CN115629127B

CN115629127B - Container defect analysis method, device and equipment and readable storage medium

Info

Publication number: CN115629127B
Application number: CN202211620867.8A
Authority: CN
Inventors: 杨雷
Original assignee: Beijing Zhonghai Xingda Construction Co ltd
Current assignee: Beijing Zhonghai Xingda Construction Co ltd
Priority date: 2022-12-16
Filing date: 2022-12-16
Publication date: 2023-04-14
Anticipated expiration: 2042-12-16
Also published as: CN115629127A

Abstract

The invention provides a container defect analysis method, a device, equipment and a readable storage medium, wherein the method comprises the steps of obtaining a first time domain signal, a second time domain signal and a third time domain signal, wherein the first time domain signal is a signal acquired after ultrasonic waves pass through a container to be detected, the second time domain signal is a signal acquired after the ultrasonic waves pass through a defect-free container, and the third time domain signal is a signal acquired after the ultrasonic waves returned from a defect position and received at an ultrasonic wave transmitting end; preprocessing the first time domain signal to obtain a first spectrogram, and preprocessing the second time domain signal to obtain a second spectrogram; according to the method and the device, whether the container to be detected has defects or not is judged according to the first frequency spectrogram and the second frequency spectrogram, and if the container to be detected has defects, the defect position of the container is determined according to the third time domain signal.

Description

Container defect analysis method, device and equipment and readable storage medium

Technical Field

The invention relates to the technical field of container detection, in particular to a container defect analysis method, device and equipment and a readable storage medium.

Background

Because the climate in some areas of our country is changeable, the temperature and humidity change range is large, and in addition, in some areas, the environmental pollution is serious, and acid rain is frequent, so that various defects are easily formed on the container due to the action of corrosive environment and the collision of the building materials to the container in the process of transporting the building materials of the automobile container, therefore, a container defect analysis method is urgently needed, the defect analysis method is used for positioning the defect on the container and identifying the defect type, corresponding repair measures are taken for the defects of different types, and the safe transportation of the building materials is ensured.

Disclosure of Invention

The invention aims to provide a container defect analysis method, a device, equipment and a readable storage medium, so as to improve the problems.

In order to achieve the above object, the embodiments of the present application provide the following technical solutions:

in one aspect, an embodiment of the present application provides a container defect analysis method, where the method includes:

acquiring a first time domain signal, a second time domain signal and a third time domain signal, wherein the first time domain signal is a signal acquired after ultrasonic waves pass through a container to be detected, the second time domain signal is a signal acquired after the ultrasonic waves pass through a flawless container, and the third time domain signal is a signal acquired by the ultrasonic waves returned from a defect position and received at an ultrasonic transmitting end;

preprocessing the first time domain signal to obtain a first spectrogram, and preprocessing the second time domain signal to obtain a second spectrogram;

and judging whether the container to be detected has defects according to the first frequency spectrogram and the second spectrogram, wherein if the container to be detected has defects, the defect position of the container is determined according to the third time domain signal.

In a second aspect, the present application provides a container defect analysis apparatus, including:

the device comprises a first acquisition module, a second acquisition module and a third acquisition module, wherein the first acquisition module is used for acquiring a first time domain signal, a second time domain signal and a third time domain signal, the first time domain signal is acquired after ultrasonic waves pass through a container to be detected, the second time domain signal is acquired after the ultrasonic waves pass through a defect-free container, and the third time domain signal is acquired after the ultrasonic waves received at an ultrasonic transmitting end return from a defect position;

the processing module is used for preprocessing the first time domain signal to obtain a first spectrogram and preprocessing the second time domain signal to obtain a second spectrogram;

and the judging module is used for judging whether the container to be detected has defects according to the first frequency spectrogram and the second frequency spectrogram, wherein if the container to be detected has defects, the position of the defect of the container is determined according to the third time domain signal.

In a third aspect, an embodiment of the present application provides a container defect analysis apparatus, which includes a memory and a processor. The memory is used for storing a computer program; the processor is used for realizing the steps of the container defect analysis method when the computer program is executed.

