Harbor container truck anti-hoisting judgment method, system and device and storage medium

Abstract

The invention discloses an anti-hoisting judgment method, system and device for a port container truck and a storage medium, and the method comprises the steps: obtaining a detection signal outputted by a weighing sensor within a set time period, and processing the detection signal to obtain a dynamic signal during the separation of a sling from the container truck and a steady-state signal during the stable rising of a container sling; inputting the steady-state signal into a slope prediction model to obtain a predicted slope of the steady-state signal, and judging whether the container truck is hoisted or not based on a predicted efficiency value; according to the slope prediction model, a steady-state signal is imaged through continuous Fourier signal transformation and filling, image details and abstract features of the imaged signal are extracted in a multi-level mode through a plurality of residual networks connected in a jumping mode, and the imaged signal is divided into a plurality of local calculation areas. And feature fusion is adopted to carry out reverse step-by-step semantic feature fusion on the image signal, a calculation result of autocorrelation of each part is output, the calculation result is an overall region slope value containing all local information, and the effect of improving an anti-hoisting judgment result is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of industrial control, and in particular, relates to a method, system, device and computer storage medium for judging whether a container truck is lifted up in a port. Background Art

[0002] When the rail crane is lifting the container on the container truck bracket, if the locking pins are not fully released due to human negligence, structural deformation, etc., there will be a safety hazard of the bracket being lifted. In the case of unmanned automated rail cranes, in order to prevent the container truck from being lifted, manual inspection is required during the unloading process of the container truck. Although it can ensure safety, it reduces the degree of automation of the production process and reduces production efficiency.

[0003] In order to increase the degree of automation and improve production efficiency, the most common automated method to prevent container trucks from being lifted is to use machine vision to conduct all-round observation and coverage of container trucks through front-end cameras, and to use artificial intelligence technology to automatically identify abnormal changes in the characteristics of the bracket when the container truck is lifted, to determine whether the operating container truck has been lifted.

[0004] Machine vision can be used to image and identify the lifting area of ​​trucks and containers, which has a certain degree of flexibility. It is non-contact and can avoid contact detection from causing damage to the product, thereby improving the reliability of the system. However, in terms of its high-precision positioning, since machine vision has extremely high requirements for the stability of lighting, when the lighting changes by 10%-20%, the measurement result may have an error of 1-2 pixels, which will cause the image to change along the specific position. Therefore, it is easy to have a high false alarm rate in severe weather conditions such as heavy rain and fog. Summary of the invention

[0005] The object of the present invention is to provide a method, system, device and storage medium for judging whether a container truck is lifted in a port, so as to solve the problem that there is error in the way machine vision judges whether a container truck is lifted.

[0006] The present invention is implemented by the following technical solutions: A method for judging whether a container truck can be lifted up in a port is proposed, comprising: Acquiring a detection signal output by a weighing sensor within a set time period; wherein the weighing sensor is installed on a container spreader; Processing the detection signal to obtain a dynamic signal and a steady-state signal; Inputting the steady-state signal into a slope prediction model to obtain a predicted slope output by the slope prediction model; When the predicted slope is greater than a set value, it is determined that the container truck is lifted; Among them, the slope prediction model is used to: visualize the steady-state signal through continuous Fourier signal transform and filling, extract the image details and abstract features of the imaged signal through multiple levels of residual networks with multiple jump connections, divide the imaged signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the imaged signal, and output the calculation results of the autocorrelation of each part.

[0007] In some embodiments of the present invention, the network structure of the slope prediction model is as follows: the left half of the model is a VGG-16 encoder, which is composed of convolution and Max pooling; the right half of the model is a decoder, which improves the calculation accuracy of the steady-state signal through deconvolution, and the feature maps within each segmentation range are cascaded and integrated with the features from the encoder, so that the output of the decoder is an overall regional slope value containing all local information.

[0008] In some embodiments of the present invention, the set value is obtained by training a deep learning model using a training set.

