CN112329782A - Raw material granularity determination method, system, terminal and medium - Google Patents
Raw material granularity determination method, system, terminal and medium Download PDFInfo
- Publication number
- CN112329782A CN112329782A CN202011217989.3A CN202011217989A CN112329782A CN 112329782 A CN112329782 A CN 112329782A CN 202011217989 A CN202011217989 A CN 202011217989A CN 112329782 A CN112329782 A CN 112329782A
- Authority
- CN
- China
- Prior art keywords
- raw material
- image
- sample image
- particle
- material particle
- 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
Links
- 239000002994 raw material Substances 0.000 title claims abstract description 283
- 238000000034 method Methods 0.000 title claims abstract description 49
- 239000002245 particle Substances 0.000 claims abstract description 231
- 230000011218 segmentation Effects 0.000 claims abstract description 28
- 238000012549 training Methods 0.000 claims abstract description 11
- 238000012545 processing Methods 0.000 claims description 34
- 230000009466 transformation Effects 0.000 claims description 23
- 238000003860 storage Methods 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 11
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000002372 labelling Methods 0.000 claims description 8
- 238000004891 communication Methods 0.000 claims description 6
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims description 3
- 238000005520 cutting process Methods 0.000 claims description 2
- 238000012216 screening Methods 0.000 abstract description 6
- 238000001514 detection method Methods 0.000 abstract description 5
- 238000010586 diagram Methods 0.000 description 11
- 230000006870 function Effects 0.000 description 7
- 238000012886 linear function Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 239000000571 coke Substances 0.000 description 4
- 238000012544 monitoring process Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 4
- 239000008188 pellet Substances 0.000 description 4
- 238000003709 image segmentation Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 229910000831 Steel Inorganic materials 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 238000006243 chemical reaction Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 238000005260 corrosion Methods 0.000 description 2
- 230000007797 corrosion Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000000644 propagated effect Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- 239000010959 steel Substances 0.000 description 2
- 230000001131 transforming effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000012798 spherical particle Substances 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/02—Affine transformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4084—Scaling of whole images or parts thereof, e.g. expanding or contracting in the transform domain, e.g. fast Fourier transform [FFT] domain scaling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/60—Rotation of whole images or parts thereof
-
- 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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06T2207/30204—Marker
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Multimedia (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Computation (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method, a system, a terminal and a medium for determining the granularity of a raw material, wherein the method comprises the steps of acquiring a real sample image, establishing a data set, training according to the data set to generate a raw material contour segmentation model, acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model to generate a raw material particle contour image, and determining the granularity of the raw material according to the raw material particle contour image; the technical problem of having solved traditional screening detection method and confirming raw materials granularity on the belt feeder, consume the manual work, it is long consuming time, inefficiency, there is great error in the raw materials granularity determination, be difficult to guarantee the quality of ironmaking is solved, reached the mode that detects through the machine, replaced the mode of artifical definite raw materials granularity, saved the manual work, the definite efficiency of raw materials granularity more rapid, promoted definite efficiency, the error is less, can promote the quality of ironmaking.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method, a system, a terminal and a medium for determining the granularity of a raw material.
Background
In the mining production, the raw materials are conveyed by the belt conveyor to enter the blast furnace, which is an important link of blast furnace ironmaking, and the quality of steel generated by blast furnace ironmaking is closely related to the granularity of the raw materials, so that the accurate determination of the granularity of the raw materials has great influence on the quality of the produced steel.
At present, the granularity of raw materials on a belt conveyor is generally determined by a traditional screening detection method, the traditional screening method needs to consume manpower, consumes long time and has low efficiency, the granularity of the raw materials is determined to have large errors, and the quality of ironmaking is difficult to ensure.
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 determining a particle size of a raw material, which are used to solve the technical problems of labor consumption, long time consumption, low efficiency, large error in determining a particle size of a raw material, and difficulty in ensuring iron-making quality in a conventional screening detection method.
In view of the above problems, the present invention provides a method for determining a particle size of a raw material, comprising:
acquiring a real sample image and establishing a data set;
training and generating a raw material contour segmentation model according to the data set;
acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model, and generating a raw material particle contour image;
and determining the granularity of the raw material according to the raw material particle outline image.
Optionally, the manner of establishing the data set includes at least one of:
marking the raw material particles in the real sample image, and establishing a data set;
and processing the real sample image to generate a false sample image, labeling the raw material particles in the false sample image, and establishing a data set.
Optionally, if the mode of establishing the data set includes processing the real sample image to generate a false sample image, labeling the raw material particles in the false sample image, and establishing the data set;
the real sample image is processed in a manner that includes at least one of: geometric transformation processing and image enhancement processing.
