CN116883413A - Visual detection method for retention of waste picking and receiving materials - Google Patents
Visual detection method for retention of waste picking and receiving materials Download PDFInfo
- Publication number
- CN116883413A CN116883413A CN202311152699.9A CN202311152699A CN116883413A CN 116883413 A CN116883413 A CN 116883413A CN 202311152699 A CN202311152699 A CN 202311152699A CN 116883413 A CN116883413 A CN 116883413A
- Authority
- CN
- China
- Prior art keywords
- retention
- inclined groove
- centroid
- gray level
- area
- 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
- 239000000463 material Substances 0.000 title claims abstract description 352
- 230000014759 maintenance of location Effects 0.000 title claims abstract description 107
- 239000002699 waste material Substances 0.000 title claims abstract description 71
- 238000001514 detection method Methods 0.000 title claims abstract description 17
- 230000000007 visual effect Effects 0.000 title claims abstract description 11
- 230000000717 retained effect Effects 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 8
- 238000000034 method Methods 0.000 claims description 55
- 230000008569 process Effects 0.000 claims description 11
- 238000011179 visual inspection Methods 0.000 claims description 9
- 230000008859 change Effects 0.000 claims description 4
- 238000010276 construction Methods 0.000 claims description 4
- 238000013507 mapping Methods 0.000 claims description 4
- 230000002950 deficient Effects 0.000 description 10
- 238000004519 manufacturing process Methods 0.000 description 4
- 238000007599 discharging Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000004806 packaging method and process Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009826 distribution Methods 0.000 description 2
- 239000011800 void material Substances 0.000 description 2
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000000979 retarding effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Geometry (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image data processing, and provides a visual detection method for retention of waste picking and receiving materials, which comprises the following steps: acquiring a gray level image; carrying out material identification processing on the material areas, acquiring a retention position influence factor of each material centroid according to the spatial position relation of each material centroid, and acquiring a two-pole gap difference index of each material centroid according to retention information of materials in the inclined chute; acquiring retention information positioning coefficients of each material centroid based on retention position influence factors of each material centroid and a bipolar gap difference index, acquiring weights in bipartite graphs of adjacent images based on Euclidean distances and the retention information positioning coefficients, and acquiring retention materials in the inclined grooves by utilizing an optimal matching algorithm according to the bipartite graphs of the adjacent images; and acquiring the serious condition of retention in the waste rejecting and receiving equipment according to the occupied area of the retained materials. The invention effectively avoids the material retention condition in the waste rejecting and receiving, and improves the waste rejecting efficiency of the waste rejecting and receiving equipment.
Description
Technical Field
The invention relates to the technical field of image data processing, in particular to a visual detection method for retention of waste picking and receiving materials.
Background
In the industrial production process, defective products are usually produced according to various factors and mixed in a product packaging line, defective waste products are removed at the moment, the conventional defective waste removing operation is to manually remove defective waste products on the packaging line by hiring workers, and along with innovation of technical equipment, more advanced defective waste removing equipment is arranged on many production lines at present, defective waste products are identified and automatically removed from the production packaging line by a machine by setting defective waste detection criteria, so that the defective waste removing efficiency is greatly improved. However, due to imperfect automation technology and equipment design requirements, the situation that materials are retained at the edge of a waste rejecting groove or on a discharge hole easily occurs in a waste rejecting receiving process, and finally, the waste rejecting efficiency is reduced, the accuracy is reduced and even the waste rejecting process is terminated because of excessive accumulation.
The current detection of the retention condition of the reject flow mainly comprises the step of judging whether the retention condition exists in the reject equipment or not by calculating the difference value between the total sum of the feeding and the discharging and the total sum of the waste, however, most of the conditions of the method are influenced by the speed of feeding and discharging materials and the statistical speed, the final result is inaccurate due to the loss of feeding and discharging materials caused by other conditions, and the error judgment is finally carried out on the retention condition in the equipment.
Disclosure of Invention
The invention provides a visual detection method for retention of waste picking and receiving materials, which aims to solve the problem of inaccurate retention detection of materials, and adopts the following specific technical scheme:
the invention provides a visual inspection method for retention of waste picking and receiving materials, which comprises the following steps:
acquiring an inclined groove gray level image in the waste rejecting and receiving equipment by using an image acquisition equipment;
acquiring a material centroid of each inclined groove gray image and a centroid of a discharge hole area according to identification processing of materials in each inclined groove gray image, acquiring a detention position influence factor of each material centroid in each inclined groove gray image according to Euclidean distance between the material centroid and the centroid of the discharge hole area and Euclidean distance between the material centroid and boundaries of two sides of the inclined groove, and acquiring a two-pole gap difference index of each material centroid in each inclined groove gray image according to differences between inclined groove gaps of two ends of line segments where the material centroid is located;
acquiring retention information positioning coefficients of mass centers of materials in each inclined groove gray image according to retention position influence factors and two-pole gap difference indexes of the mass centers of materials in each inclined groove gray image, acquiring weights in bipartite graphs of the inclined groove gray images acquired at adjacent time points according to differences between Euclidean distances among the mass centers of materials in the inclined groove gray images acquired at adjacent time points and the retention information positioning coefficients of the mass centers of materials, and taking the bipartite graphs with weights of the inclined groove gray images acquired at the adjacent time points as input of a KM optimal matching algorithm, and acquiring retention materials in the inclined grooves according to output of the KM optimal matching algorithm;
and acquiring the severity of material retention in the waste rejecting and receiving equipment according to the area of the inclined groove occupied by the retained material in the inclined groove, and acquiring a visual detection result of the waste rejecting and receiving equipment according to the severity of material retention.