In a fourth aspect, the present application provides a readable storage medium, on which a computer program is stored, where the computer program is executed by a processor to implement the steps of the container defect analysis method.

The invention has the beneficial effects that:

1. according to the invention, whether the container has defects or not can be preliminarily analyzed through the first time domain signal and the second time domain signal, so that the container with the defects is preliminarily screened, and then the container with the defects is further analyzed through the third time domain signal, so that the defect position is accurately positioned.

2. According to the method, the image at the defect position is sliced through the window with the preset size, then the data of the image information at the defect position after slicing is cleaned, the image which does not contain the defect is filtered, the expansion of a small sample set is realized, and the precision of identifying the defect image is effectively improved.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by the practice of the embodiments of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the embodiments will be briefly described below, it should be understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and for those skilled in the art, other related drawings can be obtained according to the drawings without inventive efforts.

Fig. 1 is a schematic flow chart of a container defect analysis method according to an embodiment of the present invention.

Fig. 2 is a schematic structural diagram of a container defect analysis apparatus according to an embodiment of the present invention.

Fig. 3 is a schematic structural diagram of a container defect analysis apparatus according to an embodiment of the present invention.

The drawing is marked with: 901. a first acquisition module; 902. a processing module; 903. a judgment module; 904. a second acquisition module; 905. building a module; 906. an expansion module; 907. a training module; 908. an identification module; 9031. a first processing unit; 9032. a judgment unit; 9033. a first acquisition unit; 9034. a second processing unit; 9035. a third processing unit; 9036. a fourth processing unit; 9061. a second acquisition unit; 9062. a fifth processing unit; 9063. a sixth processing unit; 9064. a seventh processing unit; 9071. an eighth processing unit; 9072. a ninth processing unit; 9073. a tenth processing unit; 800. a container defect analysis device; 801. a processor; 802. a memory; 803. a multimedia component; 804. an I/O interface; 805. a communication component.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. The components of embodiments of the present invention generally described and illustrated in the figures herein may be arranged and designed in a wide variety of different configurations. Thus, the following detailed description of the embodiments of the present invention, presented in the figures, is not intended to limit the scope of the invention, as claimed, but is merely representative of selected embodiments of the invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that: like reference numbers and letters refer to like items in the following figures, and thus, once an item is defined in one figure, it need not be further defined and explained in subsequent figures. Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only for distinguishing the description, and are not to be construed as indicating or implying relative importance.

Example 1:

the embodiment provides a container defect analysis method, and it can be understood that in the embodiment, a scene may be laid, for example, a scene in which a container loaded with a building material needs to be subjected to defect detection before the building material is transported for a long distance.

Referring to fig. 1, it is shown that the method includes step S1, step S2 and step S3, wherein specifically:

s1, acquiring a first time domain signal, a second time domain signal and a third time domain signal, wherein the first time domain signal is acquired after ultrasonic waves pass through a container to be detected, the second time domain signal is acquired after the ultrasonic waves pass through a flawless container, and the third time domain signal is acquired after the ultrasonic waves returned from a flawed position are received at an ultrasonic transmitting end;

it can be understood that the two ends of the container to be detected are respectively provided with a signal excitation point and a signal receiving point, the ultrasonic waves are transmitted at the signal excitation point, the signals acquired by the ultrasonic waves are received at the signal receiving points, the first time domain signals are the signals acquired after the ultrasonic waves pass through the container to be detected, and the second time domain signals are the signals acquired after the ultrasonic waves pass through the container without defects.

S2, preprocessing the first time domain signal to obtain a first spectrogram, and preprocessing the second time domain signal to obtain a second spectrogram;

it is understood that the spectrogram corresponding to the ultrasonic signal data can be obtained by processing the ultrasonic signal data in the time domain by using FFT (fast fourier transform), wherein the processing of the ultrasonic signal data in the time domain by using FFT (fast fourier transform) is well known to those skilled in the art and therefore will not be described herein in detail.

And S3, judging whether the container to be detected has defects according to the first frequency spectrogram and the second frequency spectrogram, wherein if the container to be detected has defects, the defect position of the container is determined according to the third time domain signal.

It can be understood that step S3 further includes step S31 and step S32, where:

step S31, determining amplitude information of a first spectrogram according to the first spectrogram, and determining amplitude information of a second spectrogram according to the second spectrogram;

it can be understood that, after converting the time-domain signal into the spectrogram, the corresponding amplitude information can be obtained according to the spectrogram, that is, the amplitude information of the first spectrogram can be obtained according to the first spectrogram, and the amplitude information of the second spectrogram can be obtained according to the second spectrogram.

Step S32, judging whether the container to be detected has defects according to the amplitude information of the first spectrogram and the amplitude information of the second spectrogram, wherein if the amplitude information of the first spectrogram is lower than the amplitude information of the second spectrogram, judging that the container to be detected has defects; and if the amplitude information of the first frequency spectrogram is not lower than that of the second frequency spectrogram, judging that the container to be detected has no defects.

It can be understood that, when the defective containers and the non-defective containers are compared, the fourier coefficients of the defective containers are reduced, that is, the amplitude information of the first frequency spectrum diagram is lower than the amplitude information of the second frequency spectrum diagram, so that whether the containers are defective or not can be determined, and the defective containers can be primarily screened.

It can be understood that step S3 further includes step S33, step S34, step S35 and step S36, where:

step S33, acquiring preset group velocity information;

step S34, determining first information according to an excitation signal corresponding to a third time domain signal, wherein the first information includes a time value corresponding to the maximum amplitude of the excitation signal, and determining second information according to the third time domain signal, wherein the second information includes a time value corresponding to the maximum amplitude of the third time domain signal;

it can be understood that, after the ultrasonic wave detects the defect, a part of the frequency components will be transmitted, and the other part of the frequency components will be reflected or scattered, so that there is a part of energy reflected to excite the signal end to generate a third time domain signal, a time value corresponding to the maximum amplitude position of the excitation signal can be obtained according to the excitation signal of the third time domain signal, and a time value corresponding to the maximum amplitude position of the third time domain signal can be obtained according to the third time domain signal;

step S35, subtracting the first information from the second information to obtain third information;

it is understood that subtracting the time value corresponding to the maximum amplitude of the excitation signal from the time value corresponding to the maximum amplitude of the third time domain signal results in a time difference between the first information and the second information.

And S36, calculating according to the third information and the preset group velocity information to obtain the defect position of the container.

It can be understood that the defect position of the container is obtained by calculating according to the third information and the preset group velocity information, specifically:

in the above formula, X represents the calculated defect position, t represents the difference between the time value corresponding to the maximum amplitude position of the excitation signal and the time value corresponding to the maximum amplitude position of the third time domain signal, i.e., the third information, and V is the preset group velocity information.

It can be understood that, after the step S3, a step S4, a step S5, a step S6, a step S7, and a step S8 are further included, where specifically:

s4, acquiring image information of at least one defect position according to the defect position of the container;

it can be understood that after determining the defect position of the container, the defect of the container needs to be identified so as to take corresponding repair measures to ensure safe transportation of the building material, and it should be noted that the image information at the defect position should include defect images with different angles and different brightness.

S5, constructing a sample set based on the image information at the defect position;

it will be appreciated that the sample set is constructed based on different angles, different intensities of defect images for subsequent training of the model.

S6, expanding the sample set to obtain an expanded sample set;

it can be understood that step S6 further includes step S61, step S62, step S63, and step S64, where:

s61, acquiring a window with a preset size;

it will be appreciated that windows of size X Y are provided, where X is the width of the sliding window and Y is the height of the sliding window, the particular size being set according to the size of the image at the defect location.

S62, performing sliding slicing on the image information at the defect position according to the window with the preset size to obtain the image information at the defect position after at least one slice;

it can be understood that, the sliding window method is used to process the image information at the defect position, and the scaling operation on the image is not required, so that the problem of missing the defect target information is effectively solved, wherein the sliding window method is used to process the image information at the defect position is a technical solution known by those skilled in the art, and is not described herein.