Optionally, if the processing manner of the real sample image includes a geometric transformation process, the geometric transformation process includes,
at least one of image turning of the real sample image, image rotation of the real sample image, image clipping of the real sample image, image scaling of the real sample image, and affine transformation of the real sample image;
if the processing mode of the real sample image comprises image enhancement processing, the image enhancement processing comprises any one of gray scale linear transformation of the real sample image and histogram equalization transformation of the real sample image.
Optionally, the method further includes:
and closing the non-closed raw material particle contour in the raw material contour image.
Optionally, the determining the granularity of the raw material according to the raw material particle profile image includes determining a raw material granularity parameter according to the raw material particle profile image, and determining the granularity of the raw material according to the raw material granularity parameter; the raw material granularity parameter comprises
At least one of the equivalent diameter of each raw material particle, the perimeter of each raw material particle, the area of each raw material particle and the average size of each raw material particle;
the perimeter of the single raw material particle is determined according to the raw material outline image;
determining the area of the single raw material particle according to the raw material outline image;
the calculation formula of the equivalent diameter of the raw material particles comprises,
the calculation formula of the average size of the raw material particles comprises,
the calculation formula of the total area of the raw material particles comprises,
wherein S istotalIs the total area S of each raw material particle in the image to be detected under the actual sceneiThe area of a single raw material particle in an image to be detected in an actual scene, D is the equivalent diameter of the single raw material particle in the image to be detected in the actual scene, P is the perimeter of the single raw material particle in the image to be detected in the actual scene, and SaveAnd k is a preset coefficient, and N is the number of the raw material particles, wherein k is the average size of the raw material particles in the image to be detected in an actual scene.
Optionally, the method further includes:
and carrying out color filling on the contour of each raw material particle in the raw material particle contour image according to the raw material particle size parameters to generate a raw material particle identification image.
The invention also provides a raw material granularity determining system, which comprises:
the establishing module is used for acquiring a real sample image and establishing a data set;
the model generation module is used for generating a raw material contour segmentation model according to the data set training;
the image generation module is used for acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model and generating a raw material particle contour image;
and the determining module is used for determining the granularity of the raw material according to the raw material particle outline image.
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 method for determining a particle size of a raw material as described in 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 perform a method of determining particle size of a feedstock as described in any one of the above embodiments.
As described above, the method, the system, the terminal and the medium for determining the granularity of the raw material provided by the present invention have the following beneficial effects:
acquiring a real sample image, establishing a data set, training according to the data set to generate a raw material contour segmentation model, acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model to generate a raw material particle contour image, and determining the granularity of a raw material according to the raw material particle contour image; the technical problem of having solved traditional screening detection method and confirming raw materials granularity on the belt feeder, consume the manual work, it is long consuming time, inefficiency, there is great error in the raw materials granularity determination, be difficult to guarantee the quality of ironmaking is solved, reached the mode that detects through the machine, replaced the mode of artifical definite raw materials granularity, saved the manual work, the definite efficiency of raw materials granularity more rapid, promoted definite efficiency, the error is less, can promote the quality of ironmaking.
Drawings
FIG. 1 is a schematic flow chart of a method for determining particle size of raw materials according to an embodiment of the present invention;
fig. 2 is a schematic diagram of an image to be detected including sintered ore particles according to an embodiment of the present invention;
fig. 3 is a schematic diagram of an image to be detected including spherical particles according to an embodiment of the present invention;
FIG. 4 is a raw material particle contour image generated by inputting FIG. 2 into a raw material contour segmentation model;
FIG. 5 is a raw material particle contour image generated by inputting FIG. 3 into a raw material contour segmentation model;
FIG. 6 is a schematic representation of a raw material particle profile image with non-occluded raw material particle profiles;
FIG. 7 is a schematic illustration of an image of the contour of the feedstock particles after the non-occluded feedstock particle contours present in FIG. 6 have been occluded;
FIG. 8 is a raw material particle identification image generated by color filling the raw material particle outline image of FIG. 4;
FIG. 9 is an alternative feedstock particle identification image generated by color filling the feedstock particle outline image of FIG. 5;
FIG. 10 is a schematic structural diagram of a raw material granularity determining system provided in the second embodiment of the present invention;
fig. 11 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 determining a particle size of a raw material according to an embodiment of the present invention includes:
s101: acquiring a real sample image and establishing a data set.
Optionally, the method for determining the particle size of the raw material is applied to the determination of the particle size of the raw material conveyed on a belt conveyor in the blast furnace ironmaking process.
Optionally, the real sample image includes, but is not limited to, blast furnace ironmaking raw material conveyed on a belt conveyor.
Optionally, the real sample image may be extracted through a video, and the video may be a real-time monitoring video, or may be a historical video, which is not limited herein.
