CN112326011A - Water body steady state detection system and method based on artificial intelligence - Google Patents
Water body steady state detection system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN112326011A CN112326011A CN202011216083.XA CN202011216083A CN112326011A CN 112326011 A CN112326011 A CN 112326011A CN 202011216083 A CN202011216083 A CN 202011216083A CN 112326011 A CN112326011 A CN 112326011A
- Authority
- CN
- China
- Prior art keywords
- water body
- turbidity
- water
- workpiece
- degree
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01H—MEASUREMENT OF MECHANICAL VIBRATIONS OR ULTRASONIC, SONIC OR INFRASONIC WAVES
- G01H3/00—Measuring characteristics of vibrations by using a detector in a fluid
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/26—Measuring arrangements characterised by the use of optical techniques for measuring angles or tapers; for testing the alignment of axes
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/59—Transmissivity
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
Abstract
The invention relates to the technical field of artificial intelligence, in particular to a water body steady-state detection system and method based on artificial intelligence. The system comprises a workpiece placement detection module, a water inlet detection module and a water outlet detection module, wherein the workpiece placement detection module is used for acquiring the mass and the quantity of workpieces, the water inlet speed when the workpieces just contact the water surface, and the mean value of the motion trail length of the workpieces in the water body; the turbidity detection module is used for acquiring the change rate of the turbidity of the water body between the first turbidity of the water body and the second turbidity of the water body and acquiring the third turbidity of the water body; the shaking detection module is used for acquiring a third shaking degree of the water body and taking the turbidity change rate as a correction factor of the third shaking degree; and the water body steady state detection module is used for weighting and summing the third turbidity and the third shaking degree and judging the water body steady state of the workpiece. The water body steady state is quantized, and the attention degree of the water body turbidity and the water surface shaking degree is determined according to the magnitude of the two induction factors in the water body steady state, so that the water body steady state calculation result is more reasonable and accurate.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, in particular to a water body steady-state detection system and method based on artificial intelligence.
Background
When the air tightness of the workpiece is detected, a water immersion method is generally adopted for detection, the workpiece is placed into a glass cylinder filled with water, air pressure is applied to the cavity of the workpiece, and then if the workpiece has air leakage holes, the air bubbles can be detected, and the characteristics of the air leakage holes can be detected according to the characteristics of the air bubbles.
When a workpiece is placed in a water body, the stable state of the water body can be damaged, the water body can be shaken and becomes turbid, the bubble detection system cannot be started due to the instability of the water body, the water body needs to be kept stand, whether the water body reaches the stable state or not is artificially observed, then whether the bubble detection system is started or not is judged, manpower is wasted, and a good system for detecting the stable state of the water body does not exist at present.
At present, in the detection of the water turbidity, the loss intensity of laser in the water is often used to judge the water turbidity.
In an anti-bubble settling time control apparatus of the type disclosed in the grant publication No. CN109663517B, it is disclosed that the settling time of the anti-bubble is controlled by the degree of fluctuation of the liquid level, and the size of the diameter of the anti-bubble is inversely related to the frequency of the vibration source.
In a water level detection method based on image recognition with an authorized notice number of CN107506798B, a method for judging the water level by machine learning according to the characteristics of a water gauge picture marking a water level area to be detected is disclosed.
In practice, the inventors found that the above prior art has the following disadvantages:
when the steady state of the water body is analyzed, the turbidity of the water body is analyzed singly, the shaking degree of the water body is analyzed singly, and the accuracy of judging the steady state of the water body is low.
Disclosure of Invention
In order to solve the technical problems, the invention aims to provide a water body steady-state detection system and method based on artificial intelligence, and the adopted technical scheme is as follows:
in a first aspect, an embodiment of the present invention provides an artificial intelligence-based water steady-state detection system, which includes a workpiece placement detection module, a turbidity detection module, a shake detection module, and a water steady-state detection module.
And the workpiece placement detection module is used for controlling the mechanical arm for grabbing the workpiece to stretch into water and sequentially placing the workpiece, acquiring the momentum of the workpiece when the workpiece just contacts the water surface from the mass and the quantity of the workpiece and the water inlet speed when the workpiece just contacts the water surface, and acquiring the mean value of the motion trail length of the workpiece after the workpiece enters the water body.
The turbidity detection module is used for acquiring a water turbidity change rate of a first turbidity of the water body before the workpiece contacts the water surface and a second turbidity of the water body when the mechanical arm moves out of the water surface and a third turbidity after the mechanical arm moves away from the water surface.
And the shake detection module is used for acquiring the height and the inclination degree of the water surface after the workpiece is moved away from the water surface, acquiring a third shake degree of the water body after the workpiece is moved away from the water surface, and taking the turbidity change rate as a correction factor of the third shake degree.
And the water body steady state detection module is used for taking the momentum of the workpiece just contacting the water surface as the weight of the third shaking degree, taking the mean value of the motion track lengths of the workpiece after the workpiece enters the water and the number of the workpieces as the weight of the third turbidity, weighting and summing the third water body turbidity and the third water body shaking degree, and judging the water body steady state after the workpiece is moved out of the water surface.
In a second aspect, another embodiment of the present invention provides a method for detecting water body steady state based on artificial intelligence, including the steps of:
extending a mechanical arm for grabbing workpieces into water to sequentially place the workpieces; acquiring a first turbidity of a water body before a workpiece contacts with the water surface; when the workpiece just contacts the water surface, acquiring the momentum of the workpiece according to the mass, the quantity and the water inlet speed of the workpiece; after the workpiece enters the water body, acquiring the mean value of the motion trail length of the workpiece; when the mechanical arm just moves out of the water surface, acquiring a second turbidity of the water body and a turbidity change rate between the first turbidity and the second turbidity; and after the mechanical arm moves out of the water surface, acquiring a third turbidity and a third shaking degree of the water body, taking the turbidity change rate as a correction factor of the third shaking degree, taking the momentum of the workpiece as the weight of the third shaking degree, taking the mean value of the motion track length of the workpiece and the number of the mean value as the weight of the turbidity, carrying out weighted summation on the third turbidity and the third shaking degree, and judging the steady state of the water body.
The invention has at least the following beneficial effects:
the invention combines the turbidity of the water body and the shaking degree of the water surface to carry out comprehensive analysis and judge the steady state of the water body. In the judgment process, the influences of the number, the quality, the water inlet speed and the mean value of the motion path length of the workpieces entering water on the shaking degree of the water body and the turbidity degree of the water body and the relation between the turbidity change of the water body and the shaking degree of the water body are also considered, so that the water body stable state is quantized, and the water body stable state calculation result is more reasonable and accurate.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions and advantages of 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 some embodiments of the present invention, and other drawings can be obtained by those skilled in the art without creative efforts.
FIG. 1 is a block diagram of a system for detecting water steady state based on artificial intelligence according to an embodiment of the present invention;
FIG. 2 is a block diagram of a turbidity detection module of an artificial intelligence-based water steady-state detection system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a turbidity detection module of an artificial intelligence-based water steady-state detection system according to an embodiment of the present invention;
FIG. 4 is a block diagram of a sway detection module of an artificial intelligence-based water steady-state detection system according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a sway detection module of an artificial intelligence-based water steady-state detection system according to an embodiment of the present invention;
FIG. 6 is a block diagram of a water steady-state detection module of an artificial intelligence-based water steady-state detection system according to an embodiment of the present invention;
fig. 7 is a flowchart of a water steady-state detection method based on artificial intelligence according to another embodiment of the present invention.
Detailed Description
In order to further illustrate the technical means and effects of the present invention adopted to achieve the predetermined invention purpose, the following detailed description of the system and method for detecting water body steady state based on artificial intelligence according to the present invention with reference to the accompanying drawings and preferred embodiments, the specific implementation, structure, features and effects thereof are described in detail as follows. In the following description, different "one embodiment" or "another embodiment" refers to not necessarily the same embodiment. Furthermore, the particular features, structures, or characteristics may be combined in any suitable manner in one or more embodiments.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following describes a specific scheme of the artificial intelligence-based water body steady-state detection system and method provided by the invention in detail with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of an artificial intelligence based water steady-state detection system according to an embodiment of the present invention is shown. The system comprises an image acquisition module 100, a workpiece placement module 200, a turbidity detection module 300, a shake detection module 400 and a water body steady state detection module 500.
The image acquisition module is used for acquiring water body laser images, acquiring float images above the water surface and acquiring workpiece placement images.
Specifically, a first camera is used for collecting a water body laser beam image, the first camera is installed on one side of a cylinder wall and is opposite to the water body, a red laser light source is installed on the side surface of the cylinder wall on the side adjacent to the first camera and emits laser into the water body, and the light beam can present a red laser channel in the water body. The first camera captures images of all red laser channels in the captured image.
Utilize the cursory image of second camera collection water surface top, the second camera is installed in one side of cylinder body, and the surface of water is looked to the second camera side, places a cursory on the surface of water, and when the mechanical arm placed the in-process of work piece to the aquatic, can make the surface of water take place to fluctuate and rock, and the cursory can be along with the fluctuation of the surface of water and produce motion characteristic. According to the motion characteristics of the water surface buoy, the shaking degree of the water surface can be judged. The second camera is used for collecting images of the part of the float exposed to the water surface.
Wherein the first camera and the second camera keep synchronous acquisition.
In the embodiment of the invention, the mechanical arm acquires a plurality of workpieces at one time in one placing process, the workpieces enter water to be placed at the bottom of the cylinder one by one, and finally the mechanical arm moves out of the water surface and is closed. Wherein, a placing process refers to the steps of placing a workpiece in front of the water surface by the mechanical arm, placing the workpiece by the mechanical arm when the workpiece is just in contact with the water surface, placing the workpiece by the mechanical arm after the workpiece is placed in the water surface, moving the mechanical arm out of the water surface and moving the mechanical arm out of the water surface.
The whole process of placing the workpiece is monitored by using a third camera, the third camera is arranged on one side of the cylinder and is right opposite to the inside of the cylinder body, the visual field covers the water body liquid level and the placing area of the workpiece in the cylinder body, and the whole placing process of the workpiece is monitored. And at the moment, the third camera starts to record each frame of image of the workpiece in the water body after the workpiece enters the water body.
The turbidity detection module is used for acquiring a water turbidity change rate of a first turbidity of the water body before the workpiece contacts the water surface and a second turbidity of the water body when the mechanical arm moves out of the water surface and a third turbidity after the mechanical arm moves away from the water surface.
As shown in FIG. 2, the turbidity detection module 300 includes a turbidity analysis unit 310 and a turbidity determination unit 320.