S63, carrying out data cleaning on the image information at the defect position after slicing to obtain the image information at the defect position after cleaning;

it can be understood that the image information at the defect position after slicing, which is generated after processing the image information at the defect position by using the sliding window method, may include some background images with very small defect proportion and without defects, so that the sample is doped with noise images in a concentrated manner, and therefore, data cleaning needs to be performed on the image information at the defect position after slicing to ensure the effect of the sample on model training, where one cleaning manner may be to preliminarily filter the image including only the background by using a classification model, and then screen out the image with very small defect proportion by using a manual cleaning manner.

And S64, expanding the sample set based on the image information of the cleaned defect position to obtain an expanded sample set.

It can be understood that, the expanding the sample set based on the image information at the defect position after the cleaning is specifically: obtaining a noise sample, wherein the noise sample comprises random noise; sending the image information at the cleaned defect position and the noise sample to a generation model to generate a countermeasure sample; judging whether the countermeasure sample can be filled into a sample set or not by using a discrimination model, if the countermeasure sample can be filled into the sample set, sending the countermeasure sample to a generation model, and expanding the sample set according to the countermeasure sample sent to the generation model to obtain an expanded sample set, wherein the generation model and the judgment model are both models after training is completed, and the corresponding training modes are schemes well known by those skilled in the art, and are not repeated here.

In this embodiment, the image at the defect position is sliced through the window with the preset size, and then the data of the image information at the defect position after slicing is cleaned, so as to filter the image without the defect, thereby realizing the expansion of the small sample set and effectively improving the precision of identifying the defect image.

S7, training a defect recognition model by using the extended sample set to obtain a trained defect recognition model;

it can be understood that step S7 further includes step S71, step S72 and step S73, where specifically:

s71, performing feature extraction on the images included in the extended sample set by using a convolutional neural network to obtain a feature map corresponding to each image;

it can be understood that, extracting the multi-scale features of the images included in the extended sample set by using convolution kernels of different scales to obtain three first feature maps, where the first feature maps include feature maps corresponding to feature information extracted by using convolution kernels of 1X1, 3X3, and 5X5 scales; the three first feature maps are spliced in a cascading mode to obtain a fused second feature map, namely the feature map corresponding to each image, and it should be noted that by fusing the feature maps with different scales, the details of the defects of the container can be effectively improved to ensure the accuracy of defect identification.

Step S72, inputting the characteristic diagram into a preset defect identification model to obtain an identification result corresponding to each image;

it can be understood that, in the defect identification model, softmax is used to calculate the probability value of each feature map identified as the corresponding preset defect classification label, where the identification result may be scratches, rust, cracks, deformations, and holes, and it should be noted that, calculating the probability value of each feature map identified as the corresponding preset defect classification label by Softmax is a technical solution well known to those skilled in the art, and therefore is not described herein again.

And S73, iteratively updating the defect recognition model according to the recognition result corresponding to each image and a preset defect classification label until the loss function meets a preset condition, and obtaining the trained defect recognition model.

It is understood that the loss function may be used to evaluate the performance of the recognition model, when the loss function is smaller, the performance of the model is better, and the performance of the model is improved by continuously optimizing the loss function, it should be noted that the trained defect recognition model is obtained by stopping the model training when the preset condition is that the loss function is smaller than the preset threshold, and in addition, the classification of the container defects may be scratches, rusts, cracks, deformations, and holes, but is not limited to the above classification.

And S8, sending the image information at the defect position to the trained defect identification model to obtain the identification result of the defect.