[0009] In some embodiments of the present invention, before obtaining the detection signal output by the weighing sensor within a set time period, the method further includes: monitoring whether a locking signal and a lifting signal of the container spreader are received; generating a trigger signal when the locking signal and the lifting signal of the container spreader are received; and starting to obtain the detection signal output by the weighing sensor according to the trigger signal.

[0010] In some embodiments of the present invention, the method further comprises: the set time period is a period from when the trigger signal is generated to when the lifting height of the container spreader is equal to a target threshold.

[0011] In some embodiments of the present invention, the target threshold is set according to the distance at which the container is completely separated from the container truck.

[0012] In some embodiments of the present invention, the method also includes: performing nonlinear filtering on the detection signal; extracting low-frequency drift in the interference signal from the detection signal after nonlinear filtering; subtracting the low-frequency drift from the interference signal to obtain a processed detection signal; and using the processed detection signal to update the detection signal.

[0013] A port container truck anti-lifting system is proposed, comprising: Weighing sensor, installed on container spreader; A signal acquisition module, used to acquire the detection signal output by the weighing sensor within a set time period; A signal analysis module, used for processing the detection signal to obtain a dynamic signal and a steady-state signal; and inputting the steady-state signal into a slope prediction model to obtain a predicted slope output by the slope prediction model; A lifting judgment module, used for judging that the container truck is lifted when the predicted efficiency is greater than a set value; The lifting processing module is used to initiate a response operation when the container truck is lifted; The slope prediction model is used to visualize the steady-state signal through continuous Fourier signal transformation and filling, extract image details and abstract features of the imaged signal through multiple layers of residual networks with multiple jump connections, divide the imaged signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the imaged signal, and output the calculation results of the autocorrelation of each part.

[0014] A port container truck anti-lifting device is proposed, comprising: A signal acquisition module, used to acquire a detection signal output by a weighing sensor within a set time period; wherein the weighing sensor is installed on the container spreader; A signal analysis module, used for processing the detection signal to obtain a dynamic signal and a steady-state signal; and inputting the steady-state signal into a slope prediction model to obtain a predicted slope output by the slope prediction model; A lifting judgment module, used for judging that the container truck is lifted when the predicted efficiency is greater than a set value; The lifting processing module is used to initiate a response operation when the container truck is lifted; The slope prediction model is used to visualize the steady-state signal through continuous Fourier signal transformation and filling, extract image details and abstract features of the imaged signal through multiple layers of residual networks with multiple jump connections, divide the imaged signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the imaged signal, and output the calculation results of the autocorrelation of each part.

[0015] A computer storage medium is proposed for storing a computer program, wherein the computer program can be executed by at least one computer processor to implement the port container truck anti-lifting judgment method as described above.

[0016] Compared with the prior art, the advantages and positive effects of the present invention are as follows: in the port container truck anti-lifting judgment method, system, device and storage medium proposed by the present invention, the detection signal output by the weighing sensor is obtained within a set time period. During the period when the container spreader and the container truck are separated, the force on the sling will undergo a large jump-like change, resulting in a large volatility in the detection signal of the weighing sensor, so that the detection signal includes several peak points; in the rising stage when the separation is completed, if the separation is achieved, even if there are fluctuations in a short range, the detection signal as a whole will remain relatively stable. If the separation is not achieved and the spreader and the container truck are lifted together, the detection signal will show a nonlinear increasing trend. In view of this, the present invention processes the detection signal within a set time period to distinguish the period when the sling is separated from the container truck. The dynamic signal and the steady-state signal of the stable rise of the container hoist are discarded, and the steady-state signal is input into the slope prediction model to obtain the predicted slope of the steady-state signal. Whether the container truck is lifted is judged based on the predicted efficiency value; the slope prediction model visualizes the steady-state signal through continuous Fourier signal transform and filling, and extracts the image details and abstract features of the visualized signal through multiple skip-connected residual networks at multiple levels, divides the visualized signal into multiple local calculation areas, and uses feature fusion to perform reverse step-by-step semantic feature fusion on the visualized signal, and outputs the calculation results of the autocorrelation of each part. The calculation result is an overall regional slope value containing all local information. Compared with the direct use of straight line fitting, the slope deviation is significantly reduced, the judgment result of anti-lifting is improved, and the false alarm rate is reduced.

[0017] Other features and advantages of the present invention will become more apparent after reading the detailed description of the embodiments of the present invention in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments and descriptions of the present invention are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the accompanying drawings described below are only some embodiments. For those skilled in the art, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0019] Figure 1 The figure is a schematic diagram of the process of the method for judging the anti-lifting of container trucks in a port proposed by the present invention; Figure 2 It is a signal diagram of the detection signal of the weighing sensor when the container is in the detached state in the present invention; Figure 3 It is a signal diagram of the detection signal of the weighing sensor in the present invention when the container is not separated; Figure 4 Schematic diagram of the model architecture of the slope prediction model proposed in the present invention; Figure 5 The figure is a schematic diagram of the system architecture of the port container truck anti-lifting system proposed by the present invention.

[0020] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. The following embodiments are used to illustrate the present invention but are not used to limit the scope of the present invention.

[0022] In the description of the present invention, it should be noted that the directions or positional relationships indicated by terms such as “upper”, “lower”, “front”, “back”, “left”, “right”, “vertical”, “inside” and “outside” are based on the directions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0023] In the description of the present invention, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected" and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or connected through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0024] The method for judging whether to prevent a container truck from lifting in a port is proposed by the present invention. Figure 1 As shown, the following steps are included: Step S1: Acquire the detection signal output by the weighing sensor within a set time period.

[0025] Compared with the prior art method of using a front-end camera to conduct all-round observation of the container truck, the present invention obtains the detection signal of a weighing sensor within a period of time. The weighing sensor is installed on the container spreader. The container spreader is used to measure the total weight, four-corner weight, overload and overweight of the container during the container lifting process. It can perform intelligent detection and has remote transmission and management functions. By adopting the technical solution of the present invention, there is no need to add additional detection devices, and the complicated installation and debugging of the observation camera is eliminated. The existing weighing sensor can be used to realize the lifting judgment, and the connection between the container and the container truck can be accurately judged in real time. It can also avoid the external factors such as network freezes and weather environment from having a greater impact on the detection.

[0026] Step S2: Process the detection signal to obtain a steady-state signal of the dynamic signal.

[0027] During the separation of the container spreader from the container truck, the force on the sling will change greatly, resulting in a large fluctuation in the detection signal of the weighing sensor, so that the detection signal includes several high peak points, such as Figure 2 and Figure 3 When the separation is completed, if the separation is achieved, even if there is a short range of fluctuations, the overall detection signal will remain relatively stable, as shown in Figure 1. Figure 2 As shown in Figure 1, if the separation is not achieved and the spreader and the container truck are lifted together, the detection signal will show a nonlinear increase trend, such as Figure 3 As shown; in view of this, the present invention processes the detection signal within a set time period, distinguishes the dynamic signal during the period when the sling is separated from the container truck, and the steady-state signal of the stable rise of the container spreader, and uses the steady-state signal to analyze the rising state of the container spreader.

[0028] In some embodiments of the present invention, the detection signal of the weighing sensor is continuously tracked within a set time period, so as to obtain a trend curve of the weight (y) of the container hanging spreader versus time (t), such as Figure 2 and Figure 3 As shown, the range of large jump changes starts from the starting point t1 of the set time period, indicating that the container and the container truck are in a detached state. The process includes several high peak points. The final peak point is the connection point of the dynamic signal and the steady-state signal, corresponding to t2 within the set time period. Since the dynamic signal is a signal generated during the separation of the container and the container truck, after the connection point of the dynamic signal and the steady-state signal is identified, that is, after determining the starting point of the steady-state signal, it can be screened out and not used.

[0029] According to the different slow lifting states of the container spreader: empty or connected to a container truck, the steady-state signal presents different trend changes until the end point t3 of the set time period.

[0030] Step S3: Input the steady-state signal into the slope prediction model.

[0031] During the lifting process of the container, slight fluctuations will occur in some ranges. If the direct curve fitting method is adopted, the obtained result will have a large deviation from the overall curve change trend. In view of this, the present invention inputs the steady-state signal into the slope prediction model, such as Figure 4As shown in the figure, the network structure of the slope prediction model is as follows: the left half of the model is a VGG-16 encoder, which is composed of convolution and Max pooling; the right half of the model is a decoder, which improves the calculation accuracy of the steady-state signal through deconvolution, and the feature maps within each segmentation range are cascaded and integrated with the features from the encoder, so that the output of the decoder is an overall regional slope value containing all local information.

[0032] The model adopts a U-shaped network algorithm based on feature cross-scale jump connection. First, the image construction of the steady-state signal is completed by continuous Fourier signal transform and filling to obtain a 256*256 high-bit data image with 1 channel number. Then, the single-channel input image is changed and reduced in size to obtain a signal image size of 128*128. Then, through 8 (2,2,2,2) skip-connected residual networks, the image details and abstract features of the signal are extracted at multiple levels, so that the steady-state signal is divided into multiple local calculation areas of various sizes, and feature fusion is used to perform reverse step-by-step semantic feature fusion on the signal to output the calculation results of the autocorrelation of each part. Based on the role of the slope prediction model, through intensive cross-scale adjustment, the local features and the overall features are fully integrated, and the steady-state signal can be divided into local calculation areas of various sizes, so that the calculation result is a slope value of the overall area containing all local information. Compared with the direct use of linear fitting, the slope deviation is significantly reduced, the judgment result of anti-lifting is improved, and the false alarm rate is reduced.

[0033] Step S4: Obtain the predicted slope output by the slope prediction model, and determine that the container truck is lifted when the predicted slope is greater than a set value.

[0034] During the slow lifting process of the container spreader, if there is no container truck hanging, the detection signal of the weighing sensor will remain relatively stable as a whole after passing through the dynamic signal stage, although there will be short-term fluctuations in the steady-state signal stage. If there is a container truck hanging, the detection signal of the weighing sensor will show a nonlinear increasing trend in the steady-state signal stage.

[0035] The present invention uses a deep learning model to pre-train a training set to obtain a set value K1. If the predicted slope value output by the slope prediction model is less than K1, it is judged that there is no connection between the container spreader and the container truck. If the slope prediction value is greater than K1, it is judged that there is a connection between the container spreader and the container truck.

[0036] When it is determined that the container truck is lifted, the control system can make different response actions under different conditions, including but not limited to issuing a connection alarm, braking the container lifter, etc.

[0037] In some embodiments of the present invention, in order to reduce calculation time, before obtaining the detection signal output by the weighing sensor within a set time period, a prerequisite for anti-lifting detection and judgment is set, that is, when the container spreader implements the operation of grabbing the container from the container truck, it is first monitored whether the locking signal of the container spreader is received, and then it is monitored whether the lifting signal of the container spreader is received after receiving the locking signal. When both the locking signal and the lifting signal are monitored, it fully indicates that the container spreader is about to prepare for the grabbing operation. At this time, a trigger signal is generated. The trigger signal is used to indicate the start of obtaining the detection signal output by the weighing sensor. The system implements the anti-lifting judgment method based on the trigger signal, which can effectively obtain calculation data, avoid complicated data calculation process, and improve calculation efficiency.

[0038] The above-mentioned set time period is at least able to capture the dynamic signal and steady-state signal part of the weighing sensor in the container truck grabbing operation. Considering that after the container spreader completes the locking action, it receives the lifting command signal to perform the spreader lifting action. Due to the contact between the bottom corner of the container and the container truck lock, the spreader grabbing the box may cause the container spreader to rub against the container truck and separate. At this time, the force on the sling will undergo a large jump change, resulting in a large volatility in the detection signal of the weighing sensor, which can be determined as a dynamic signal output during the loading process of the container spreader. In view of this, in order to detect more valid data and obtain the output signal of the container in a steady state, in some embodiments of the present invention, the set time period is set to: from the generation of the trigger signal to the time when the lifting height of the container spreader is equal to the target threshold, and the target threshold is determined according to the distance at which the container is completely separated from the container truck.

[0039] In some embodiments of the present invention, after obtaining the detection signal of the weighing sensor, the detection signal is first preprocessed. Since the output signal of the weighing sensor is affected by external or internal accidental time and may cause large errors, large errors will obviously distort the measurement results. Their existence will seriously affect the precision and accuracy of the measurement. In the embodiments of the present invention, in order to filter out large errors, the detection signal is first nonlinearly filtered. For example, a single large jump in the steady-state signal x(n) is equivalent to an impulse function, and the spectrum contains various frequencies. Because the signal spectrum Y(ω) after linear filtering is equal to the input signal spectrum X(ω) and the filter frequency The product of the corresponding H(ω) cannot filter out all frequency components of the impulse function for a specific designed filter (such as a low-pass filter), that is, it cannot completely eliminate the influence of large errors. However, the nonlinear filtering method can filter out large errors. After nonlinear filtering, the interference and high-frequency noise in the steady-state signal are effectively suppressed, but the low-frequency drift still exists. The drift suppression can be based on the characteristics of the drift and signal after filtering, and some algorithms are used to extract the drift from the interference signal, and then the drift is subtracted from the interference signal to obtain a relatively more accurate detection signal. The processed detection signal is used to perform steps S1 to S4, which can significantly improve the judgment accuracy.

[0040] Based on the above-mentioned port container truck anti-lifting method, the present invention also proposes a port container truck anti-lifting system, such as Figure 5 As shown, including: The weighing sensor 51 is installed on the container spreader.

[0041] The signal acquisition module 52 is used to acquire the detection signal output by the weighing sensor within a set time period.

[0042] The signal analysis module 53 is used to process the detection signal to obtain a dynamic signal and a steady-state signal; and input the steady-state signal into the slope prediction model 54 to obtain the predicted slope output by the slope prediction model 54.

[0043] The lifting judgment module 55 is used to judge whether the container truck is lifted when the predicted efficiency is greater than a set value.

[0044] A lifting processing module 56, used to initiate a response operation when the container truck is lifted; The slope prediction model 54 is used to visualize the steady-state signal through continuous Fourier signal transform and filling, extract the image details and abstract features of the visualized signal through multiple layers of residual networks with multiple jump connections, divide the visualized signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the visualized signal, and output the calculation results of the autocorrelation of each part.

[0045] A port container truck anti-lifting device is proposed, comprising: The signal acquisition module is used to acquire the detection signal output by the weighing sensor within a set time period; wherein the weighing sensor is installed on the container spreader.

[0046] The signal analysis module is used to process the detection signal to obtain a dynamic signal and a steady-state signal; and input the steady-state signal into the slope prediction model to obtain the predicted slope output by the slope prediction model.

[0047] The lifting judgment module is used to judge whether the container truck is lifted when the predicted efficiency is greater than the set value.

[0048] The lifting processing module is used to initiate a response operation when the container truck is lifted.

[0049] The slope prediction model is used to visualize the steady-state signal through continuous Fourier signal transform and filling, extract the image details and abstract features of the visualized signal through multiple levels of residual networks with multiple jump connections, divide the visualized signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the visualized signal, and output the calculation results of the autocorrelation of each part.

[0050] A computer storage medium is proposed for storing a computer program. The computer program can be executed by at least one computer processor to implement the above-mentioned method for determining whether a container truck is lifted up in a port.

[0051] It should be pointed out that the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by ordinary technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. A method for judging whether a container truck can be lifted up in a port, characterized in that: include: Acquiring a detection signal output by a weighing sensor within a set time period; wherein the weighing sensor is installed on a container spreader; Processing the detection signal to obtain a dynamic signal and a steady-state signal; Inputting the steady-state signal into a slope prediction model to obtain a predicted slope output by the slope prediction model; When the predicted slope is greater than a set value, it is determined that the container truck is lifted; Among them, the slope prediction model is used to: visualize the steady-state signal through continuous Fourier signal transform and filling, extract the image details and abstract features of the imaged signal through multiple levels of residual networks with multiple jump connections, divide the imaged signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the imaged signal, and output the calculation results of the autocorrelation of each part.

2. The method for judging whether a container truck is lifted up in a port according to claim 1, characterized in that: The network structure of the slope prediction model is as follows: the left half of the model is a VGG-16 encoder, which consists of convolution and Max pooling; the right half of the model is a decoder, which improves the calculation accuracy of the steady-state signal through deconvolution, and the feature maps within each segmentation range are cascaded and integrated with the features from the encoder, so that the output of the decoder is an overall regional slope value containing all local information.

3. The method for judging whether a container truck is lifted up in a port according to claim 1, characterized in that: The set value is obtained by training a deep learning model using a training set.

4. The method for determining whether a container truck is lifted up in a port according to claim 1, characterized in that: Before acquiring the detection signal output by the weighing sensor within a set time period, the method further includes: Monitor whether the locking signal and lifting signal of the container spreader are received; When receiving the locking signal and lifting signal of the container spreader, a trigger signal is generated; The detection signal output by the weighing sensor starts to be acquired according to the trigger signal.

5. The method for judging whether a container truck is lifted up in a port according to claim 4, characterized in that: The method further comprises: The set time period is a period from when the trigger signal is generated to when the lifting height of the container spreader is equal to the target threshold.

6. The method for judging whether a container truck is lifted up in a port according to claim 5, characterized in that: The target threshold is set according to the distance at which the container is completely separated from the container truck.

7. The method for determining whether a container truck is lifted up in a port according to claim 1, characterized in that: The method further comprises: performing nonlinear filtering on the detection signal; Extracting low-frequency drift in the interference signal from the nonlinearly filtered detection signal; Subtracting the low frequency drift from the interference signal to obtain a processed detection signal; The processed detection signal is used to update the detection signal.

8. A port container truck anti-lifting system, characterized in that: include: Weighing sensor, installed on container spreader; A signal acquisition module, used to acquire the detection signal output by the weighing sensor within a set time period; A signal analysis module, used for processing the detection signal to obtain a dynamic signal and a steady-state signal; and inputting the steady-state signal into a slope prediction model to obtain a predicted slope output by the slope prediction model; A lifting judgment module, used for judging that the container truck is lifted when the predicted efficiency is greater than a set value; The lifting processing module is used to initiate a response operation when the container truck is lifted; The slope prediction model is used to visualize the steady-state signal through continuous Fourier signal transformation and filling, extract image details and abstract features of the imaged signal through multiple layers of residual networks with multiple jump connections, divide the imaged signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the imaged signal, and output the calculation results of the autocorrelation of each part.

9. A port container truck anti-lifting device, characterized in that: include: A signal acquisition module, used to acquire a detection signal output by a weighing sensor within a set time period; wherein the weighing sensor is installed on the container spreader; A signal analysis module, used for processing the detection signal to obtain a dynamic signal and a steady-state signal; and inputting the steady-state signal into a slope prediction model to obtain a predicted slope output by the slope prediction model; A lifting judgment module, used for judging that the container truck is lifted when the predicted efficiency is greater than a set value; The lifting processing module is used to initiate a response operation when the container truck is lifted; The slope prediction model is used to visualize the steady-state signal through continuous Fourier signal transformation and filling, extract image details and abstract features of the imaged signal through multiple layers of residual networks with multiple jump connections, divide the imaged signal into multiple local calculation areas, and use feature fusion to perform reverse step-by-step semantic feature fusion on the imaged signal, and output the calculation results of the autocorrelation of each part.

10. A computer storage medium, characterized in that: Used to store a computer program, which can be executed by at least one computer processor to implement the port container truck anti-lifting judgment method as described in any one of claims 1-7.

Images