The turbidity analysis unit is used for acquiring an image of a mask area of a laser beam passing through a water body, acquiring a minimum external rectangular frame of the mask area, and acquiring the height and width of the minimum external rectangular frame.
Specifically, the first camera is used for collecting RGB images of laser beams passing through the water body, and the laser beams are red, so that the gray value of the laser beams in a red channel is high, and the gray value is between green and blueThe grey value on the channel is low, so that the red channel I of the image is acquiredr. However, considering that an exposure area may appear in the image I, and the gray value of the exposure area is relatively high, since the gray values of the exposure area in the three channels of red, green and blue are equal, the area of the laser beam is obtained by the following method:
here, mask1 is a region of the laser beam. The mask1 region is binarized and an opening operation is performed to remove small isolated noise.
Since a laser beam may exhibit multiple beam trajectories in a body of water, mask1 may have multiple connected components that correspond to the multiple beam trajectories. The minimum circumscribed rectangular frames of each connected domain in mask1 are obtained, the minimum circumscribed rectangular frames represent the ROI area of each laser beam track, the average value of the heights of the minimum circumscribed rectangular frames is recorded as H, the H reflects the thickness of the laser beam, the sum of the widths of the circumscribed rectangular frames is recorded as W, and the W reflects the track length of the laser beam.
The turbidity judging unit is used for judging the turbidity degree of the water body according to a perception model of the water body turbidity, which is constructed by the positive correlation relationship between the width and the height of the mask area and the turbidity of the water body.
Specifically, since the width and length of the laser beam and the turbidity of the water body are in a positive correlation relationship, the determination model for constructing the width W and height H of the laser beam and the turbidity L of the water body is as follows:
L＝ln(W+1)+exp(H-H′) (1)
in the formula (1), H' represents the true width of the laser beam.
Referring to fig. 3, before the mechanical arm places the workpiece in water, the first camera acquires an image of the current RGB laser passing through the water body, and acquires and records the first turbidity L of the water body before the workpiece enters the water through the turbidity determination model0. When the mechanical arm just moves out of the water surface, the first camera collects the current RGB laser image, and the current RGB laser image is detected through turbidityThe measurement module is used for acquiring and recording the second turbidity L of the water body when the mechanical arm just moves out of the water surface1. The first turbidity L of the water body before the workpiece enters the water0And a second turbidity L of the water body when the mechanical arm moves out of the water surface1Calculating and recording the turbidity change rate delta L of the water body of the workpiece before and after placement0Comprises the following steps:
when the mechanical arm is completely moved out of the water surface, the first camera acquires images at a fixed frequency, and the water body turbidity L of the current frame is acquired through the turbidity judgment model when one frame of the image of the laser passing through the water body is acquired.
And the shake detection module is used for acquiring the height and the inclination degree of the water surface after the workpiece is moved away from the water surface, acquiring a third shake degree of the water body after the workpiece is moved away from the water surface, and taking the turbidity change rate as a correction factor of the third shake degree.
As shown in fig. 4, the shake detection module 400 includes a shake analysis unit 410 and a shake determination unit 420.
The device comprises a shake analysis unit, a first image acquisition unit, a second image acquisition unit, a first image acquisition unit and a second image acquisition unit, wherein the shake analysis unit is used for acquiring a plurality of frames of water body shake images after a workpiece is moved away from the water surface, acquiring the height and the inclination degree of the water surface in the k frame of image and the height and the inclination degree of the water surface in the k-1 frame of image; acquiring height change and inclination change of the water surface in the kth frame image and the kth-1 frame image; and acquiring the first turbidity and the second turbidity, and acquiring a turbidity change rate serving as a correction factor of the third shaking degree.
The specific process of the shake detection module is shown in fig. 5. In the embodiment of the invention, the float uses the suspension mark on the fishing gear, one section of the float is immersed in water, and the other section of the float is exposed out of the water surface, when the water body is static, the length of the part exposed out of the water surface is fixed, when the water surface fluctuates, the part exposed out of the water surface can be changed by length and inclination angle, and the length change and the inclination angle change can reflect the shaking degree of the water body.
The second camera is used for acquiring a floating RGB image above the water surface, a DNN network is used for acquiring a semantic area mask2 exposed out of the water surface, the DNN network is a mask-RCNN network, a DeepLabv3 network and other DNN networks, and the DeepLabv3 network is used in the embodiment of the invention. Since the method for acquiring the semantic area mask2 floating out of the water surface is conventional, no specific description is given in the embodiment of the present invention.
After the semantic area mask2 of the float exposed out of the water surface is obtained, binarization processing is carried out on the semantic area mask2, the connected domain of the mask2 of the float exposed out of the water surface is obtained, the minimum circumscribed rectangular frame of the connected domain of the mask2 is obtained, the height of the minimum circumscribed rectangular frame is obtained, and the sine value of the angle of the height direction axis of the rectangular frame deviating from the vertical direction is obtained.