Example 2:

as shown in fig. 2, the present embodiment provides a container defect analysis apparatus, which includes a first obtaining module 901, a processing module 902, and a determining module 903, where:

a first obtaining module 901, configured to obtain a first time domain signal, a second time domain signal, and a third time domain signal, where the first time domain signal is a signal acquired after an ultrasonic wave passes through a container to be detected, the second time domain signal is a signal acquired after the ultrasonic wave passes through a defect-free container, and the third time domain signal is a signal acquired by the ultrasonic wave returned from a defect position and received at an ultrasonic wave transmitting end;

a processing module 902, configured to pre-process the first time domain signal to obtain a first spectrogram, and pre-process the second time domain signal to obtain a second spectrogram;

a determining module 903, configured to determine whether the container to be detected has a defect according to the first spectrogram and the second spectrogram, where if the container to be detected has a defect, the position of the defect of the container is determined according to the third time domain signal.

In a specific embodiment of the present disclosure, the determining module 903 further includes a first processing unit 9031 and a determining unit 9032, where:

a first processing unit 9031, configured to determine amplitude information of a first spectrogram according to the first spectrogram and determine amplitude information of a second spectrogram according to the second spectrogram;

a judging unit 9032, configured to judge whether the container to be detected has a defect according to the amplitude information of the first frequency spectrogram and the amplitude information of the second frequency spectrogram, where if the amplitude information of the first frequency spectrogram is lower than the amplitude information of the second frequency spectrogram, it is judged that the container to be detected has a defect; and if the amplitude information of the first frequency spectrogram is not lower than that of the second frequency spectrogram, judging that the container to be detected has no defects.

In a specific embodiment of the present disclosure, the determining module 903 further includes a first obtaining unit 9033, a second processing unit 9034, a third processing unit 9035, and a fourth processing unit 9036, where:

a first obtaining unit 9033, configured to obtain preset group velocity information;

a second processing unit 9034, configured to determine first information according to an excitation signal corresponding to a third time domain signal, where the first information includes a time value corresponding to a maximum amplitude of the excitation signal, and determine second information according to the third time domain signal, where the second information includes a time value corresponding to the maximum amplitude of the third time domain signal;

a third processing unit 9035, configured to subtract the first information from the second information to obtain third information;

and the fourth processing unit 9036 is configured to calculate to obtain a defect position of the container according to the third information and the preset group velocity information.

In a specific embodiment of the present disclosure, the determining module 903 further includes a second obtaining module 904, a constructing module 905, an expanding module 906, a training module 907, and an identifying module 908, where:

a second obtaining module 904, configured to obtain image information at least one defect position according to the defect position of the container;

a constructing module 905 configured to construct a sample set based on the image information at the defect position;

an expansion module 906, configured to expand the sample set to obtain an expanded sample set;

a training module 907 for training a defect recognition model by using the extended sample set to obtain a trained defect recognition model;

and an identifying module 908, configured to send the image information at the defect position to the trained defect identifying model to obtain a defect identifying result.

In a specific embodiment of the present disclosure, the extension module 906 further includes a second obtaining unit 9061, a fifth processing unit 9062, a sixth processing unit 9063, and a seventh processing unit 9064, where:

a second obtaining unit 9061, configured to obtain a window of a preset size;

a fifth processing unit 9062, configured to slide-slice the image information at the defect position according to the window with the preset size, to obtain image information at the defect position after at least one slice is obtained;

a sixth processing unit 9063, configured to perform data cleaning on the image information at the defect position after slicing, to obtain image information at the defect position after cleaning;

a seventh processing unit 9064, configured to expand the sample set based on the image information at the cleaned defect position, to obtain an expanded sample set.

In a specific embodiment of the present disclosure, the training module 907 further includes an eighth processing unit 9071, a ninth processing unit 9072, and a tenth processing unit 9073, where specifically:

an eighth processing unit 9071, configured to perform feature extraction on the images included in the extended sample set by using a convolutional neural network, to obtain a feature map corresponding to each image;

a ninth processing unit 9072, configured to input the feature map to a preset defect identification model, so as to obtain an identification result corresponding to each image;

a tenth processing unit 9073, configured to iteratively update the defect identification model according to the identification result corresponding to each image and a preset defect classification label, until the loss function meets a preset condition, to obtain a trained defect identification model.

It should be noted that, regarding the apparatus in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment related to the method, and will not be elaborated herein.

Example 3:

corresponding to the above method embodiment, the present embodiment further provides a container defect analyzing apparatus, and a container defect analyzing apparatus described below and a container defect analyzing method described above may be referred to in correspondence with each other.

Fig. 3 is a block diagram of a container defect analysis apparatus 800 shown in accordance with an exemplary embodiment. As shown in fig. 3, the container defect analyzing apparatus 800 may include: a processor 801, a memory 802. The container defect analysis device 800 may further include one or more of a multimedia component 803, an I/O interface 804, and a communication component 805.

The processor 801 is configured to control the overall operation of the container defect analyzing apparatus 800 to complete all or part of the steps of the container defect analyzing method. Memory 802 is used to store various types of data to support operation at the container defect analysis device 800, which may include, for example, instructions for any application or method operating on the container defect analysis device 800, as well as application-related data, such as contact data, messaging, pictures, audio, video, and so forth. The Memory 802 may be implemented by any type of volatile or non-volatile Memory device or combination thereof, such as Static Random Access Memory (SRAM), electrically Erasable Programmable Read-Only Memory (EEPROM), erasable Programmable Read-Only Memory (EPROM), programmable Read-Only Memory (PROM), read-Only Memory (ROM), magnetic Memory, flash Memory, magnetic disk or optical disk. The multimedia components 803 may include screen and audio components. Wherein the screen may be, for example, a touch screen and the audio component is used for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving an external audio signal. The received audio signal may further be stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, such as a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the container defect analysis apparatus 800 and other apparatuses. Wireless communication, such as Wi-Fi, bluetooth, near Field Communication (NFC), 2G, 3G, or 4G, or a combination of one or more of them, so that the corresponding communication component 805 may include: wi-Fi module, bluetooth module, NFC module.

In an exemplary embodiment, the container defect analyzing apparatus 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital Signal Processing Devices (DSPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing the above-described container defect analyzing method.

In another exemplary embodiment, a computer readable storage medium comprising program instructions which, when executed by a processor, implement the steps of the container defect analysis method described above is also provided. For example, the computer readable storage medium may be the memory 802 described above including program instructions executable by the processor 801 of the bin defect analysis device 800 to perform the bin defect analysis method described above.

Example 4:

corresponding to the above method embodiment, a readable storage medium is also provided in this embodiment, and a readable storage medium described below and a container defect analysis method described above may be referred to in correspondence.

A readable storage medium, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the method for analyzing a defect of a container according to the above-mentioned method embodiment.

The readable storage medium may be a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk, and various other readable storage media capable of storing program codes.

The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention, and various modifications and changes may be made by those skilled in the art. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. A method for analyzing defects of a container, comprising:

acquiring a first time domain signal, a second time domain signal and a third time domain signal, wherein the first time domain signal is a signal acquired after ultrasonic waves pass through a container to be detected, the second time domain signal is a signal acquired after the ultrasonic waves pass through a defect-free container, and the third time domain signal is a signal acquired by the ultrasonic waves returned from a defect position and received at an ultrasonic wave transmitting end;

judging whether the container to be detected has defects according to the first frequency spectrogram and the second spectrogram, wherein if the container to be detected has defects, the defect position of the container is determined according to the third time domain signal;

judging whether the container to be detected has defects according to the first frequency spectrogram and the second frequency spectrogram, wherein the judging comprises the following steps of:

determining amplitude information of a first spectrogram according to the first spectrogram and determining amplitude information of a second spectrogram according to the second spectrogram;

judging whether the container to be detected has defects according to the amplitude information of the first frequency spectrogram and the amplitude information of the second frequency spectrogram, wherein if the amplitude information of the first frequency spectrogram is lower than the amplitude information of the second frequency spectrogram, the container to be detected is judged to have defects; if the amplitude information of the first frequency spectrogram is not lower than that of the second frequency spectrogram, judging that the container to be detected has no defects;

if the container is defective, determining the defective position of the container according to the third time domain signal, and then:

acquiring image information of at least one defect position according to the defect position of the container;

constructing a sample set based on the image information at the defect location;

expanding the sample set to obtain an expanded sample set;

training a defect recognition model by using the extended sample set to obtain a trained defect recognition model;

sending the image information at the defect position to the trained defect identification model to obtain the identification result of the defect;

wherein, expanding the sample set to obtain an expanded sample set comprises:

acquiring a window with a preset size;

performing sliding slicing on the image information at the defect position according to the window with the preset size to obtain the image information at the defect position after at least one slice is cut;

performing data cleaning on the image information at the defect position after slicing to obtain the image information at the defect position after cleaning;

and expanding the sample set based on the image information of the cleaned defect position to obtain an expanded sample set.

2. The container defect analysis method of claim 1, wherein determining a defect location of the container based on the third time domain signal if a defect exists comprises:

acquiring preset group velocity information;

determining first information according to an excitation signal corresponding to a third time domain signal, the first information including a time value corresponding to a maximum amplitude of the excitation signal, and determining second information according to the third time domain signal, the second information including a time value corresponding to the maximum amplitude of the third time domain signal;

subtracting the first information from the second information to obtain third information;

and calculating according to the third information and the preset group velocity information to obtain the defect position of the container.

3. A container defect analysis device, characterized in that includes:

the judging module is used for judging whether the container to be detected has defects according to the first frequency spectrogram and the second frequency spectrogram, wherein if the container to be detected has defects, the position of the defect of the container is determined according to the third time domain signal;

wherein, the judging module comprises:

the first processing unit is used for determining amplitude information of a first spectrogram according to the first spectrogram and determining amplitude information of a second spectrogram according to the second spectrogram;

the judging unit is used for judging whether the container to be detected has defects or not according to the amplitude information of the first frequency spectrogram and the amplitude information of the second frequency spectrogram, wherein if the amplitude information of the first frequency spectrogram is lower than the amplitude information of the second frequency spectrogram, the container to be detected is judged to have defects; if the amplitude information of the first frequency spectrogram is not lower than that of the second frequency spectrogram, judging that the container to be detected has no defects;

wherein, behind the judging module, still include:

the second acquisition module is used for acquiring image information of at least one defect position according to the defect position of the container;

a construction module for constructing a sample set based on the image information at the defect location;

the expansion module is used for expanding the sample set to obtain an expanded sample set;

the training module is used for training a defect recognition model by using the extended sample set to obtain a trained defect recognition model;

the recognition module is used for sending the image information at the defect position to the trained defect recognition model to obtain a defect recognition result;

wherein, the expansion module comprises:

the second acquisition unit is used for acquiring a window with a preset size;

a fifth processing unit, configured to slide-slice the image information at the defect position according to the window with the preset size, to obtain image information at the defect position after at least one slice;

a sixth processing unit, configured to perform data cleaning on the image information at the defect position after slicing, to obtain image information at the defect position after cleaning;

and the seventh processing unit is used for expanding the sample set based on the image information at the cleaned defect position to obtain an expanded sample set.

4. The container defect analysis device of claim 3, wherein the determining module further comprises:

a first obtaining unit, configured to obtain preset group velocity information;

a second processing unit, configured to determine first information according to an excitation signal corresponding to a third time domain signal, where the first information includes a time value corresponding to a maximum amplitude of the excitation signal, and determine second information according to the third time domain signal, where the second information includes a time value corresponding to the maximum amplitude of the third time domain signal;

the third processing unit is used for subtracting the first information from the second information to obtain third information;

and the fourth processing unit is used for calculating according to the third information and the preset group velocity information to obtain the defect position of the container.

5. A container defect analysis apparatus, comprising:

a memory for storing a computer program;

a processor for implementing the steps of the container defect analysis method of any one of claims 1 to 2 when executing the computer program.

6. A readable storage medium, characterized by: the readable storage medium has stored thereon a computer program which when executed by a processor carries out the steps of the method of analyzing a defect of a container as claimed in any one of claims 1 to 2.