CN112288716A - Steel coil bundling state detection method, system, terminal and medium - Google Patents
Steel coil bundling state detection method, system, terminal and medium Download PDFInfo
- Publication number
- CN112288716A CN112288716A CN202011174991.7A CN202011174991A CN112288716A CN 112288716 A CN112288716 A CN 112288716A CN 202011174991 A CN202011174991 A CN 202011174991A CN 112288716 A CN112288716 A CN 112288716A
- Authority
- CN
- China
- Prior art keywords
- steel coil
- target
- binding band
- detection result
- image
- 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.)
- Granted
Links
- 229910000831 Steel Inorganic materials 0.000 title claims abstract description 421
- 239000010959 steel Substances 0.000 title claims abstract description 421
- 238000001514 detection method Methods 0.000 title claims abstract description 219
- 238000012549 training Methods 0.000 claims abstract description 55
- 238000013528 artificial neural network Methods 0.000 claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 29
- 238000002372 labelling Methods 0.000 claims description 20
- 230000002159 abnormal effect Effects 0.000 claims description 19
- 238000013135 deep learning Methods 0.000 claims description 17
- 238000003860 storage Methods 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004891 communication Methods 0.000 claims description 6
- 208000037805 labour Diseases 0.000 abstract description 8
- 238000010586 diagram Methods 0.000 description 9
- 230000000694 effects Effects 0.000 description 9
- 238000004519 manufacturing process Methods 0.000 description 7
- 238000012544 monitoring process Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 239000007787 solid Substances 0.000 description 4
- 238000012360 testing method Methods 0.000 description 4
- 238000013527 convolutional neural network Methods 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000005098 hot rolling Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30136—Metal
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a steel coil bundling state detection method, a steel coil bundling state detection system, a terminal and a medium, wherein the method comprises the steps of acquiring a sample image, respectively establishing a steel coil target detection model and a steel coil binding band target detection model based on a deep neural network according to the sample image, training, acquiring current image information to be detected, inputting the current image information to be detected into the steel coil target detection model to generate a first detection result, inputting the current image information to be detected into the steel coil binding band target detection model to generate a second detection result, and determining the steel coil bundling state according to the first detection result and the second detection result; the technical problems that the bundling state of the hot-rolled wire steel coil still needs to be detected on site by workers, the detection is not timely, accurate and comprehensive enough, the labor risk of the workers is large, and the cost is high are solved, the detection is more timely and accurate, the steel coil bundling state detection is realized through a machine, the labor risk of the workers is reduced, and the cost is reduced.
Description
Technical Field
The invention relates to the technical field of image processing in the field of steel, in particular to a method, a system, a terminal and a medium for detecting the bundling state of a steel coil.
Background
In the production of steel products, hot rolled wire steel coils are an important product for convenient storage and transportation. And binding the hot rolled wire steel coil by using a binding belt before transporting the hot rolled wire steel coil. The hot rolled wire steel coil is seriously loosened due to abnormal bundling conditions such as breakage of a binding band or missing binding of the hot rolled wire steel coil.
The bundling state of the existing steel coil is usually recognized manually, the abnormal steel coil in the whole bundling state cannot be detected timely, accurately and comprehensively, the labor risk of workers is large, the danger is high, and the detection cost of the bundling state of the hot-rolled wire steel coil is high.
Disclosure of Invention
In view of the above disadvantages of the prior art, an object of the present invention is to provide a method, a system, a terminal and a medium for detecting a bundling state of a steel coil, which are used to solve the technical problems that the bundling state of a hot-rolled wire steel coil still needs to be detected by a worker on site, and is not timely, accurate and comprehensive enough to be detected, and the worker has a high labor risk, high risk and high cost.
In view of the above problems, the present invention provides a method for detecting a bundling state of a steel coil, including:
acquiring a sample image, wherein the sample image comprises at least one of a steel coil and a steel coil binding band;
respectively establishing a steel coil target detection model and a steel coil binding band target detection model based on the deep neural network according to the sample image, and training;
acquiring current image information to be detected, inputting the current image information to the steel coil target detection model, and generating a first detection result;
inputting the current image information to be detected to the steel coil binding band target detection model to generate a second detection result;
and determining the bundling state of the steel coil according to the first detection result and the second detection result.
Optionally, before the current image information to be detected is input to the steel coil strap target detection model and a second detection result is generated, the method further includes:
acquiring range information of an interested area;
and if the target steel coil is determined to be positioned in the region of interest according to the first detection result and the range information of the region of interest, inputting the current image information to be detected to the steel coil binding band target detection model.
Optionally, the field area that the area of interest corresponds is provided with supplementary discernment background board, if the target coil of strip is located in the area of interest, the current image of waiting to detect includes the target coil of strip with supplementary discernment background board, the target coil of strip is located before the supplementary discernment background board, just the target coil of strip is located supplementary discernment background board within range.
Optionally, determining an interested rectangular selection frame corresponding to the interested region in the current image to be detected; determining a current steel coil rectangular target frame corresponding to the steel coil in a current image to be detected;
when the target steel coil is positioned in the region of interest, the following conditions are met:
ROIxmin<Holexmin；Holexmax<ROIxmax
ROIymin<Holeymin；Holeymax<ROIymax
wherein, ROIxminSelect the upper left coordinate, ROI, of the box for the rectangle of interestyminSelect the upper left ordinate, ROI, of the box for the rectangle of interestxmaxFor selecting a frame for a rectangle of interestLower right corner abscissa, ROIymaxFor the lower right-hand ordinate of the rectangle of interest, HolexminIs the coordinate of the upper left corner of the current rectangular target frame of the steel coil, HoleyminIs the vertical coordinate of the upper left corner of the current rectangular target frame of the steel coil, HolexmaxIs the horizontal coordinate of the lower right corner of the current rectangular target frame of the steel coil, HoleymaxIs the vertical coordinate of the lower right corner of the current rectangular target frame of the steel coil.
Optionally, the method further comprises the step of,
establishing a steel coil target detection model based on a deep neural network according to the sample image, and training the steel coil target detection model, wherein the steel coil target detection model comprises the following steps: marking the steel coil in the obtained sample image to obtain a steel coil data set, obtaining steel coil effective information in the steel coil data set, constructing a steel coil deep learning neural network, training the steel coil deep learning neural network by combining the steel coil effective information, and generating a steel coil target detection model;
establishing a steel coil binding belt target detection model based on a deep neural network according to the sample image, and training the steel coil binding belt target detection model, wherein the training comprises the following steps: and marking the obtained binding bands in the sample image to obtain a binding band data set, obtaining binding band effective information in the binding band data set, constructing a binding band deep learning neural network, training the binding band deep learning neural network by combining the binding band effective information, and generating a steel coil binding band target detection model.
The optional labeling of the steel coil in the sample image includes labeling the position of a hole in the side face of the steel coil in the sample image through a steel coil rectangular selection frame, the effective information of the steel coil includes basic attributes of the steel coil image and steel coil labeling information, the basic attributes of the steel coil image include at least one of a file name of the steel coil image, a width of the steel coil, a height of the steel coil and a depth of the steel coil image, and the steel coil labeling information includes an upper left-corner abscissa of the steel coil rectangular selection frame, an upper left-corner ordinate of the steel coil rectangular selection frame, a lower right-corner abscissa of the steel coil rectangular selection frame and a lower right-corner ordinate of the steel coil rectangular selection frame;
labeling the position of a bandage in the sample image, wherein the bandage in the sample image is obtained through a bandage rectangular selection frame, the effective bandage information comprises bandage image basic attributes and bandage labeling information, the bandage image basic attributes comprise at least one of a bandage image file name, a bandage width, a bandage height and a bandage image depth, and the bandage labeling information comprises an upper left-corner abscissa of the bandage rectangular selection frame, an upper left-corner ordinate of the bandage rectangular selection frame, a lower right-corner abscissa of the bandage rectangular selection frame and a lower right-corner ordinate of the bandage rectangular selection frame.
Optionally, the determining, according to the first detection result and the second detection result, a bundling state of the steel coil includes any one of:
if the position information of the current steel coil binding band in the second detection result is empty, the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, acquiring the number of the steel coil binding bands in the second detection result, and if the number is less than a preset number threshold value, determining that the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, acquiring the number of the steel coil binding bands in the second detection result, if the number is not less than a preset number threshold value, acquiring the position information of each steel coil binding band, matching the position information of the steel coil binding band with preset position information, and if the matching fails, determining that the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, the number of the steel coil binding bands in the second detection result is obtained, if the number is not less than a preset number threshold value, the position information of each steel coil binding band is obtained, the position information of the steel coil binding bands is matched with the preset position information, and if the matching is successful, the steel coil bundling state is abnormal.
The invention also provides a steel coil bundling state detection system, which comprises:
the image acquisition module is used for acquiring a sample image, and the sample image comprises at least one of a steel coil and a steel coil binding band;
the model building training module is used for building a steel coil target detection model and a steel coil binding band target detection model based on the deep neural network according to the sample image and training;
the first generation module is used for acquiring the current image information to be detected, inputting the current image information to be detected into the steel coil target detection model and generating a first detection result;
the second generation module is used for inputting the current image information to be detected to the steel coil binding band target detection model to generate a second detection result;
and the determining module is used for determining the bundling state of the steel coil according to the first detection result and the second detection result.
The invention also provides a terminal, which comprises a processor, a memory and a communication bus;
the communication bus is used for connecting the processor and the memory;
the processor is configured to execute the computer program stored in the memory to implement the steel coil bundling state detection method according to one or more of the above embodiments.
The present invention also provides a computer-readable storage medium, having stored thereon a computer program,
the computer program is for causing the computer to execute the steel coil bundling state detection method according to any one of the above-described embodiments.
As described above, the method, system, terminal and medium for detecting the bundling state of steel coils provided by the present invention have the following beneficial effects:
the method comprises the steps of establishing a steel coil target detection model and a steel coil binding belt target detection model based on a deep neural network according to a sample image by obtaining the sample image, training, obtaining current image information to be detected, inputting the current image information to be detected to the steel coil target detection model to generate a first detection result, inputting the current image information to be detected to the steel coil binding belt target detection model to generate a second detection result, and determining the steel coil bundling state according to the first detection result and the second detection result; the technical problems that the bundling state of the hot-rolled wire steel coil still needs to be detected on site by workers, timely, accurate and comprehensive detection is not enough, the labor risk of the workers is large, the danger is high, and the cost is high are solved, the detection is more timely and accurate, the steel coil bundling state detection is realized through a machine, the labor risk and the danger of the workers are reduced, and the cost is reduced.
Drawings
Fig. 1 is a schematic flow chart of a steel coil bundling state detection method according to an embodiment of the present invention;
fig. 2 is a schematic diagram of relative positions of a lens, a hole on a side surface of a steel coil, and an auxiliary identification background plate of the image acquisition device according to the first embodiment of the present invention;
fig. 3 is a specific flowchart of a method for detecting a bundling state of a hot-rolled steel coil according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a current image to be detected according to an embodiment of the present invention;
fig. 5 is a schematic structural diagram of a steel coil bundling state detection system according to a second embodiment of the present invention;
fig. 6 is a schematic structural diagram of a terminal according to a second embodiment of the present invention.
Detailed Description
The embodiments of the present invention are described below with reference to specific embodiments, and other advantages and effects of the present invention will be easily understood by those skilled in the art from the disclosure of the present specification. The invention is capable of other and different embodiments and of being practiced or of being carried out in various ways, and its several details are capable of modification in various respects, all without departing from the spirit and scope of the present invention. It is to be noted that the features in the following embodiments and examples may be combined with each other without conflict.
It should be noted that the drawings provided in the following embodiments are only for illustrating the basic idea of the present invention, and the components related to the present invention are only shown in the drawings rather than drawn according to the number, shape and size of the components in actual implementation, and the type, quantity and proportion of the components in actual implementation may be changed freely, and the layout of the components may be more complicated.
Example one
Referring to fig. 1, a method for detecting a bundling state of a steel coil according to an embodiment of the present invention includes:
s101: a sample image is acquired.
The sample image may be acquired by using a camera or an image acquisition device, or may be retrieved from an image or a video stored in another database, which is not limited herein.
Optionally, the sample image includes at least one of a steel coil and a steel coil strap, that is, the sample image includes a plurality of sub-images, and the sub-images may include only the steel coil or both the steel coil and the steel coil strap.
Optionally, the sample image includes an image of a strap normally tied to the steel coil.
Alternatively, coils include, but are not limited to, hot rolled wire coils, and the like.
S102: and respectively establishing a steel coil target detection model and a steel coil binding band target detection model based on the deep neural network according to the sample image, and training.
In some embodiments, establishing a steel coil target detection model based on a deep neural network according to the sample image, and the training includes: marking the steel coil in the obtained sample image to obtain a steel coil data set, obtaining effective information of the steel coil in the steel coil data set, constructing a steel coil deep learning neural network, training the steel coil deep learning neural network by combining the effective information of the steel coil, and generating a steel coil target detection model.
Optionally, marking a steel coil in the obtained sample image to obtain a steel coil data set, and dividing the steel coil data set into a training set, a testing set, and a verification set according to a certain proportion, for example, according to 8:1:1, and the like. And training the steel coil deep learning neural network by using the data in the training set to generate a steel coil target detection model.
Optionally, the effective information of the steel coil in the steel coil data set includes effective information of the steel coil in a training set in the steel coil data set.
Optionally, the obtained sample graph is markedThe coil of strip in the image includes the position of coil of strip side hole in the sample image through coil of strip rectangle select frame mark, and coil of strip effective information includes coil of strip image basis attribute and coil of strip mark information, and coil of strip image basis attribute includes at least one in coil of strip image file name, coil of strip width, coil of strip height and coil of strip image degree of depth, and coil of strip mark information includes the upper left corner abscissa yHole of coil of strip rectangle select framexminThe vertical coordinate yHole of the upper left corner of the rectangular selection frame of the steel coilyminAnd the lower right corner abscissa yHole of the rectangular steel coil selection framexmaxAnd the ordinate yHole of the lower right corner of the rectangular selection frame of the steel coilymax。
Optionally, the marking of the steel coil in the acquired sample image to obtain the steel coil data set includes:
and taking the rectangular selection frame of the image marking tool as a steel coil rectangular selection frame, marking the position of the steel coil side hole in the sample image by using the steel coil rectangular selection frame, recording the position of the steel coil rectangular selection frame, and forming a steel coil data set.
Optionally, the effective information of the steel coil further includes a class of the target object.
Optionally, training the steel coil deep learning neural network by combining the effective information of the steel coil, and generating a steel coil target detection model comprises:
and obtaining a steel coil target detection model by learning the steel coil target characteristics in the range of the steel coil identification frame in each steel coil training set image.
Optionally, the steel coil deep learning neural network includes, but is not limited to, any one of an SSD-MobileNet neural network, an R-CNN neural network, a Faster-RCNN neural network, a YOLO series neural network, and the like.
In some embodiments, establishing a steel coil bandage target detection model based on a deep neural network according to the sample image, and performing training includes: and marking the binding bands in the obtained sample images to obtain a binding band data set, obtaining effective binding band information in the binding band data set, constructing a deep learning neural network of the binding bands, training the deep learning neural network of the binding bands by combining the effective binding band information, and generating a steel coil binding band target detection model.
Optionally, the bands in the obtained sample image are labeled to obtain a band data set, and the band data set is divided into a training set, a testing set, and a verification set according to a certain proportion, for example, according to 8:1:1, etc. And training the binding belt deep learning neural network by using the data in the training set to generate a steel coil binding belt target detection model.
Optionally, the effective information of the steel coil in the binding band data set includes effective information of the binding band in the training set of the binding band data set.
Optionally, the bandage in the sample image obtained by labeling includes labeling the position of the bandage in the sample image through a rectangular frame of the bandage, the effective information of the bandage includes basic attributes of the bandage image and the annotation information of the bandage, the basic attributes of the bandage image include at least one of the file name of the bandage image, the width of the bandage, the height of the bandage and the depth of the bandage image, and the annotation information of the bandage includes the upper left-hand abscissa yBand of the rectangular frame of the bandagexminAnd the ordinate yBand of the upper left corner of the rectangular selection frame of the binding bandyminAnd the lower right corner abscissa yBand of the rectangular selection frame of the binding bandxmaxAnd the lower right corner ordinate yBand of rectangular selection frame of binding bandymax。
Optionally, labeling the bands in the acquired sample image to obtain a band data set includes:
and taking the rectangular frame of the image marking tool as a rectangular frame of the binding band, marking the position of the binding band in the sample image by using the rectangular frame of the binding band, recording the position of the rectangular frame of the binding band, and forming a binding band data set.
Optionally, the band valid information further includes a class of the target object.
Optionally, combine bandage effective information training bandage degree of depth study neural network, generate coil of strip bandage target detection model and include:
and obtaining a steel coil bandage target detection model by learning the bandage target characteristics in the bandage identification frame range in each bandage training set image.
Optionally, the bandage deep learning neural network includes, but is not limited to, any one of an SSD-MobileNet neural network, an R-CNN neural network, a Faster-RCNN neural network, a YOLO series neural network, and the like.
Optionally, if the sample image includes a steel coil and a binding band, the standard of the steel coil and the binding band may be performed twice respectively during labeling, or the steel coil and the binding band may be labeled in the same labeling process, which is not limited herein.
S103: and acquiring current image information to be detected, inputting the current image information to the steel coil target detection model, and generating a first detection result.
It should be noted that the current image information to be detected may be obtained from a real-time monitoring image. At the moment, the steel coil bundling state is monitored according to the real-time monitoring image information, and the bundling state of the target steel coil passing through the monitoring equipment can be acquired in real time. If the current image information to be detected is provided by a real-time image, the method for monitoring the bundling state of the steel coil provided by the embodiment can realize real-time monitoring.
Optionally, the first detection result includes, but is not limited to, whether the image to be detected includes the target steel coil or not, and if the image includes the target steel coil, the first detection result further includes position information of a current steel coil rectangular target frame corresponding to the target steel coil.
In some embodiments, if the first detection result is empty, that is, the steel coil target detection model does not detect the target steel coil in the current image information to be detected, the detection of the bundling state of the target steel coil is finished. Optionally, if the current image to be detected is a real-time image, the target steel coil may not move to the optimal recognition position at this time, and new current image information to be detected may be acquired again at intervals of preset time, so as to perform bundling state detection on the target steel coil again. Optionally, the preset time may be determined according to the motion state of the target steel coil.
In some embodiments, the first detection result includes position information of all holes on the side surface of the target steel coil in the current image to be detected.
Optionally, the format and content of the first detection result include:
[Holexmin,Holeymin,Holexmax,Holeymax],
the four coordinates are respectively corresponding to the side holes of the target steel coilAnd coordinates of upper left, lower right and upper right points of the rectangular target frame of the front steel coil. Wherein, Holexmin,HoleyminRespectively the horizontal and vertical coordinates of the upper left corner point; holexmax,HoleymaxRespectively the horizontal and vertical coordinates of the lower right corner point.
It should be noted that, the first detection result may include one or more target steel coil side holes, and if the first detection result includes a plurality of target steel coil side holes, the format and content of the first detection result are as follows:
if the target steel coil is not detected in the current image information to be detected by the steel coil target detection model, the first detection result is empty, and the identification can be marked as [ 0,0,0,0 ], and the identification method can be other modes, and is not limited herein.
In some embodiments, after generating the first detection result, inputting the current image information to be detected to the steel coil binding belt target detection model, and before generating the second detection result, the method further includes:
acquiring range information of an interested area;
and if the target steel coil is determined to be positioned in the region of interest according to the first detection result and the range information of the region of interest, inputting the current image information to be detected into the steel coil binding band target detection model.