Optionally, the real sample image may also be acquired by a snapshot device, the snapshot device is arranged at a certain preset position of the raw material conveying device, the snapshot is performed according to a preset snapshot rule, and a plurality of images obtained by the snapshot are used as the real sample image. The preset snapshot rule includes, but is not limited to, if the raw material is identified, the raw material is snapshot at certain time intervals; and if the raw material is not identified, stopping the snapshot until the raw material is identified again and starting the snapshot.
Optionally, the real sample image may also be acquired by a high-definition industrial camera or a high-definition video camera from raw material sample image information on the belt conveyor in the target area, where the raw material type on the belt conveyor includes, but is not limited to, ore, sinter, pellet, coke, and the like.
Optionally, the data set is created from a plurality of real sample images, wherein the real sample images comprise a plurality of raw material particles. Optionally, the feedstock particles include, but are not limited to, at least one of sinter particles, pellet particles, ore particles, coke particles, and the like.
Optionally, the way of building the data set from the real sample image includes, but is not limited to, at least one of the following:
marking the raw material particles in the real sample image, and establishing a data set;
and processing the real sample image to generate a false sample image, labeling the raw material particles in the false sample image, and establishing a data set.
Optionally, one or more processing methods may be performed on the same real sample image to obtain one or more false sample images. By processing the real sample images, the number and diversity of the sample images can be increased, and overfitting of a subsequently generated raw material contour segmentation model is avoided. Therefore, more sample data can be obtained by collecting a smaller number of real sample images based on multiple processing modes, so that the data in the data set is richer, and the credibility and the accuracy of the raw material contour segmentation model can be further improved.
Optionally, the data set may be obtained by labeling only the raw material particles in the real sample image, may also be obtained by labeling only the raw material particles in the false sample image, and may also be obtained by labeling the raw material particles in the real sample image and the raw material particles in the false sample image, respectively, which is not limited herein.
Optionally, in this embodiment, the raw material particles in the real sample image are labeled, and edges of each raw material particle that can be seen in the real sample image can be smoothly outlined by lines with a pixel value of 255 and a width value of 5 pixels on a background plate with pixel values of 0. Of course, the pixel value of the background plate, the pixel value of the line and the line width are not limited herein, and those skilled in the art can select the background plate, the line and the line width according to the needs.
Optionally, if the manner of creating the data set includes processing the real sample image to generate a false sample image, labeling the raw material particles in the false sample image, and creating the data set, the manner of processing the real sample image includes, but is not limited to, at least one of: geometric transformation processing and image enhancement processing.
And generating a plurality of false sample images by performing geometric transformation processing on the same real sample image through different parameters. Similarly, a plurality of false sample images can be generated for the same real sample image by image enhancement processing modes with different parameters. Therefore, by introducing the false sample image, data in a data set can be greatly enriched, so that the generated raw material contour segmentation model can be suitable for the images to be detected shot at various angles and the raw material particles at various angles in the images to be detected.
Optionally, if the processing manner of the real sample image includes a geometric transformation process, the geometric transformation process includes but is not limited to at least one of the following:
the method comprises the steps of turning over a real sample image, rotating the real sample image, cutting the real sample image, scaling the real sample image and carrying out affine transformation on the real sample image.
Wherein the image scaling of the real sample image comprises: and scaling the real sample image in an equal proportion or performing deformation scaling on the real sample image.
Optionally, if the processing mode of the real sample image includes image enhancement processing, the image enhancement processing includes any one of the following: and carrying out gray scale linear transformation on the real sample image and carrying out histogram equalization transformation on the real sample image.
Optionally, histogram equalization transformation is performed on the real sample image, that is, histogram equalization processing is performed on the real sample image, so that the global contrast of the real sample image can be increased.
Optionally, if the manner of processing the real sample image includes performing gray scale linear transformation on the real sample image, the gray scale linear transformation includes transforming a pixel value of the real sample image by a preset linear function, where the preset linear function includes:
g (x, y) ═ m × f (x, y) + b equation (1)
Wherein g (x, y) is the pixel value of the transformed false sample image, f (x, y) is the pixel value of the real sample image, m is the slope of the preset linear function straight line, b is the intercept of the preset linear function straight line on the y axis, and g (x, y) is more than or equal to 0 and less than or equal to 255.
Optionally, the gray scale linear transformation includes transforming pixel values of the real sample image by a preset linear function, so as to enhance or reduce brightness or contrast of the image. When the m value is larger than 1, the image contrast is increased, when the m value is 1, the adjustment of the image brightness is realized by adjusting the b value, when the m value is 0-1, the image contrast is weakened, when the m value is smaller than 0, the area with high image brightness is darkened, and the area with darker image is changed.
S102: and training according to the data set to generate a raw material contour segmentation model.
Optionally, training is performed according to the data set by using an image segmentation network based on deep learning to perform model training, so as to generate a raw material contour segmentation model. A U-Net image segmentation network may be used, or other image segmentation networks may be used instead, including but not limited to DeepLab series, SegNet, U-Net, FCNs, etc.
Optionally, the raw material contour segmentation model in this embodiment mainly consists of two parts, a compression network part and an expansion network part. The compressed network part is mainly used for capturing context information in the real sample image and/or the false sample image, and the symmetrical expanded network part is used for accurately positioning a part which needs to be segmented in the real sample image and/or the false sample image. In the network part, every two 3 × 3 convolutional layers are followed by a 2 × 2 max pooling layer with a step size of 2, and each convolutional layer is followed by a ReLU activation function to perform down-sampling on the original picture, in addition to which, each down-sampling is multiplied by the number of channels. In upsampling of the extended network portion, there is one 2 x 2 convolutional layer per step, the activation function is also ReLU, and two 3 x 3 convolutional layers, while the upsampling of each step adds a feature map from the corresponding compressed network portion, optionally tailored to maintain the same shape. At the last layer of the network is a 1 x 1 convolutional layer, by which the 64-channel feature vectors can be converted to the number of classification results required. Finally, the entire network has 23 convolutional layers.
S103: and acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model, and generating a raw material particle contour image.
Optionally, the image to be detected comprises a plurality of raw material particles.
Optionally, before the image to be detected is input to the raw material contour segmentation model, the method further includes: and performing image enhancement, such as image denoising and the like, on the image to be detected.
Optionally, the image to be detected may be obtained by monitoring a video of the real-time monitoring image, and the image to be detected may also be obtained by capturing an image captured by the capturing device. The method for acquiring the image to be detected is not limited herein.
Optionally, the image to be detected includes a real-time image on a belt conveyor according to blast furnace ironmaking.
Alternatively, referring to fig. 2, fig. 2 is a schematic view of an image to be detected including sintered ore particles. Referring to fig. 3, fig. 3 is a schematic diagram of an image to be detected including pellet particles.
Alternatively, referring to fig. 4, fig. 4 is a raw material particle contour image generated by inputting fig. 2 to the raw material contour segmentation model. Referring to fig. 5, fig. 5 is a raw material particle contour image generated by inputting fig. 3 to a raw material contour segmentation model.
In some embodiments, the feedstock particle size determination method further comprises:
the closed material particle outline in the closed material outline image is not closed.
Optionally, the generated raw material contour image includes a raw material particle contour of each raw material particle in the image to be detected. Due to problems such as imaging light, gaps may exist in the partial raw material particle outlines, and in this case, the non-closed raw material particle outlines may be closed. Optionally, the broken edges of the contour of the non-closed raw material particles are connected. Fig. 6 is a schematic diagram of a raw material particle profile image with non-occluded raw material particle profiles, and in fig. 6, there are 3 non-occluded raw material particle profiles A, B and C. Fig. 7 is a schematic diagram of a raw material particle contour image obtained by closing the non-closed raw material particle contour existing in fig. 6, wherein the non-closed raw material particle contour a and the non-closed raw material particle contour B directly connect two break points by a straight line, and the non-closed raw material particle contour C connects two break points by a curved line.
Optionally, the closing mode may be to connect the two closest break points, and the connection mode may be a straight line connection, or a curved line connection is adopted, and the curvature of the curved line may be determined according to the curvatures of the curved lines corresponding to the two break points. The closing means may be other means determined by those skilled in the art, and is not limited herein.
Optionally, the method for determining the particle size of the raw material further includes:
carrying out image expansion or image corrosion on the raw material particle outline image;
and determining the granularity parameter of the raw material according to the raw material particle outline image subjected to image expansion or image corrosion, and further determining the granularity of the raw material.
S104: and determining the granularity of the raw material according to the raw material particle outline image.
In some embodiments, determining the feedstock particle size from the feedstock particle profile image comprises:
and determining the granularity parameter of the raw material according to the contour image of the raw material particles, and determining the granularity of the raw material according to the granularity parameter of the raw material.
Optionally, the method for calculating the granularity parameter of the raw material according to the raw material profile image includes:
processing the raw material contour image by adopting an image processing method to obtain the raw material contour imageCalculating the area S of a single raw material particle, the number N of the raw material particles and the perimeter P of the single raw material particle to obtain the equivalent diameter D of the single raw material particle and the total area S of the raw material particlestotalAverage size S of raw material particlesave。
In some embodiments, the image processing methods include methods of finding contours, calculating contour areas, calculating contour perimeters, and the like.
In some embodiments, the feedstock particle size parameter comprises at least one of an equivalent diameter of a single feedstock particle, a perimeter of a single feedstock particle, an area of a single feedstock particle, and an average size of feedstock particles;
determining the perimeter of a single raw material particle according to the raw material contour image;
determining the area of a single raw material particle according to the raw material contour image;
the formula for calculating the equivalent diameter of the feedstock particles includes,
the formula for calculating the average size of the feedstock particles includes,
the calculation formula of the total area of the raw material particles comprises,
wherein S istotalIs the total area of the raw material particles, SiIs the area of a single raw material particle, D is the equivalent diameter of the single raw material particle, P is the perimeter of the single raw material particle, SaveThe average size of the raw material particles, k is a preset coefficient, and N is the number of the raw material particles.
It should be noted that k is a preset coefficient for converting a pixel into an actual unit in a real scene, and therefore StotalThe total area of each raw material particle in the image to be detected in a real scene; d is the equivalent diameter of a single raw material particle in an image to be detected in a real scene, SaveFor the original position in the image to be detected in the real sceneAverage particle size of the feedstock; siThe area of a single raw material particle in the raw material outline image is shown in unit of pixel; p is the perimeter of a single raw material particle in the raw material profile image, and the unit is pixel.
Optionally, k is a preset coefficient of the pixel converted into an actual unit in the real scene, and the equivalent diameter (D) of a single raw material particle and the average size (S) of the raw material particle in the real scene can be calculated according to the preset coefficientave)。
In some embodiments, the predetermined coefficients are determined as follows:
the width (width) of the sample image under an actual scene is obtained through field measurement, the number n of pixel points contained in the wide side of the sample image is determined, and then an actual unit coefficient k of pixel conversion is obtained, and the calculation mode of the preset coefficient is as follows:
wherein, width is the width of the sample image in the actual scene, k is the actual unit coefficient of pixel conversion, and n is the number of pixel points included in the wide side of the sample image.
Optionally, the preset coefficient may also be determined by using other relevant manners, which are not limited herein.
In some embodiments, the feedstock particle size determination method further comprises:
and carrying out color filling on the contour of each raw material particle in the raw material particle contour image according to the raw material particle size parameters to generate a raw material particle identification image.
Optionally, if the raw material particle size parameter includes the area of a single raw material particle, the raw material particle size determining method further includes:
and carrying out color filling on the outline of each raw material particle in the raw material particle outline image according to the area of the single raw material particle to generate a raw material particle identification image.
Optionally, if the raw material particle size parameter includes the area of a single raw material particle, a color threshold is defined based on the area of the single raw material particle, and the raw material particle is color-filled based on the threshold.
Optionally, if the raw material particle size parameter includes the area of a single raw material particle, an area grade is defined according to the area of each single raw material particle, and colors of different depths are filled according to the area grade. The color may be darker as the area is larger, or may be darker as the area is smaller, and the color may be filled according to other filling rules set by those skilled in the art, which is not limited herein.
Optionally, when the particle size parameter of the raw material includes the equivalent diameter of a single raw material particle, the color filling manner is similar to the above manner, and is not described herein again.
Optionally, if the raw material granularity parameter includes multiple parameters, color filling may be performed according to one of the parameters, or a comprehensive rating may be obtained by selecting multiple parameters to perform weighted average, and then color filling may be performed according to the comprehensive rating.
Optionally, the color fill may be a black, gray, and white color transition fill. The color filling may also be a color filling, for example, it is determined that a certain area range of a single raw material particle is filled with red, and the other area ranges of the single raw material particle are filled with other colors respectively.
Referring to fig. 8 and 9, fig. 8 is a raw material particle recognition image generated by color filling the raw material particle outline image of fig. 4; fig. 9 is another raw material particle recognition image generated by color filling the raw material particle outline image of fig. 5.
Through the raw material particle recognition image, the particle size condition of each raw material particle in the image to be detected can be visually seen.
In some embodiments, determining the feedstock particle size comprises:
acquiring a statistical rule, and respectively carrying out statistics according to the statistical rule and the granularity parameter dimension of each raw material to generate a statistical result;
and determining the granularity of the raw materials according to the statistical result.
Alternatively, the statistical rules may be pre-designed. Different statistical rules can be set for different types of raw material particles, taking the statistical calculation of the equivalent diameter dimension of the raw material particles as an example, if the raw material particles are sinter particles or pellet particles, the number of the raw material particles with the equivalent diameter of less than 5mm, between 5mm and mm, between 10mm and 25mm, between 25mm and 40mm and more than 40mm needs to be counted; when the raw material particles are coke particles or ore particles, the number of coke particles or ore particles with equivalent diameters of less than 40mm, 40 mm-60 mm, 60 mm-80 mm and more than 80mm needs to be counted. And determining the granularity of the raw material according to the statistical result, wherein the statistical result comprises the number ratio of the raw material particles in each value interval. It should be noted that the above value ranges are only an exemplary illustration, and those skilled in the art can adjust the value ranges as needed.
Optionally, the method of performing statistics by using the average size of the raw material particles, the area of a single raw material particle, and the perimeter of a single raw material particle as dimensions is similar to the above method of performing statistics by using the equivalent diameter dimension of the raw material particles, and is not described herein again.
Optionally, the statistical result may include the number of the raw material particles in each value interval in each raw material particle size parameter dimension, and the statistical result may also include the number ratio of the raw material particles in each value interval in each raw material particle size parameter dimension.
Optionally, the particle size of the raw material can be determined according to the distribution of each color in the raw material particle identification image.
The embodiment of the invention provides a raw material particle size determining method, which comprises the steps of obtaining a real sample image, establishing a data set, training according to the data set to generate a raw material contour segmentation model, obtaining an image to be detected, inputting the image to be detected into the raw material contour segmentation model to generate a raw material particle contour image, and determining the raw material particle size according to the raw material particle contour image; the technical problem of having solved traditional screening detection method and confirming raw materials granularity on the belt feeder, consume the manual work, it is long consuming time, inefficiency, there is great error in the raw materials granularity determination, be difficult to guarantee the quality of ironmaking is solved, reached the mode that detects through the machine, replaced the mode of artifical definite raw materials granularity, saved the manual work, the definite efficiency of raw materials granularity more rapid, promoted definite efficiency, the error is less, can promote the quality of ironmaking.
Optionally, if the image to be detected includes a real-time monitoring image of the raw material conveyed on the belt conveyor, the raw material granularity determining method based on the embodiment can realize real-time continuous on-line determination of the raw material granularity, and further ensure the quality of the ironmaking finished product.
Optionally, the method can be realized based on equipment with corresponding functions, so that the method can be suitable for severe operation sites such as high temperature and the like, and is wide in application.
Example two
Referring to fig. 10, a system 1000 for determining particle size of raw material includes:
an establishing module 1001, configured to obtain a real sample image and establish a data set;
a model generation module 1002, configured to generate a raw material contour segmentation model according to data set training;
the image generation module 1003 is used for acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model, and generating a raw material particle contour image;
a determining module 1004 for determining a feedstock particle size from the feedstock particle profile image.
In this embodiment, the raw material granularity determining system is substantially provided with a plurality of modules for executing the raw material granularity determining method in the above embodiment, and specific functions and technical effects are only referred to in the first embodiment, which is not described herein again.
Referring to fig. 6, an embodiment of the present invention further provides a terminal 1100, including a processor 1101, a memory 1102, and a communication bus 1103;
a communication bus 1103 is used to connect the processor 1101 with the memory 1102;
the processor 1101 is configured to execute a computer program stored in the memory 1102 to implement the method for determining the granularity of raw material as described in one or more of the first embodiment above.
An embodiment of the present invention also provides a computer-readable storage medium, characterized in that, a computer program is stored thereon,
a computer program for causing a computer to perform a method of determining particle size of a feedstock as described in any one of the above embodiments.
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 method for determining a particle size of a feedstock, comprising:
acquiring a real sample image and establishing a data set;
training and generating a raw material contour segmentation model according to the data set;
acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model, and generating a raw material particle contour image;
and determining the granularity of the raw material according to the raw material particle outline image.
2. The feedstock particle size determination method according to claim 1, wherein the manner in which the data set is established includes at least one of:
marking the raw material particles in the real sample image, and establishing a data set;
and processing the real sample image to generate a false sample image, labeling the raw material particles in the false sample image, and establishing a data set.
3. The raw material granularity determination method according to claim 2, wherein if the data set is established in a manner that the real sample image is processed to generate a false sample image, raw material particles in the false sample image are labeled to establish a data set;
the real sample image is processed in a manner that includes at least one of: geometric transformation processing and image enhancement processing.
4. The method of determining the particle size of a feedstock according to claim 3,
if the processing mode of the real sample image comprises geometric transformation processing, the geometric transformation processing comprises at least one of image turning of the real sample image, image rotation of the real sample image, image cutting of the real sample image, image scaling of the real sample image and affine transformation of the real sample image;
if the processing mode of the real sample image comprises image enhancement processing, the image enhancement processing comprises any one of gray scale linear transformation of the real sample image and histogram equalization transformation of the real sample image.
5. The feedstock particle size determination method according to any one of claims 1-4, further comprising:
and closing the non-closed raw material particle contour in the raw material contour image.
6. The feedstock particle size determination method according to claim 5, wherein said determining a feedstock particle size from said feedstock particle profile image comprises determining a feedstock particle size parameter from said feedstock particle profile image, determining a feedstock particle size from said feedstock particle size parameter; the raw material particle size parameter comprises at least one of the equivalent diameter of a single raw material particle, the perimeter of the single raw material particle, the area of the single raw material particle and the average size of the raw material particle;
the perimeter of the single raw material particle is determined according to the raw material outline image;
determining the area of the single raw material particle according to the raw material outline image;
the calculation formula of the equivalent diameter of the raw material particles comprises,
the calculation formula of the average size of the raw material particles comprises,
the calculation formula of the total area of the raw material particles comprises,
wherein S istotalIs the total area S of each raw material particle in the image to be detected under the actual sceneiThe area of a single raw material particle in an image to be detected in an actual scene, D is the equivalent diameter of the single raw material particle in the image to be detected in the actual scene, P is the perimeter of the single raw material particle in the image to be detected in the actual scene, and SaveAnd k is a preset coefficient, and N is the number of the raw material particles, wherein k is the average size of the raw material particles in the image to be detected in an actual scene.
7. The method for determining the particle size of a feedstock according to claim 6, further comprising:
and carrying out color filling on the contour of each raw material particle in the raw material particle contour image according to the raw material particle size parameters to generate a raw material particle identification image.
8. A feedstock particle size determination system, comprising:
the establishing module is used for acquiring a real sample image and establishing a data set;
the model generation module is used for generating a raw material contour segmentation model according to the data set training;
the image generation module is used for acquiring an image to be detected, inputting the image to be detected into the raw material contour segmentation model and generating a raw material particle contour image;
and the determining module is used for determining the granularity of the raw material according to the raw material particle outline image.
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 adapted to execute a computer program stored in the memory to implement the method of determining particle size of a raw material as claimed in 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 feedstock particle size determination method according to any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011217989.3A CN112329782A (en) | 2020-11-04 | 2020-11-04 | Raw material granularity determination method, system, terminal and medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011217989.3A CN112329782A (en) | 2020-11-04 | 2020-11-04 | Raw material granularity determination method, system, terminal and medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112329782A true CN112329782A (en) | 2021-02-05 |
Family
ID=74323628
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011217989.3A Pending CN112329782A (en) | 2020-11-04 | 2020-11-04 | Raw material granularity determination method, system, terminal and medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112329782A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113192037A (en) * | 2021-05-06 | 2021-07-30 | 中冶赛迪重庆信息技术有限公司 | Belt conveyor monitoring method, system, medium and electronic terminal |
| CN113591740A (en) * | 2021-08-04 | 2021-11-02 | 清华大学 | Deep learning-based method and device for identifying silt particles in complex river environment |
| CN115436238A (en) * | 2022-09-22 | 2022-12-06 | 中冶南方工程技术有限公司 | Blast furnace raw fuel granularity online detection method and system |
| TWI829366B (en) * | 2022-09-28 | 2024-01-11 | 中國鋼鐵股份有限公司 | System and method for analyzing particle size of raw material |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104063866A (en) * | 2014-06-26 | 2014-09-24 | 中国矿业大学(北京) | Method for detecting granularity in ore transmission process |
| CN104075965A (en) * | 2014-07-02 | 2014-10-01 | 北京机械设备研究所 | Method for measuring granularity of microscopic image particles segmented on basis of watershed |
| CN109785378A (en) * | 2018-03-21 | 2019-05-21 | 威海晶合数字矿山技术有限公司 | A kind of ore grain size online detection instrument based on atlas image algorithm analytical technology |
| CN109859224A (en) * | 2019-01-31 | 2019-06-07 | 东北大学 | A kind of conveyer belt rock or ore particle image partition method |
| US20190251397A1 (en) * | 2018-02-14 | 2019-08-15 | Nvidia Corporation | Generation of Synthetic Images For Training a Neural Network Model |
| CN110852395A (en) * | 2019-11-15 | 2020-02-28 | 鞍钢集团矿业有限公司 | Ore granularity detection method and system based on autonomous learning and deep learning |
| CN111047555A (en) * | 2019-11-13 | 2020-04-21 | 鞍钢集团矿业有限公司 | Ore image granularity detection algorithm based on image processing technology |
-
2020
- 2020-11-04 CN CN202011217989.3A patent/CN112329782A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104063866A (en) * | 2014-06-26 | 2014-09-24 | 中国矿业大学(北京) | Method for detecting granularity in ore transmission process |
| CN104075965A (en) * | 2014-07-02 | 2014-10-01 | 北京机械设备研究所 | Method for measuring granularity of microscopic image particles segmented on basis of watershed |
| US20190251397A1 (en) * | 2018-02-14 | 2019-08-15 | Nvidia Corporation | Generation of Synthetic Images For Training a Neural Network Model |
| CN109785378A (en) * | 2018-03-21 | 2019-05-21 | 威海晶合数字矿山技术有限公司 | A kind of ore grain size online detection instrument based on atlas image algorithm analytical technology |
| CN109859224A (en) * | 2019-01-31 | 2019-06-07 | 东北大学 | A kind of conveyer belt rock or ore particle image partition method |
| CN111047555A (en) * | 2019-11-13 | 2020-04-21 | 鞍钢集团矿业有限公司 | Ore image granularity detection algorithm based on image processing technology |
| CN110852395A (en) * | 2019-11-15 | 2020-02-28 | 鞍钢集团矿业有限公司 | Ore granularity detection method and system based on autonomous learning and deep learning |
Non-Patent Citations (4)
| Title |
|---|
| 周永章 等: "《地球科学大数据挖掘与机器学习》", 30 September 2018, 中山大学出版社 * |
| 柳小波 等: "基于U-Net和Res_UNet模型的传送带矿石图像分割方法", 《东北大学学报（自然科学版）》 * |
| 青十五: "《策略产品经理 模型与方法论》", 30 September 2020, 机械工业出版社 * |
| 黄习敏: "基于图像识别的在线粒度检测方法研究与检测系统设计", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113192037A (en) * | 2021-05-06 | 2021-07-30 | 中冶赛迪重庆信息技术有限公司 | Belt conveyor monitoring method, system, medium and electronic terminal |
| CN113591740A (en) * | 2021-08-04 | 2021-11-02 | 清华大学 | Deep learning-based method and device for identifying silt particles in complex river environment |
| CN113591740B (en) * | 2021-08-04 | 2024-05-14 | 清华大学 | Deep learning-based sediment particle identification method and device in complex river environment |
| CN115436238A (en) * | 2022-09-22 | 2022-12-06 | 中冶南方工程技术有限公司 | Blast furnace raw fuel granularity online detection method and system |
| CN115436238B (en) * | 2022-09-22 | 2023-11-21 | 中冶南方工程技术有限公司 | Blast furnace raw fuel granularity online detection method and system |
| TWI829366B (en) * | 2022-09-28 | 2024-01-11 | 中國鋼鐵股份有限公司 | System and method for analyzing particle size of raw material |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112329782A (en) | Raw material granularity determination method, system, terminal and medium | |
| CN111598091A (en) | Image recognition method and device, electronic equipment and computer readable storage medium | |
| CN115115644B (en) | Vehicle welding defect detection method based on artificial intelligence | |
| CN111008961B (en) | Transmission line equipment defect detection method and system, equipment and medium thereof | |
| CN108898132A (en) | A kind of terahertz image dangerous material recognition methods based on Shape context description | |
| CN112712513A (en) | Product defect detection method, device, equipment and computer readable storage medium | |
| CN115131354A (en) | Laboratory plastic film defect detection method based on optical means | |
| CN115631116B (en) | Aircraft power inspection system based on binocular vision | |
| CN115019160B (en) | Method for controlling image quality of warehouse goods in factory | |
| CN114772208B (en) | Non-contact belt tearing detection system and method based on image segmentation | |
| CN112419397A (en) | Ore particle size grading method and system based on image and deep neural network | |
| CN111402250A (en) | Machine vision defect detection method and platform based on edge calculation | |
| CN113822110B (en) | Target detection method and device | |
| CN113674277A (en) | Unsupervised domain adaptive surface defect region segmentation method and device and electronic equipment | |
| Nikolaev et al. | Estimation of The Surface Quality Of Galvanazed Steel: The Method Of Decomposing The Image Into Layers | |
| CN109447911B (en) | Image restoration method, device, storage medium and terminal equipment | |
| CN112906611A (en) | Well lid detection method and device, electronic equipment and storage medium | |
| CN112712077B (en) | Steel flow contour determination method, system, terminal and medium | |
| CN116977190A (en) | Image processing method, apparatus, device, storage medium, and program product | |
| CN114863277A (en) | Machine vision-based method for rapidly detecting irregular particle group overrun particles | |
| JP3578321B2 (en) | Image normalizer | |
| CN112669207A (en) | Method for enhancing resolution of face image based on television camera | |
| CN113011268A (en) | Intelligent vehicle navigation method and device, electronic equipment and storage medium | |
| CN117808808B (en) | Ore granularity detection method, system, electronic equipment and storage medium | |
| CN111008604A (en) | Prediction image acquisition method and device, computer equipment and storage medium |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210205 |
|
| RJ01 | Rejection of invention patent application after publication |