Before the workpiece is placed in the water, the second camera and the first camera perform synchronous image acquisition to acquire RGB (red, green and blue) images of the floating surface, the RGB images above the floating surface are analyzed by the shake detection module at the moment, and the length H of the floating surface of the workpiece before entering the water is acquired0. When the mechanical arm just moves out of the water surface, the second camera and the first camera perform synchronous image acquisition, the RGB image of the float exposed out of the water surface at the moment is acquired as the initial frame image of the third shaking degree of the water body, and the length H of the float exposed out of the water surface when the mechanical arm just moves out of the water surface can be acquired1And the inclination degree A of the upper end axis of the float deviating from the vertical direction1。
After the mechanical arm is completely moved out of the water surface, the second camera and the first camera synchronously acquire images at a fixed frequency, and the length H of the water surface exposed by the float in the current frame image can be acquired every time one frame of image is acquiredk+1The inclination degree A deviating from the vertical direction of the axis of the upper end of the floatk+1. When the frame number k is 0, it represents that the initial frame image of the third shaking degree of the water body is collected, and then the value of k is increased by 1 every time one frame image is collected.
Length H of the current frame image floating out of the water surfacek+1The inclination degree A deviating from the vertical direction of the axis of the upper end of the floatk+1Length H of the floating water surface in the previous frame imagekThe inclination degree A deviating from the vertical direction of the axis of the upper end of the floatkComparing, and judging the length change delta H of the front and back frames of imageskChange delta A of inclination degree deviating from vertical direction of upper end axis of floatkWherein the length change Δ Hk＝Hk+1-HkK is not less than 0, and the change of inclination degree is delta Ak＝Ak+1-Ak，k≥0。ΔHk、ΔAkThe larger the water body is, the larger the shaking degree of the water body is reflected.
The shaking degree of the water body can be represented by the motion characteristics of the float and is also related to the turbidity change of the water body, and when the workpiece is placed into the water to cause the water body to shake, the workpiece also causes deposited impurities to float so as to cause the turbidity of the water body to change, so that the turbidity change of the water body reflects the shaking of the water body to a certain degree. The turbidity change Delta L of the water body of the workpiece before and after the workpiece is placed and stored by the turbidity detection module0Changing the turbidity of the water body by Delta L0As a correction factor for the water body sloshing degree. When the mechanical arm is completely moved out of the water surface, the turbidity change delta L of the obtained water body0Then the turbidity change Delta L of the water body is used as a fixed value0The larger the water is, the greater the degree of water sloshing is reflected.
And the shaking judgment unit is used for judging the third shaking degree of the water body in the image according to a third shaking degree perception model established by the positive correlation relationship between the water surface height change and the inclination degree change and the shaking degree of the water body.
After the mechanical arm moves out of the water surface, the float height change Delta H in the collected images of the adjacent frameskChange in degree of inclination Δ AkAnd the change Delta L of the turbidity of the water body before and after the workpiece is placed0And forming a positive correlation relation with the shaking degree of the water surface, and constructing a perception model of the shaking degree R of the water surface as follows:
in the formula (2), H0The length of the workpiece which floats out of the water surface before entering the water is adopted.
The third shaking degree of the water body after the mechanical arm moves out of the water surface can be obtained through the sensing model of the shaking degree R of the water surface.
And the workpiece placement detection module is used for controlling the mechanical arm for grabbing the workpieces to extend into the water and sequentially placing the workpieces, acquiring the momentum of the workpieces when the workpieces are just in contact with the water by the mass and the quantity of the workpieces and the water inlet speed when the workpieces just in contact with the water, and acquiring the mean value of the motion trail lengths of the workpieces after the workpieces enter the water.
As shown in fig. 6, the water steady-state detection module 500 includes a determination unit 510.
The judgment unit is used for obtaining the water body stable state by a judgment model of the water body stable state, which is constructed by the positive correlation relationship between the third turbidity and the third shaking degree and the water body stable state.
Specifically, in the embodiment of the present invention, the boundary frame of the workpiece and the model of the workpiece are obtained through the yolov3DNN network, the workpieces of different models correspond to different qualities, the database is searched according to the model of the workpiece to obtain the quality M corresponding to the workpiece, and the yolov3DNN network can also output the boundary frame where the liquid level is located.
The third camera is used for collecting the image of the workpiece after entering the water body, and the mechanical arm is used for placing the workpiece, so that the placing speed, the water entering time and the water exiting time of the mechanical arm in the placing process are very easy to obtain. The water inlet speed V of the workpiece just entering the water surface, namely the placing speed V of the mechanical arm is recorded and stored, and in the process of placing the workpiece by the mechanical arm, the number of workpiece boundary frames appearing in the camera view field is counted, so that the number N of the workpieces placed by the mechanical arm can be obtained. What the income water speed V of work piece, quantity N of work piece and the quality M of work piece characterize is the momentum that the work piece was gone into the water, is the important inducement that the water rocked, consequently can acquire that water rocks induction factor beta and is: β NMV (3).
Mean value of motion trail length of workpiece
The motion path of the workpiece in the water body is represented, the longer the path is, the greater the influence degree on impurities in the water body is, and the water body isOne of the major inducers of volume opacification. The more the number N of the workpieces, the more the impurities are introduced, so the number of the workpieces is one of the causes of the water body becoming turbid. Therefore, the induction factor alpha of the water body turbidity can be obtained as follows:
specifically, in the embodiment of the invention, the mean value of the motion trail lengths of the workpieces is obtained
Comprises the following steps:
when the workpiece enters the water, the camera starts to record the position of the center of the bounding box of the workpiece on each frame of image, wherein the position represents the position of the workpiece in each frame of image: p ═ Pn，1，Pn，2，......，Pn，k,... wherein Pn，kIndicated as the position of the nth workpiece in the image of the k frame. The moving distance of the nth workpiece between two adjacent frames of images is obtained according to the position of the central point of the workpiece boundary frame on each frame of image
Wherein, Δ Dn，kIndicated as the distance of movement of the workpiece in the image of the nth workpiece in the k frame and k-1 frame. The motion track total length D of the nth workpiece is obtained by accumulative additionnComprises the following steps: dn＝∑k∈PDn，k. The mean value of the lengths of the motion tracks of the N workpieces can be obtained
Comprises the following steps:
and the water body steady state detection module is used for taking the momentum of the workpiece just contacting the water surface as the weight of the third shaking degree, taking the mean value of the motion track lengths of the workpiece after the workpiece enters the water and the number of the workpieces as the weight of the third turbidity, weighting and summing the third water body turbidity and the third water body shaking degree, and judging the water body steady state after the workpiece is moved out of the water surface.
Specifically, the water body shaking induction factor obtained after entering water is used as the weight of the water body shaking degree, and the water body turbidity induction factor obtained after entering water is used as the weight of the water body turbidity, and weighted summation is carried out, and by combining the formula (1), the formula (2), the formula (3) and the formula (4), the model for constructing the steady state S of the water body is as follows:
S＝α(L-L0)+βR。
when the mechanical arm moves out of the water surface, images acquired by the first camera and the second camera are analyzed, the water body stable state after the mechanical arm moves out of the water surface is analyzed through the water body stable state model, when the water body stable state is smaller than a preset threshold th, th is a super parameter, the water body turbidity and the water body shaking degree are reduced to a certain value, the water body reaches a stable state at the moment, and bubble detection can be carried out.
In conclusion, the method establishes the turbidity model of the water body by detecting the width and the length of the track of the laser in the turbid water body; the water surface shaking degree is obtained by detecting the length of the leaked water surface of the water surface float and the change characteristics of the inclination angle of the float, and a correction factor is introduced through the change of the turbidity of the water body before and after the workpiece is placed, and is used for correcting the shaking degree of the water body. The method comprises the steps of obtaining a water turbidity induction factor and a water body shaking induction factor by obtaining the number, the quality, the water inlet speed and the movement path length of devices of a mechanical arm in the placing process, and carrying out weighted operation on the turbidity and the water surface shaking degree of a water body by utilizing the two induction factors to finally obtain the water body stable state. The invention not only quantifies the water body steady state, but also determines the attention degree of the water body turbidity and the water surface shaking degree according to the magnitude of the two induction factors, so that the water body steady state calculation result is more reasonable and accurate.
Based on the same inventive concept as the system, another embodiment of the invention also provides a water body steady-state detection method based on artificial intelligence.
Referring to fig. 7, a flowchart of a method for detecting water body steady state based on artificial intelligence according to another embodiment of the present invention is shown. The method comprises the following steps:
and step S1, extending the mechanical arm for grabbing the workpieces into the water to sequentially place the workpieces.
Specifically, the robot arm is used for placing the workpieces, in the embodiment of the invention, the robot arm acquires a plurality of workpieces at one time in one placing process, the workpieces enter water to be placed at the bottom of the cylinder one by one, and finally the robot arm moves out of the water surface and is closed. Wherein, a placing process refers to the steps of placing a workpiece in front of the water surface by the mechanical arm, placing the workpiece by the mechanical arm when the workpiece is just in contact with the water surface, placing the workpiece by the mechanical arm after the workpiece is placed in the water surface, moving the mechanical arm out of the water surface and moving the mechanical arm out of the water surface.
Step S2, before the workpiece contacts the water surface, a first turbidity of the water body is obtained.
Collecting the RGB image of the laser beam passing through the water body before the workpiece contacts the water surface, wherein the gray value of the laser beam in the red channel is higher and the gray value of the laser beam in the green and blue channels is lower due to the fact that the laser beam is red, so that the red channel I of the image is obtainedr. However, considering that an exposure area may appear in the image I, and the gray value of the exposure area is relatively high, since the gray values of the exposure area in the three channels of red, green and blue are equal, the area of the laser beam is obtained by the following method:
here, mask1 is a region of the laser beam. The mask1 region is binarized and an opening operation is performed to remove small isolated noise.
Since a laser beam may exhibit multiple beam trajectories in a body of water, mask1 may have multiple connected components that correspond to the multiple beam trajectories. The minimum bounding rectangle of each connected domain in mask1 is obtained, the average value of the heights of the minimum bounding rectangles is recorded as H, and the sum of the widths of the bounding rectangles is recorded as W.
Because the width and the length of the laser form a positive correlation with the turbidity of the water body, a judgment model for constructing the width W and the height H of the laser beam and the turbidity L of the water body is as follows:
L＝ln(W+1)+exp(H-H′) (1)
in the formula (1), H' represents the true width of the laser beam.
Before the workpiece contacts the water surface, the first turbidity L of the water body at the moment is obtained0。
And step S3, when the workpiece just contacts the water surface, acquiring the momentum of the workpiece according to the mass, the quantity and the water inlet speed of the workpiece.
Because the mechanical arm is used for placing the workpiece, the placing speed, the water inlet time and the water outlet time of the mechanical arm in the placing process are very easy to obtain. The water inlet speed V of the workpiece just entering the water surface, namely the placing speed V of the mechanical arm is recorded and stored, and in the process of placing the workpiece by the mechanical arm, the number of workpiece boundary frames appearing in the camera view field is counted, so that the number N of the workpieces placed by the mechanical arm can be obtained. The workpiece entry velocity V, the number of workpieces N, and the mass of the workpieces M are indicative of the momentum of the workpiece entry NMV.
And step S4, acquiring the mean value of the motion trail length of the workpiece after the workpiece enters the water body.
Specifically, when the workpiece enters the water, the camera starts to record the position of the center of the workpiece bounding box on each frame of image, which represents the position of the workpiece in each frame of image: p ═ Pn，1，Pn，2，......，Pn，k,... wherein Pn，kIndicated as the position of the nth workpiece in the image of the k frame. The moving distance of the nth workpiece between two adjacent frames of images is obtained according to the position of the central point of the workpiece boundary frame on each frame of image
Wherein, Δ Dn，kIndicated as the distance of movement of the workpiece in the image of the nth workpiece in the k frame and k-1 frame. The motion track total length D of the nth workpiece is obtained by accumulative additionnComprises the following steps: dn＝∑k∈PDn，k. The mean value of the lengths of the motion tracks of the N workpieces can be obtained
Comprises the following steps:
and step S5, when the mechanical arm just moves out of the water, acquiring a second turbidity of the water, and acquiring a turbidity change rate between the first turbidity and the second turbidity.
When the mechanical arm just moves out of the water surface, collecting RGB images of the laser beam passing through the water body at the moment, and acquiring a second turbidity L of the water body when the mechanical arm just moves out of the water surface through the turbidity detection process in the step S21. The turbidity change rate Delta L of the water body caused by the placed workpiece can be obtained according to the first turbidity and the second turbidity0Comprises the following steps:
and step S6, after the mechanical arm moves out of the water surface, acquiring a third turbidity and a third shaking degree of the water body, and correcting the third shaking degree by taking the turbidity change rate as a correction factor of the third shaking degree.
After the mechanical arm moves out of the water surface, obtaining a plurality of frames of RGB images of the laser beam passing through the water body, and obtaining a third turbidity L of the water body of the current frame through the turbidity detection process or the turbidity detection process of the step S2.
Acquiring a plurality of frames of water body shaking images, and recording the height H of the water surface in the current k frame of imagekAnd degree of inclination AkAnd the height H of the water surface in the k-1 image of the frame preceding itk-1And degree of inclination Ak-1Obtaining the height change delta H of the water surface in the k frame image and the k-1 imagekAnd change of inclination degree Delta Ak. Wherein the height change Δ HkComprises the following steps: Δ Hk＝Hk-Hk-1K is not less than 2, and the change of the inclination degree is delta AkComprises the following steps: delta Ak＝Ak-Ak-1，k≥2。
The turbidity change rate Δ L obtained in step S50As a correction factor for the third degree of sloshing; the turbidity change rate DeltaL0Comprises the following steps:
and then changes by the height of the water surfacekAnd change of inclination degree Delta AkAnd forming a positive correlation relation with the shaking degree of the water body, and constructing a third shaking degree R perception model as follows:
wherein H0Representing the height of the water surface at rest;
and judging the third shaking degree R of the water body in the kth frame of image by the water body shaking degree perception model.
And step S7, acquiring momentum of the workpiece as the weight of the third shaking degree, acquiring the mean value of the motion track length of the workpiece and the number of the mean value as the weight of the turbidity, performing weighted summation on the third turbidity and the third shaking degree, and judging the steady state of the water body.
Because the momentum of the workpiece entering water is an important inducement for water body shaking, the water body shaking induction factor beta can be obtained as follows: β NMV (3).
Mean value of motion trail length of workpiece
The moving distance of the workpiece in the water body is represented, and the influence degree of impurities in the water body is larger as the moving distance is longer, so that the moving distance is one of main induction factors for the turbidity of the water body. The more the number N of the workpieces, the more the impurities are introduced, so the number of the workpieces is one of the causes of the water body becoming turbid. Therefore, the induction factor alpha of the water body turbidity can be obtained as follows:
taking the obtained water body shaking induction factor after entering water as the weight of the water body shaking degree and the obtained water body turbidity induction factor after entering water as the weight of the water body turbidity, carrying out weighted summation, and combining the formula (1), the formula (2), the formula (3) and the formula (4), wherein the model for constructing the steady state S of the water body is as follows:
S＝α(L-L0)+βR。
it should be noted that: the precedence order of the above embodiments of the present invention is only for description, and does not represent the merits of the embodiments. And specific embodiments thereof have been described above. Other embodiments are within the scope of the following claims. In some cases, the actions or steps recited in the claims may be performed in a different order than in the embodiments and still achieve desirable results. In addition, the processes depicted in the accompanying figures do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing may also be possible or may be advantageous.
The embodiments in the present specification are described in a progressive manner, and the same and similar parts among the embodiments are referred to each other, and each embodiment focuses on the differences from the other embodiments.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (10)
1. The water body steady state detection system based on artificial intelligence is characterized by comprising a workpiece placement detection module, a turbidity detection module, a shaking detection module and a water body steady state detection module;
the workpiece placement detection module is used for controlling a mechanical arm for grabbing workpieces to stretch into water and sequentially placing the workpieces, acquiring the momentum of the workpieces when the workpieces just contact the water surface from the mass and the quantity of the workpieces and the water inlet speed when the workpieces just contact the water surface, and acquiring the mean value of the motion trail lengths of the workpieces after the workpieces enter the water body;
the turbidity detection module is used for acquiring a change rate of the turbidity of the water body between a first turbidity of the water body before the workpiece contacts the water surface and a second turbidity of the water body when the mechanical arm moves out of the water surface, and acquiring a third turbidity of the water body after the mechanical arm moves out of the water surface;
the shaking detection module is used for acquiring the change of the height and the inclination degree of the water surface after the workpiece is removed from the water surface, acquiring a third shaking degree of the water body after the workpiece is removed from the water surface, and taking the turbidity change rate as a correction factor of the third shaking degree;
and the water body steady state detection module is used for taking the momentum of the workpiece just contacting the water surface as the weight of the third shaking degree, taking the mean value of the motion track length of the workpiece and the number of the workpieces as the weight of the third turbidity, weighting and summing the third water body turbidity and the third water body shaking degree, and judging the water body steady state after the workpiece is moved out of the water surface.
2. The artificial intelligence based water body steady-state detection system of claim 1, wherein the turbidity detection module further comprises a turbidity analysis unit and a turbidity determination unit:
the turbidity analysis unit is used for acquiring an image of a mask area of a laser beam passing through a water body, acquiring a minimum circumscribed rectangular frame of the mask area, and acquiring the height H and the width W of the minimum circumscribed rectangular frame;
the turbidity judging unit is used for judging the turbidity degree of the water body according to a perception model of the water body turbidity L constructed by the positive correlation relationship between the width and the height of the mask area and the turbidity of the water body; the perception model of the water body turbidity L is as follows: l ═ ln (W +1) + exp (H-H '), where H' is the true width of the laser beam.
3. The artificial intelligence based water steady-state detection system of claim 1, wherein the sway detection module further comprises a sway analysis unit and a sway determination unit:
the shaking analysis unit is used for acquiring a plurality of frames of water body shaking images after the workpiece is moved away from the water surface; acquiring the height H of the water surface in the k frame imagekAnd degree of inclination AkAnd the height H of the water surface in the k-1 th frame imagek-1And degree of inclination Ak-1(ii) a Acquiring the height change delta H of the water surface in the k frame image and the k-1 frame imagekAnd change of inclination degree Delta Ak(ii) a Obtaining the first turbidity L0And the second turbidity L1Obtaining the turbidity change rate DeltaL0As a correction factor for the third degree of sloshing; the turbidity change rate DeltaL0Comprises the following steps:
the shake determination unit is used for determining the height change Delta H of the water surfacekAnd change of inclination degree Delta AkJudging a third shaking degree of the water body in the kth frame image by using a third shaking degree R perception model which is constructed in positive correlation with the shaking degree of the water body;
the third shaking degree R perception model is as follows:
wherein H0Indicating the height of the water surface at rest.
4. The artificial intelligence based water body steady-state detection system according to claim 1, wherein the weight α of the third turbidity degree is:
whereinN is the number of the workpieces,
the mean value of the motion trail of the workpiece in the water body is obtained.
5. The artificial intelligence based water body steady state detection system of claim 1, wherein the water body steady state detection module further comprises a determination unit:
the judging unit is used for judging the steady state of the water body by a judging model of the water body steady state S, which is constructed by the positive correlation relationship between the third turbidity L and the third shaking degree R and the water body steady state S;
the judgment model of the water body steady state S is as follows:
wherein, L is0A first turbidity of the water body; the above-mentioned
Is the weight of the third turbidity L; NMV, which is the weight of the third degree of sloshing R; n is the number of the workpieces; the above-mentioned
The mean value of the motion trail of the workpiece in the water body is obtained; m is the mass of the workpiece; and V is the water inlet speed of the workpiece.
6. The water body steady state detection method based on artificial intelligence is characterized by comprising the following steps:
extending a mechanical arm for grabbing workpieces into water to sequentially place the workpieces;
acquiring a first turbidity of the water body before the workpiece contacts the water surface;
when the workpiece just contacts the water surface, acquiring the momentum of the workpiece according to the mass, the quantity and the water inlet speed of the workpiece;
after the workpiece enters a water body, acquiring the mean value of the motion trail length of the workpiece;
when the mechanical arm moves out of the water surface, acquiring a second turbidity of the water body and a turbidity change rate between the first turbidity and the second turbidity;
after the mechanical arm moves out of the water surface, acquiring a third turbidity and a third shaking degree of the water body, and correcting the third shaking degree by taking the turbidity change rate as a correction factor of the third shaking degree;
carrying out weighted summation on the third turbidity and the third shaking degree, and judging the steady state of the water body; and the weight of the third shaking degree is the momentum of the workpiece, and the weight of the third turbidity degree is the mean value of the length of the motion trail of the workpiece and the number of the motion trail of the workpiece.
7. The artificial intelligence based water body steady-state detection method according to claim 6, wherein the step of obtaining the first turbidity of the water body, the step of obtaining the second turbidity of the water body or the step of obtaining the third turbidity of the water body comprises:
acquiring a mask area image of a laser beam passing through the current water body;
acquiring a minimum external rectangular frame of the mask region image, and acquiring the height H and the width W of the minimum external rectangular frame;
judging the turbidity of the current water body according to a perception model of the water body turbidity L constructed by the positive correlation relationship between the width and the height of the mask region image and the turbidity of the water body; the perception model of the water body turbidity L is as follows: l ═ ln (W +1) + exp (H-H '), where H' is the true width of the laser beam.
8. The artificial intelligence based water body steady-state detection method according to claim 6, wherein the step of obtaining the third shaking degree of the water body comprises:
after the workpiece is moved away from the water surface, acquiring a plurality of frames of water body shaking images;
acquiring the height H of the water surface in the k frame imagekAnd degree of inclination AkAnd the height H of the water surface in the k-1 th frame imagek-1And degree of inclination Ak-1；
Acquiring the height change delta H of the water surface in the k frame image and the k-1 frame imagekAnd change of inclination degree Delta Ak；
Obtaining the first turbidity L0And the second turbidity L1Obtaining the turbidity change rate DeltaL0As a correction factor for the third degree of sloshing; the turbidity change rate DeltaL0Comprises the following steps:
according to height change Δ H from the water surfacekAnd change of inclination degree Delta AkThe third shaking degree R perception model is constructed and used for judging the third shaking degree of the water body in the k frame image; the perception model of the third shaking degree R is as follows:
wherein H0Indicating the height of the water surface at rest.
9. The artificial intelligence based water body steady-state detection method according to claim 6, wherein the weight α of the third turbidity degree is:
wherein N is the number of the workpieces,
the mean value of the motion trail of the workpiece in the water body is obtained.
10. The artificial intelligence based water body steady-state detection method according to claim 6, wherein the step of judging the steady state of the water body is:
acquiring a first turbidity L of a water body before the workpiece just contacts the water surface0；
Judging that the water body reaches a steady state according to a judgment model of the water body steady state S, wherein the judgment model of the water body steady state S is constructed by the positive correlation between the third turbidity L and the third shaking degree R and the water body steady state S; the judgment model is as follows:
wherein, the
Is the weight of the third turbidity L; NMV, which is the weight of the third degree of sloshing R; n is the number of the workpieces; the above-mentioned
The mean value of the motion trail of the workpiece in the water body is obtained; m is the mass of the workpiece; and V is the water inlet speed of the workpiece.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011216083.XA CN112326011A (en) | 2020-11-04 | 2020-11-04 | Water body steady state detection system and method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011216083.XA CN112326011A (en) | 2020-11-04 | 2020-11-04 | Water body steady state detection system and method based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112326011A true CN112326011A (en) | 2021-02-05 |
Family
ID=74324629
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011216083.XA Withdrawn CN112326011A (en) | 2020-11-04 | 2020-11-04 | Water body steady state detection system and method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112326011A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114062320A (en) * | 2021-11-17 | 2022-02-18 | 北京市自来水集团有限责任公司技术研究院 | Turbidity determination method and device of desk-top turbidity meter |
-
2020
- 2020-11-04 CN CN202011216083.XA patent/CN112326011A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114062320A (en) * | 2021-11-17 | 2022-02-18 | 北京市自来水集团有限责任公司技术研究院 | Turbidity determination method and device of desk-top turbidity meter |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109460753B (en) | Method for detecting floating object on water | |
| JP6697561B2 (en) | Method and system for determining the mass of raw material on a conveyor | |
| CN109460754A (en) | A kind of water surface foreign matter detecting method, device, equipment and storage medium | |
| CN112326011A (en) | Water body steady state detection system and method based on artificial intelligence | |
| CN103196789A (en) | Diesel vehicle tail gas smoke intensity detecting method | |
| CN103096117B (en) | Video noise detection method and device | |
| CN112465895A (en) | Bubble volume calculation method in air tightness detection based on computer vision | |
| CN111626225B (en) | License plate recognition method, device and equipment for station vehicle and storage medium | |
| CN110880184A (en) | Method and device for carrying out automatic camera inspection based on optical flow field | |
| CN115205363A (en) | Conveyor belt real-time judgment detection method and system based on improved ResNet network | |
| CN109784145B (en) | Target detection method based on depth map and storage medium | |
| CN110992415A (en) | Water surface floater pollution evaluation system and method based on big data | |
| Denissenko et al. | How waves affect the distribution of particles that float on a liquid surface | |
| CN112580634A (en) | Air tightness detection light source adjusting method and system based on computer vision | |
| CN109028238B (en) | A kind of intelligent shift method with vision-based detection module kitchen ventilator | |
| CN112329620A (en) | Water change amount prediction system of tightness detection device based on visual perception | |
| KR101519975B1 (en) | Water quality measuting apparatus and method using sensing the number of live fish in water | |
| CN112330691A (en) | Water body steady-state time prediction method based on visual perception | |
| CN114862786A (en) | Retinex image enhancement and Ostu threshold segmentation based isolated zone detection method and system | |
| Dinda et al. | Gas bubble digital data generation by image analysis using reduced-scale water modeling of a slab continuous caster mold | |
| CN113284162A (en) | Water body steady state detection method and system in air tightness detection based on artificial intelligence | |
| CN112801072A (en) | Elevator non-flat-layer door opening fault recognition device and method based on computer vision | |
| EP1595134A1 (en) | Method and instrument for measuring surface tension | |
| CN110533698A (en) | A kind of foundation pit construction stake detection control method of view-based access control model detection | |
| CN116593489A (en) | River water quality model prediction method and system based on big data |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210205 |
|
| WW01 | Invention patent application withdrawn after publication |