Optionally, determining an interested rectangular selection frame corresponding to the interested region in the current image to be detected; determining a current steel coil rectangular target frame corresponding to a steel coil in a current image to be detected, wherein the steel coil meets the following conditions when being positioned in an interested area:
ROIxmin<Holexmin；Holexmax<ROIxmax
ROIymin<Holeymin；Holeymax<ROIymax
wherein, ROIxminSelect the upper left coordinate, ROI, of the box for the rectangle of interestyminSelect the upper left ordinate, ROI, of the box for the rectangle of interestxmaxSelect the lower right-hand abscissa, ROI, of the box for the rectangle of interestymaxFor the lower right-hand ordinate of the rectangle of interest, HolexminIs the coordinate of the upper left corner of the current rectangular target frame of the steel coil, HoleyminIs the vertical coordinate of the upper left corner of the current rectangular target frame of the steel coil, HolexmaxIs the horizontal coordinate of the lower right corner of the current rectangular target frame of the steel coil, HoleymaxIs the vertical coordinate of the lower right corner of the current rectangular target frame of the steel coil.
If the position of the current steel coil rectangular target frame does not meet the conditions, the target steel coil is located outside the region of interest and does not reach the optimal identification position, and the target steel coil needs to wait to move to a proper position.
Optionally, an auxiliary identification background plate is arranged in a field area corresponding to the region of interest, if the target steel coil is located in the region of interest, the current image to be detected includes the target steel coil and the auxiliary identification background plate, the target steel coil is located in front of the auxiliary identification background plate, and the target steel coil is located within the range of the auxiliary identification background plate.
Optionally, the auxiliary identification background plate comprises a solid background plate. The detection effect can be enhanced by arranging a solid background plate in the field area, such as a brighter color like blue, red or yellow. Such colors generally have a high contrast with the factory production environment, which can enhance the recognition effect; in addition, the auxiliary recognition background plate has a certain shielding effect, the influence of a complex industrial scene in the background on the training process of the steel coil target detection model and the steel coil binding band target detection model can be avoided, and the consistency of the target is enhanced.
Optionally, referring to fig. 2, in the process of acquiring the sample image and the current image to be detected, the viewing angle of the image acquisition device is perpendicular to the auxiliary identification background plate, and the lens 1 of the image acquisition device, the circle center of the hole 2 on the side surface of the steel coil, and the center of the auxiliary identification background plate 3 are substantially located on the same straight line. Of course, the lens of the image acquisition device, the circle center of the hole in the side face of the steel coil and the center of the auxiliary identification background plate can be not located on the same straight line, and the three are inclined at a certain angle.
Optionally, if the detection effect is better when the circular side of the hot-rolled wire steel coil faces the camera, the area of the imaging range of the image acquired by the image acquisition device of the auxiliary recognition background plate is set as an area of Interest (ROI). In an actual industrial production scene, the circular side hole of the hot-rolled wire steel coil faces the lens of the image acquisition equipment, and the steel coil target detection model also realizes target detection by identifying the side hole of the steel coil.
Optionally, the shape of the side hole of the hot-rolled wire steel coil is close to a perfect circle, the target is obvious under the support of the pure-color background plate, the side hole of the hot-rolled wire steel coil is used as the identification object of the steel coil target identification model, and the steel coil can be judged to be identified when the hole is identified.
Optionally, the hole in the side of the steel coil may be the hole edge close to one side of the camera of the image acquisition device, or the hole edge far away from one side of the camera of the image acquisition device, which is not limited herein.
S104: and inputting the current image information to be detected to a steel coil binding band target detection model to generate a second detection result.
Optionally, the format and content of the second detection result include:
the four coordinates respectively correspond to the coordinates of the upper left point, the lower right point and the upper right point of the current rectangular target frame with the binding band. Wherein, Bandxmin,BandyminRespectively the horizontal and vertical coordinates of the upper left corner point; band (R)xmax,BandymaxRespectively the horizontal and vertical coordinates of the lower right corner point.
It should be noted that the format of the second detection result is only an exemplary format, and the format of the second detection result may be flexibly set according to the number of the straps.
It should be noted that a coil of steel may be bound by a plurality of straps, and the second detection result may include position information of the plurality of straps.
Optionally, the second detection result includes position information of at least one binding band, and if the steel coil target detection model does not detect a binding band in the current image information to be detected, the second detection result is empty and may be identified as [ 0,0,0,0 ], and the identification method may be in another manner, which is not limited herein.
S105: and determining the bundling state of the steel coil according to the first detection result and the second detection result.
Optionally, determining the bundling state of the steel coil according to the first detection result and the second detection result includes any one of the following:
if the position information of the current steel coil binding band in the second detection result is empty, the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, acquiring the number of the steel coil binding bands in the second detection result, and if the number is less than a preset number threshold value, determining that the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, acquiring the number of the steel coil binding bands in the second detection result, if the number is not less than a preset number threshold value, acquiring the position information of each steel coil binding band, matching the position information of each steel coil binding band with the preset position information, and if the matching fails, judging that the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, the number of the steel coil binding bands in the second detection result is obtained, if the number is not less than a preset number threshold value, the position information of each steel coil binding band is obtained, the position information of the steel coil binding band is matched with the preset position information, and if the matching is successful, the steel coil bundling state is normal.
Optionally, the method for detecting the bundling state of the steel coil further includes:
and if the steel coil bundling state is normal, outputting the position information of each binding band and the normal bundling state information of the hot-rolled wire steel coil.
Optionally, the method for detecting the bundling state of the steel coil further includes:
and if the steel coil bundling state is abnormal, sending an abnormal prompt.
Wherein, the abnormal prompt can be sent out by means of information, voice, warning lights and the like.
Optionally, the preset number threshold and the preset position information may be set by those skilled in the art as needed, and specific numerical values are not limited herein. The preset position information can be set into a plurality of position ranges as required, if the position information of a certain binding band meets one of the position ranges, the position information of the binding band is successfully matched, and if the position information of each binding band is successfully matched, or the position information of the binding bands exceeding a certain number is successfully matched, the steel coil bundling state is normal.
Optionally, if the current hot-rolled wire steel coil is located in the region of interest, the trained steel coil binding band target detection model is called: and if the hot rolled wire steel coil binding band is detected to exist in the current image to be detected, outputting the position information of each binding band target and the normal binding state information of the hot rolled wire steel coil, and if the hot rolled wire steel coil binding band is not detected in the picture, outputting the abnormal binding state information of the hot rolled wire steel coil.
Optionally, if the hot-rolled wire steel coil does not enter or leaves the region of interest, waiting until a hot-rolled wire steel coil enters the region of interest, and then inputting the current image information to be detected to the steel coil binding belt target detection model.
