and detection method, the image is subjected to the sharpening processing, the definition of the image acquired under weak light and mechanical vibration can be improved, the damage identification is performed through the sharpened image, the identification accuracy is improved, the possibility of false alarm is reduced, and the detection efficiency is improved.

According to the embodiment of the steel wire rope damage identification and detection method, the image is subjected to definition processing by adopting the trained DeblurGAN network, the method is high in processing speed and good in effect, and the method is particularly suitable for an image processing mode in which the image is blurred due to a main factor of a certain specific reason.

According to the embodiment of the steel wire rope damage identification and detection method, the fuzzy image close to a real application scene is generated through simulation of the image acquisition device and is used as a training set to train the DeblurgAN network, so that a large number of images can be conveniently acquired, and the training effect is better compared with the fuzzy image processed through software.

According to the embodiment of the steel wire rope damage identification and detection method, the defects of the steel wire rope are identified by adopting the YOLOV3 network, and the images in the training set are expanded in a turning and/or rotating mode, so that the number of samples in the training set is more, and the training effect is better. By means of overturning and/or rotating, the adopted images are expanded, and identification accuracy is improved.

It should be understood that the sequence numbers of the steps in the above embodiments do not mean the execution sequence, and the execution sequence of each process should be determined by the function and the inherent logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

The following are apparatus embodiments of the invention, and for details not described in detail therein, reference may be made to the corresponding method embodiments described above.

Fig. 4 is a functional block diagram of a wire rope damage detection device according to an embodiment of the present invention, and referring to fig. 4, the wire rope damage detection device 4 includes: an image acquisition module 401, an image processing module 402, and a damage determination module 403;

the image acquisition module 401 is used for acquiring a steel wire rope image;

an image processing module 402, configured to perform sharpening on the image to generate a preprocessed image;

and a damage determining module 403, configured to determine whether the steel wire rope is damaged according to the preprocessed image.

Fig. 5 is a functional block diagram of a terminal according to an embodiment of the present invention. As shown in fig. 5, the terminal 5 of this embodiment includes: a processor 500, a memory 501 and a computer program 502 stored in said memory 501 and executable on said processor 500. The processor 500 executes the computer program 502 to implement the above-mentioned wire rope damage detection method and the steps of the embodiments of the wire rope damage detection method, such as the steps 101 to 103 shown in fig. 1.

Illustratively, the computer program 502 may be partitioned into one or more modules/units that are stored in the memory 501 and executed by the processor 500 to implement the present invention.

The terminal 5 may be a desktop computer, a notebook, a palm computer, a cloud server, or other computing devices. The terminal 5 may include, but is not limited to, a processor 500, a memory 501. It will be appreciated by those skilled in the art that fig. 5 is only an example of a terminal 5 and does not constitute a limitation of the terminal 5 and may include more or less components than those shown, or some components in combination, or different components, for example the terminal may also include input output devices, network access devices, buses, etc.

The Processor 500 may be a Central Processing Unit (CPU), other general purpose Processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA) or other Programmable logic device, discrete Gate or transistor logic device, discrete hardware component, or the like. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 501 may be an internal storage unit of the terminal 5, such as a hard disk or a memory of the terminal 5. The memory 501 may also be an external storage device of the terminal 5, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) and the like provided on the terminal 5. Further, the memory 501 may also include both an internal storage unit and an external storage device of the terminal 5. The memory 501 is used for storing the computer program and other programs and data required by the terminal. The memory 501 may also be used to temporarily store data that has been output or is to be output.

It will be apparent to those skilled in the art that, for convenience and brevity of description, only the above-mentioned division of the functional units and modules is illustrated, and in practical applications, the above-mentioned function distribution may be performed by different functional units and modules according to needs, that is, the internal structure of the apparatus is divided into different functional units or modules to perform all or part of the above-mentioned functions. Each functional unit and module in the embodiments may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit, and the integrated unit may be implemented in a form of hardware, or may be implemented in a form of software functional unit. In addition, specific names of the functional units and modules are only for convenience of distinguishing from each other, and are not used for limiting the protection scope of the present application. The specific working processes of the units and modules in the system may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

In the above embodiments, the description of each embodiment is focused on, and for parts that are not described or illustrated in detail in a certain embodiment, reference may be made to the description of other embodiments.

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

In the embodiments provided in the present invention, it should be understood that the disclosed apparatus/terminal and method may be implemented in other manners. For example, the above-described apparatus/terminal embodiments are merely illustrative, and for example, the division of the modules or units is only one logical division, and there may be other divisions in actual implementation, for example, multiple units or components may be combined or integrated into another system, or some features may be omitted, or not executed. In addition, the shown or discussed mutual coupling or direct coupling or communication connection may be an indirect coupling or communication connection through some interfaces, devices or units, and may be in an electrical, mechanical or other form.

The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment.

In addition, functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may exist alone physically, or two or more units are integrated into one unit. The integrated unit can be realized in a form of hardware, and can also be realized in a form of a software functional unit.

The integrated modules/units, if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium. Based on such understanding, the present invention may also be implemented by implementing all or part of the flow in the method according to the above embodiment, and instructing relevant hardware to complete the implementation through a computer program, where the computer program may be stored in a computer readable storage medium, and when the computer program is executed by a processor, the steps of the embodiments of the wire rope damage detection method and the wire rope damage detection device may be implemented. Wherein the computer program comprises computer program code, which may be in the form of source code, object code, an executable file or some intermediate form, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, usb disk, removable hard disk, magnetic disk, optical disk, computer Memory, Read-Only Memory (ROM), Random Access Memory (RAM), electrical carrier wave signals, telecommunications signals, software distribution medium, and the like. It should be noted that the computer readable medium may contain other components which may be suitably increased or decreased as required by legislation and patent practice in jurisdictions, for example, in some jurisdictions, computer readable media which may not include electrical carrier signals and telecommunications signals in accordance with legislation and patent practice.

The above embodiments are only used for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; such modifications and substitutions do not substantially depart from the spirit and scope of the embodiments of the present invention, and are intended to be included within the scope of the present invention.

Claims

1. A steel wire rope damage detection method is characterized by comprising the following steps:

acquiring a steel wire rope image;

performing sharpening processing on the image to generate a preprocessed image;

2. The method for detecting damage to a steel wire rope according to claim 1, wherein the step of performing sharpness processing on the image to generate a preprocessed image comprises:

acquiring a trained DeblurgAN network;

3. The method according to claim 2, wherein the obtaining of the trained DeblurGAN network comprises:

constructing a generator and a discriminator;

4. The method for detecting damage to a steel wire rope according to claim 3, wherein the acquiring of the blurred image and the sharp image includes:

the first image is a blurred image, and the second image is a sharp image.

5. The method for detecting damage to a steel wire rope according to any one of claims 1 to 4, wherein the determining whether the steel wire rope is damaged or not according to the preprocessed image comprises:

acquiring a trained YOLOV3 network;

6. The method according to claim 5, wherein the obtaining of the trained YOLOV3 network comprises:

acquiring a steel wire rope damage image set;

7. The method for detecting damage to a steel wire rope according to claim 5, wherein the expanding the preprocessed image to form an expanded image set includes:

8. A wire rope damage detection device, characterized by includes:

the image acquisition module is used for acquiring a steel wire rope image;

9. A terminal comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any of the preceding claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, in which a computer program is stored which, when being executed by a processor, carries out the steps of the method according to any one of claims 1 to 7.