CN112700419A - Yield measuring method, device and system based on image segmentation - Google Patents
Yield measuring method, device and system based on image segmentation Download PDFInfo
- Publication number
- CN112700419A CN112700419A CN202011640633.0A CN202011640633A CN112700419A CN 112700419 A CN112700419 A CN 112700419A CN 202011640633 A CN202011640633 A CN 202011640633A CN 112700419 A CN112700419 A CN 112700419A
- Authority
- CN
- China
- Prior art keywords
- ore
- video
- conveyor belt
- image
- segmentation
- 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
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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
-
- 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
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
- G06T2207/30104—Vascular flow; Blood flow; Perfusion
Abstract
The invention provides an image segmentation-based yield measuring method, device and system, wherein the method comprises the following steps: acquiring a video of ore conveyed by a conveyor belt in a target time period, and performing image segmentation on a plurality of image frames of the video to obtain an ore segmentation result; determining the total area of the ore corresponding to the video according to the ore segmentation result; and estimating to obtain the ore yield of the target time period according to the total area of the ore and a preset estimation model. The method comprises the steps of carrying out image segmentation on image frames in the ore video conveyed by the conveyor belt through an image segmentation method, then determining the total area of the ore in a target time interval based on an image segmentation result, obtaining the ore yield in the target time interval according to the total area of the ore and an estimation model, detecting the yield in real time through a non-contact mode of video acquisition and identification, and being not easily influenced by the complex environment of ore mining, high in metering precision and low in yield metering cost.
Description
Technical Field
The invention relates to the technical field of yield measurement, in particular to a yield measurement method, device and system based on image segmentation.
Background
Coal is one of the main energy sources widely used in recent times, and is in a very important strategic position in an energy supply system. The coal yield data is reasonably and accurately acquired, so that data information service can be provided for government departments and industrial enterprises, and the method can play an important role in the aspects of enterprise safety production management, coal yield regulation and control, tax collection and management enhancement, overrun overload control and the like.
The coal mining process mainly depends on belt transportation, and the continuous weighing modes such as an electronic belt scale are generally adopted for weighing the coal mining quantity. However, the belt is susceptible to changes such as deformation and relaxation caused by changes of downhole temperature and humidity, so that the measurement data is inaccurate and needs to be regularly adjusted; moreover, coal is unevenly distributed on the belt, so that the impact on the sensor is large, the abrasion is serious, and the metering data is inaccurate.
Disclosure of Invention
The invention solves the problem of inaccurate yield measurement in the existing belt weighing method.
In order to solve the above problems, the present invention provides an image segmentation-based yield measurement method, comprising: acquiring a video of ore conveyed by a conveyor belt in a target time period, and performing image segmentation on a plurality of image frames extracted from the video to obtain an ore segmentation result; determining the total area of ores corresponding to the video according to the ore segmentation result; and estimating to obtain the ore yield of the target time period according to the total ore area and a preset estimation model.
Optionally, the image segmentation of the plurality of image frames extracted from the video to obtain an ore segmentation result includes: determining the running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range; determining the number of interval frames according to the operation duration and the frame rate of the video; extracting a plurality of image frames in the video at intervals of the interval frame number; and inputting the extracted image frames into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
Optionally, the determining a running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range specifically includes: acquiring the actual running speed of the conveyor belt and the actual length of the conveyor belt in the video range; calculating to obtain operation duration according to the actual length and the actual operation speed; or, dividing the pixel distance moved by the target point in any adjacent image frame of the video by the time interval of the adjacent image frame, and calculating to obtain the pixel speed of the conveyor belt; acquiring the pixel length of a conveyor belt in the video range; and calculating to obtain the running time according to the pixel length and the pixel speed. Optionally, the determining the total area of the ore corresponding to the video according to the ore segmentation result includes: and summing the ore areas corresponding to the ore segmentation results of the image frames to obtain the total ore area corresponding to the video.
Optionally, the method further comprises: and performing linear regression fitting according to the total ore area and the ore yield corresponding to the video in the historical time period to obtain the preset estimation model.
Optionally, the preset estimation model is as follows:
M＝β*S+α
wherein M is the ore yield, S is the total area of the ore, alpha is the intercept, and beta is the slope.
The invention provides a yield metering device based on image segmentation, which comprises: the image segmentation module is used for acquiring a video of ore conveyed by a conveyor belt in a target time period and performing image segmentation on a plurality of image frames of the video to obtain an ore segmentation result; the area determining module is used for determining the total area of the ores corresponding to the video according to the ore segmentation result; and the yield estimation module is used for estimating and obtaining the ore yield of the target time period according to the total ore area and a preset estimation model.
Optionally, the image segmentation module is specifically configured to: determining the running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range; determining the number of interval frames according to the operation duration and the frame rate of the video; extracting a plurality of image frames in the video at intervals of the interval frame number; and inputting the extracted image frames into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
Optionally, the image segmentation module is specifically configured to: acquiring the actual running speed of the conveyor belt and the actual length of the conveyor belt in the video range; calculating to obtain operation duration according to the actual length and the actual operation speed; or, dividing the pixel distance moved by the target point in any adjacent image frame of the video by the time interval of the adjacent image frame, and calculating to obtain the pixel speed of the conveyor belt; acquiring the pixel length of a conveyor belt in the video range; and calculating to obtain the running time according to the pixel length and the pixel speed.
The invention provides a yield metering system based on image segmentation, which is characterized by comprising a camera device and a server; the camera device is used for collecting videos of ore conveyed by the conveyor belt and sending the videos to the server; the server is used for executing the yield measuring method based on image segmentation.
The method comprises the steps of carrying out image segmentation on image frames in the ore video conveyed by the conveyor belt through an image segmentation method, then determining the total area of the ore in a target time interval based on an image segmentation result, obtaining the ore yield in the target time interval according to the total area of the ore and an estimation model, detecting the yield in real time through a non-contact mode of video acquisition and identification, and being not easily influenced by the complex environment of ore mining, high in metering precision and low in yield metering cost.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
FIG. 1 is a schematic diagram of an application environment of an image segmentation-based throughput measurement method according to the present invention;
FIG. 2 is a schematic flow chart of an image segmentation based throughput method in an embodiment of the present invention;
FIG. 3 is a diagram illustrating a coal flow segmentation result of an image frame according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of a throughput metrology device based on image segmentation according to an embodiment of the present invention.
Description of reference numerals:
401-an image segmentation module; 402-an area determination module; 403-yield estimation module.
Detailed Description
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail below. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Coal mine resources mainly come from underground coal mining, while coal mining is mainly carried out by a conveyor belt, if the actual yield of coal can be detected in real time in the conveyor belt transportation process in a non-contact mode of images, the yield metering cost can be compressed to a great extent, the influence of an underground external environment on the metering precision is reduced, and the method can play an important role in the coal mine industry.
Referring to fig. 1, an application environment diagram of the image segmentation-based yield measurement method shows a network explosion-proof camera, a coal mine brain computing terminal, a coal mine brain platform, a cloud server, and an application terminal.
The network anti-explosion camera is arranged above the conveyor belt and is used for collecting a video of ore conveyed by the conveyor belt; the coal mine brain computing terminal comprises a video metering yield model, and video data are analyzed and yield is estimated based on the model; the coal mine brain platform is used for receiving and storing videos, intelligently monitoring and analyzing and the like; the cloud server comprises an Object Storage Service (OBS) for storing data such as videos and the like; the application terminal is provided with coal mine brain application software and can execute functions such as inquiry, control and the like.
The system is simple in structure, not easy to be influenced by complex underground environments and high in precision. The equipment and the installation requirements on the mine are simple, the explosion-proof camera and the communication network are only needed to be installed to the coal mine brain computing terminal, the coal mine brain platform and the like, the service can be provided for the mine enterprises in a service form, the construction cost is low, and the operation, the operation and the maintenance are simple.
FIG. 2 is a schematic flow chart of an image segmentation based throughput method in an embodiment of the present invention, including:
s202, obtaining a video of ore conveyed by a conveyor belt in a target time period, and carrying out image segmentation on a plurality of image frames extracted from the video to obtain an ore segmentation result.
The ore comprises coal mine, iron ore and other solid ores which can be conveyed by using a conveyor belt.
Taking coal as an example, in the coal mining process, coal mined from an underground coal face can be generally conveyed by a conveyor belt. A camera is positioned above the conveyor belt to capture video of the coal stream. Then, the server may segment a plurality of image frames of the video based on an image segmentation algorithm to obtain an ore segmentation result, where the ore segmentation result includes information such as an outline, a position, and a category of each target in the image frames.
The image segmentation algorithm may, for example, use an example segmentation algorithm, a panorama segmentation algorithm, or the like to perform image segmentation on the ore, the conveyor belt, and other backgrounds in the image frame, so as to obtain the position and pixel information of the ore flow. It should be noted that, in consideration of computational efficiency, image segmentation may be performed on only a part of image frames in the video, and the selection of the image frames needs to meet the requirement of calculating the area of the mineral flow in the video.
In consideration of the need for calculating the total area of ore in the subsequent steps, image segmentation is performed only on the extracted partial image frames, in which the ore flows are non-overlapping and continuous. Optionally, the method is performed according to the following steps:
firstly, the running time of any pixel point on a conveyor belt in a video from entering a video range to leaving the video range is determined.
Secondly, determining the number of interval frames according to the operation duration and the frame rate of the video, and extracting a plurality of image frames in the video by taking the number of interval frames as an interval. And periodically extracting a plurality of image frames in the video by using the interval frame number determined by the running time length. The number of interval frames can be obtained by multiplying the running duration by the frame rate of the video.
And then, inputting each extracted image frame into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
And S204, determining the total area of the ore corresponding to the video according to the ore segmentation result.
After the ore segmentation results of a plurality of image frames are obtained through segmentation, the area of the ore in each image frame can be determined based on the position and pixel information of the ore in the ore segmentation results. The area here may be a pixel area, that is, an area of all pixels of a region occupied by a mineral in an image frame, or may be expressed in terms of the number of all pixels in the occupied region.
After the ore areas of a plurality of image frames are obtained, the total area of the ores appearing in the whole video can be calculated through a reasonable summation mode. Specifically, the total area of the ore corresponding to the video is obtained by summing the ore areas corresponding to the ore segmentation results of each image frame. For example, if the number of the interval frames is 20, the total area S of the coal flowGeneral assemblyThe calculation formula of (a) is as follows:
Sgeneral assembly＝Sn+Sn+20*1……+Sn+20*m (1)
Wherein S isnIs the coal flow area of the nth frame, Sn+20*1Area of coal flow, S, for the n +20 th framen+20*mThe area of the coal flow of the (n + 20) th frame is m, and m and n are positive integers.
And S206, estimating the ore yield in the target time period according to the total area of the ore and a preset estimation model.
Before the ore production estimation, estimation model training is performed. Firstly, data of a historical period is obtained, wherein the data comprises the total ore area of a video of the historical period and the ore yield obtained by weighing in the historical period, and then linear regression fitting is carried out according to the total ore area and the ore yield corresponding to the video in the historical period to obtain a preset estimation model. Illustratively, the preset estimation model is as follows:
M＝β*S+α (2)
wherein M is the ore yield, S is the total area of the ore, α is the intercept, and β is the slope (regression coefficient).
The linear regression estimation model adopts a linear function to fit the total ore area and the ore yield in the historical time period. Alternatively, assume that the function (Hypothesis) is as follows:
h(S)＝β*S+α (3)
where h (S) is a hypothetical function and α and β are parameters of a linear function.
The loss function J is as follows:
where M is the actual value, i is the serial number of the historical time period, and M is the total number of the historical time periods.
Firstly, setting a linear function h (S), then selecting a loss function J, and searching corresponding parameters alpha and beta when the loss function J takes the minimum value according to a gradient descent method. The smaller the value of the loss function J, the closer the estimated value is to the actual value. Min (J (alpha, beta)) is taken as an optimization target.
According to the yield measurement method based on image segmentation, the image segmentation is carried out on the image frames in the ore video conveyed by the conveyor belt through the image segmentation method, then the total area of the ore in the target time interval is determined based on the image segmentation result, the ore yield in the target time interval can be obtained according to the total area of the ore and the estimation model, the yield is detected in real time through a non-contact mode of video acquisition and identification, the method is not easily influenced by the complex environment of ore mining, the measurement precision is high, and the yield measurement cost is low.
Considering two situations of known conveyor belt speed or unknown conveyor belt speed, the following two ways can be adopted for calculating the running time of any pixel point on the conveyor belt from entering the video range to leaving the video range:
(1) acquiring the actual running speed of the conveyor belt and the actual length of the conveyor belt in a video range; and then, calculating to obtain the operation time length according to the actual length and the actual operation speed.
If the conveyor belt device can feed back the running speed, the running speed can be directly obtained; the actual length of the conveyor belt in the video range can be obtained through field measurement; and dividing the actual length by the running speed to obtain the running time length.
(2) Dividing the pixel distance of the movement of the target point in any adjacent image frame of the video by the time interval of the adjacent image frame, and calculating to obtain the pixel speed of the conveyor belt; then, the pixel length of the conveyor belt in the video range is obtained; and finally, calculating to obtain the running time according to the pixel length and the pixel speed.
If the conveyor belt device cannot feed back the running speed information, the running speed information can be determined based on the moving distance of a certain point in an adjacent image frame in the image frame of the video divided by the time interval of the adjacent image frame. The speed of the conveyor belt can be measured through the video collected by the camera. The formula for the conveyor speed v is as follows:
v＝Δd/Δt (5)
and delta d is the coordinate moving distance of the target point in the two adjacent image frames, and delta t is the time interval of the two adjacent image frames. Note that the unit of this speed is "pixel/second". Since only its speed ratio is needed, its actual speed is not known.
Then, the pixel length of the conveyor belt in the image frame is obtained through identification, and the running time length can be obtained by dividing the pixel length by the pixel speed.
Referring to the schematic diagram of the coal flow segmentation result of the image frame shown in fig. 3, in the white frame region in the segmentation result, i.e., the region occupied by the segmented coal flow, the pixel area occupied by the coal flow can be calculated according to the segmentation result, and the conveyor belt length (length in the up-down direction) in the image frame can be identified.
Taking an example segmentation algorithm as an example, the following steps included in S102 are introduced: and inputting each extracted image frame into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
For example, image frames of coal conveyed by a conveyor belt in a historical period are obtained, the image frames are manually labeled to obtain training samples, and an instance segmentation model is trained.
Illustratively, a YOLACT model is adopted, image frames of videos in historical periods are manually labeled to obtain a training data set, the image frames in the training data set are used as input, and the YOLACT model is trained by using the manually labeled categories of the image frames as output. And after the convergence condition is met, the model training is finished, and the obtained model parameters are suitable for the current yield estimation environment for conveying coal by the conveyor belt. The example segmentation algorithm model combines target detection and semantic segmentation, and inputs image frames into the trained model to detect targets (target detection), and then labels each target (semantic segmentation), where the labels may include: coal flow, conveyor belt, background, etc.
FIG. 4 is a schematic structural diagram of an image segmentation-based throughput metrology apparatus, in one embodiment of the present invention, the apparatus comprising:
the image segmentation module 401 is configured to obtain a video of ore conveyed by a conveyor belt in a target time period, and perform image segmentation on a plurality of image frames extracted from the video to obtain an ore segmentation result;
an area determining module 402, configured to determine a total area of the ore corresponding to the video according to the ore segmentation result;
and a yield estimation module 403, configured to estimate, according to the total area of the ores and a preset estimation model, a yield of the ores in the target time period.
The yield metering device based on image segmentation provided by the embodiment performs image segmentation on image frames in ore video conveyed by a conveyor belt through an image segmentation method, then determines the total area of ore in a target time interval based on an image segmentation result, can obtain the ore yield in the target time interval according to the total area of ore and an estimation model, detects the yield in real time through a non-contact mode of video acquisition and identification, is not easily influenced by the complex environment of ore mining, and has high metering precision and low yield metering cost.
Optionally, as an embodiment, the image segmentation module 401 is specifically configured to: determining the running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range; determining the number of interval frames according to the operation duration and the frame rate of the video; extracting a plurality of image frames in the video at intervals of the interval frame number; and inputting the extracted image frames into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
Optionally, as an embodiment, the image segmentation module 401 is specifically configured to: acquiring the actual running speed of the conveyor belt and the actual length of the conveyor belt in the video range; calculating to obtain operation duration according to the actual length and the actual operation speed; or, dividing the pixel distance moved by the target point in any adjacent image frame of the video by the time interval of the adjacent image frame, and calculating to obtain the pixel speed of the conveyor belt; acquiring the pixel length of a conveyor belt in the video range; and calculating to obtain the running time according to the pixel length and the pixel speed.
Optionally, as an embodiment, the area determining module 402 is specifically configured to: and summing the ore areas corresponding to the ore segmentation results of the image frames to obtain the total ore area corresponding to the video.
Optionally, as an embodiment, the apparatus further includes a model fitting module, configured to:
and performing linear regression fitting according to the total ore area and the ore yield corresponding to the video in the historical time period to obtain the preset estimation model.
Optionally, as an embodiment, the preset estimation model is as follows:
M＝β*S+α
wherein M is the ore yield, S is the total area of the ore, alpha is the intercept, and beta is the slope.
The yield metering device based on image segmentation provided by the above embodiment can implement each process in the above embodiment of the yield metering method based on image segmentation, and is not described herein again in order to avoid repetition.
The embodiment of the invention also provides an image segmentation-based yield metering system, which comprises a camera device and a server; the camera device is used for collecting videos of ore conveyed by the conveyor belt and sending the videos to the server; the server is used for the yield measuring method based on image segmentation.
The embodiment of the present invention further provides a computer-readable storage medium, where a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, the computer program implements each process of the above-mentioned embodiment of the yield measurement method based on image segmentation, and can achieve the same technical effect, and in order to avoid repetition, the computer program is not described herein again. The computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk or an optical disk.
Of course, those skilled in the art will understand that all or part of the processes in the methods of the above embodiments may be implemented by instructing the control device to perform operations through a computer, and the programs may be stored in a computer-readable storage medium, and when executed, the programs may include the processes of the above method embodiments, where the storage medium may be a memory, a magnetic disk, an optical disk, and the like.
In this document, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. An image segmentation based throughput measurement method, the method comprising:
acquiring a video of ore conveyed by a conveyor belt in a target time period, and performing image segmentation on a plurality of image frames extracted from the video to obtain an ore segmentation result;
determining the total area of ores corresponding to the video according to the ore segmentation result;
and estimating to obtain the ore yield of the target time period according to the total ore area and a preset estimation model.
2. The method of claim 1, wherein the image segmentation of the plurality of image frames extracted from the video results in an ore segmentation result, comprising:
determining the running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range;
determining the number of interval frames according to the operation duration and the frame rate of the video;
extracting a plurality of image frames in the video at intervals of the interval frame number;
and inputting the extracted image frames into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
3. The method according to claim 2, wherein the determining the running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range specifically comprises:
acquiring the actual running speed of the conveyor belt and the actual length of the conveyor belt in the video range;
calculating to obtain operation duration according to the actual length and the actual operation speed;
alternatively, the first and second electrodes may be,
dividing the pixel distance moved by the target point in any adjacent image frame of the video by the time interval of the adjacent image frame, and calculating to obtain the pixel speed of the conveyor belt;
acquiring the pixel length of a conveyor belt in the video range;
and calculating to obtain the running time according to the pixel length and the pixel speed.
4. The method of claim 1, wherein determining the total ore area corresponding to the video according to the ore segmentation result comprises:
and summing the ore areas corresponding to the ore segmentation results of the image frames to obtain the total ore area corresponding to the video.
5. The method according to any one of claims 1-4, further comprising:
and performing linear regression fitting according to the total ore area and the ore yield corresponding to the video in the historical time period to obtain the preset estimation model.
6. The method of claim 5, wherein the preset estimation model is as follows:
M＝β*S+α
wherein M is the ore yield, S is the total area of the ore, alpha is the intercept, and beta is the slope.
7. An image segmentation based throughput metering apparatus, the apparatus comprising:
the image segmentation module is used for acquiring a video of ore conveyed by a conveyor belt in a target time period and performing image segmentation on a plurality of image frames extracted from the video to obtain an ore segmentation result;
the area determining module is used for determining the total area of the ores corresponding to the video according to the ore segmentation result;
and the yield estimation module is used for estimating and obtaining the ore yield of the target time period according to the total ore area and a preset estimation model.
8. The apparatus of claim 7, wherein the image segmentation module is specifically configured to:
determining the running time of any pixel point on the conveyor belt in the video from entering the video range to leaving the video range;
determining the number of interval frames according to the operation duration and the frame rate of the video;
extracting a plurality of image frames in the video at intervals of the interval frame number;
and inputting the extracted image frames into a pre-trained example segmentation algorithm model to obtain an ore segmentation result of each image frame.
9. The apparatus of claim 8, wherein the image segmentation module is specifically configured to:
acquiring the actual running speed of the conveyor belt and the actual length of the conveyor belt in the video range; calculating to obtain operation duration according to the actual length and the actual operation speed; or, dividing the pixel distance moved by the target point in any adjacent image frame of the video by the time interval of the adjacent image frame, and calculating to obtain the pixel speed of the conveyor belt; acquiring the pixel length of a conveyor belt in the video range; and calculating to obtain the running time according to the pixel length and the pixel speed.
10. A production metering system based on image segmentation is characterized by comprising a camera device and a server;
the camera device is used for collecting videos of ore conveyed by the conveyor belt and sending the videos to the server;
the server for performing the image segmentation based throughput method of any one of claims 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011640633.0A CN112700419A (en) | 2020-12-31 | 2020-12-31 | Yield measuring method, device and system based on image segmentation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011640633.0A CN112700419A (en) | 2020-12-31 | 2020-12-31 | Yield measuring method, device and system based on image segmentation |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112700419A true CN112700419A (en) | 2021-04-23 |
Family
ID=75513970
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011640633.0A Pending CN112700419A (en) | 2020-12-31 | 2020-12-31 | Yield measuring method, device and system based on image segmentation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112700419A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283344A (en) * | 2021-05-27 | 2021-08-20 | 中国矿业大学 | Mining conveying belt deviation detection method based on semantic segmentation network |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08218126A (en) * | 1995-02-14 | 1996-08-27 | Kawasaki Steel Corp | Operation of sintering together with lime production |
| CN101413822A (en) * | 2008-11-28 | 2009-04-22 | 西安大唐电信有限公司 | Mining remote production quantity monitoring dynamic weighing calibration method |
| CN107424143A (en) * | 2017-04-13 | 2017-12-01 | 中国矿业大学 | A kind of mine belt conveyor coal quantity measuring method based on binocular stereo vision depth perception |
| CN108510539A (en) * | 2018-05-09 | 2018-09-07 | 中国矿业大学(北京) | A kind of coal production monitoring method based on image analysis |
-
2020
- 2020-12-31 CN CN202011640633.0A patent/CN112700419A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH08218126A (en) * | 1995-02-14 | 1996-08-27 | Kawasaki Steel Corp | Operation of sintering together with lime production |
| CN101413822A (en) * | 2008-11-28 | 2009-04-22 | 西安大唐电信有限公司 | Mining remote production quantity monitoring dynamic weighing calibration method |
| CN107424143A (en) * | 2017-04-13 | 2017-12-01 | 中国矿业大学 | A kind of mine belt conveyor coal quantity measuring method based on binocular stereo vision depth perception |
| CN108510539A (en) * | 2018-05-09 | 2018-09-07 | 中国矿业大学(北京) | A kind of coal production monitoring method based on image analysis |
Non-Patent Citations (1)
| Title |
|---|
| 李萍: "基于机器视觉的散状物料动态计量系统研究", 《中国博士学位论文全文数据库（电子期刊）》, vol. 2018, no. 02 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113283344A (en) * | 2021-05-27 | 2021-08-20 | 中国矿业大学 | Mining conveying belt deviation detection method based on semantic segmentation network |
| CN113283344B (en) * | 2021-05-27 | 2024-03-12 | 中国矿业大学 | Mining conveyor belt deviation detection method based on semantic segmentation network |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108230352B (en) | Target object detection method and device and electronic equipment | |
| CN111242108B (en) | Belt transfer point coal blockage identification method based on target detection | |
| US20210342593A1 (en) | Method and apparatus for detecting target in video, computing device, and storage medium | |
| CN113723190A (en) | Multi-target tracking method for synchronous moving target | |
| Memarzadeh et al. | Real-time and automated recognition and 2D tracking of construction workers and equipment from site video streams | |
| Chen et al. | Multivariate video analysis and Gaussian process regression model based soft sensor for online estimation and prediction of nickel pellet size distributions | |
| CN111046832B (en) | Retrograde judgment method, device, equipment and storage medium based on image recognition | |
| CN112700419A (en) | Yield measuring method, device and system based on image segmentation | |
| CN109508621B (en) | Method, device and equipment for monitoring state of loading and unloading port and storage medium thereof | |
| CN106023249A (en) | Moving object detection method based on local binary similarity pattern | |
| CN109658128A (en) | A kind of shops based on yolo and centroid tracking enters shop rate statistical method | |
| CN110519566B (en) | Belt motion state monitoring method based on video processing | |
| Xu et al. | Measuring bulk material flow—Incorporating RFID and point cloud data processing | |
| CN115223105A (en) | Big data based risk information monitoring and analyzing method and system | |
| US11393091B2 (en) | Video image processing and motion detection | |
| CN113674322A (en) | Motion state detection method and related device | |
| Pratama | Edge computing implementation for action recognition systems | |
| Srilekha et al. | A novel approach for detection and tracking of vehicles using Kalman filter | |
| CN116245838A (en) | Monitoring method, monitoring device, equipment and medium for rock-soil exploration behaviors | |
| CN115526878A (en) | Coal flow detection method and device of belt conveyor | |
| CN104020467A (en) | Tracking and detecting device and method for invading foreign substances of power transmission line | |
| CN108696722B (en) | Target monitoring method, system and device and storage medium | |
| Ren et al. | An intelligent monitoring method of underground unmanned electric locomotive loading process based on deep learning method | |
| CN112767438A (en) | Multi-target tracking method combining space-time motion | |
| Kolditz et al. | Flexible aerodynamic part feeding using high-speed image processing |
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 |