Preferably, the method for obtaining the mass center of the material and the mass center of the discharge hole area of each inclined groove gray level image according to the identification processing of the material in each inclined groove gray level image comprises the following steps:
for any inclined groove gray level image, acquiring the sum of the areas of the dividing areas and the number of the dividing areas in the inclined groove gray level image by using an Ojin threshold dividing algorithm, and taking the ratio of the sum of the areas of the dividing areas to the number of the dividing areas as an average material area;
taking the divided areas with the area smaller than or equal to the average material area as material areas;
taking the dividing region with the dividing region area larger than the average material area as a discharge hole region;
and taking the center of the smallest circumscribed rectangle of each material area and each discharge hole area as the mass center of the material in the material area and the mass center in the discharge hole area respectively.
Preferably, the method for obtaining the retention position influence factor of each material centroid in each inclined groove gray level image according to the euclidean distance between the material centroid and the centroid of the discharge hole area and the euclidean distance between the material centroid and the boundaries of two sides of the inclined groove comprises the following steps:
for any one inclined groove gray level image, taking the accumulated sum of Euclidean distance negative mapping results between each material centroid and all material outlet area centroids as a first summation factor of each material centroid, and acquiring a second summation factor of each material centroid according to Euclidean distance between each material centroid and the boundary of two sides of the inclined groove;
and taking the sum value of the first summation factor and the second summation factor of each material centroid as a detention position influence factor of each material centroid.
Preferably, the method for obtaining the second summation factor of each material centroid according to the Euclidean distance between each material centroid and the boundary of two sides of the inclined groove comprises the following steps:
for any one inclined slot gray level image, the accumulated sum of Euclidean distances between each material centroid and all inclined slot boundary points with the same ordinate is taken as a second summation factor of each material centroid.
Preferably, the method for obtaining the two-pole gap difference index of each material centroid in each inclined slot gray level image according to the difference between the inclined slot gaps at the two ends of the line segment where the material centroid is located comprises the following steps:
in the method, in the process of the invention,is the two-pole gap difference index of the mass center of the material, +.>Is the area of the upper part u of the material line segment,area of the lower part d of the material line segment, +.>The area of the ith material area on the upper part of the material line segment, p is the number of the material areas on the upper part of the material line segment,/->The area of the j-th material area at the lower part of the material line segment, and q is the number of the material areas at the lower part of the material line segment.
Preferably, the method for obtaining the retention information positioning coefficient of each material centroid in each inclined groove gray level image according to the retention position influence factor and the two-pole gap difference index of each material centroid in each inclined groove gray level image comprises the following steps:
and for any inclined groove gray level image, taking the product of the detention position influence factor of each material centroid in the inclined groove gray level image and the gap difference index of the two poles as the detention information positioning coefficient of each material centroid in the inclined groove gray level image.
Preferably, the method for obtaining the weight value in the bipartite graph of the inclined groove gray level image collected at the adjacent time point according to the difference between the Euclidean distance between the material centroids in the inclined groove gray level image collected at the adjacent time point and the retention information positioning coefficient of the material centroids comprises the following steps:
for the gray level images of the inclined grooves acquired at adjacent time points, a first product factor is acquired according to the difference between the retention information positioning coefficients of the mass centers of the materials;
the Euclidean distance between the mass centers of materials in the inclined groove gray level images acquired by adjacent time points is used as a second product factor;
and taking a negative mapping result of the product between the first product factor and the second product factor as a weight value in a bipartite graph of the inclined slot gray level image acquired at the adjacent time point.
Preferably, the method for obtaining the first product factor according to the difference between the retention information positioning coefficients of the mass centers of the materials comprises the following steps:
and taking the absolute value of the difference value between the retention information positioning coefficients of the mass centers of two materials in different inclined groove gray level images as a first product factor of the inclined groove gray level images acquired at the adjacent time points for the inclined groove gray level images acquired at the adjacent time points.
Preferably, the weighted bipartite graph of the gray level image of the inclined slot collected at the adjacent time point is used as the input of the KM optimal matching algorithm, and the method for obtaining the retained materials in the inclined slot according to the output of the KM optimal matching algorithm comprises the following steps:
taking all material centroids in the inclined groove gray level image of the current acquisition time as a current set, taking all material centroids in the inclined groove gray level image of the next acquisition time as a matching set, and carrying out connection construction edges on each element of the current set and each element in the matching set to construct a bipartite graph of the inclined groove gray level image acquired at the adjacent time point;
giving weight values in bipartite graphs of the inclined slot gray images acquired at adjacent time points to each edge of the bipartite graphs, taking the bipartite graph with the weight values of the inclined slot gray images acquired at the adjacent time points as input of a KM optimal matching algorithm, and taking output of KM optimal matching as a matching set;
and acquiring position information in the gray level images of the inclined grooves, acquired at different time points, of each material according to the matching set, and taking the material which does not change in position within the preset time as a retention material.
Preferably, the method for obtaining the severity of material retention in the waste rejecting and material receiving device according to the area of the inclined chute occupied by the retained material in the inclined chute and obtaining the visual detection result of the waste rejecting and material receiving device according to the severity of material retention comprises the following steps:
and acquiring the material retention ratio of the total area of the waste rejecting inclined groove area occupied by the material retention area according to the material retention area, and judging that the material retention condition is serious in the visual detection result of the waste rejecting and receiving when the material retention ratio is larger than a preset parameter, wherein the retained material in the waste rejecting and receiving equipment is required to be removed.
The beneficial effects of the invention are as follows: in the conventional process flow production, the reject receiving device usually generates retention of materials in the device due to various factors, and the conventional retention detection method is to calculate the total difference of the fed and discharged materials to judge the retention of the materials in the device, which is limited by the statistical speed and the influence of external factors. According to the invention, a plurality of images are shot in the same time interval according to the movement condition of materials in the waste rejecting and receiving equipment, a retention position influence factor and a two-pole gap difference index are constructed according to the characteristics of the materials in the equipment, finally, a retention information positioning coefficient is constructed for each material, the Euclidean distance of the materials between the images is combined to represent the weight of each material in each image, a bipartite graph with the weight is constructed according to the obtained images and the material positioning coefficient, the optimal matching between the images is carried out by utilizing a KM algorithm, the materials which have no change in a short time are obtained, and finally, the material retention condition of the equipment is judged by counting the numerical value and the area ratio of the waste rejecting equipment. The method has the beneficial effects that the material retention detection accuracy is improved, the material retention condition in the waste picking and receiving process is effectively avoided, and meanwhile, the waste picking efficiency of the waste picking and receiving equipment is improved.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, it being obvious that the drawings in the description below are only some embodiments of the invention, and that other drawings can be obtained according to these drawings without inventive faculty for a person skilled in the art.
FIG. 1 is a schematic flow chart of a visual inspection method for retention of rejected material according to an embodiment of the invention;
fig. 2 is a schematic structural diagram of an internal inclined groove of the waste rejecting and receiving device according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, a flow chart of a visual inspection method for retention of rejected material according to an embodiment of the invention is shown, the method comprises the following steps:
step S001, acquiring a series of inclined groove gray level images inside the waste rejecting and receiving equipment.
Arranging a small CMOS camera in an inclined groove in the waste rejecting and receiving equipment, photographing the surface of the inclined groove at a overlooking angle, starting an illumination device, photographing once every 0.2s, obtaining a series of RGB images of the inclined groove in the waste rejecting and receiving equipment, and converting gray level images of the RGB images by using a weighted average method, wherein the weighted average method is a known technology and is not redundant. In order to avoid the influence of noise on analysis results, the obtained inclined groove gray level image is subjected to denoising by means of mean value filtering, and an operator can select other denoising methods according to actual conditions.
So far, a series of inclined groove gray level images inside the waste rejecting and receiving equipment are obtained.
Step S002, carrying out material identification processing on each inclined groove gray level image, and obtaining the retention position influence factor of each material centroid and the two-pole gap difference index of each material centroid in each inclined groove gray level image.
In the invention, the structure schematic diagram of the internal inclined groove is shown in fig. 2 when the waste rejecting and receiving equipment operates, the automatic waste rejecting and receiving equipment generally adopts the form of the inclined groove, the finished material is poured from the feed inlet, the movement of the material from top to bottom is realized through the gradient of the inclined groove, the defective waste in the material is removed through the waste rejecting device arranged on the inclined groove, finally the defective waste is received into the waste groove, the finished material reaching the standard is received into the standard groove, and then the next production operation is carried out.
Because need reject and connect the waste material to the waste material groove in, reject receiving equipment can set up the discharge gate in inclined chute usually for reject the device and reject it fast when detecting the waste material, and can not be because the chute surface is too smooth and fall down fast and lead to unable detection in order to the material, still can set up certain retarding effect texture at the chute surface. However, the material retention most likely occurs near the discharge port, and the material is easily stuck near the discharge port due to the opening and closing of the discharge port, and the local friction force is large due to the speed reduction texture on the surface of the inclined groove and the boundary baffles on the two sides of the inclined groove, so that the speed of the material in the falling process cannot be maintained, and the material is retained in the inclined groove.
Based on the characteristics, firstly, materials in the waste rejecting and receiving device are required to be distinguished, and the invention adopts an Ojin threshold segmentation algorithm, the materials are used as the foreground, other areas are used as the background to be segmented, and the foreground area containing the materials is extracted. However, as the discharge hole on the chute is mixed with the material, the obtained foreground area comprises a discharge hole area besides the material area after being divided by adopting an Ojin threshold dividing algorithm. Therefore, the spout region needs to be divided from the foreground region.
Specifically, in order to ensure that the material can be discharged normally, the discharge opening is generally much larger in size than the material. Counting the pixel area of each divided area and the number of the divided areas for all the divided foreground areas, obtaining the average pixel area of each divided area by dividing the sum of the pixel areas of the divided areas on all the divided areas by the number of the divided areas, recording the average pixel area as the average material area, comparing the pixel area of each divided area with the average material area, and if the pixel area of a certain divided area is larger than the average material area, obtaining the area as a discharge hole area; if the pixel area of a certain divided area is smaller than or equal to the average material area, the area is a material area. And according to the obtained material outlet areas and the obtained material outlet areas, carrying out minimum circumscribed rectangles on each material area and each material outlet area, and taking the center of the minimum circumscribed rectangle as the mass center of the material in the material area or the mass center of the material outlet area.
Further, the coordinates of the mass center of the discharge hole area are recorded asThe coordinate record coordinate of the mass center of the material is +.>Calculating the Euclidean distance between the mass center of each material and the mass center of the discharge hole area to be +.>. The closer the material is to the discharge port area, the greater the likelihood of stagnation. In addition, the coordinates of boundary points on two sides of the rejected material receiving inclined groove, which are the same as the ordinate of the mass center of the material, are obtained, and the Euclidean distance between the coordinates of the boundary points and the mass center of the material is calculated>The closer the material is to the boundary of the inclined trough sides, the greater the likelihood of stagnation.
Calculating the retention position influence factor of the mass center of each mass center in any inclined groove gray level image:
in the method, in the process of the invention,is the retention position influence factor of mass center of the material, +.>Is an exponential function based on natural constant, m is the number of mass centers of the discharge hole area, ++>Is the Euclidean distance between the mass center of the material and the mass center of the ith discharge hole area,is the Euclidean distance between the mass center of the material and the boundary point of the inclined groove with the j-th same ordinate of the mass center>The number of inclined slot boundary points representing the same ordinate of the mass center of the material.
Euclidean distance between material centroid and ith discharge port area centroidThe smaller the first summation factor of the mass center of the material +.>The larger the material is, the closer the material is to the discharge hole in the inclined groove, the more likely the material is to be detained, and the larger the material detaining position influence factor of the mass center of the material is. Euclidean distance of material centroid and its j-th inclined groove boundary point with same ordinate>The smaller the second summation factor of the mass center of the material +.>The larger the material is, the closer the material is to the edge of the inclined groove in the inclined groove, and the more likely the material is to be detained, the smaller the influence factor of the material detained position of the mass center of the material is.
Further, due to the fluidity of the materials in the process of rejecting waste, a part of the materials are retained in the rejecting waste inclined groove, and possibly collide with the upper materials to a normal state due to falling of the materials, and the material retention condition belongs to frictional retention; another part of the retained material may be due to mechanical retention at the discharge opening, i.e. to a certain position in the inclined chute or to the discharge opening, in which case the retention is difficult to re-hit by other material to a normal state. For frictional retention, the possible existence time is shorter, because the friction retention is possible at the moment or the next moment, the friction retention is not impacted and separated by other materials falling from the upper part, the distribution of the other materials is uniform, and the overall gap of the inclined groove is less; for mechanical retention, the upper material is difficult to separate by collision, so that the upper material can be blocked from falling at the same position in a long period of time, and finally, the distribution difference of other materials at the upper part and the lower part of the material is larger, the lower material is more dispersed, the upper material is more densely distributed, and the whole gap of the inclined groove is larger.
Specifically, for any material centroid, searching edge points on two sides of a waste removal inclined groove with the same ordinate as the current coordinate, connecting the two points with the material centroid point to form a line segment, taking the material line segment as the reference, taking a region with the ordinate larger than the ordinate of the line segment as the upper part of the material line segment, and taking the ordinate smaller than the ordinate of the line segment as the lower part of the material line segment. For the upper area of the material line segment, counting the mass center quantity of the material and the corresponding material area, and recording asAnd the area of the upper part of the line segment of the material was recorded as +.>The method comprises the steps of carrying out a first treatment on the surface of the For the lower area of the line segment, counting the mass center quantity of the material and the corresponding material area, and recording as +.>And the lower part of the material line segmentThe area is recorded as +.>。
For each material mass center in any one inclined groove gray level image, calculating a two-pole gap difference index of the material mass center:
in the method, in the process of the invention,is the two-pole gap difference index of the mass center of the material, +.>Is the area of the upper part u of the material line segment,area of the lower part d of the material line segment, +.>The area of the ith material area on the upper part of the material line segment, p is the number of the material areas on the upper part of the material line segment,/->The area of the j-th material area at the lower part of the material line segment, and q is the number of the material areas at the lower part of the material line segment.
Area of the ith material area on the upper part of the material line segmentThe larger the upper gap of the material line section is, the more->The smaller the area of the j-th material area at the lower part of the material line section is +.>The smaller the material line segment, the lower gap +.>The larger the absolute value of the difference between the first item and the second item is, which means that the larger the difference of the gap areas of the inclined grooves at the upper part and the lower part is, the larger the two-pole gap difference index of the mass center of the material is, i.e. the more likely the material is to stay; area of the ith material area on the upper part of the material line segment +.>The smaller the upper gap of the material line section is, the more->The larger the area of the j-th material area at the lower part of the material line section is +.>The larger the material line segment, the lower gap +.>The smaller the difference between the first term and the second term, i.e., the larger the absolute value of the difference, the larger the difference between the void areas of the inclined grooves at the upper and lower parts, the larger the two-pole void difference index of the mass center of the material, i.e., the more likely the material is to stagnate.
Step S003, acquiring retention information positioning coefficients of mass centers of materials in each inclined groove gray level image, and acquiring weights in bipartite graphs of the inclined groove gray level images acquired at adjacent time points by combining Euclidean distances between the mass centers of materials to acquire retention materials in the inclined grooves.
And taking the product of the detention position influence factor of the material centroid and the gap difference index of the two poles as the detention information positioning coefficient of the material centroid for each material centroid in any inclined groove gray level image.
When the value of the retention information positioning coefficient is larger, the position of the material is closer to the boundary of the two sides of the discharge hole or the inclined groove, and the gap difference between the upper inclined groove and the lower inclined groove of the material is larger; on the contrary, the smaller the value of the retention information positioning coefficient, the farther the material is from the edges of the two sides of the discharge hole or the inclined groove, and the smaller the gap difference between the upper inclined groove and the lower inclined groove of the material is. The material retention state of the material in the inclined groove can be positioned through the coefficient so as to be used for judging the retention condition of the material in a plurality of images.
Further, for each material mass center in any one inclined groove gray level image, the retention information positioning coefficient and the mass center coordinate position of the material mass center are obtained. For each material centroid in the current acquisition time point image, calculating the weight of the material centroid of the current acquisition time point image and the material centroid in the image of the next adjacent acquisition time point, namely the weight in the bipartite graph of the inclined slot gray level image acquired by the adjacent time point:
in the method, in the process of the invention,weight in bipartite graph of oblique slot gray level image collected for adjacent time point, +.>For Euclidean distance between the mass center of the material in the current inclined groove gray level image and the i-th mass center in the inclined groove gray level image of the next acquisition time, +.>Locating coefficients for retention information of mass centers of materials in gray level images of current inclined grooves, +.>And locating coefficients for retention information of the mass center of the ith material in the gray level image of the inclined slot at the next acquisition time.
The smaller the difference between the retention information positioning coefficient of the mass center of the material in the current inclined groove gray level image and the retention information positioning coefficient of the i-th mass center in the inclined groove gray level image of the next acquisition time is, the first product factorSmaller, then adjacent time point acquisitionThe larger the weight in the bipartite graph of the inclined-groove gray-scale image. The smaller the Euclidean distance between the mass center of the material in the current inclined groove gray level image and the i-th mass center in the inclined groove gray level image of the next acquisition time is, the second product factor is->The smaller the weight value in the bipartite graph of the inclined slot gray scale image acquired at the adjacent time point is, the larger the weight value is.
Thus, the weight value in the bipartite graph of the inclined slot gray level image acquired at the adjacent time point is acquired.
The construction mode of the bipartite graph of the inclined groove gray level image acquired by the adjacent time points is as follows: taking all material centroids in the inclined groove gray level image of the current acquisition time as a current set, taking all material centroids in the inclined groove gray level image of the next acquisition time as a matching set, and carrying out connection construction edges on each element of the current set and each element in the matching set to construct a bipartite graph of the inclined groove gray level image acquired at the adjacent time point. The bipartite graph is a known technology and will not be described in detail.
The weight in the bipartite graph of the inclined slot gray level image acquired at the adjacent time point is actually calculated as the weight on the edge formed by all material centroids in the current inclined slot gray level image and all material centroids in the inclined slot gray level image at the next acquisition time, and each calculated weight is given to the corresponding edge to obtain the bipartite graph with the weight of the inclined slot gray level image acquired at the adjacent time point.
So far, the weighted bipartite graph of the inclined groove gray level image acquired at every two adjacent time points is acquired.
The weighted bipartite graph of the inclined slot gray level image acquired by the adjacent time point is used as the input of the KM optimal matching algorithm, the output M is a matching set containing n sides, and each side in the matching set represents the optimal matching between each element of the weighted bipartite graph. After the KM algorithm is matched, the position of each material in each image can be confirmed from the first image, the position of each material is recorded, and the material which has no position change in a short time is searched, namely the retained material.
Thus, the retained material in the inclined groove is obtained.
Step S004, the severity of material retention in the waste rejecting and receiving device is obtained based on the area of the inclined groove occupied by the retained material in the inclined groove.
According to the method, the ratio of the area of the retained material to the total area of the waste removal inclined groove is used as the material retention duty ratio according to the retained material in the inclined groove in the waste removal receiving equipment, and when the material retention duty ratio is larger than a preset parameter T (the empirical value is 0.5), the condition that the waste removal receiving equipment is seriously retained in the visual detection result is judged, and the internal retained material of the waste removal receiving equipment is required to be removed.
In this specification, each embodiment is described in a progressive manner, and identical and similar parts of each embodiment are all referred to each other, and each embodiment mainly describes differences from other embodiments. The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. The visual inspection method for the retention of the rejected material is characterized by comprising the following steps of:
acquiring an inclined groove gray level image in the waste rejecting and receiving equipment by using an image acquisition equipment;
acquiring a material centroid of each inclined groove gray image and a centroid of a discharge hole area according to identification processing of materials in each inclined groove gray image, acquiring a detention position influence factor of each material centroid in each inclined groove gray image according to Euclidean distance between the material centroid and the centroid of the discharge hole area and Euclidean distance between the material centroid and boundaries of two sides of the inclined groove, and acquiring a two-pole gap difference index of each material centroid in each inclined groove gray image according to differences between inclined groove gaps of two ends of line segments where the material centroid is located;
acquiring retention information positioning coefficients of mass centers of materials in each inclined groove gray image according to retention position influence factors and two-pole gap difference indexes of the mass centers of materials in each inclined groove gray image, acquiring weights in bipartite graphs of the inclined groove gray images acquired at adjacent time points according to differences between Euclidean distances among the mass centers of materials in the inclined groove gray images acquired at adjacent time points and the retention information positioning coefficients of the mass centers of materials, and taking the bipartite graphs with weights of the inclined groove gray images acquired at the adjacent time points as input of a KM optimal matching algorithm, and acquiring retention materials in the inclined grooves according to output of the KM optimal matching algorithm;
and acquiring the severity of material retention in the waste rejecting and receiving equipment according to the area of the inclined groove occupied by the retained material in the inclined groove, and acquiring a visual detection result of the waste rejecting and receiving equipment according to the severity of material retention.
2. The visual inspection method for inspecting retention of waste picking and receiving materials according to claim 1, wherein the method for acquiring the mass center of the material and the mass center of the discharge hole area of each inclined groove gray level image according to the identification processing of the material in each inclined groove gray level image comprises the following steps:
for any inclined groove gray level image, acquiring the sum of the areas of the dividing areas and the number of the dividing areas in the inclined groove gray level image by using an Ojin threshold dividing algorithm, and taking the ratio of the sum of the areas of the dividing areas to the number of the dividing areas as an average material area;
taking the divided areas with the area smaller than or equal to the average material area as material areas;
taking the dividing region with the dividing region area larger than the average material area as a discharge hole region;
and taking the center of the smallest circumscribed rectangle of each material area and each discharge hole area as the mass center of the material in the material area and the mass center in the discharge hole area respectively.
3. The method for visually detecting the retention of the waste picking and receiving materials according to claim 1, wherein the method for obtaining the retention position influence factor of each material centroid in each inclined groove gray level image according to the euclidean distance between the material centroid and the centroid of the discharge hole area and the euclidean distance between the material centroid and the boundaries of the two sides of the inclined groove is as follows:
for any one inclined groove gray level image, taking the accumulated sum of Euclidean distance negative mapping results between each material centroid and all material outlet area centroids as a first summation factor of each material centroid, and acquiring a second summation factor of each material centroid according to Euclidean distance between each material centroid and the boundary of two sides of the inclined groove;
and taking the sum value of the first summation factor and the second summation factor of each material centroid as a detention position influence factor of each material centroid.
4. A method for visually inspecting retention of rejected material according to claim 3, wherein the method for obtaining the second summation factor of each material centroid according to the euclidean distance between each material centroid and the boundary of two sides of the inclined slot is as follows:
for any one inclined slot gray level image, the accumulated sum of Euclidean distances between each material centroid and all inclined slot boundary points with the same ordinate is taken as a second summation factor of each material centroid.
5. The visual inspection method for detecting the retention of the rejected material according to claim 1, wherein the method for obtaining the two-pole gap difference index of each material centroid in each inclined slot gray level image according to the difference between the inclined slot gaps at the two ends of the line segment where the material centroid is located is as follows:
in the method, in the process of the invention,is the two-pole gap difference index of the mass center of the material, +.>Area of upper part u of material line segment, +.>Area of the lower part d of the material line segment, +.>The area of the ith material area on the upper part of the material line segment, p is the number of the material areas on the upper part of the material line segment,/->The area of the j-th material area at the lower part of the material line segment, and q is the number of the material areas at the lower part of the material line segment.
6. The method for visually detecting the retention of the waste picking and receiving materials according to claim 1, wherein the method for obtaining the retention information positioning coefficient of each material centroid in each inclined groove gray level image according to the retention position influence factor of each material centroid in each inclined groove gray level image and the two-pole gap difference index is characterized by comprising the following steps:
and for any inclined groove gray level image, taking the product of the detention position influence factor of each material centroid in the inclined groove gray level image and the gap difference index of the two poles as the detention information positioning coefficient of each material centroid in the inclined groove gray level image.
7. The method for visually detecting the retention of the waste picking and receiving materials according to claim 1, wherein the method for obtaining the weight value in the bipartite graph of the inclined groove gray level image acquired at the adjacent time point according to the Euclidean distance between the material centroids in the inclined groove gray level image acquired at the adjacent time point and the difference between the retention information positioning coefficients of the material centroids is as follows:
for the gray level images of the inclined grooves acquired at adjacent time points, a first product factor is acquired according to the difference between the retention information positioning coefficients of the mass centers of the materials;
the Euclidean distance between the mass centers of materials in the inclined groove gray level images acquired by adjacent time points is used as a second product factor;
and taking a negative mapping result of the product between the first product factor and the second product factor as a weight value in a bipartite graph of the inclined slot gray level image acquired at the adjacent time point.
8. The method for visually detecting the retention of the rejected material according to claim 7, wherein the method for obtaining the first product factor according to the difference between the retention information positioning coefficients of the mass center of the material is as follows:
and taking the absolute value of the difference value between the retention information positioning coefficients of the mass centers of two materials in different inclined groove gray level images as a first product factor of the inclined groove gray level images acquired at the adjacent time points for the inclined groove gray level images acquired at the adjacent time points.
9. The visual inspection method for retention of rejected material according to claim 1, wherein the weighted bipartite graph of the gray level image of the inclined slot collected at adjacent time points is used as input of a KM optimal matching algorithm, and the method for obtaining the retention material in the inclined slot according to output of the KM optimal matching algorithm comprises the following steps:
taking all material centroids in the inclined groove gray level image of the current acquisition time as a current set, taking all material centroids in the inclined groove gray level image of the next acquisition time as a matching set, and carrying out connection construction edges on each element of the current set and each element in the matching set to construct a bipartite graph of the inclined groove gray level image acquired at the adjacent time point;
giving weight values in bipartite graphs of the inclined slot gray images acquired at adjacent time points to each edge of the bipartite graphs, taking the bipartite graph with the weight values of the inclined slot gray images acquired at the adjacent time points as input of a KM optimal matching algorithm, and taking output of KM optimal matching as a matching set;
and acquiring position information in the gray level images of the inclined grooves, acquired at different time points, of each material according to the matching set, and taking the material which does not change in position within the preset time as a retention material.
10. The visual inspection method for retention of waste picking and receiving materials according to claim 1, wherein the method for obtaining the severity of retention of materials in the waste picking and receiving equipment according to the area of the inclined chute occupied by the retained materials in the inclined chute and obtaining the visual inspection result of the waste picking and receiving materials according to the severity of retention of materials is as follows:
and acquiring the material retention ratio of the total area of the waste rejecting inclined groove area occupied by the material retention area according to the material retention area, and judging that the material retention condition is serious in the visual detection result of the waste rejecting and receiving when the material retention ratio is larger than a preset parameter, wherein the retained material in the waste rejecting and receiving equipment is required to be removed.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311152699.9A CN116883413B (en) | 2023-09-08 | 2023-09-08 | Visual detection method for retention of waste picking and receiving materials |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311152699.9A CN116883413B (en) | 2023-09-08 | 2023-09-08 | Visual detection method for retention of waste picking and receiving materials |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116883413A true CN116883413A (en) | 2023-10-13 |
| CN116883413B CN116883413B (en) | 2023-12-01 |
Family
ID=88257236
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311152699.9A Active CN116883413B (en) | 2023-09-08 | 2023-09-08 | Visual detection method for retention of waste picking and receiving materials |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116883413B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117474924A (en) * | 2023-12-28 | 2024-01-30 | 山东鲁抗医药集团赛特有限责任公司 | Label defect detection method based on machine vision |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110007945A1 (en) * | 2008-03-14 | 2011-01-13 | Gerhard Youssefi | Fast algorithm for streaming wavefront |
| CN105678810A (en) * | 2016-01-12 | 2016-06-15 | 上海交通大学 | Cell tracking method based on global and local optimum |
| CN106683077A (en) * | 2016-12-07 | 2017-05-17 | 华南理工大学 | Escalator floor board large-object retention detection method |
| CN108053427A (en) * | 2017-10-31 | 2018-05-18 | 深圳大学 | A kind of modified multi-object tracking method, system and device based on KCF and Kalman |
| JP2018162997A (en) * | 2017-03-24 | 2018-10-18 | 株式会社ミツトヨ | Information processing device and program |
| CN115170608A (en) * | 2022-06-27 | 2022-10-11 | 中冶京诚工程技术有限公司 | Material tracking method and device |
| CN115294123A (en) * | 2022-10-08 | 2022-11-04 | 淄博永丰环保科技有限公司 | Corrugated board quality detection method based on image vision |
| CN115830536A (en) * | 2022-11-28 | 2023-03-21 | 湖南大学 | Method and device for detecting staying object and storage medium |
| CN116245813A (en) * | 2023-01-03 | 2023-06-09 | 中冶南方工程技术有限公司 | Double-layer shear waste retention detection method, terminal equipment and storage medium |
| CN116363127A (en) * | 2023-05-31 | 2023-06-30 | 临沂中科芯华新材料科技有限公司 | Image processing-based quality detection method for fully-degradable plastic product |
| CN116703913A (en) * | 2023-08-07 | 2023-09-05 | 山东大拇指喷雾设备有限公司 | Spraying quality detection method of sprayer |
-
2023
- 2023-09-08 CN CN202311152699.9A patent/CN116883413B/en active Active
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110007945A1 (en) * | 2008-03-14 | 2011-01-13 | Gerhard Youssefi | Fast algorithm for streaming wavefront |
| CN105678810A (en) * | 2016-01-12 | 2016-06-15 | 上海交通大学 | Cell tracking method based on global and local optimum |
| CN106683077A (en) * | 2016-12-07 | 2017-05-17 | 华南理工大学 | Escalator floor board large-object retention detection method |
| JP2018162997A (en) * | 2017-03-24 | 2018-10-18 | 株式会社ミツトヨ | Information processing device and program |
| CN108053427A (en) * | 2017-10-31 | 2018-05-18 | 深圳大学 | A kind of modified multi-object tracking method, system and device based on KCF and Kalman |
| CN115170608A (en) * | 2022-06-27 | 2022-10-11 | 中冶京诚工程技术有限公司 | Material tracking method and device |
| CN115294123A (en) * | 2022-10-08 | 2022-11-04 | 淄博永丰环保科技有限公司 | Corrugated board quality detection method based on image vision |
| CN115830536A (en) * | 2022-11-28 | 2023-03-21 | 湖南大学 | Method and device for detecting staying object and storage medium |
| CN116245813A (en) * | 2023-01-03 | 2023-06-09 | 中冶南方工程技术有限公司 | Double-layer shear waste retention detection method, terminal equipment and storage medium |
| CN116363127A (en) * | 2023-05-31 | 2023-06-30 | 临沂中科芯华新材料科技有限公司 | Image processing-based quality detection method for fully-degradable plastic product |
| CN116703913A (en) * | 2023-08-07 | 2023-09-05 | 山东大拇指喷雾设备有限公司 | Spraying quality detection method of sprayer |
Non-Patent Citations (1)
| Title |
|---|
| 陈子文;李伟;张文强;李云伍;李明生;李慧;: "基于自动Hough变换累加阈值的蔬菜作物行提取方法研究", 农业工程学报, no. 22 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117474924A (en) * | 2023-12-28 | 2024-01-30 | 山东鲁抗医药集团赛特有限责任公司 | Label defect detection method based on machine vision |
| CN117474924B (en) * | 2023-12-28 | 2024-03-15 | 山东鲁抗医药集团赛特有限责任公司 | Label defect detection method based on machine vision |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116883413B (en) | 2023-12-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116883413B (en) | Visual detection method for retention of waste picking and receiving materials | |
| CN113989279B (en) | Plastic film quality detection method based on artificial intelligence and image processing | |
| CN115351598A (en) | Numerical control machine tool bearing detection method | |
| CN115082418B (en) | Precise identification method for automobile parts | |
| EP2548147B1 (en) | Method to recognize and classify a bare-root plant | |
| CN110120042B (en) | Crop image pest and disease damage area extraction method based on SLIC super-pixel and automatic threshold segmentation | |
| CN108711148B (en) | Tire defect intelligent detection method based on deep learning | |
| CN109636824B (en) | Multi-target counting method based on image recognition technology | |
| CN116205919B (en) | Hardware part production quality detection method and system based on artificial intelligence | |
| CN115018838B (en) | Method for identifying pitting defects on surface of oxidized steel pipe material | |
| CN115100199B (en) | Method for detecting wafer low-texture defects | |
| CN106485708A (en) | A kind of round log method of counting based on image recognition | |
| CN109684938A (en) | It is a kind of to be taken photo by plane the sugarcane strain number automatic identifying method of top view based on crop canopies | |
| CN110443278B (en) | Method, device and equipment for detecting thickness abnormality of grid line of solar cell | |
| CN109215009A (en) | Continuous casting billet surface image defect inspection method based on depth convolutional neural networks | |
| CN116309577B (en) | Intelligent detection method and system for high-strength conveyor belt materials | |
| CN111398287A (en) | Battery pole piece scratch detection system and detection method | |
| CN115601307A (en) | Automatic cell detection method | |
| CN117115148B (en) | Chip surface defect intelligent identification method based on 5G technology | |
| CN108765426A (en) | automatic image segmentation method and device | |
| CN117237747A (en) | Hardware defect classification and identification method based on artificial intelligence | |
| CN115797361A (en) | Aluminum template surface defect detection method | |
| CN110866917A (en) | Tablet type and arrangement mode identification method based on machine vision | |
| CN115254655A (en) | Multi-index passion fruit quality grading method based on machine vision | |
| CN110694940A (en) | Control method and system for adjusting blowing of spray valve in real time based on dead pixel and size |
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 |