CN117612124A - Anti-collision detection method, device, equipment and medium for intelligent car dumper - Google Patents

Anti-collision detection method, device, equipment and medium for intelligent car dumper Download PDF

Info

Publication number
CN117612124A
CN117612124A CN202311682863.7A CN202311682863A CN117612124A CN 117612124 A CN117612124 A CN 117612124A CN 202311682863 A CN202311682863 A CN 202311682863A CN 117612124 A CN117612124 A CN 117612124A
Authority
CN
China
Prior art keywords
carriage
hook
car
video image
position information
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.)
Pending
Application number
CN202311682863.7A
Other languages
Chinese (zh)
Inventor
刘翔
周鹏
姚林发
刘照林
付青俊
刘斯洛
郑袁
闫印强
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.)
Golden Port Power Generation Co ltd
Original Assignee
Golden Port Power Generation 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 Golden Port Power Generation Co ltd filed Critical Golden Port Power Generation Co ltd
Priority to CN202311682863.7A priority Critical patent/CN117612124A/en
Publication of CN117612124A publication Critical patent/CN117612124A/en
Pending legal-status Critical Current

Links

Classifications

    • 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
    • G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
    • 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
    • G06N3/0464—Convolutional networks [CNN, ConvNet]
    • 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
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/07—Target detection

Abstract

The invention relates to the technical field of target detection, in particular to an anti-collision detection method, device, equipment and medium for an intelligent car dumper. The method comprises the following steps: acquiring video stream data of point position cameras on the front side and the rear side of the car dumper in real time; performing real-time frame extraction processing on video stream data, and performing, for each video image extracted from each side: inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; wherein, the front side and the rear side of the carriage are provided with hooks; judging whether the carriage is stationary or not based on the position information of the carriage in each two adjacent video images after frame extraction; after the car is stationary, it is determined whether the car and the hook exceed specified limits on the side based on the position information of the car and the position information of the hook to give an alarm when the car and the hook exceed the specified limits. The scheme can achieve the intellectualization, high efficiency and low cost of the anti-collision detection of the carriage of the dumper through a computer vision algorithm.

Description

Anti-collision detection method, device, equipment and medium for intelligent car dumper
Technical Field
The embodiment of the invention relates to the technical field of target detection, in particular to an anti-collision detection method, device, equipment and medium for an intelligent car dumper.
Background
Safety is a primary principle in industrial operations. The dumper is large mechanical equipment for dumping bulk materials (coal, ore sand and the like) of railway open cars, and the dumper is started to overturn when a carriage does not reach a designated position right below the dumper, so that the carriage, a hook and the ground edge near the dumper collide, and serious safety accidents occur.
The existing anti-collision detection method of the car dumper is to artificially detect whether a carriage reaches the appointed position of the car dumper and whether a hook exceeds the appointed limit so as to avoid accidents. However, the human detection method is not only inefficient but also costly in terms of labor.
Therefore, a method for detecting collision of the intelligent car dumper is needed.
Disclosure of Invention
In order to solve the problems of low efficiency, low intelligent degree and high labor cost of a mode for artificially detecting and preventing collision edges of a carriage and a hook of an intelligent car dumper, the embodiment of the invention provides an anti-collision detection method, device, equipment and medium of the intelligent car dumper.
In a first aspect, an embodiment of the present invention provides a method for detecting an anti-collision of an intelligent car dumper, where the method includes:
acquiring video stream data of point position cameras on the front side and the rear side of the car dumper in real time;
performing real-time frame extraction processing on the video stream data, and performing, for each video image extracted from each side:
inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; the hooks are arranged on the front side and the rear side of the carriage;
judging whether the carriage is stationary or not based on the position information of the carriage in every two adjacent video images after frame extraction;
after the carriage is stationary, based on the position information of the carriage and the position information of the hook, it is judged whether the carriage and the hook exceed the specified limits of the side or not, so as to give an alarm when the carriage and the hook exceed the specified limits.
In a second aspect, an embodiment of the present invention further provides an apparatus for detecting an anti-collision of an intelligent car dumper, where the apparatus includes:
the acquisition unit is used for acquiring video stream data of the point position cameras at the front side and the rear side of the car dumper in real time;
the identifying unit is used for carrying out real-time frame extraction processing on the video stream data, and for each video image extracted from each side, the identifying unit is used for executing the following steps: inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; the hooks are arranged on the front side and the rear side of the carriage;
the static unit is used for judging whether the carriage is static or not based on the position information of the carriage in every two adjacent video images after frame extraction;
and the judging unit is used for judging whether the carriage and the hook exceed the specified limit of the side or not based on the position information of the carriage and the position information of the hook after the carriage is stationary so as to give an alarm when the carriage and the hook exceed the specified limit.
In a third aspect, an embodiment of the present invention further provides a computing device, including a memory and a processor, where the memory stores a computer program, and the processor implements a method according to any embodiment of the present specification when executing the computer program.
In a fourth aspect, embodiments of the present invention also provide a computer-readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform a method according to any of the embodiments of the present specification.
The embodiment of the invention provides an anti-collision detection method, device, equipment and medium for an intelligent car dumper, which are characterized in that video stream data shot by point cameras at the front side and the rear side of the car dumper are detected in real time by utilizing a detection model, the position information of carriages and hooks at the front side and the rear side is identified, after the carriage is judged to be stationary, whether the carriages and the hooks at the front side and the rear side exceed a specified limit is judged according to the position information, so that an alarm is sent when the carriage is exceeded. According to the scheme, the intelligent, efficient and low-cost anti-collision detection of the car of the dumper can be achieved through a computer vision algorithm.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. The drawings in the following description are illustrative of certain embodiments of the invention and other drawings may be made by those skilled in the art without undue burden.
FIG. 1 is a flowchart of an anti-collision detection method for an intelligent car dumper according to an embodiment of the invention;
FIG. 2 is a diagram illustrating a video image according to an embodiment of the present invention;
FIG. 3 is a view of another video image according to an embodiment of the present invention;
FIG. 4 is a view of still another video image according to an embodiment of the present invention;
FIG. 5 is a hardware architecture diagram of a computing device according to one embodiment of the invention;
fig. 6 is a structural diagram of an anti-collision detection device of an intelligent car dumper according to an embodiment of the invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. It is apparent that the described embodiments are some, but not all embodiments of the present invention, and that all other embodiments obtained by persons of ordinary skill in the art without making creative efforts based on the embodiments in the present invention are within the protection scope of the present invention.
Specific implementations of the above concepts are described below.
Referring to fig. 1, an embodiment of the present invention provides a method for generating a media image of a non-cooperative target, including:
step 100, video stream data of point position cameras on the front side and the rear side of the car dumper are obtained in real time;
step 102, performing real-time frame extraction processing on the video stream data, and for each video image extracted from each side, performing: inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; wherein, the front side and the rear side of the carriage are provided with hooks;
104, judging whether the carriage is stationary or not based on the position information of the carriage in each two adjacent video images after frame extraction;
and step 106, after the carriage is stationary, judging whether the carriage and the hook exceed the specified limit of the side or not based on the carriage position information and the hook position information so as to give an alarm when the carriage and the hook exceed the specified limit.
In the embodiment of the invention, the position information of the carriage and the hook at the front side and the rear side is identified by detecting the video stream data shot by the point position cameras at the front side and the rear side of the car dumper in real time by using the detection model, and after the carriage is judged to be stationary, whether the carriage and the hook at the front side and the rear side exceed the specified limit is judged according to the position information so as to give an alarm when the carriage and the hook exceed the specified limit. According to the scheme, the intelligent, efficient and low-cost anti-collision detection of the car of the dumper can be achieved through a computer vision algorithm.
For steps 100 and 102:
in some embodiments, step 102 may include:
carrying out illumination enhancement on the current video image of the side by using a CSNorm module of the detection model to obtain an enhanced image;
and identifying and detecting the enhanced image by using a target detection module of the detection model to obtain the position information of the carriage and the hook in the current video image.
In this embodiment, since the conventional target detection algorithm is greatly affected by illumination. Particularly, at night, the brightness of the picture is low, the target characteristics are not obvious, the detection rate of the model is greatly influenced, the position information of the carriage and the hook cannot be determined, and finally whether the carriage and the hook exceed the appointed limit of the car dumper cannot be accurately judged. The CSNorm is a technology capable of improving illumination adaptability of the model, and can generalize unknown illumination conditions while maintaining reconstruction capability of known illumination conditions, so that data illumination conditions of the car dumper at night are enhanced, and the detection rate of the model to a carriage and a hook at night is improved.
Therefore, the detection rate of the dumper, the carriage and the hook at night can be improved by combining the target detection module with CSNorm, and the judgment of the position information of the dumper, the carriage and the hook is facilitated, so that the detection of the anti-collision of the dumper carriage is more accurately completed. Reference may be made to fig. 2 and 3, where fig. 2 is a model output image of a tippler point when CSNorm is not used, fig. 3 is a model output image of a tippler point when CSNorm is used, and it can be seen that fig. 3 and 2 are at night 22: the image of the same frame of 56 minutes is obviously brighter and clearer than that of fig. 2 in fig. 3 after CSNorm treatment, and the carriage and the hook are also easier to identify.
In some embodiments, the target detection module is a Yolov5 network.
In this embodiment, the target detection module is a Yolov5 network, and the detection model of this embodiment is based on a CSNorm-Yolov5 detection model, and on the basis that the Yolov5 target detection model can detect the car and the hook in high precision and real time, combining with CSNorm can improve the generalization ability of the model under unknown illumination conditions, improve the detection rate of the Yolov5 on the car dumper, the car and the hook at night, and finally accurately judge whether the car and the hook exceed the edge of the car dumper and impact the ground. Compared with the traditional manual detection mode, the method realizes the intellectualization, high efficiency and low cost of the anti-collision of the car dumper. With continued reference to fig. 2 and 3, it can be seen that fig. 3 and 2 are at night 22:56 minutes of the same frame image, the carriage and the hook are detected in the figure 3 after the CSNor treatment, the carriage is detected in the figure 2 without the CSNor treatment, and the confidence is not higher than that in the figure 3.
For step 104:
in some embodiments, step 104 may include:
for every two adjacent video images after frame extraction, executing:
determining intersection and union of a position frame of a carriage in a video image of a previous frame and a position frame of a carriage in a video image of a next frame;
determining an intersection ratio of two adjacent video images based on the intersection and the union;
and when the intersection ratio is larger than the set threshold value, determining that the carriage is stationary.
In the step, the point positions on both sides can be respectively judged whether to be static or not, and the association processing is not carried out. It can be understood that after both side point positions are judged to be stationary, whether the specified limit is exceeded or not can be judged, or only one side point position is judged to be stationary, and the device can be set according to actual conditions.
In the embodiment of the invention, the blending ratio of two adjacent video images can be calculated by the following formula:
wherein IOU is the intersection ratio of two adjacent video images, A is the position frame of the carriage in the video image of the previous frame, and B is the position frame of the carriage in the video image of the next frame.
It will be appreciated that when the car is stationary, the position frames of the car in the front and rear frames of video images are theoretically coincident, the closer the IOU should be to 1, and when the car is moving, the smaller the intersection of the position frames of the car in the front and rear frames of video images should be, the closer the IOU should be to 0, so that when the intersection ratio is greater than the set threshold, it can be determined that the car is stationary.
For step 106:
in some embodiments, the step of "determining whether the car and the hook exceed the specified limits of the side based on the position information of the car and the position information of the hook to issue an alarm when exceeded" may include:
acquiring a position function of a specified limit of the side in the video image; the position function is an abscissa-ordinate relation function of a specified limit in the video image;
determining the abscissa of the midpoint of the bottom edge of the carriage position frame and the abscissa of the midpoint of the bottom edge of the hook position frame based on the position information of the carriage and the hook;
substituting the abscissa of the midpoint of the bottom edge of the carriage position frame and the abscissa of the midpoint of the bottom edge of the hook position frame into a position function respectively to obtain a first corresponding ordinate and a second corresponding ordinate;
and judging whether the carriage and the hook exceed the appointed limit of the side or not based on the magnitude relation between the ordinate of the midpoint of the bottom edge of the carriage position frame and the first corresponding ordinate and the magnitude relation between the ordinate of the midpoint of the bottom edge of the hook position frame and the second corresponding ordinate, so as to give an alarm when the carriage and the hook exceed the appointed limit of the side.
In this embodiment, the specified limit may be a ground step line of the pit slot, may be a warning line drawn in the pit slot of the dumper, and the position function may be a linear function or a curve function, which is determined according to the actual situation and the detection precision.
Because each extracted frame of image needs to be detected in real time, the detection time cannot be too long, in order to improve the real-time monitoring speed, in the embodiment, only the midpoint of the bottom edge of the carriage position frame and the midpoint of the bottom edge of the hook position frame are used for comparison with the appointed limit, but the midpoint and the bottom edge end point of the carriage and the position frame of the hook are used for comparison with the appointed limit, so that the detection speed is improved.
In some embodiments, the step of "determining whether the car and the hook exceed the specified limits of the side to issue an alarm when exceeded" based on the magnitude relationship of the ordinate of the midpoint of the bottom edge of the car position frame and the first corresponding ordinate and the magnitude relationship of the ordinate of the midpoint of the bottom edge of the hook position frame and the second corresponding ordinate may include:
when the ordinate of the midpoint of the bottom edge of the carriage position frame is greater than or equal to the first corresponding ordinate, determining that the carriage exceeds the appointed limit of the side;
when the ordinate of the midpoint of the bottom edge of the hook position frame is greater than or equal to the second corresponding ordinate, determining that the hook exceeds the appointed limit of the side;
and when at least one of the carriage and the hook exceeds the specified limit, an alarm is sent to a worker.
Referring to fig. 4, in this embodiment, the coordinates of the image generally select the top left corner vertex as the origin, the horizontal direction is the x axis, the vertical direction is the y axis, and the designated limit must be located at the lower side of the image than the carriage and the hook, when the ordinate of the midpoint of the bottom edge of the carriage position frame is greater than or equal to the first corresponding ordinate, the position of the abscissa corresponding to the designated limit in the bottom edge representing the carriage position frame is hidden, so that the carriage is easily collided on the ground of the edge of the dumper when the dumper is started, and therefore, when the ordinate of the midpoint of the bottom edge of the carriage position frame is greater than or equal to the first corresponding ordinate, it can be determined that the carriage exceeds the designated limit of the side. The hooks are the same.
In addition, the hook is generally close to the lower edge of the carriage, and in the output position frame of the detection model, the bottom edge of the position frame of the hook is extremely likely to be arranged below the bottom edge of the position frame of the carriage, so that if only whether the carriage exceeds a specified limit is judged, whether the hook exceeds the specified limit is not judged, and the situation that judgment errors easily occur is easy to judge, therefore, whether the carriage and the hook exceed the specified limit is judged, and if at least one of the carriage and the hook exceeds the specified limit, an alarm is sent to a worker.
In some embodiments, the location function of the specified limits on each side in the video image is determined by:
fixing the mounting position and shooting angle of the side point position camera of the car dumper;
and when a preset specified limit in the video image of the camera is not blocked, determining a position function of the specified limit in the video image.
In this embodiment, reference can be made to fig. 3 and 4, in which fig. 3 the specified limits are obscured by the preceding car, and fig. 4 the preceding no car, the specified limits preset for this embodiment can be seen. In actual detection, since the specified limits in the images are most likely to be blocked, if the specified limits of each video image are detected in real time by using the detection model, the detection results are most likely to be affected if the specified limits are not detected due to the blocking.
Therefore, in this embodiment, the installation position and the shooting angle of the side point position camera of the car dumper need to be fixed in advance, so that the position function of the specified limit in the image in each video image is unchanged, and when the preset specified limit in the video image is not blocked, the position function of the specified limit in the video image can be identified by using the detection model, and the position function of the specified limit is set as a fixed function, and before the installation position and the shooting angle of the side point position camera are not changed, the position function of the specified limit is directly used as the fixed function, and real-time re-detection is not needed.
In this embodiment, the location function specifying the limit can be expressed as:
y＝kx+b
where y is the ordinate corresponding to the specified limit, x is a plurality of abscissas corresponding to the specified limit, k is the slope, and b is a constant.
It will be appreciated that the ground steps of the dumper pit can be considered as a straight line, but in practice it is also possible to have a curve, so that the function of the position of the specified limit can be determined according to the practice.
As shown in fig. 5 and 6, the embodiment of the invention provides an anti-collision detection device of an intelligent car dumper. The apparatus embodiments may be implemented by software, or may be implemented by hardware or a combination of hardware and software. From the hardware level, as shown in fig. 5, a hardware architecture diagram of a computing device where an anti-collision detection device of an intelligent car dumper provided by an embodiment of the invention is located is shown. In addition to the processor, memory, network interfaces, and non-volatile storage shown in fig. 5, the computing device in which the apparatus is located in an embodiment may generally include other hardware, such as a forwarding chip or the like that is responsible for processing messages. Taking a software implementation as an example, as shown in fig. 6, as a device in a logic sense, the device is formed by reading a corresponding computer program in a nonvolatile memory into a memory by a CPU of a computing device where the device is located. The embodiment provides a detection device that intelligent tipper anticollision, the device includes:
the acquiring unit 601 is configured to acquire video stream data of point cameras on the front side and the rear side of the car dumper in real time;
an identifying unit 602, configured to perform real-time frame extraction processing on video stream data, and perform, for each video image extracted from each side: inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; wherein, the front side and the rear side of the carriage are provided with hooks;
a stationary unit 603, configured to determine whether the carriage is stationary based on the position information of the carriage in each two adjacent video images after frame extraction;
and a judging unit 604 for judging whether the car and the hook exceed the specified limits of the side based on the position information of the car and the position information of the hook after the car is stationary, so as to give an alarm when the car and the hook exceed the specified limits.
In one embodiment of the invention, the identification unit 602 is configured to perform:
carrying out illumination enhancement on the current video image of the side by using a CSNorm module of the detection model to obtain an enhanced image;
and identifying and detecting the enhanced image by using a target detection module of the detection model to obtain the position information of the carriage and the hook in the current video image.
In one embodiment of the present invention, the object detection module in the identification unit 602 is a Yolov5 network.
In one embodiment of the invention, the stationary unit 603 is configured to perform:
for every two adjacent video images after frame extraction, executing:
determining intersection and union of a position frame of a carriage in a video image of a previous frame and a position frame of a carriage in a video image of a next frame;
determining an intersection ratio of two adjacent video images based on the intersection and the union;
and when the intersection ratio is larger than the set threshold value, determining that the carriage is stationary.
In one embodiment of the present invention, the determining unit 604 is configured to perform:
acquiring a position function of a specified limit of the side in the video image; the position function is an abscissa-ordinate relation function of a specified limit in the video image;
determining the abscissa of the midpoint of the bottom edge of the carriage position frame and the abscissa of the midpoint of the bottom edge of the hook position frame based on the position information of the carriage and the hook;
substituting the abscissa of the midpoint of the bottom edge of the carriage position frame and the abscissa of the midpoint of the bottom edge of the hook position frame into a position function respectively to obtain a first corresponding ordinate and a second corresponding ordinate;
and judging whether the carriage and the hook exceed the appointed limit of the side or not based on the magnitude relation between the ordinate of the midpoint of the bottom edge of the carriage position frame and the first corresponding ordinate and the magnitude relation between the ordinate of the midpoint of the bottom edge of the hook position frame and the second corresponding ordinate, so as to give an alarm when the carriage and the hook exceed the appointed limit of the side.
In one embodiment of the present invention, the determining unit 604 is configured to determine whether the car and the hook exceed the specified limits of the side based on the magnitude relation between the ordinate of the midpoint of the bottom edge of the car position frame and the first corresponding ordinate and the magnitude relation between the ordinate of the midpoint of the bottom edge of the hook position frame and the second corresponding ordinate, so as to issue an alarm when the car and the hook exceed the specified limits:
when the ordinate of the midpoint of the bottom edge of the carriage position frame is greater than or equal to the first corresponding ordinate, determining that the carriage exceeds the appointed limit of the side;
when the ordinate of the midpoint of the bottom edge of the hook position frame is greater than or equal to the second corresponding ordinate, determining that the hook exceeds the appointed limit of the side;
and when at least one of the carriage and the hook exceeds the specified limit, an alarm is sent to a worker.
In one embodiment of the present invention, the location function of the specified limits on each side of the determination unit 604 in the video image is determined by:
fixing the mounting position and shooting angle of the side point position camera of the car dumper;
and when a preset specified limit in the video image of the camera is not blocked, determining a position function of the specified limit in the video image.
It will be appreciated that the structure illustrated in the embodiments of the present invention does not constitute a specific limitation on an anti-collision detection device for an intelligent car dumper. In other embodiments of the invention, an intelligent dumper collision avoidance detection device may include more or fewer components than shown, or may combine certain components, or split certain components, or may have a different arrangement of components. The illustrated components may be implemented in hardware, software, or a combination of software and hardware.
The content of information interaction and execution process between the units in the device is based on the same conception as the embodiment of the method of the present invention, and specific content can be referred to the description in the embodiment of the method of the present invention, which is not repeated here.
The embodiment of the invention also provides a computing device which comprises a memory and a processor, wherein the memory stores a computer program, and the processor realizes the anti-collision detection method of the intelligent car dumper in any embodiment of the invention when executing the computer program.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium is stored with a computer program which, when being executed by a processor, causes the processor to execute the anti-collision detection method of the intelligent car dumper in any embodiment of the invention.
Specifically, a system or apparatus provided with a storage medium on which a software program code realizing the functions of any of the above embodiments is stored, and a computer (or CPU or MPU) of the system or apparatus may be caused to read out and execute the program code stored in the storage medium.
In this case, the program code itself read from the storage medium may realize the functions of any of the above-described embodiments, and thus the program code and the storage medium storing the program code form part of the present invention.
Examples of the storage medium for providing the program code include a floppy disk, a hard disk, a magneto-optical disk, an optical disk (e.g., CD-ROM, CD-R, CD-RW, DVD-ROM, DVD-RAM, DVD-RW, DVD+RW), a magnetic tape, a nonvolatile memory card, and a ROM. Alternatively, the program code may be downloaded from a server computer by a communication network.
It should be clear that the functions of any of the above-described embodiments may be implemented not only by executing the program code read out by the computer, but also by causing an operating system or the like operating on the computer to perform a part or all of the actual operations based on the instructions of the program code.
Further, it is understood that the program code read out by the storage medium is written into a memory provided in an expansion board inserted into a computer or into a memory provided in an expansion module connected to the computer, and then a CPU or the like mounted on the expansion board or the expansion module is caused to perform part and all of actual operations based on instructions of the program code, thereby realizing the functions of any of the above embodiments.
It is noted that relational terms such as first and second, and the like, are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: various media in which program code may be stored, such as ROM, RAM, magnetic or optical disks.
Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present invention, and are not limiting; although the invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims (10)

1. The anti-collision detection method for the intelligent car dumper is characterized by comprising the following steps of:
acquiring video stream data of point position cameras on the front side and the rear side of the car dumper in real time;
performing real-time frame extraction processing on the video stream data, and performing, for each video image extracted from each side:
inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; the hooks are arranged on the front side and the rear side of the carriage;
judging whether the carriage is stationary or not based on the position information of the carriage in every two adjacent video images after frame extraction;
after the carriage is stationary, based on the position information of the carriage and the position information of the hook, it is judged whether the carriage and the hook exceed the specified limits of the side or not, so as to give an alarm when the carriage and the hook exceed the specified limits.
2. The method according to claim 1, wherein the step of inputting the current video image of the side into a pre-trained detection model to identify the position information of the car and the hook in the current video image includes:
carrying out illumination enhancement on the current video image of the side by using a CSNorm module of the detection model to obtain an enhanced image;
and identifying and detecting the enhanced image by utilizing a target detection module of the detection model to obtain the position information of the carriage and the hook in the current video image.
3. The method of claim 2, wherein the object detection module is a Yolov5 network.
4. The method of claim 1, wherein determining whether the car is stationary based on the position information of the car in each two adjacent video images after the frame extraction comprises:
for every two adjacent video images after frame extraction, executing:
determining intersection and union of the position frame of the carriage in the video image of the previous frame and the position frame of the carriage in the video image of the next frame;
determining an intersection ratio of two adjacent video images based on the intersection and the union;
and when the intersection ratio is larger than a set threshold value, determining that the carriage is stationary.
5. The method of claim 1, wherein the determining whether the car and the hook exceed a specified limit on the side based on the car's position information and the hook's position information to issue an alarm when exceeded, comprises:
acquiring a position function of a specified limit of the side in the video image; the position function is an abscissa-ordinate relation function of the specified limit in the video image;
determining the abscissa of the midpoint of the bottom edge of the carriage position frame and the abscissa of the midpoint of the bottom edge of the hook position frame based on the position information of the carriage and the hook;
substituting the abscissa of the midpoint of the bottom edge of the carriage position frame and the abscissa of the midpoint of the bottom edge of the hook position frame into the position function respectively to obtain a first corresponding ordinate and a second corresponding ordinate;
and judging whether the carriage and the hook exceed the appointed limit of the side or not based on the magnitude relation between the ordinate of the midpoint of the bottom edge of the carriage position frame and the first corresponding ordinate and the magnitude relation between the ordinate of the midpoint of the bottom edge of the hook position frame and the second corresponding ordinate, so as to give an alarm when the carriage and the hook exceed the appointed limit of the side.
6. The method of claim 5, wherein said determining whether the car and the hook exceed the specified limit of the side based on the magnitude relationship of the ordinate of the midpoint of the bottom edge of the car position frame and the first corresponding ordinate and the magnitude relationship of the ordinate of the midpoint of the bottom edge of the hook position frame and the second corresponding ordinate, to alert if exceeded, comprises:
determining that the car exceeds the specified limit of the side when the ordinate of the midpoint of the bottom edge of the car position frame is greater than or equal to the first corresponding ordinate;
when the ordinate of the midpoint of the bottom edge of the hook position frame is greater than or equal to the second corresponding ordinate, determining that the hook exceeds the specified limit of the side;
and when at least one of the carriage and the hook exceeds the specified limit, alarming to staff.
7. The method of claim 5, wherein the location function of the specified limits on each side in the video image is determined by:
fixing the mounting position and shooting angle of the side point position camera of the car dumper;
and when a preset specified limit in the video image of the camera is not blocked, determining a position function of the specified limit in the video image.
8. An intelligent car dumper anti-collision detection device, which is characterized by comprising:
the acquisition unit is used for acquiring video stream data of the point position cameras at the front side and the rear side of the car dumper in real time;
the identifying unit is used for carrying out real-time frame extraction processing on the video stream data, and for each video image extracted from each side, the identifying unit is used for executing the following steps: inputting the current video image of the side to a pre-trained detection model, and identifying to obtain the position information of the carriage and the hook in the current video image; the hooks are arranged on the front side and the rear side of the carriage;
the static unit is used for judging whether the carriage is static or not based on the position information of the carriage in every two adjacent video images after frame extraction;
and the judging unit is used for judging whether the carriage and the hook exceed the specified limit of the side or not based on the position information of the carriage and the position information of the hook after the carriage is stationary so as to give an alarm when the carriage and the hook exceed the specified limit.
9. A computing device comprising a memory and a processor, the memory having stored therein a computer program, the processor implementing the method of any of claims 1-7 when the computer program is executed.
10. A computer readable storage medium having stored thereon a computer program which, when executed in a computer, causes the computer to perform the method of any of claims 1-7.
CN202311682863.7A 2023-12-08 2023-12-08 Anti-collision detection method, device, equipment and medium for intelligent car dumper Pending CN117612124A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202311682863.7A CN117612124A (en) 2023-12-08 2023-12-08 Anti-collision detection method, device, equipment and medium for intelligent car dumper

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202311682863.7A CN117612124A (en) 2023-12-08 2023-12-08 Anti-collision detection method, device, equipment and medium for intelligent car dumper

Publications (1)

Publication Number Publication Date
CN117612124A true CN117612124A (en) 2024-02-27

Family

ID=89949766

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202311682863.7A Pending CN117612124A (en) 2023-12-08 2023-12-08 Anti-collision detection method, device, equipment and medium for intelligent car dumper

Country Status (1)

Country Link
CN (1) CN117612124A (en)

Similar Documents

Publication Publication Date Title
JP3503230B2 (en) Nighttime vehicle recognition device
JP6234063B2 (en) Detection of stationary objects on intersections of paths (methods, systems, and programs)
US10860869B2 (en) Time to collision using a camera
Almagambetov et al. Robust and computationally lightweight autonomous tracking of vehicle taillights and signal detection by embedded smart cameras
US20200026936A1 (en) Vehicle lamp detection methods and apparatuses, methods and apparatuses for implementing intelligent driving, media and devices
US6476806B1 (en) Method and apparatus for performing occlusion testing while exploiting frame to frame temporal coherence
CN109766867B (en) Vehicle running state determination method and device, computer equipment and storage medium
CN111178119A (en) Intersection state detection method and device, electronic equipment and vehicle
WO2023039781A1 (en) Method for detecting abandoned object, apparatus, electronic device, and storage medium
KR101239718B1 (en) System and method for detecting object of vehicle surroundings
CN110544271B (en) Parabolic motion detection method and related device
US20220108552A1 (en) Method and Apparatus for Determining Drivable Region Information
CN113838125A (en) Target position determining method and device, electronic equipment and storage medium
CN113569812A (en) Unknown obstacle identification method and device and electronic equipment
CN117612124A (en) Anti-collision detection method, device, equipment and medium for intelligent car dumper
Munajat et al. Vehicle detection and tracking based on corner and lines adjacent detection features
Lee An accident detection system on highway through CCTV with calogero-moser system
JP6723492B2 (en) Fog identification device, fog identification method, and fog identification program
JP6533244B2 (en) Object detection device, object detection method, and object detection program
JP2013149177A (en) Optical flow processor
WO2017077261A1 (en) A monocular camera cognitive imaging system for a vehicle
CN115761668A (en) Camera stain recognition method and device, vehicle and storage medium
JP2001091246A (en) Obstacle detecting device
CN113674311A (en) Abnormal behavior detection method and device, electronic equipment and storage medium
KR100853444B1 (en) Method of sensing car accidents

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