The invention provides a steel coil bundling state detection method, which comprises the steps of establishing a steel coil target detection model and a steel coil binding belt target detection model based on a deep neural network respectively according to a sample image by obtaining the sample image, training, obtaining current image information to be detected, inputting the current image information to be detected into the steel coil target detection model to generate a first detection result, inputting the current image information to be detected into the steel coil binding belt target detection model to generate a second detection result, and determining the steel coil bundling state according to the first detection result and the second detection result; the technical problems that the bundling state of the hot-rolled wire steel coil still needs to be detected on site by workers, timely, accurate and comprehensive detection is not enough, the labor risk of the workers is large, the danger is high, and the cost is high are solved, the detection is more timely and accurate, the steel coil bundling state detection is realized through a machine, the labor risk and the danger of the workers are reduced, and the cost is reduced.
Optionally, when there are many production lines and need carry out coil of strip bundling state detection, can realize the detection that coil of strip bundled state detected at each production line through setting up the device that distributes and can realize this embodiment coil of strip bundling state detection, the suitability is strong, detects more comprehensively, in time.
A steel coil bundling state detection method provided by the present embodiment is exemplarily described below by a specific embodiment, and referring to fig. 3, a hot-rolled wire steel coil bundling state detection method includes:
s301: and arranging an auxiliary identification background plate in a specified industrial scene, and acquiring images of a plurality of hot-rolled wire steel coils bundled with binding bands as sample images by using an industrial camera.
Optionally, the auxiliary identification background plate includes a solid background plate, and the detection effect is enhanced by setting the solid background plate on site, such as a brighter color like blue, red, or yellow. Such colors generally have higher contrast with the factory production environment, enhancing the recognition effect of the algorithm; the auxiliary recognition background plate has a certain shielding effect, the influence of a complex industrial scene in the background on the training process of the steel coil target detection model and the steel coil binding band target detection model can be avoided, and the consistency of the target is enhanced.
S302: and respectively labeling the steel coil and the binding band in the sample image, and manufacturing a hot-rolled wire steel coil data set and a hot-rolled wire steel coil binding band data set.
Optionally, image annotation is performed on a sample image obtained by shooting in a specific industrial scene, and a rectangular selection frame of an image annotation tool is used as a steel coil rectangular selection frame, so that the position of the round side hole of the hot-rolled wire steel coil in the image is marked, the position information of the steel coil rectangular selection frame is recorded, and a hot-rolled wire steel coil data set is manufactured and is divided into three parts: training set, testing set, verifying set, training the hot rolling wire steel coil target detection model by using the data of the training set.
Effective information which can be used for training in the hot-rolled wire steel coil training set after image marking comprises steel coil image basic attributes and steel coil marking information. The basic properties of the steel coil picture are as follows: the filename, the width, the height and the depth of the steel coil are the file name, the width, the height and the image depth of the steel coil. The steel coil labeling information comprises: xmin, ymin, xmax and ymax respectively represent the horizontal coordinate of the upper left corner, the vertical coordinate of the upper left corner, the horizontal coordinate of the lower right corner and the vertical coordinate of the lower right corner of the rectangular steel coil selection frame corresponding to the round side hole of the hot rolled wire steel coil in the sample image; class, class of the target object.
The method comprises the following steps of carrying out image annotation on a sample image shot and obtained in a specific industrial scene, using a rectangular selection frame of an image annotation tool as a rectangular selection frame of a binding band, marking the positions of all binding bands in the sample image, recording the position information of the rectangular selection frame of the binding band, manufacturing a hot-rolled wire steel coil binding band data set, and dividing the hot-rolled wire steel coil binding band data set into three parts: training set, testing set, verifying set, training the target detection model of the hot-rolled wire steel coil binding band by using the data of the training set.
Effective information which can be used for training in the hot-rolled wire steel coil bandage training set after image labeling comprises basic attributes of bandage images and bandage labeling information. The basic properties of the bandage picture are as follows: filename-strap filename, width-strap width, height-strap height, depth-strap image depth. The bandage labeling information includes: xmin, ymin, xmax and ymax respectively represent the horizontal coordinate of the upper left corner, the vertical coordinate of the upper left corner, the horizontal coordinate of the lower right corner and the vertical coordinate of the lower right corner of each rectangular binding band selection frame of the hot rolled wire steel coil in the sample image; class, class of the target object.
S303: and building a neural network, and training by using the hot-rolled wire steel coil data set and the hot-rolled wire steel coil binding band respectively to obtain a hot-rolled wire steel coil target detection model and a hot-rolled wire steel coil binding band target detection model.
And finally obtaining a hot-rolled wire steel coil target detection model by learning the target characteristics in the rectangular frame selection range of the steel coil in each hot-rolled wire steel coil training set image. And finally obtaining a target detection model of the hot-rolled wire steel coil binding band by learning the target characteristics in the rectangular frame selection range of the binding band in each hot-rolled wire steel coil binding band training set image. The hot-rolled wire steel coil training set image is a sample image marked with a steel coil, and the hot-rolled wire steel coil bandage training set image is a sample image marked with a bandage.
Optionally, the neural network includes SSD-MobileNet neural network, R-CNN neural network, Faster-RCNN neural network, YOLO series neural network, and other target recognition type neural networks.
S304: and acquiring current image information to be detected, inputting the current image information to the hot-rolled wire steel coil target detection model, and generating a first detection result.
The first detection result comprises the position information of all hot rolled wire steel coils in the current image to be detected.
The format and content of the first detection result comprise:
[Holexmin,Holeymin,Holexmax,Holeymax],
the four coordinates in the list correspond to the upper left, lower right and upper right points of the current rectangular target frame of the steel coil respectively. Holexmin,HoleyminRespectively the horizontal and vertical coordinates of the upper left corner point; holexmax,HoleymaxRespectively the horizontal and vertical coordinates of the lower right corner point.
Referring to fig. 4, fig. 4 is a schematic diagram of a current image to be detected, which includes a current steel coil rectangular target frame a and four current bandage rectangular target frames B.
S305: and comparing the position information of the current steel coil rectangular target frame in the first detection result with the region of interest, and judging whether the hot-rolled wire steel coil is positioned in the region of interest.
Optionally, whether the current hot-rolled wire steel coil is located in the range suitable for detection may be determined by comparing the position information of the hole on the circular side of the hot-rolled wire steel coil (the position information of the current steel coil rectangular target frame) with the range of the region of interest. Optionally, the range of the region of interest in the current image to be detected is determined by coordinates of upper left corner and lower right corner points of an interested rectangular frame of the auxiliary recognition background board in the current image to be detected, and the content and format of the position information include:
[ROIymin,ROIxmin,ROIymax,ROIxmax]
the conditions for judging whether the hot rolled wire steel coil is positioned in the region of interest at this time are as follows:
ROIxmin<Holexmin；Holexmax<ROIxmax
ROIymin<Holeymin；Holeymax<ROIymax
wherein, ROIxminSelect the upper left coordinate, ROI, of the box for the rectangle of interestyminSelect the upper left ordinate, ROI, of the box for the rectangle of interestxmaxSelect the lower right-hand abscissa, ROI, of the box for the rectangle of interestymaxFor the lower right-hand ordinate of the rectangle of interest, HolexminIs the coordinate of the upper left corner of the current rectangular target frame of the steel coil, HoleyminIs the vertical coordinate of the upper left corner of the current rectangular target frame of the steel coil, HolexmaxIs the horizontal coordinate of the lower right corner of the current rectangular target frame of the steel coil, HoleymaxIs the vertical coordinate of the lower right corner of the current rectangular target frame of the steel coil.
If the conditions are met, the hot-rolled wire steel coil is located in the region of interest, and a hot-rolled wire steel coil binding band target detection model can be called to perform the next step of identification; if any condition is not met, the hot-rolled wire steel coil is positioned outside the region of interest, does not reach the optimal identification position, and needs to wait to move to a proper position.
S306: and inputting the current image information to be detected to a hot-rolled wire steel coil binding band target detection model to generate a second detection result.
Optionally, when the hot-rolled wire steel coil is located in the region of interest, the hot-rolled wire steel coil binding band target detection model is called, the current image information to be detected is input into the hot-rolled wire steel coil binding band target detection model, and a second detection result is generated.
Optionally, if the second detection result includes the position information of at least one hot-rolled wire steel coil binding band, the detected position information of the hot-rolled wire steel coil binding band and the normal bundling state information of the hot-rolled wire steel coil are output together.
Optionally, if the second detection result is empty, that is, if the hot-rolled wire steel coil binding band is not detected, the information of the abnormal binding state of the hot-rolled wire steel coil is output.
Optionally, the second detection result includes position information of all hot-rolled wire steel coil straps in the current image to be detected. The format and content of the second detection result output position information include:
the four coordinates in the list correspond to the upper left, lower right, and upper right points of the current rectangular target frame with a bandage, respectively. Band (R)xmin,BandyminRespectively the horizontal and vertical coordinates of the upper left corner point; band (R)xmax,BandymaxRespectively the horizontal and vertical coordinates of the lower right corner point.
Optionally, when the hot-rolled wire steel coil is located in the region of interest, the trained hot-rolled wire steel coil binding band target detection model is called: and if the hot rolled wire steel coil binding band is detected to exist in the current image to be detected, outputting the position information of each binding band and the normal bundling state information of the hot rolled wire steel coil, and if the hot rolled wire steel coil binding band is not detected in the picture of the current image to be detected, outputting the abnormal bundling state information of the hot rolled wire steel coil. If the hot rolled coil of wire has not entered or has left the area of interest, wait until there is a hot rolled coil of wire entering the area of interest.
Example two
Referring to fig. 5, a system 500 for detecting a bundling status of a steel coil includes:
an image obtaining module 501, configured to obtain a sample image, where the sample image includes at least one of a steel coil and a steel coil binding band;
a model building training module 502, configured to build a steel coil target detection model and a steel coil binding band target detection model based on a deep neural network according to the sample image, and perform training;
the first generating module 503 is configured to acquire current image information to be detected, input the current image information to the steel coil target detection model, and generate a first detection result;
the second generating module 504 is configured to input the current image information to be detected to the steel coil strap target detection model, and generate a second detection result;
and a determining module 505, configured to determine a steel coil bundling state according to the first detection result and the second detection result.
In this embodiment, the long nozzle replacing device based on the visual servo is substantially provided with a plurality of modules for executing the steel coil bundling state detection method in the above embodiment, and specific functions and technical effects are only required by referring to the first embodiment, which is not described herein again.
Referring to fig. 6, an embodiment of the present invention further provides a terminal 600, which includes a processor 601, a memory 602, and a communication bus 603;
a communication bus 603 is used to connect the processor 601 and the memory 602;
the processor 601 is used for executing the computer program stored in the memory 602 to implement the steel coil bundling state detection and replacement method as described in one or more of the above embodiments.
An embodiment of the present invention also provides a computer-readable storage medium, characterized in that, a computer program is stored thereon,
the computer program is for causing a computer to execute the steel coil bundling state detection method according to any one of the above-described first embodiment.
Embodiments of the present application also provide a non-transitory readable storage medium, where one or more modules (programs) are stored in the storage medium, and when the one or more modules are applied to a device, the device may execute instructions (instructions) included in an embodiment of the present application.
It should be noted that the computer readable medium in the present disclosure can be a computer readable signal medium or a computer readable storage medium or any combination of the two. A computer readable storage medium may be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the foregoing. More specific examples of the computer readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present disclosure, a computer readable storage medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution system, apparatus, or device. In contrast, in the present disclosure, a computer readable signal medium may comprise a propagated data signal with computer readable program code embodied therein, either in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may also be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: electrical wires, optical cables, RF (radio frequency), etc., or any suitable combination of the foregoing.
The computer readable medium may be embodied in the electronic device; or may exist separately without being assembled into the electronic device.
Computer program code for carrying out operations for aspects of the present disclosure may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + +, and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of methods and computer program products according to various embodiments of the present disclosure. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The foregoing embodiments are merely illustrative of the principles and utilities of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present invention. Accordingly, it is intended that all equivalent modifications or changes which can be made by those skilled in the art without departing from the spirit and technical spirit of the present invention be covered by the claims of the present invention.
Claims (10)
1. A steel coil bundling state detection method is characterized by comprising the following steps:
acquiring a sample image, wherein the sample image comprises at least one of a steel coil and a steel coil binding band;
respectively establishing a steel coil target detection model and a steel coil binding band target detection model based on the deep neural network according to the sample image, and training;
acquiring current image information to be detected, inputting the current image information to the steel coil target detection model, and generating a first detection result;
inputting the current image information to be detected to the steel coil binding band target detection model to generate a second detection result;
and determining the bundling state of the steel coil according to the first detection result and the second detection result.
2. The method for detecting the steel coil bundling state according to claim 1, wherein before inputting the current image information to be detected into the steel coil bundling band target detection model and generating a second detection result, the method further comprises:
acquiring range information of an interested area;
and if the target steel coil is determined to be positioned in the region of interest according to the first detection result and the range information of the region of interest, inputting the current image information to be detected to the steel coil binding band target detection model.
3. The method for detecting the bundling state of steel coil according to claim 2, wherein an auxiliary recognition background plate is disposed in a field area corresponding to the region of interest, if the target steel coil is located in the region of interest, the current image to be detected includes the target steel coil and the auxiliary recognition background plate, the target steel coil is located in front of the auxiliary recognition background plate, and the target steel coil is located within the range of the auxiliary recognition background plate.
4. The method for detecting the bundling status of steel coils according to claim 2, further comprising determining an interested rectangle selection frame corresponding to the interested region in the current image to be detected; determining a current steel coil rectangular target frame corresponding to the target steel coil in a current image to be detected;
when the target steel coil is positioned in the region of interest, the following conditions are met:
ROIxmin<Holexmin；Holexmax<ROIxmax
ROIymin<Holeymin；Holeymax<ROIymax
wherein, ROIxminSelect the upper left coordinate, ROI, of the box for the rectangle of interestyminSelect the upper left ordinate, ROI, of the box for the rectangle of interestxmaxSelect the lower right-hand abscissa, ROI, of the box for the rectangle of interestymaxFor the lower right-hand ordinate of the rectangle of interest, HolexminIs the coordinate of the upper left corner of the current rectangular target frame of the steel coil, HoleyminIs the vertical coordinate of the upper left corner of the current rectangular target frame of the steel coil, HolexmaxIs the horizontal coordinate of the lower right corner of the current rectangular target frame of the steel coil, HoleymaxIs the vertical coordinate of the lower right corner of the current rectangular target frame of the steel coil.
5. The method for detecting the steel coil bundling state according to any one of claims 1 to 4, further comprising the steps of establishing a steel coil target detection model based on a deep neural network according to the sample image, and performing training including: marking the steel coil in the obtained sample image to obtain a steel coil data set, obtaining steel coil effective information in the steel coil data set, constructing a steel coil deep learning neural network, training the steel coil deep learning neural network by combining the steel coil effective information, and generating a steel coil target detection model;
establishing a steel coil binding belt target detection model based on a deep neural network according to the sample image, and training the steel coil binding belt target detection model, wherein the training comprises the following steps: and marking the obtained binding bands in the sample image to obtain a binding band data set, obtaining binding band effective information in the binding band data set, constructing a binding band deep learning neural network, training the binding band deep learning neural network by combining the binding band effective information, and generating a steel coil binding band target detection model.
6. The steel coil bundling state detection method according to claim 5,
marking the position of a hole in the side face of the steel coil in the sample image by a steel coil rectangular selection frame, wherein the steel coil effective information comprises steel coil image basic attributes and steel coil marking information, the steel coil image basic attributes comprise at least one of a steel coil image file name, a steel coil width, a steel coil height and a steel coil image depth, and the steel coil marking information comprises an upper left-corner horizontal coordinate of the steel coil rectangular selection frame, an upper left-corner vertical coordinate of the steel coil rectangular selection frame, a lower right-corner horizontal coordinate of the steel coil rectangular selection frame and a lower right-corner vertical coordinate of the steel coil rectangular selection frame;
labeling the position of a bandage in the sample image, wherein the bandage in the sample image is obtained through a bandage rectangular selection frame, the effective bandage information comprises bandage image basic attributes and bandage labeling information, the bandage image basic attributes comprise at least one of a bandage image file name, a bandage width, a bandage height and a bandage image depth, and the bandage labeling information comprises an upper left-corner abscissa of the bandage rectangular selection frame, an upper left-corner ordinate of the bandage rectangular selection frame, a lower right-corner abscissa of the bandage rectangular selection frame and a lower right-corner ordinate of the bandage rectangular selection frame.
7. The method according to any one of claims 1 to 4, wherein the determining the steel coil bundling state according to the first and second detection results includes any one of:
if the position information of the current steel coil binding band in the second detection result is empty, the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, acquiring the number of the steel coil binding bands in the second detection result, and if the number is less than a preset number threshold value, determining that the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, acquiring the number of the steel coil binding bands in the second detection result, if the number is not less than a preset number threshold value, acquiring the position information of each steel coil binding band, matching the position information of the steel coil binding band with preset position information, and if the matching fails, determining that the steel coil binding state is abnormal;
if the position information of the current steel coil binding band in the second detection result is not empty, the number of the steel coil binding bands in the second detection result is obtained, if the number is not less than a preset number threshold value, the position information of each steel coil binding band is obtained, the position information of the steel coil binding bands is matched with the preset position information, and if the matching is successful, the steel coil bundling state is abnormal.
8. The utility model provides a steel coil bundling state detecting system which characterized in that includes:
the image acquisition module is used for acquiring a sample image, and the sample image comprises at least one of a steel coil and a steel coil binding band;
the model building training module is used for building a steel coil target detection model and a steel coil binding band target detection model based on the deep neural network according to the sample image and training;
the first generation module is used for acquiring the current image information to be detected, inputting the current image information to be detected into the steel coil target detection model and generating a first detection result;
the second generation module is used for inputting the current image information to be detected to the steel coil binding band target detection model to generate a second detection result;
and the determining module is used for determining the bundling state of the steel coil according to the first detection result and the second detection result.
9. A terminal comprising a processor, a memory, and a communication bus;
the communication bus is used for connecting the processor and the memory;
the processor is configured to execute the computer program stored in the memory to implement the steel coil bundling state detection method according to one or more of claims 1-7.
10. A computer-readable storage medium, having stored thereon a computer program,
the computer program is for causing the computer to execute the steel coil bundling state detection method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011174991.7A CN112288716B (en) | 2020-10-28 | 2020-10-28 | Method, system, terminal and medium for detecting bundling state of steel coil |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011174991.7A CN112288716B (en) | 2020-10-28 | 2020-10-28 | Method, system, terminal and medium for detecting bundling state of steel coil |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112288716A true CN112288716A (en) | 2021-01-29 |
| CN112288716B CN112288716B (en) | 2023-10-27 |
Family
ID=74372406
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011174991.7A Active CN112288716B (en) | 2020-10-28 | 2020-10-28 | Method, system, terminal and medium for detecting bundling state of steel coil |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112288716B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113537199A (en) * | 2021-08-13 | 2021-10-22 | 上海淇玥信息技术有限公司 | Image bounding box screening method, system, electronic device and medium |
| CN114638783A (en) * | 2022-02-10 | 2022-06-17 | 中冶南方工程技术有限公司 | Steel coil height centering system based on image recognition |
| CN115049620A (en) * | 2022-06-17 | 2022-09-13 | 中轻长泰(长沙)智能科技股份有限公司 | Paper stack appearance detection method, device, equipment and storage medium |
| TWI787748B (en) * | 2021-03-03 | 2022-12-21 | 中國鋼鐵股份有限公司 | Method and system for predicting defects of holes in strip product |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017049163A1 (en) * | 2015-09-18 | 2017-03-23 | Auris Surgical Robotics, Inc. | Navigation of tubular networks |
| CN110197170A (en) * | 2019-06-05 | 2019-09-03 | 北京科技大学 | Coil of strip scroll defects detection recognition methods based on target detection |
| CN110220914A (en) * | 2019-07-16 | 2019-09-10 | 中冶赛迪技术研究中心有限公司 | High-temperature steel coil Bundling strap on-line detecting system and method |
| CN110363761A (en) * | 2019-07-22 | 2019-10-22 | 上海眼控科技股份有限公司 | A kind of start-stop Mark Detection system and method for vehicle chassis dynamic detection |
| CN110781834A (en) * | 2019-10-28 | 2020-02-11 | 上海眼控科技股份有限公司 | Traffic abnormality image detection method, device, computer device and storage medium |
| CN111160824A (en) * | 2019-12-05 | 2020-05-15 | 深圳市铭华航电工艺技术有限公司 | Engineering data detection method and device, computer equipment and storage medium |
| CN111428729A (en) * | 2019-01-09 | 2020-07-17 | 北京京东尚科信息技术有限公司 | Target detection method and device |
| CN111553950A (en) * | 2020-04-30 | 2020-08-18 | 中冶赛迪重庆信息技术有限公司 | Steel coil centering judgment method, system, medium and electronic terminal |
| CN111652834A (en) * | 2019-03-02 | 2020-09-11 | 周口师范学院 | Steel coil strapping tape identification method based on support vector machine |
-
2020
- 2020-10-28 CN CN202011174991.7A patent/CN112288716B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2017049163A1 (en) * | 2015-09-18 | 2017-03-23 | Auris Surgical Robotics, Inc. | Navigation of tubular networks |
| CN111428729A (en) * | 2019-01-09 | 2020-07-17 | 北京京东尚科信息技术有限公司 | Target detection method and device |
| CN111652834A (en) * | 2019-03-02 | 2020-09-11 | 周口师范学院 | Steel coil strapping tape identification method based on support vector machine |
| CN110197170A (en) * | 2019-06-05 | 2019-09-03 | 北京科技大学 | Coil of strip scroll defects detection recognition methods based on target detection |
| CN110220914A (en) * | 2019-07-16 | 2019-09-10 | 中冶赛迪技术研究中心有限公司 | High-temperature steel coil Bundling strap on-line detecting system and method |
| CN110363761A (en) * | 2019-07-22 | 2019-10-22 | 上海眼控科技股份有限公司 | A kind of start-stop Mark Detection system and method for vehicle chassis dynamic detection |
| CN110781834A (en) * | 2019-10-28 | 2020-02-11 | 上海眼控科技股份有限公司 | Traffic abnormality image detection method, device, computer device and storage medium |
| CN111160824A (en) * | 2019-12-05 | 2020-05-15 | 深圳市铭华航电工艺技术有限公司 | Engineering data detection method and device, computer equipment and storage medium |
| CN111553950A (en) * | 2020-04-30 | 2020-08-18 | 中冶赛迪重庆信息技术有限公司 | Steel coil centering judgment method, system, medium and electronic terminal |
Non-Patent Citations (3)
| Title |
|---|
| ZHENAN JIANG 等: "AC loss measurements in HTS coil assemblies with hybrid coil structures", 《SUPERCONDUCTOR SCIENCE AND TECHNOLOGY》, vol. 29, no. 9, pages 1 - 13 * |
| 付英双 等: "热轧带钢卷打捆机控制系统的研究及应用", 《冶金自动化》, pages 297 - 301 * |
| 骆志坚 等: "基于计算机视觉检测技术自动计数系统的研究与应用", 仪表技术与传感器, no. 3, pages 19 - 20 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI787748B (en) * | 2021-03-03 | 2022-12-21 | 中國鋼鐵股份有限公司 | Method and system for predicting defects of holes in strip product |
| CN113537199A (en) * | 2021-08-13 | 2021-10-22 | 上海淇玥信息技术有限公司 | Image bounding box screening method, system, electronic device and medium |
| CN114638783A (en) * | 2022-02-10 | 2022-06-17 | 中冶南方工程技术有限公司 | Steel coil height centering system based on image recognition |
| CN115049620A (en) * | 2022-06-17 | 2022-09-13 | 中轻长泰(长沙)智能科技股份有限公司 | Paper stack appearance detection method, device, equipment and storage medium |
| CN115049620B (en) * | 2022-06-17 | 2024-03-26 | 中轻长泰(长沙)智能科技股份有限公司 | Paper stack appearance detection method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112288716B (en) | 2023-10-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112288716B (en) | Method, system, terminal and medium for detecting bundling state of steel coil | |
| US11263807B2 (en) | Method for the automatic material classification and texture simulation for 3D models | |
| CN110705405B (en) | Target labeling method and device | |
| JP5764238B2 (en) | Steel pipe internal corrosion analysis apparatus and steel pipe internal corrosion analysis method | |
| CN102930296B (en) | A kind of image-recognizing method and device | |
| CN103914802A (en) | Image selection and masking using imported depth information | |
| CN112649436A (en) | Container image acquisition and inspection system for non-fixed scene | |
| CN113393448A (en) | Deformation detection method, device and equipment and computer readable storage medium | |
| WO2023236371A1 (en) | Visual analysis method for cable element identification | |
| CN110458108A (en) | Method for real-time monitoring, system, terminal device and storage medium hand-manipulated | |
| CN112989768B (en) | Method and device for correcting connection questions, electronic equipment and storage medium | |
| CN111126417A (en) | Data center machine room management method and device | |
| CN112749746A (en) | Method, system and device for iteratively updating defect sample | |
| CN108171205A (en) | For identifying the method and apparatus of face | |
| CN112037128A (en) | Panoramic video splicing method | |
| CN104899854A (en) | Detection method and detection device of grain piling height line | |
| KR20210037199A (en) | Apparatus for dividing, tagging an image and for detecting defect of facilities using the same | |
| CN114926429A (en) | Method, device and equipment for detecting trace length and readable storage medium | |
| CN114861946A (en) | Breeding equipment management method and device, electronic equipment and storage medium | |
| CN116543327A (en) | Method, device, computer equipment and storage medium for identifying work types of operators | |
| CN112967224A (en) | Electronic circuit board detection system, method and medium based on artificial intelligence | |
| CN111553210B (en) | Training method of lane line detection model, lane line detection method and device | |
| JP2021015101A (en) | Image inspection device and image inspection program | |
| AU2015275308B2 (en) | A method for automatic material classification and texture simulation for 3d models | |
| CN104700396B (en) | The method and system of the parameter for estimating the volume of traffic is determined from image |
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 | ||
| CB02 | Change of applicant information |
Address after: 401329 No. 5-6, building 2, No. 66, Nongke Avenue, Baishiyi Town, Jiulongpo District, Chongqing Applicant after: MCC CCID information technology (Chongqing) Co.,Ltd. Address before: 20-24 / F, No.7 Longjing Road, North New District, Yubei District, Chongqing Applicant before: CISDI CHONGQING INFORMATION TECHNOLOGY Co.,Ltd. |
|
| CB02 | Change of applicant information | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |