CN111832415B - Truck safety intelligent protection system for container hoisting operation - Google Patents

Truck safety intelligent protection system for container hoisting operation Download PDF

Info

Publication number
CN111832415B
CN111832415B CN202010541326.0A CN202010541326A CN111832415B CN 111832415 B CN111832415 B CN 111832415B CN 202010541326 A CN202010541326 A CN 202010541326A CN 111832415 B CN111832415 B CN 111832415B
Authority
CN
China
Prior art keywords
truck
neural network
training
lifting
deep neural
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202010541326.0A
Other languages
Chinese (zh)
Other versions
CN111832415A (en
Inventor
张韶越
何志成
尚继辉
张宾
陈小虎
时金利
张骏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beiyu Technology Shanghai Co ltd
Aerospace Intelligent Manufacturing Shanghai Technology Co ltd
Original Assignee
Beiyu Technology Shanghai Co ltd
Aerospace Intelligent Manufacturing Shanghai Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beiyu Technology Shanghai Co ltd, Aerospace Intelligent Manufacturing Shanghai Technology Co ltd filed Critical Beiyu Technology Shanghai Co ltd
Priority to CN202010541326.0A priority Critical patent/CN111832415B/en
Publication of CN111832415A publication Critical patent/CN111832415A/en
Application granted granted Critical
Publication of CN111832415B publication Critical patent/CN111832415B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/40—Scenes; Scene-specific elements in video content
    • G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
    • B66C13/00—Other constructional features or details
    • B66C13/16—Applications of indicating, registering, or weighing devices
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B66—HOISTING; LIFTING; HAULING
    • B66C—CRANES; LOAD-ENGAGING ELEMENTS OR DEVICES FOR CRANES, CAPSTANS, WINCHES, OR TACKLES
    • B66C13/00—Other constructional features or details
    • B66C13/18—Control systems or devices
    • B66C13/48—Automatic control of crane drives for producing a single or repeated working cycle; Programme control
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/084—Backpropagation, e.g. using gradient descent
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Software Systems (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Molecular Biology (AREA)
  • General Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Multimedia (AREA)
  • Mechanical Engineering (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Automation & Control Theory (AREA)
  • Control And Safety Of Cranes (AREA)

Abstract

The invention relates to an intelligent safety protection system for a truck, which is used for container hoisting operation. The system mainly comprises hardware systems composed of parts such as field industrial control equipment, a network switch, a signal trigger, a digital camera, a laser radar and the like; the software system is composed of algorithms such as video analysis, signal processing, communication control and the like. In the process that the goods carried by the truck are hoisted, a hardware system collects in real time, and a software system analyzes in real time, so that three accidents are mainly prevented. Firstly, whether the truck is erroneously lifted, secondly, whether the truck head is hit or not, and thirdly, whether the truck is pulled or not. If the lifting exceeds a certain preset height, or the truck head enters a lifting operation area, or the load moves transversely in the lifting appliance unloading process, the system sends an alarm to the crane so as to prevent the damage of the truck and the injury and death of a driver. Compared with the prior art, the system has the advantages that the artificial intelligence algorithm based on the deep neural network is adopted, various load cargoes, truck trays and truck heads in videos can be intelligently identified, and therefore the functions of preventing lifting, smashing and dragging of the trucks are achieved through one set of system. The system can intelligently distinguish the load condition of the truck and the type of the pallet, is little affected by the environment, has low misjudgment rate, and can be used for safety production applications such as unmanned monitoring in the process of lifting the container of the port tire crane.

Description

Truck safety intelligent protection system for container hoisting operation
Technical Field
The invention belongs to the technical field of machine vision based on an artificial intelligence algorithm, and relates to an automatic video tracking and monitoring technology for specific interested objects.
Background
The truck safety protection system (ATL system) is mainly used for avoiding the occurrence of accidents of lifting the truck, crashing the head of the truck or dragging the lifting appliance when the wharf tyre crane or the track crane is used for unloading the truck. Taking the typical truck load as an example of a container: the handling of containers is a special type of work, which may lead to two serious accidents when lifting the container, because the locking feet of the truck are not fully opened: one is that the spreader lifts the container and truck as a whole or at one end, and the other is that the truck is started without actually separating from the load. Both of these accidents can lead to damage to containers, spreaders and trucks, and more seriously to casualties for the truck driver. Another accident that may occur in the actual lifting process is that the parking position is incorrect when the truck is unloaded, so that the truck head enters the original container lifting area. At this time, if the crane driver is distracted from the observation or cannot observe due to the limited conditions, the hoist is still lowered according to the original operation flow, and then the heavy hoist bumps into the truck head.
Conventional truck safety protection systems only focus on anti-pick protection, with two technical routes, laser radar alone, and surveillance camera alone. The former uses the laser scanning distance measurement principle, only can obtain real-time data on a scanning line, and then judges whether a truck is hoisted along with a load after comparing with a pre-stored template. In practice, however, the truck and load conditions are all the more so, and when combined with each other, there are hundreds or thousands of possible conditions, the data obtained from a single laser scan line is very non-intuitive, requiring significant time and management costs for the developer, and in the case of a crane out-of-production, manual debugging is performed to produce all the possible template data. In addition, the interference ratio of the laser radar under the outdoor complex meteorological conditions such as rain, snow, haze and the like is large, the lifting system of the laser radar is independently used, the misjudgment rate is high, and the smashing prevention function of the truck head cannot be realized.
The technology of using the monitoring camera alone has the biggest advantage that real-time video can be obtained and is very visual compared with the technology of using the laser radar alone. And the single image capturing is equivalent to the improvement from one dimension of one line to two dimensions of one surface, so that the data volume is greatly increased. However, an increase in the total data amount brings about a simultaneous increase in effective data and interference data. Although it is possible for a monitoring person to quickly determine whether a lifting accident exists at a glance, it also presents a great challenge to the conventional automatic recognition algorithm. The existing typical digital image anti-lifting identification is that a plurality of sub-blocks covering a truck bracket are manually divided in an image, and then the image of each frame in the sub-blocks is tracked. And judging whether the truck bracket is lifted or not by calculating the vector value of the vertical upward motion of the current frame relative to the previous frame in the sub-block. The greatest hidden danger of the method is that the sub-blocks calculated by tracking are artificially divided. Specifically, changes in truck model, truck load, and truck parking position all cause the position of the tray in the video to change greatly, with the result that the sub-blocks may not always track exactly the portion of the truck tray, possibly with the truck load resulting in false positives, or with the background resulting in false negatives. Similarly, because the sub-block area is small, the existing system cannot realize the functions of preventing the head from being crashed and dragged.
SUMMARY OF THE PATENT FOR INVENTION
The invention aims to overcome the defects of the prior art, and provides an automatic monitoring system for preventing lifting, smashing and dragging of a truck by adopting an artificial intelligence algorithm to perform real-time video analysis. The system has the self-learning function, high recognition rate, wide adaptation environment, quick installation and deployment and high cost performance.
The system utilizes the laser-assisted vision measurement detection principle, integrates artificial intelligence and automatic control technology, can realize real-time detection of the positions of a lifting appliance, a load and a truck, and establishes a series of multiple safety control strategies through a software algorithm, so that the accidents that the truck is lifted, smashed and dragged due to human errors can be avoided, the on-site operation efficiency is ensured, and meanwhile, the dangerous occurrence is avoided. Typically, when the port tire crane works, if accident hidden danger is detected, the system can realize the operation control of the lifting mechanism, and the safety of the site bridge work is ensured.
The functions of the invention can be realized by the following technical proposal:
a truck safety intelligent protection system for container hoisting operation mainly comprises hardware devices such as field industrial control equipment, a network switch, a signal trigger, a digital camera, a laser radar and the like. The acquisition subsystem is an algorithm processing system composed of an image processing software module, a communication module and trigger control logic. After the system is electrified, the image and the laser radar acquire data in real time, perform mode identification, enter a monitoring state by default according to the state of the crane PLC signal and the mode identification result of the system, and transmit as required. When the system detects that the truck is mistakenly hoisted, if the hoisting height exceeds a first threshold value, a first alarm is sent to the PLC; if the lifting height continues to rise, an alarm is sent again to the PLC and the alarm continues until there is a human intervention.
The system hardware core is a detection component, and two types of detection components are digital cameras and laser radars. Both are installed on the side, and the anti-lifting alarm is carried out on the truck on one side of the lane. The digital cameras are respectively arranged at two ends of the tire crane, and the arrangement positions of the digital cameras can ensure that the truck tray and the truck load can enter the camera view field at different truck unloading positions; at the same time, when the truck is in the normal unloading position, the truck head does not occupy the range of more than 2/3 of the camera field of view. The laser radar is arranged at the middle lower part of the tire crane and is 200 mm-1000 mm away from the ground. The included angle between the scanning plane of the laser radar and the horizontal plane is within +/-15 degrees, so that the truck can sweep a plurality of tires on the side surface of the truck when the truck is in a normal operation position. The two components are complementary comprehensively, so that the separation condition of the box car can be identified hundred percent correctly.
The system software consists of the following parts: the system comprises an acquisition service module, a separation identification module, a report data module, a PLC communication module and a fault alarm module. The acquisition service module is an algorithm processing system composed of an image processing software module, a communication module and trigger control logic. In the loading and unloading process, the real-time acquisition of images and the real-time transmission of laser radar data are realized; the separation and identification module is mainly used for realizing an ATL function, namely, monitoring is carried out in real time in the case unloading process so as to avoid accidents; the report data module is mainly used for recording historical data, including partial original images and laser radar readings, and can be used for regularly training an artificial intelligent recognition system to improve recognition accuracy and keeping a file for later use; and the final PLC communication and fault alarm module is used for realizing the data interaction between the system and the crane PLC so as to alarm and indicate the abnormal state.
The core of the system software is the algorithm writing of the separation identification module. The method is divided into a laser radar data pattern recognition algorithm and a camera image pattern recognition algorithm. The laser radar data pattern recognition algorithm is used for automatically recognizing one of several typical light bar patterns, so that the state of the current truck and the container is determined, and whether the truck is lifted by the following container or not is judged. When the truck container is normally separated, the waveform is basically unchanged; once the header is lifted, the waveform data at the portion of the tire may change significantly. Depending on the position to be lifted, the waveform of a certain tire on the total waveform data may disappear. After the system monitors the change, the system compares the camera data to find that the condition that the integrated card is lifted happens, namely, an alarm is given.
The digital camera image pattern recognition algorithm monitors the front and rear areas, each field of view area being about 2.5m by 1.9m. Three types of object detection capabilities of interest are formed for truck head, truck tray and truck load by artificial intelligence algorithms. If the truck head is detected in the image and the proportion of the field of view occupied by the truck head exceeds a preset value (generally 2/3 of the field of view area), the anti-smashing alarm is started.
If a truck load and truck tray are detected, then it is further detected whether the lower edge of the load is being separated from the upper edge of the tray, whether the load is moving laterally, and whether the upper edge of the tray is rising, to determine whether the truck is currently normally separated, or whether the truck is pulling the load, or whether the truck is being lifted.
In order to reduce misjudgment, the invention also determines the system state and part of system parameters by detecting whether a load exists in a view field and whether a baffle exists in a truck tray and integrating a plurality of indexes transmitted to the system by a crane PLC. For example, if the truck is unloaded, or the truck tray is a blind, non-tapered inner tray, or the spreader is in an unlocked state, no anti-lifting or towing accidents are possible in these states, the system will be in a surveillance-only state and will not be false-alarmed.
The invention uses artificial intelligence technology to judge the anti-lifting and anti-smashing of the truck. Specifically, an image segmentation algorithm implemented by a deep neural network is adopted. A neural network is a mathematical model that applies structures similar to brain nerve synapses for information processing. If the number of layers of the neural network is very large, it can be called a deep neural network. The deep neural network segmentation algorithm used in the invention consists of a plurality of convolution layers, a pooling layer and a full connection layer. After the input image enters the network, the input image is changed into a series of feature vector data after a plurality of convolution and pooling operations, and the selection probability of each label class is finally calculated for each pixel after the feature vector is subjected to full-connection layer operation. The label classes are divided into four classes, truck head, truck tray, truck load and background, respectively. And (3) taking the label with the maximum selection probability, giving the label to the current pixel, traversing all pixels, and obtaining the segmentation result of the object of interest on the picture. This process of inputting an image to a deep neural network of existing ready parameters and then outputting a classified label image is referred to as forward propagation.
In order to obtain parameters of each layer of the deep neural network with the best effect, the best fitting is required according to training data. The optimization fitting process adopts a gradient descent method, and the purpose of minimizing the loss function is achieved through gradual iteration. And in the process of calculating the gradient once, calculating from the output end of the deep neural network, and reversely obtaining the gradient of each layer of parameters step by step from the output layer to the input layer by using a chain method. This process is called back propagation. Training data for counter-propagation is obtained from raw data acquired by a digital camera and a lidar for manual tagging. The more training data, the better the parameters of the network are optimized, and the higher the final recognition rate of the whole algorithm.
The large amount of training data for back propagation is stored on an offsite cloud server, which is called a training end. All marking and training processes are performed at the training end. The high-performance industrial personal computer used in the system site is called a deployment terminal, and only the structure and optimized parameters of the deep neural network model are saved. The deployment end realizes image segmentation through forward propagation on one hand, and stores new effective operation data on the other hand, and regularly and remotely transmits the data back to the training end. The training end adds the operation data obtained after manual cleaning, sorting and marking into the original training data to form a large truck anti-lifting video database. The training end also inherits training at random, optimizes parameters of all layers of the deep neural network, updates the parameters to the deployment end, and improves the recognition rate of various heads, trays and loads. Therefore, the software composition of the system has certain machine learning capability, and the intelligent expression is higher and higher along with the accumulation of big data until the accuracy of the complete and manual judgment is not comparable.
Drawings
FIG. 1 is a schematic diagram of a hardware connection of the present invention;
FIG. 2 is a block diagram of a deep neural network employed in the software system of the present invention;
FIG. 3 is a software functional judgment flow chart of the present invention;
FIG. 4 is a schematic diagram showing the installation position of the main hardware of the system in the embodiment 1;
fig. 5 is an actual picture taken by the system digital camera assembly of example 1, after the intelligent video analysis by the software system of the present invention. 5a is a front camera picture and 5b is a rear camera picture.
The figure indicates:
1-math camera, 2-laser radar, 3-high performance industrial personal computer, 4-intelligent video analysis module, 5-tyre crane, 6-truck (including headstock and bracket), 7-truck load (here container).
Detailed Description
The invention will now be described in detail with reference to the drawings and specific examples. The present embodiment is implemented on the premise of the technical scheme of the present invention, and a detailed implementation manner and a specific operation process are given, but the protection scope of the present invention is not limited to the following examples.
Example 1:
the technical scheme of the invention is applied to the automatic monitoring of the operation of the actual container terminal tire crane, and is used for preventing the container truck with the bracket lock head (called as an external collecting card) which is unloaded before from being crashed by the collecting truck head and from being mistakenly lifted due to the locking of the lock head.
As shown in fig. 4, a digital camera 1 is installed at the front and rear of the tire crane 5 at a height of about 1.4m from the ground; a laser radar 2 is arranged at the height of about 50cm from the ground at the lower middle part, and the scanning plane of the laser radar is 15 degrees with the horizontal plane; in the electric room of the tire crane 5, the high-performance industrial personal computer 3 of the present invention is installed. The industrial personal computer 3 is connected with the digital camera 1 and the laser radar 2 by optical fibers to receive data, and is connected with a tire crane PLC in an electric room by a network cable.
The invention works in that after the collection truck 6 enters the field of view of the front and rear digital cameras 1, the system software can automatically identify and divide out the region of the object of interest in the field of view, as shown in fig. 5a and 5 b. Fig. 5a shows the front camera segmentation result, and it can be seen that the system software automatically identifies three types of objects, namely a vehicle head (sector), a container (box) and a truck carriage (outer), and the areas thereof. In the example, the head of the vehicle does not occupy more than 1/3 of the left side of the picture, so that the anti-smashing alarm of the system cannot be triggered. Fig. 5b is a rear camera segmentation result, and the system automatically recognizes two types of objects, namely a container (box) and a truck tray (after), and areas thereof. When the tyre crane lifting tool starts to unload the container, the lifting tool firstly contacts the upper edge of the container and then locks the lifting lug on the container, and at the moment, the PLC firstly sends a lifting tool locking box signal to the system, and the process lasts for a plurality of seconds. The system can enter an alarm preparation state according to the PLC signal and the self intelligent analysis result, and continuously updates the real-time card collecting bracket height and the real-time container transverse position into the reference value in the system until the locking box signal sent by the PLC disappears. Then the tyre crane enters the lifting process, the locking box signal disappears, at the moment, the lifting appliance keeps the locking state of the lifting lug lock head on the container, the lifting appliance does not contact the upper edge of the container any more, and the PLC correspondingly sends the locking box signal to the system. During the period of blocking the unset signal, the system continues to do two processes. Firstly, in the current video, the bracket height which is intelligently analyzed is compared with a previously stored reference value, so that the function of preventing the truck from being lifted is realized. And secondly, comparing the transverse position of the current container with a previously stored reference value, thereby realizing the function of preventing dragging. If the current height of the collecting card bracket exceeds the reference value by 20cm or the transverse displacement of the container exceeds the reference value by 20cm, sending an alarm signal to the PLC; if the height exceeds the standard value by 25cm or the displacement exceeds 25cm, the lifting process of the lifting appliance is cut off except for an alarm signal until manual intervention.
Through innovation and improvement of the invention, the application example can automatically adapt to various truck head models and bracket models, and the misjudgment rate is continuously reduced along with the accumulation of a system database until the misjudgment rate is completely different from the accuracy rate of manual monitoring judgment. Therefore, the embodiment can completely replace manual monitoring and ensure the safety of the tire crane operation process.
The previous description of the embodiments is provided to facilitate a person of ordinary skill in the art in order to make and use the present invention. It will be apparent to those skilled in the art that various modifications can be readily made to these embodiments and the generic principles described herein may be applied to other embodiments without the use of the inventive faculty. Therefore, the present invention is not limited to the above-described embodiments, and those skilled in the art, based on the present disclosure, should make improvements and modifications without departing from the scope of the present invention.

Claims (3)

1. The intelligent truck safety protection system for container hoisting operation is characterized in that the system hardware comprises a digital camera (1), a laser radar (2) and a high-performance industrial personal computer (3), wherein the digital camera (1) and the laser radar (2) are connected with the high-performance industrial personal computer (3) through optical fibers, the system software comprises an intelligent video analysis module (4), the intelligent video analysis module (4) adopts a deep neural network algorithm, three interested objects of a truck head, a truck tray and a truck load in an image are found out through a pixel discrimination method, the intelligent video analysis module (4) adopts a deep neural network to realize an image segmentation algorithm, the segmentation algorithm distributes a label for each pixel on an input image, and the label is selected from the four types of truck head, truck tray, truck load and background, and the specific selection process is as follows: the picture is used as an input item to be transmitted into a neural network, the output item is the selection probability of each pixel for each label class, the label with the largest selection probability is taken, the label is assigned to the current pixel, the process is called forward propagation,
the digital cameras (1) are arranged at two ends of the tire crane, and the installation positions of the digital cameras can ensure that the truck tray and the truck load can enter the camera view fields at different truck unloading positions; meanwhile, when the truck is at the normal unloading position, the truck head does not occupy the range of more than 2/3 of the camera field of view,
the laser radar (2) is arranged at the lower middle part of the tire crane and is 200-1000 mm away from the ground, and the included angle between the scanning plane of the laser radar and the horizontal plane is within +/-15 degrees, so that the truck can sweep a plurality of tires on the side surface of the truck when the truck is in a normal operation position.
2. The intelligent truck safety protection system for container hoisting operation according to claim 1, characterized in that the video analysis module (4) adopts a deep neural network in an image processing algorithm, parameters of each layer of the network are optimally fitted according to training data, a gradient descent method is adopted in the process of optimizing the fitting, the purpose of minimizing a loss function is achieved through gradual iteration, a single gradient calculation process is called back propagation, training data for back propagation is obtained by manually marking raw data acquired by the digital camera (1) and the laser radar (2).
3. The intelligent truck safety protection system for container hoisting operation according to claim 2, characterized in that the video analysis module (4) uses a deep neural network in an image processing algorithm to reversely spread a large amount of training data, the training data is stored on an offsite cloud server, the server is called a training end, all marking and training processes are carried out on the training end, a high-performance industrial personal computer (3) used in the system on site is called a deployment end, only the structure and optimized parameters of a deep neural network model are stored, the deployment end realizes image segmentation through forward spreading, on the other hand, new effective operation data are stored, the data are transmitted back to the training end remotely at regular intervals, the training end adds the operation data obtained after manual cleaning, finishing and marking into original training data to form a large truck anti-hoisting video database, the training end does not inherit training at regular intervals, the parameters of all layers of the deep neural network are optimized, and the parameters of the deep neural network are updated to the deployment end.
CN202010541326.0A 2020-06-15 2020-06-15 Truck safety intelligent protection system for container hoisting operation Active CN111832415B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010541326.0A CN111832415B (en) 2020-06-15 2020-06-15 Truck safety intelligent protection system for container hoisting operation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010541326.0A CN111832415B (en) 2020-06-15 2020-06-15 Truck safety intelligent protection system for container hoisting operation

Publications (2)

Publication Number Publication Date
CN111832415A CN111832415A (en) 2020-10-27
CN111832415B true CN111832415B (en) 2023-12-26

Family

ID=72897666

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010541326.0A Active CN111832415B (en) 2020-06-15 2020-06-15 Truck safety intelligent protection system for container hoisting operation

Country Status (1)

Country Link
CN (1) CN111832415B (en)

Families Citing this family (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112883750A (en) * 2020-11-26 2021-06-01 航天智造（上海）科技有限责任公司 Dynamic two-dimensional code dynamic generation and reading system for AGV positioning navigation and communication
CN112580517A (en) * 2020-12-22 2021-03-30 上海振华重工(集团)股份有限公司 Anti-smashing protection system and method for truck head, computer storage medium and gantry crane
CN113184707B (en) * 2021-01-15 2023-06-02 福建电子口岸股份有限公司 Method and system for preventing lifting of collection card based on laser vision fusion and deep learning
CN112978579B (en) * 2021-05-13 2021-07-23 新乡职业技术学院 Crane with anti-collision control system
CN113291975B (en) * 2021-05-24 2023-10-24 万雄集团(沈阳)智能科技有限公司 A collection card parks and aims at and prevents pounding and prevent lifting device for tire crane
CN113420646B (en) * 2021-06-22 2023-04-07 天津港第二集装箱码头有限公司 Lock station connection lock detection system and method based on deep learning
CN114104980B (en) * 2021-10-15 2023-06-02 福建电子口岸股份有限公司 Safe operation control method and system for quay crane based on combination of AI and vision
KR102454058B1 (en) * 2022-05-17 2022-10-14 주식회사 이노메트릭스 Crane safety management system

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1594791A1 (en) * 2002-12-31 2005-11-16 Abb Ab A system and a method for container identification
CN203976241U (en) * 2014-06-12 2014-12-03 上海海镭激光科技有限公司 The anti-system of slinging of tyre crane walking positioning and rectifying and truck contraposition
CN106829742A (en) * 2017-03-20 2017-06-13 北京国泰星云科技有限公司 A kind of anti-detecting system and method for slinging based on laser
CN106986272A (en) * 2017-02-24 2017-07-28 北京航天自动控制研究所 It is a kind of to prevent slinging method and system based on the container container car that machine vision is tracked
CN206427928U (en) * 2016-11-10 2017-08-22 北京国泰星云科技有限公司 Operation auxiliary system under bank bridge
CN109858573A (en) * 2019-03-14 2019-06-07 上海西井信息科技有限公司 Truck neural network based is anti-to sling method
CN111196559A (en) * 2018-11-19 2020-05-26 邵筱芳 Gantry crane container lifting control system and method

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP1594791A1 (en) * 2002-12-31 2005-11-16 Abb Ab A system and a method for container identification
CN203976241U (en) * 2014-06-12 2014-12-03 上海海镭激光科技有限公司 The anti-system of slinging of tyre crane walking positioning and rectifying and truck contraposition
CN206427928U (en) * 2016-11-10 2017-08-22 北京国泰星云科技有限公司 Operation auxiliary system under bank bridge
CN106986272A (en) * 2017-02-24 2017-07-28 北京航天自动控制研究所 It is a kind of to prevent slinging method and system based on the container container car that machine vision is tracked
CN106829742A (en) * 2017-03-20 2017-06-13 北京国泰星云科技有限公司 A kind of anti-detecting system and method for slinging based on laser
CN111196559A (en) * 2018-11-19 2020-05-26 邵筱芳 Gantry crane container lifting control system and method
CN109858573A (en) * 2019-03-14 2019-06-07 上海西井信息科技有限公司 Truck neural network based is anti-to sling method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
集卡防吊起技术方案探讨;陈鹏元;李书强;余珍胜;;港工技术(第02期) *

Also Published As

Publication number Publication date
CN111832415A (en) 2020-10-27

Similar Documents

Publication Publication Date Title
CN111832415B (en) Truck safety intelligent protection system for container hoisting operation
CN110532889B (en) Track foreign matter detection method based on rotor unmanned aerial vehicle and YOLOv3
CN105744232B (en) A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology
CN114132842A (en) Real-time monitoring system and monitoring method for operation state of container gantry crane storage yard
CN107578012A (en) A kind of drive assist system based on clustering algorithm selection sensitizing range
CN111046797A (en) Oil pipeline warning method based on personnel and vehicle behavior analysis
CN107200274A (en) A kind of anti-container truck based on machine vision is lifted method
CN112661013B (en) Automatic wharf bridge crane legacy lock pad detection method and system
CN115205796B (en) Rail line foreign matter intrusion monitoring and risk early warning method and system
CN105321289B (en) A kind of round-the-clock monitoring image intellectual analysis and warning system and method
Srinivas et al. Image processing edge detection technique used for traffic control problem
CN111539344A (en) Control system and method for preventing container truck from being lifted based on video stream and artificial intelligence
CN113093746A (en) Working machine environment sensing method, device and system and working machine
CN111667655A (en) Infrared image-based high-speed railway safety area intrusion alarm device and method
KR102296471B1 (en) Artificial intelligence-based vehicle load failure detection system
CN105046223A (en) Device for detecting severity of ''black-hole effect'' at tunnel entrance and method thereof
CN111646349B (en) Elevator protection method and device based on TOF image
CN215439237U (en) Intelligent truck safety protection system for container hoisting operation
CN116092003A (en) Automatic identification method for preventing lifting of collection card based on multi-view visual detection
CN216613787U (en) Real-time monitoring system for operation state of container gantry crane storage yard
EP3894349A1 (en) Container-loading system and method for monitoring operation therein
CN114902282A (en) System and method for efficient sensing of collision threats
CN114613098B (en) Tray stacking out-of-range detection method
CN117622228B (en) Remote control method and device for unmanned automatic operation automobile in automobile
CN219351804U (en) Monitoring and alarming integrated device for anti-hanging of collection card

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant