CN113751365A - Nuclear waste detecting and sorting system and method based on double optical cameras - Google Patents
Nuclear waste detecting and sorting system and method based on double optical cameras Download PDFInfo
- Publication number
- CN113751365A CN113751365A CN202111144641.0A CN202111144641A CN113751365A CN 113751365 A CN113751365 A CN 113751365A CN 202111144641 A CN202111144641 A CN 202111144641A CN 113751365 A CN113751365 A CN 113751365A
- Authority
- CN
- China
- Prior art keywords
- nuclear waste
- image
- thermal radiation
- radiation image
- pixel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/3416—Sorting according to other particular properties according to radiation transmissivity, e.g. for light, x-rays, particle radiation
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
- B07C5/362—Separating or distributor mechanisms
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/38—Collecting or arranging articles in groups
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0054—Sorting of waste or refuse
Abstract
The invention discloses a nuclear waste detecting and sorting system and method based on a dual-optical camera, wherein the system comprises a thermal infrared imager, a gripping device, a mechanical arm, a combustible classification container, a conveyor belt, a zero-crossing point detection switch and a non-combustible classification container; the infrared thermal imaging instrument is mainly used for sensing the change of the gray level image pixel value of a thermal radiation image corresponding to the change of temperature. The depth camera is mainly used for obtaining more texture information and depth information of the substance. Because the radioactivity of the nuclear waste is large, even if people wear protective clothing, the people can still cause unthinkable injury to the human body in places with strong radioactivity for a long time, the system can automatically distinguish different nuclear waste substances, and people can know which class the nuclear waste belongs to only by identifying the nuclear waste substances through a camera, so that zero contact between the people and the radioactive substances is realized in the classification process.
Description
Technical Field
The invention belongs to the field of nuclear robots, and particularly relates to a nuclear waste detecting and sorting system and method based on a dual-optical camera.
Background
The out-of-reactor nuclear instrumentation system is a safety-level device that continuously monitors reactor power, power level changes, and power distribution by measuring the neutron fluence rate leaking from the reactor core, and is an important input parameter for reactor protection systems and five major control systems of power plants. The out-of-core nuclear measurement detector is an eye of an out-of-core nuclear measurement instrument system, is arranged at the periphery of the reactor pressure vessel and directly detects the neutron fluence rate level of reactor core leakage.
At present, the nuclear waste is mainly sorted by adopting an artificial method. Because of the high radioactivity of nuclear waste, and the manual sorting requires a certain time to sort. Even if the sorting personnel wear the whole set of radiation protection suit, the radiation protection suit can be irradiated by a large dose for a long time when being in contact with nuclear waste, and the radiation protection suit can cause unthinkable human bodies.
Disclosure of Invention
Aiming at the defects in the prior art, the nuclear waste detecting and sorting system and method based on the dual-optical camera provided by the invention solve the problem of sorting nuclear waste with strong radioactivity.
In order to achieve the purpose of the invention, the invention adopts the technical scheme that: a nuclear waste detecting and sorting system based on a dual-optical camera comprises a thermal infrared imager, a grabbing device, a mechanical arm, a combustible classifying container, a conveyor belt, a zero-crossing point detecting switch and a non-combustible classifying container;
the zero crossing point detection switch is arranged in the middle of a conveyor provided with the conveyor belt, the thermal infrared imager is arranged at a position 1 m above the nuclear waste, the mechanical arm is arranged at a position 20 cm away from the nuclear waste through the base and connected with the grabbing device, the combustible classification container is arranged on the right side of the mechanical arm, and the incombustible classification container is arranged on the side edge of the conveyor belt.
Further: a depth camera is arranged in the grabbing device and is arranged in the positive center of the grabbing device.
The beneficial effects of the above further scheme are: and the depth camera acquires texture information and depth information of the grabbed object.
A nuclear waste detection and sorting method based on a dual-optical camera comprises the following steps:
s1, acquiring a first thermal radiation image of the nuclear waste through a thermal infrared imager, denoising the first thermal radiation image to obtain a first gray scale image, and generating a first difference average gradient image according to the first gray scale image;
s2, fitting a linear relation between pixel values of pixel points of the first difference average gradient image and the temperature change rate through a least square method, and obtaining a first rough region set according to a fitting result;
s3, processing the first rough region set by a clustering method to obtain a first region;
s4, acquiring a second thermal radiation image of the nuclear waste through a thermal infrared imager, obtaining a second difference average gradient image according to the second thermal radiation image, and obtaining a second area through a least square method and a clustering method;
s5, comparing the second area with the first area and taking the same area to obtain a segmentation area;
and S6, inputting the segmentation areas into the grabbing model based on the grabbing model trained by the convolutional neural network, and controlling a grabbing device to grab the nuclear waste to finish the sorting of the nuclear waste.
Further: in the step S1, the first thermal radiation image of the nuclear waste includes a thermal radiation image at a time t and a thermal radiation image at a time t +5, and the first gray scale map includes a gray scale map at the time t and a gray scale map at the time t + 5;
the step S1 includes the following sub-steps:
s11, acquiring a thermal radiation image at the time t and a thermal radiation image at the time t +5 by using a thermal infrared imager, and denoising the thermal radiation images and converting the thermal radiation images into a grayscale image at the time t and a grayscale image at the time t + 5;
and S12, subtracting the gray scale image at the time t from the gray scale image at the time t +5, and dividing the difference result by 5 to obtain a first difference average gradient image.
The beneficial effects of the above further scheme are: the thermal infrared imager can sense the change of the gray level image pixel value of the thermal radiation image corresponding to the change of the temperature, and can obtain the linear relation between the pixel value of the pixel point and the change rate of the temperature.
Further: the step S2 includes the following sub-steps:
s21, reading the temperature value of each pixel point in the first difference average gradient image through the thermal infrared imager, and obtaining the temperature change rate of each pixel point according to the interval time for collecting the first thermal radiation image;
s22, reading the pixel value of each pixel point in the first difference average gradient image;
s23, fitting the relationship between the pixel value of each pixel point in the first difference average gradient image and the temperature change rate through a least square method to obtain the slope a and the offset b corresponding to each pixel point, and further obtain the pixel value average change rate of each fitted temperature change;
the expression of the least square method is specifically as follows:
in the formula, f (x)i) Average rate of change of pixel value, x, to fit temperature changesiThe pixel value of the ith pixel point is shown, wherein i is the serial number of the pixel point, i is 1,2, … n, and n is the total number of the pixel points; y isiIs the temperature value of the ith pixel point in the first difference average gradient image,
is an error;
the specific expression of the average change rate of the pixel values of the fitting temperature change is as follows:
f(xi)＝axi+b
and S24, dividing the first difference average gradient image by taking the part with the sudden average change rate of the pixel values fitting the temperature change as a boundary, dividing the part with the same average change rate of the pixel values fitting the temperature change into the same region, generating a plurality of first rough regions, and taking the first rough regions as a first rough region set.
The beneficial effects of the above further scheme are: the regions where the average rate of change of pixel values fitted to the temperature change is the same kind of nuclear waste, and the nuclear waste can be classified by the average rate of change of pixel values.
Further: the step S3 includes the following sub-steps:
s31, determining a K value according to the divided areas in the first rough area;
s32, taking all the pixel values of the fitting temperature changes in the current first rough region set as a data set, and selecting K data points from the data set as a centroid, wherein the centroid is the average value of the pixel values of the fitting temperature changes in each first rough region;
s33, calculating the distance from each data point in the data set to all centroids, and combining the data points to a data point set corresponding to a first rough area to which the closest centroid belongs;
s34, recalculating the centroid of each set after the sets to which all centroids belong are merged, and judging whether the difference between the recalculated centroid and the original centroid is smaller than a preset threshold value;
if yes, go to step S35;
if not, returning to the step S32;
and S35, taking the new first rough area set as a first area.
The beneficial effects of the above further scheme are: the first region may better classify nuclear waste than the first coarse region.
Further: in the step S4, the second thermal radiation image includes a thermal radiation image at time m and a thermal radiation image at time m + 10; the step S4 includes the following sub-steps:
s41, acquiring a thermal radiation image at the m moment and a thermal radiation image at the m +10 moment by using a thermal infrared imager, converting the thermal radiation images into a gray scale image, subtracting the gray scale image at the m moment from the gray scale image at the m +10 moment, and dividing the difference by 10 to obtain a second difference average gradient image;
s42, obtaining a second rough region set through a least square method;
and S43, dividing the second rough region set by a clustering method to obtain a second region.
The beneficial effects of the above further scheme are: the second area is used for comparing the intersection of the same area with the first area, so that the area represented by the same nuclear waste can be acquired more accurately.
Further: the step S6 includes the following sub-steps:
s61, training a grasping model through a convolutional neural network according to the data set for manufacturing the nuclear waste;
and S62, acquiring the partition areas through the grabbing models, controlling the grabbing device to grab the nuclear waste, and finishing the sorting of the nuclear waste.
The beneficial effects of the above further scheme are: the convolutional neural network can train out the center point of the grab, the width, the angle and the confidence of the grab.
The invention has the beneficial effects that:
(1) according to the invention, the sorting of the nuclear waste can be realized through the thermal infrared imager and the depth camera.
(2) The invention realizes the segmentation of the areas among different objects through the thermal infrared imager and has the characteristics of high precision, strong noise resistance and high positioning speed.
(3) The data set which is made by the user is trained through the network structure designed by the invention, and the width of a grabbing frame of a grabbing central point and a grabbing position, the grabbing angle and the grabbing confidence coefficient of the data set can be obtained.
Drawings
FIG. 1 is a system diagram of the present invention;
FIG. 2 is a view of the robot arm of the present invention;
FIG. 3 is a flow chart of the present invention;
FIG. 4 is a network architecture diagram of a convolutional neural network of the present invention;
wherein: 1. a thermal infrared imager; 2. a gripping device; 3. a mechanical arm; 4. a combustible classifying container; 5. a conveyor belt; 6. a zero-crossing point detection switch; 7. a depth camera; 8. a non-combustible classification vessel.
Detailed Description
The following description of the embodiments of the present invention is provided to facilitate the understanding of the present invention by those skilled in the art, but it should be understood that the present invention is not limited to the scope of the embodiments, and it will be apparent to those skilled in the art that various changes may be made without departing from the spirit and scope of the invention as defined and defined in the appended claims, and all matters produced by the invention using the inventive concept are protected.
As shown in fig. 1, in one embodiment of the present invention, a dual-optical-camera-based nuclear waste detection and sorting system includes a thermal infrared imager 1, a gripping device 2, a mechanical arm 3, a combustible sorting container 4, a conveyor belt 5, a zero-crossing point detection switch 6 and a non-combustible sorting container 8;
the zero crossing point detection switch 6 is arranged in the middle of a conveyor provided with the conveyor belt 5, the thermal infrared imager 1 is arranged at a position 1 m above the nuclear waste, the mechanical arm 3 is arranged at a position 20 cm away from the nuclear waste through a base and is connected with the gripping device 2, the combustible classification container 4 is arranged on the right side of the mechanical arm 3, and the incombustible classification container 8 is arranged on the side edge of the conveyor belt.
The thermal infrared imager 1 is used for sensing the change of the gray scale image pixel value of the thermal radiation image corresponding to the change of the temperature, the gripping device 2 is used for sorting nuclear waste, the mechanical arm 3 is used for matching with the gripping device 2 to sort the nuclear waste, the combustible classification container 4 and the incombustible classification container 8 are used for containing the nuclear waste, the conveyor belt 5 is used for conveying the nuclear waste, the zero-crossing detection switch 6 is used for detecting the passing of the nuclear waste,
a depth camera 7 is arranged in the grabbing device 2, the depth camera 7 is arranged at the right center of the grabbing device 2, and the depth camera 7 is used for acquiring more texture information and depth information of the nuclear waste;
the working process of the system of the invention is as follows: the method comprises the steps that a conveyor belt 5 transports nuclear waste, a first thermal radiation image of the nuclear waste is collected by a thermal infrared imager 1 through a zero detection switch 6, denoising is carried out to generate a first gray scale image, a difference result of the first gray scale image is divided by 5 to obtain a first difference average gradient image, the thermal infrared imager 1 is used for reading out the temperature change rate of each pixel point in the first gray scale image, a Python language is used for obtaining the pixel value of each pixel point in the first difference average gradient image, the least square method is used for fitting the relation between the pixel value and the temperature value of each pixel point in the first difference average gradient image, the regions with the same pixel value average change rate of fitting temperature change are divided into the same region and are regarded as the same type of nuclear waste, a first rough region set is generated, and then clustering is carried out to obtain a first region; continuously acquiring a second thermal radiation image of the nuclear waste, converting the second thermal radiation image into a gray level graph, calculating a difference value of the gray level graph, dividing the result by 10 to obtain a second difference average gradient image, obtaining a second rough area set by a least square method, and dividing the second rough area set by clustering to generate a second area; comparing the second area with the first area and taking the same area to generate divided areas, wherein different areas in the divided areas represent different types of nuclear waste;
the data set of the nuclear waste is manufactured through the depth camera 7, the model of the depth camera 7 is RGB-D in the embodiment, the data set of the nuclear waste is trained through the convolutional neural network to obtain a grabbing model, the segmented areas are input into the grabbing model, the grabbing model is calculated to obtain the central point of each area and is used as the central pixel point of one-time grabbing, the grabbing device 2 is controlled to grab the nuclear waste and put the nuclear waste into the combustible classification container 4 or the non-combustible classification container 8, and sorting of the nuclear waste is completed.
As shown in fig. 2, in one embodiment of the present invention, a dual-optical camera based nuclear waste detection and sorting method includes the following steps:
s1, acquiring a first thermal radiation image of the nuclear waste through the thermal infrared imager 1, denoising the first thermal radiation image to obtain a first gray scale image, and generating a first difference average gradient image according to the first gray scale image;
s2, fitting a linear relation between pixel values of pixel points of the first difference average gradient image and the temperature change rate through a least square method, and obtaining a first rough region set according to a fitting result;
s3, processing the first rough region set by a clustering method to obtain a first region;
s4, acquiring a second thermal radiation image of the nuclear waste through the thermal infrared imager 1, obtaining a second difference average gradient image according to the second thermal radiation image, and obtaining a second area through a least square method and a clustering method;
s5, comparing the second area with the first area and taking the same area to obtain a segmentation area;
and S6, inputting the segmentation areas into the grabbing model based on the grabbing model trained by the convolutional neural network, and controlling the grabbing device 2 to grab the nuclear waste to finish the sorting of the nuclear waste.
In the step S1, the first thermal radiation image of the nuclear waste includes a thermal radiation image at a time t and a thermal radiation image at a time t + 5, and the first gray scale map includes a gray scale map at the time t and a gray scale map at the time t + 5;
the step S1 includes the following sub-steps:
s11, acquiring a thermal radiation image at the time t and a thermal radiation image at the time t +5 through the thermal infrared imager 1, and denoising and converting the thermal radiation images into a grayscale image at the time t and a grayscale image at the time t + 5;
and S12, subtracting the gray scale image at the time t from the gray scale image at the time t + 5, and dividing the difference result by 5 to obtain a first difference average gradient image.
The value of the pixel point in the difference average gradient image reflects the information of the temperature change.
The step S2 includes the following sub-steps:
s21, reading out the temperature value of each pixel point in the first difference average gradient image through the thermal infrared imager 1, and obtaining the temperature change rate of each pixel point according to the interval time for collecting the first thermal radiation image;
s22, reading the pixel value of each pixel point in the first difference average gradient image;
s23, fitting the relationship between the pixel value of each pixel point in the first difference average gradient image and the temperature change rate through a least square method to obtain the slope a and the offset b corresponding to each pixel point, and further obtain the pixel value average change rate of each fitted temperature change;
the expression of the least square method is specifically as follows:
in the formula, f (x)i) Average rate of change of pixel value, x, to fit temperature changesiThe pixel value of the ith pixel point is shown, wherein i is the serial number of the pixel point, i is 1,2, … n, and n is the total number of the pixel points; y isiIs the temperature value of the ith pixel point in the first difference average gradient image,
is an error;
when in use
The closer the minimum value is taken, the closer the true value is.
The specific expression of the average change rate of the pixel values of the fitting temperature change is as follows:
f(xi)＝axi+b
according to the cloud calculus, when a and b take different values, the values are different
And only have
Partial derivative of a and
when the partial derivative for b is equal to 0,
taking the minimum value, i.e. the following formula:
when in use
When the value is the minimum value, the values of a and b can be determined, and the temperature change rate of each pixel point can be further obtained.
And S24, dividing the first difference average gradient image by taking the part with the sudden average change rate of the pixel values fitting the temperature change as a boundary, dividing the part with the same average change rate of the pixel values fitting the temperature change into the same region, generating a plurality of first rough regions, and taking the first rough regions as a first rough region set.
The same substance is fitted to the regions with the same average change rate of the pixel values of the temperature change, and the nuclear waste can be classified by dividing the regions.
The step S3 includes the following sub-steps:
s31, determining a K value according to the divided areas in the first rough area;
s32, taking all the pixel values of the fitting temperature changes in the current first rough region set as a data set, and selecting K data points from the data set as a centroid, wherein the centroid is the average value of the pixel values of the fitting temperature changes in each first rough region;
s33, calculating the distance from each data point in the data set to all centroids, and combining the data points to a data point set corresponding to a first rough area to which the closest centroid belongs;
s34, recalculating the centroid of each set after the sets to which all centroids belong are merged, and judging whether the difference between the recalculated centroid and the original centroid is smaller than a preset threshold value;
if yes, go to step S35;
if not, returning to the step S32;
and S35, taking the new first rough area set as a first area.
In the step S4, the second thermal radiation image of the nuclear waste includes a thermal radiation image at a time m and a thermal radiation image at a time m + 10; the step S4 includes the following sub-steps:
s41, acquiring a thermal radiation image at the moment m and a thermal radiation image at the moment m +10 by the thermal infrared imager 1, converting the thermal radiation images into a gray scale image, subtracting the gray scale image at the moment m from the gray scale image at the moment m +10, and dividing the difference by 10 to obtain a second difference average gradient image;
s42, obtaining a second rough region set through a least square method;
and S43, dividing the second rough region set by a clustering method to obtain a second region.
The least square method used in step S4 is the same as the least square method used in step S2, and the clustering method used is the same as the clustering method used in step S3.
The step S6 includes the following sub-steps:
s61, training a grasping model through a convolutional neural network according to the data set for manufacturing the nuclear waste;
and S62, acquiring the segmentation areas through the grabbing models, and controlling the grabbing device 2 to grab the nuclear waste to finish the sorting of the nuclear waste.
As shown in fig. 3, a data set of nuclear waste is produced by an RGB-D camera, the format of the data set adopts the cannel data set format, and the produced data set of nuclear waste is trained by using the convolutional neural network of the present invention, and a capture model is obtained by training.
The invention has the beneficial effects that: according to the invention, the sorting of the nuclear waste can be realized through the thermal infrared imager 1 and the depth camera 7.
The invention realizes the segmentation of the areas among different objects through the thermal infrared imager 1 and has the characteristics of high precision, strong noise resistance and high positioning speed.
The data set which is made by the user is trained through the network structure designed by the invention, and the width of a grabbing frame of a grabbing central point and a grabbing position, the grabbing angle and the grabbing confidence coefficient of the data set can be obtained.
In the description of the present invention, it is to be understood that the terms "center", "thickness", "upper", "lower", "horizontal", "top", "bottom", "inner", "outer", "radial", and the like, indicate orientations and positional relationships based on the orientations and positional relationships shown in the drawings, and are used merely for convenience in describing the present invention and for simplicity in description, and do not indicate or imply that the referenced devices or elements must have a particular orientation, be constructed and operated in a particular orientation, and thus, are not to be construed as limiting the present invention. Furthermore, the terms "first," "second," and "third" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or an implicit indication of the number of technical features. Thus, features defined as "first", "second", "third" may explicitly or implicitly include one or more of the features.
Claims (8)
1. A nuclear waste detecting and sorting system based on a dual-optical camera is characterized by comprising a thermal infrared imager (1), a grabbing device (2), a mechanical arm (3), a combustible classification container (4), a conveyor belt (5), a zero-crossing point detection switch (6) and a non-combustible classification container (8);
the zero crossing point detection switch (6) is arranged in the middle of a conveyor provided with the conveyor belt (5), the thermal infrared imager (1) is arranged at a position 1 m above the nuclear waste, the mechanical arm (3) is arranged at a position 20 cm away from the nuclear waste through the base and is connected with the gripping device (2), the combustible classification container (4) is arranged on the right side of the mechanical arm (3), and the incombustible classification container (8) is arranged on the side edge of the conveyor belt.
2. A dual camera based nuclear waste detection sorting system as claimed in claim 1 wherein a depth camera (7) is provided in the gripper (2), the depth camera (7) being provided at the very centre of the gripper (2).
3. A nuclear waste detection and sorting method based on a dual-optical camera is characterized by comprising the following steps:
s1, acquiring a first thermal radiation image of the nuclear waste through a thermal infrared imager (1), denoising the first thermal radiation image to obtain a first gray scale image, and generating a first difference average gradient image according to the first gray scale image;
s2, fitting a linear relation between pixel values of pixel points of the first difference average gradient image and the temperature change rate through a least square method, and obtaining a first rough region set according to a fitting result;
s3, processing the first rough region set by a clustering method to obtain a first region;
s4, acquiring a second thermal radiation image of the nuclear waste through the thermal infrared imager (1), obtaining a second difference average gradient image according to the second thermal radiation image, and obtaining a second area through a least square method and a clustering method;
s5, comparing the second area with the first area and taking the same area to obtain a segmentation area;
and S6, inputting the segmentation areas into the grabbing model based on the grabbing model trained by the convolutional neural network, and controlling the grabbing device (2) to grab the nuclear waste to finish the sorting of the nuclear waste.
4. A dual optical camera-based nuclear waste detection sorting method according to claim 3, wherein in the step S1, the first thermal radiation image of the nuclear waste includes a thermal radiation image at time t and a thermal radiation image at time t +5, and the first gray scale map includes a gray scale map at time t and a gray scale map at time t + 5;
the step S1 includes the following sub-steps:
s11, acquiring a thermal radiation image at the time t and a thermal radiation image at the time t +5 by the thermal infrared imager (1), and denoising the thermal radiation images and converting the thermal radiation images into a grayscale image at the time t and a grayscale image at the time t + 5;
and S12, subtracting the gray scale image at the time t from the gray scale image at the time t +5, and dividing the difference result by 5 to obtain a first difference average gradient image.
5. A dual optical camera-based nuclear waste detection sorting method according to claim 3, wherein the step S2 includes the following substeps:
s21, reading the temperature value of each pixel point in the first difference average gradient image through the thermal infrared imager (1), and obtaining the temperature change rate of each pixel point according to the interval time for collecting the first thermal radiation image;
s22, reading the pixel value of each pixel point in the first difference average gradient image;
s23, fitting the relationship between the pixel value of each pixel point in the first difference average gradient image and the temperature change rate through a least square method to obtain the slope a and the offset b corresponding to each pixel point, and further obtain the pixel value average change rate of each fitted temperature change;
the expression of the least square method is specifically as follows:
in the formula, f (x)i) Average rate of change of pixel value, x, to fit temperature changesiThe pixel value of the ith pixel point is shown, wherein i is the serial number of the pixel point, i is 1,2, … n, and n is the total number of the pixel points; y isiIs firstThe temperature value of the ith pixel point in the difference average gradient image,
is an error;
the specific expression of the average change rate of the pixel values of the fitting temperature change is as follows:
f(xi)＝axi+b
and S24, dividing the first difference average gradient image by taking the part with the sudden average change rate of the pixel values fitting the temperature change as a boundary, dividing the part with the same average change rate of the pixel values fitting the temperature change into the same region, generating a plurality of first rough regions, and taking the first rough regions as a first rough region set.
6. A dual optical camera-based nuclear waste detection sorting method according to claim 3, wherein the step S3 includes the following substeps:
s31, determining a K value according to the divided areas in the first rough area;
s32, taking all the pixel values of the fitting temperature changes in the current first rough region set as a data set, and selecting K data points from the data set as a centroid, wherein the centroid is the average value of the pixel values of the fitting temperature changes in each first rough region;
s33, calculating the distance from each data point in the data set to all centroids, and combining the data points to a data point set corresponding to a first rough area to which the closest centroid belongs;
s34, recalculating the centroid of each set after the sets to which all centroids belong are merged, and judging whether the difference between the recalculated centroid and the original centroid is smaller than a preset threshold value;
if yes, go to step S35;
if not, returning to the step S32;
and S35, taking the new first rough area set as a first area.
7. A dual optical camera-based nuclear waste detection sorting method according to claim 6, wherein in the step S4, the second thermal radiation image of the nuclear waste includes a thermal radiation image at a time m and a thermal radiation image at a time m + 10; the step S4 includes the following sub-steps:
s41, acquiring a thermal radiation image at the m moment and a thermal radiation image at the m +10 moment through the thermal infrared imager (1), converting the thermal radiation images into a gray scale image, subtracting the gray scale image at the m moment from the gray scale image at the m +10 moment, and dividing the difference by 10 to obtain a second difference average gradient image;
s42, obtaining a second rough region set through a least square method;
and S43, dividing the second rough region set by a clustering method to obtain a second region.
8. A dual optical camera-based nuclear waste detection sorting method according to claim 3, wherein the step S6 includes the following substeps:
s61, training a grasping model through a convolutional neural network according to the data set for manufacturing the nuclear waste;
and S62, acquiring the segmentation areas through the grabbing models, and controlling the grabbing device (2) to grab the nuclear waste to finish the sorting of the nuclear waste.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111144641.0A CN113751365B (en) | 2021-09-28 | 2021-09-28 | Nuclear waste detecting and sorting system and method based on double optical cameras |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111144641.0A CN113751365B (en) | 2021-09-28 | 2021-09-28 | Nuclear waste detecting and sorting system and method based on double optical cameras |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113751365A true CN113751365A (en) | 2021-12-07 |
| CN113751365B CN113751365B (en) | 2023-03-17 |
Family
ID=78797945
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111144641.0A Active CN113751365B (en) | 2021-09-28 | 2021-09-28 | Nuclear waste detecting and sorting system and method based on double optical cameras |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113751365B (en) |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070029232A1 (en) * | 2003-09-20 | 2007-02-08 | Qinetiq Limited | Apparatus for, and method of, classifying objects in a waste stream |
| FR2895138A1 (en) * | 2005-12-20 | 2007-06-22 | Electricite De France | Plant for sorting and packing toxic nuclear waste has sealed sorting cell with input conveyor and airlock and drum manipulating chamber with sealed trapdoor |
| US20140379588A1 (en) * | 2013-03-15 | 2014-12-25 | Compology, Inc. | System and method for waste managment |
| CN106807651A (en) * | 2017-03-23 | 2017-06-09 | 江苏云光智慧信息科技有限公司 | A kind of bastard coal method for sorting and its device based on temperature and area modeling |
| CN108339761A (en) * | 2018-01-24 | 2018-07-31 | 江苏理工学院 | A kind of sorting welder and its method based on thermal infrared imager detection |
| CN109848073A (en) * | 2019-02-22 | 2019-06-07 | 浙江大学滨海产业技术研究院 | A kind of apparatus and method for sorting coal and gangue |
| CN110400345A (en) * | 2019-07-24 | 2019-11-01 | 西南科技大学 | Radioactive waste based on deeply study, which pushes away, grabs collaboration method for sorting |
| CN210208068U (en) * | 2019-07-11 | 2020-03-31 | 北京铮实环保工程有限公司 | Intelligent garbage sorting system based on thermal imaging |
| CN112017259A (en) * | 2020-08-27 | 2020-12-01 | 河北工业大学 | Indoor positioning and image building method based on depth camera and thermal imager |
| CN112170233A (en) * | 2020-09-01 | 2021-01-05 | 燕山大学 | Small part sorting method and system based on deep learning |
| CN113083721A (en) * | 2019-12-23 | 2021-07-09 | 深圳万讯自控股份有限公司 | Multi-joint sorting robot based on structured light and control method thereof |
| CN213855792U (en) * | 2020-07-17 | 2021-08-03 | 长安大学 | Coal gangue identification and sorting device |
-
2021
- 2021-09-28 CN CN202111144641.0A patent/CN113751365B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070029232A1 (en) * | 2003-09-20 | 2007-02-08 | Qinetiq Limited | Apparatus for, and method of, classifying objects in a waste stream |
| FR2895138A1 (en) * | 2005-12-20 | 2007-06-22 | Electricite De France | Plant for sorting and packing toxic nuclear waste has sealed sorting cell with input conveyor and airlock and drum manipulating chamber with sealed trapdoor |
| US20140379588A1 (en) * | 2013-03-15 | 2014-12-25 | Compology, Inc. | System and method for waste managment |
| CN106807651A (en) * | 2017-03-23 | 2017-06-09 | 江苏云光智慧信息科技有限公司 | A kind of bastard coal method for sorting and its device based on temperature and area modeling |
| CN108339761A (en) * | 2018-01-24 | 2018-07-31 | 江苏理工学院 | A kind of sorting welder and its method based on thermal infrared imager detection |
| CN109848073A (en) * | 2019-02-22 | 2019-06-07 | 浙江大学滨海产业技术研究院 | A kind of apparatus and method for sorting coal and gangue |
| CN210208068U (en) * | 2019-07-11 | 2020-03-31 | 北京铮实环保工程有限公司 | Intelligent garbage sorting system based on thermal imaging |
| CN110400345A (en) * | 2019-07-24 | 2019-11-01 | 西南科技大学 | Radioactive waste based on deeply study, which pushes away, grabs collaboration method for sorting |
| CN113083721A (en) * | 2019-12-23 | 2021-07-09 | 深圳万讯自控股份有限公司 | Multi-joint sorting robot based on structured light and control method thereof |
| CN213855792U (en) * | 2020-07-17 | 2021-08-03 | 长安大学 | Coal gangue identification and sorting device |
| CN112017259A (en) * | 2020-08-27 | 2020-12-01 | 河北工业大学 | Indoor positioning and image building method based on depth camera and thermal imager |
| CN112170233A (en) * | 2020-09-01 | 2021-01-05 | 燕山大学 | Small part sorting method and system based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 肖贤鹏: "基于机器视觉的核废物分拣方法研究", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113751365B (en) | 2023-03-17 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109344894B (en) | Garbage classification and identification method and device based on multi-sensor fusion and deep learning | |
| CN109785337A (en) | Mammal counting method in a kind of column of Case-based Reasoning partitioning algorithm | |
| CN109214373B (en) | Face recognition system and method for attendance checking | |
| CN105488456B (en) | Method for detecting human face based on adaptive threshold adjustment rejection sub-space learning | |
| CN109215806A (en) | A kind of public place health monitoring systems and method based on recognition of face | |
| CN108303748A (en) | The method for checking equipment and detecting the gun in luggage and articles | |
| CN111724355B (en) | Image measuring method for abalone body type parameters | |
| CN107230203A (en) | Casting defect recognition methods based on human eye vision attention mechanism | |
| CN108564085A (en) | A kind of method of automatic reading pointer type instrument reading | |
| CN111209822A (en) | Face detection method of thermal infrared image | |
| CN110728225A (en) | High-speed face searching method for attendance checking | |
| TWI712724B (en) | Method and device of automatically tracking urine | |
| CN113111817B (en) | Semantic segmentation face integrity measurement method, system, equipment and storage medium | |
| CN109520418A (en) | A kind of pallet method for recognizing position and attitude based on two dimensional laser scanning instrument | |
| CN109101943A (en) | It is a kind of for detecting the machine vision method of Falls Among Old People | |
| CN112151167A (en) | Intelligent screening method for six-age dental caries of children based on deep learning | |
| CN111784688A (en) | Flower automatic grading method based on deep learning | |
| CN113751365B (en) | Nuclear waste detecting and sorting system and method based on double optical cameras | |
| CN108154513A (en) | Cell based on two photon imaging data detects automatically and dividing method | |
| Liu et al. | Automatic detection and segmentation of laser stripes for industrial measurement | |
| CN109978937A (en) | A kind of ox body measurement system detected based on deep learning and characteristic portion | |
| CN209487160U (en) | It is a kind of can to use oil product and historical data to carry out Health Category division kitchen ventilator | |
| CN108288273A (en) | The automatic detection method of railway contact line abnormal object based on multi-scale coupling convolution net | |
| CN113963199B (en) | Medical waste identification method based on multiple sensor feature fusion and machine learning | |
| CN109631111A (en) | It is a kind of that the automatic cigarette stove system that food data bank compares can be sought with health |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |