CN117671014B - Mechanical arm positioning grabbing method and system based on image processing - Google Patents
Mechanical arm positioning grabbing method and system based on image processing Download PDFInfo
- Publication number
- CN117671014B CN117671014B CN202410145369.5A CN202410145369A CN117671014B CN 117671014 B CN117671014 B CN 117671014B CN 202410145369 A CN202410145369 A CN 202410145369A CN 117671014 B CN117671014 B CN 117671014B
- Authority
- CN
- China
- Prior art keywords
- pixel point
- fuzzy
- value
- target area
- reference image
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 79
- 239000013598 vector Substances 0.000 claims abstract description 163
- 230000008859 change Effects 0.000 claims abstract description 52
- 238000012216 screening Methods 0.000 claims abstract description 7
- 230000001133 acceleration Effects 0.000 claims description 19
- 230000008569 process Effects 0.000 claims description 14
- 238000004364 calculation method Methods 0.000 claims description 11
- 238000006073 displacement reaction Methods 0.000 claims description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000001105 regulatory effect Effects 0.000 claims description 4
- 238000012163 sequencing technique Methods 0.000 claims description 3
- 230000007547 defect Effects 0.000 description 43
- 239000003814 drug Substances 0.000 description 34
- 229940079593 drug Drugs 0.000 description 6
- 230000002950 deficient Effects 0.000 description 5
- 230000000007 visual effect Effects 0.000 description 5
- 230000004438 eyesight Effects 0.000 description 4
- 239000000758 substrate Substances 0.000 description 4
- 230000000694 effects Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000010606 normalization Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 230000001276 controlling effect Effects 0.000 description 1
- 239000006185 dispersion Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000016776 visual perception Effects 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image enhancement, in particular to a mechanical arm positioning and grabbing method and system based on image processing. The method comprises the steps of obtaining a preset number of frame gray level images; obtaining a motion vector according to the position change of pixel points in a target area of two adjacent frames of gray images, and screening out a fuzzy area; taking the gray level image of the fuzzy area and all gray level images before the gray level image as a reference image; acquiring a fuzzy value of a pixel point according to the motion vectors in the reference image and the reference image of the previous frame; according to the change of the fuzzy value of the same pixel point in the reference image, acquiring an actual fuzzy value, adjusting a gray value, acquiring an enhanced fuzzy region, and carrying out mechanical arm positioning grabbing. According to the invention, the actual fuzzy value of the pixel points in the fuzzy area is obtained, the gray value of each pixel point in the fuzzy area is adjusted, the fuzzy area is enhanced, the target object is accurately positioned and identified, and the accuracy of positioning and grabbing the target object by the mechanical arm is improved.
Description
Technical Field
The invention relates to the technical field of image enhancement, in particular to a mechanical arm positioning and grabbing method and system based on image processing.
Background
The mechanical arm positioning and grabbing refers to acquiring the position information of a target object through visual perception or other sensors, and then using the mechanical arm to execute accurate grabbing actions. In the existing method, a vision sensor is arranged at the tail end of a mechanical arm to acquire a shooting picture, and a movement path of the mechanical arm is planned according to the position of a target object in the shooting picture so as to accurately grasp the target object.
In practical situations, the target object is in a moving state on the conveyor belt, and when the vision sensor analyzes the captured image of each frame of target object, the motion blur phenomenon of the target object easily occurs in the image of each frame of target object, so that the positioning of the target object is inaccurate, and the mechanical arm can not accurately grasp the target object.
Disclosure of Invention
In order to solve the technical problem that the target object is positioned inaccurately due to the fact that the motion blur phenomenon of the target object easily occurs in the image of each frame of the target object, the invention aims to provide a mechanical arm positioning and grabbing method and system based on image processing, and the adopted technical scheme is as follows:
In a first aspect, an embodiment of the present invention provides a method for positioning and grabbing a mechanical arm based on image processing, where the method includes the following steps:
Acquiring gray images of a target object of a continuous preset number of frames;
Acquiring a target area in each frame of gray level image according to the size of the connected domain in the gray level image; obtaining a motion vector of each pixel point in a target area according to the position change of the same pixel point in the target area of two adjacent frames of gray images; obtaining an enhancement value of each target area according to the change condition of the length of the motion vector in each target area, and screening out a fuzzy area;
Taking the gray level image of the fuzzy area and all gray level images before the gray level image as a reference image of the fuzzy area; according to the change speed of the moving vector angle of each pixel point in the target area of each frame of reference image, the angle between the moving vector of each pixel point in the target area of each frame of reference image and the moving vector of the same pixel point in the previous frame of reference image is changed, the enhancement value of the target area of each frame of reference image and the fluctuation of the moving vector module length difference between each pixel point in the target area of each frame of reference image and surrounding pixel points are obtained, and the fuzzy value of each pixel point in the target area of each frame of reference image is obtained;
Correcting the fuzzy value of each pixel point in the fuzzy area according to the change trend of the fuzzy value of the same pixel point in the corresponding reference image of each pixel point in the fuzzy area, and obtaining the actual fuzzy value of each pixel point in the fuzzy area;
And adjusting the gray value of each pixel point in the fuzzy area according to the actual fuzzy value to obtain an enhanced fuzzy area, and carrying out mechanical arm positioning grabbing.
Further, the method for acquiring the target area comprises the following steps:
Acquiring a connected domain in each frame of gray level image through a connected domain algorithm;
Taking the connected domain with the largest area in each frame of gray level image as a target area in each frame of gray level image; the target areas in each frame of gray level image are the areas where the target objects are located, and the number of pixel points in each target area is the same.
Further, the method for obtaining the enhancement value of each target area and screening out the fuzzy area according to the change condition of the length of the motion vector in each target area comprises the following steps:
For any target area, acquiring the length of a motion vector of each pixel point in the target area;
obtaining the addition result of the maximum movement vector length and the minimum movement vector length in the target area as the change weight of the movement vector length in the target area;
Adjusting the difference value between the maximum motion vector length and the minimum motion vector length in the target area through the change weight to obtain the overall change value of the target area;
acquiring an enhancement value of the target area according to the overall change value of the target area and the variance of the length of the motion vector in the target area;
and when the normalized enhancement value is larger than a preset enhancement value threshold, taking the corresponding target area as a fuzzy area.
Further, the calculation formula of the enhancement value is:
In the method, in the process of the invention, Enhancement value for the q-th target region; /(I)The maximum motion vector length in the q-th target area; The minimum motion vector length in the q-th target area; /(I) A change weight for the q-th target region; is the variance of the length of the motion vector in the q-th target region.
Further, according to the change speed of the moving vector angle of each pixel point in the target area of each frame of reference image, the angle between the moving vector of each pixel point in the target area of each frame of reference image and the moving vector of the same pixel point in the previous frame of reference image changes, and the fluctuation of the mode length difference of the moving vector between each pixel point in the target area of each frame of reference image and the surrounding pixel points, the method for obtaining the fuzzy value of each pixel point in the target area of each frame of reference image comprises the following steps:
Acquiring the size of a time interval between two adjacent frames of reference images as a first duration; wherein, the time interval between every two adjacent frames of reference images is equal;
Acquiring the ratio of the moving vector angle of each pixel point in the target area of the reference image of the k frame to the first duration, and taking the ratio as the angular speed of the moving vector of the corresponding pixel point in the target area of the reference image of the k frame;
Acquiring the angular acceleration of the motion vector of each pixel point in the target area of the kth frame reference image according to the difference of the motion vector angle between each pixel point in the target area of the kth frame reference image and the same pixel point in the (k-1) frame reference image, the first duration and the angular velocity of the motion vector of each pixel point in the target area of the (k-1) frame reference image;
And acquiring a fuzzy value of each pixel point in the target area of the k-th frame reference image according to the angular speed and the angular acceleration of the motion vector of each pixel point in the target area of the k-th frame reference image, the enhancement value of the target area of the k-th frame reference image and the fluctuation of the motion vector mode length difference between each pixel point in the target area of the k-th frame reference image and the four-adjacent pixel points.
Further, the calculation formula of the angular acceleration is as follows:
In the method, in the process of the invention, Angular acceleration of a motion vector of an ith pixel point in a target area of a kth frame reference image; the moving vector angle of the ith pixel point in the target area of the kth frame reference image; /(I) The size of the displacement vector angle of the ith pixel point in the target area of the (k-1) th frame reference image; /(I)An angular velocity of a motion vector for an i-th pixel point within a target region of a (k-1) -th frame reference image; t is a first duration.
Further, the calculation formula of the fuzzy value is as follows:
In the method, in the process of the invention, The fuzzy value of the ith pixel point in the target area of the kth frame reference image is obtained; /(I)Enhancement values for a target region of a kth frame reference image; /(I)The moving vector angle of the ith pixel point in the target area of the kth frame reference image; t is a first duration; /(I)The angular velocity of the motion vector of the ith pixel point in the target area of the kth frame reference image; /(I)Angular acceleration of a motion vector of an ith pixel point in a target area of a kth frame reference image; n is the total number of the neighborhood pixel points of the ith pixel point in the target area of the kth frame reference image; /(I)The motion vector module length of the ith pixel point in the target area of the kth frame reference image is the motion vector module length; /(I)The motion vector module length of the n neighborhood pixel point of the i pixel point in the target area of the k frame reference image is the motion vector module length of the n neighborhood pixel point; /(I)And the average value of the motion vector modular length difference value of the ith pixel point and each neighborhood pixel point in the target area of the kth frame reference image.
Further, the method for acquiring the actual fuzzy value comprises the following steps:
For any fuzzy region, taking the j-th pixel point in the fuzzy region as a target pixel point;
Sequencing the fuzzy values of the pixel points which are the same as the target pixel point in the reference image of the fuzzy region according to the time sequence to obtain a fuzzy value sequence;
Subtracting the adjacent previous fuzzy value from the last fuzzy value of the fuzzy value sequence, and taking the obtained difference value as a target difference value;
stopping obtaining the target difference value when the target difference value is smaller than a preset target difference value threshold;
the result of accumulating the obtained target difference is used as a first result;
taking the quantity of fuzzy values participating in acquiring a first result as a first quantity;
Taking the product of the first result and the first number as an adjustment value;
taking the normalized regulating value as a first weight of a j-th pixel point in the fuzzy area;
and taking the product of the first weight of the jth pixel point in the fuzzy area and the fuzzy value as the actual fuzzy value of the jth pixel point in the fuzzy area.
Further, the method for adjusting the gray value of each pixel point in the fuzzy area according to the actual fuzzy value comprises the following steps:
Normalizing the actual fuzzy value of each pixel point in the fuzzy area to obtain a fuzzy characteristic value of the corresponding pixel point;
the result of the negative correlation of the fuzzy characteristic value is used as a non-fuzzy characteristic value of the corresponding pixel point;
when the fuzzy characteristic value is larger than a preset fuzzy characteristic value threshold value, taking the product of the gray value of the corresponding pixel point and the non-fuzzy characteristic value as an adjustment gray value of the corresponding pixel point;
When the fuzzy demarcation value is smaller than or equal to a preset demarcation value threshold value, obtaining the product of the gray value of the corresponding pixel point and the non-fuzzy characteristic value as a gray adjustment value of the corresponding pixel point; and taking the addition result of the gray value of the corresponding pixel point and the gray adjustment value as the adjustment gray value of the corresponding pixel point.
In a second aspect, another embodiment of the present invention provides an image processing-based mechanical arm positioning and grabbing system, which includes: a memory, a processor and a computer program stored in the memory and executable on the processor, the processor implementing the steps of any one of the methods described above when executing the computer program.
The invention has the following beneficial effects:
Acquiring a target area in each frame of gray level image according to the size of the connected domain in the gray level image, determining the position of a target object in each frame of gray level image, and preliminarily determining the target object in each frame of gray level image for positioning and identifying; according to the position change of the same pixel point in the target area of two adjacent frames of gray images, a motion vector of each pixel point in the target area is obtained, preparation is made for analyzing whether motion blur phenomenon exists in each target area, further, according to the change condition of the motion vector length in each target area, an enhancement value of each target area is obtained, a blur area is screened out, analysis is carried out on the blur area, and the efficiency of processing motion blur defects is improved; in order to accurately strengthen the fuzzy area, a gray level image in which the fuzzy area is positioned and all gray level images in front of the fuzzy area are used as reference images of the fuzzy area, the actual fuzzy value of each pixel point in the fuzzy area is accurately and efficiently obtained based on the change and distribution condition of the motion vector of each pixel point in the target area in the reference images, the gray level value of each pixel point in the fuzzy area is accurately adjusted, the fuzzy area is accurately strengthened, and then the target object in each frame of gray level image is accurately positioned and identified, so that the mechanical arm accurately and efficiently grabs the target object.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions and advantages of the prior art, the following description will briefly explain the drawings used in the embodiments or the description of the prior art, and it is obvious that the drawings in the following description are only some embodiments of the invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
Fig. 1 is a flowchart of a mechanical arm positioning and grabbing method based on image processing according to an embodiment of the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following is a detailed description of a specific implementation, structure, characteristics and effects of the mechanical arm positioning and grabbing method and system based on image processing according to the invention with reference to the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The invention provides a specific scheme of a mechanical arm positioning and grabbing method based on image processing, which is specifically described below with reference to the accompanying drawings.
The embodiment of the invention has the following specific scene: in order to enable the mechanical arm to accurately grasp the medicine, the embodiment of the invention sets the background of the conveyor belt to be black, the medicine is in an obvious bright area on the conveyor belt, and in order to accurately position and identify the medicine to be grasped, the embodiment of the invention carries out gray processing on each frame of acquired image to acquire a gray image, and identifies and positions the medicine to be grasped on the basis of the gray image. The graying process is the prior art, and will not be described in detail.
The aim of the embodiment of the invention is as follows: because the medicine to be grabbed is in dynamic change on the conveyor belt, motion blurring phenomenon easily exists in each frame of gray level image acquired by the vision sensor, and then double image phenomenon exists in the medicine to be grabbed, so that the medicine to be grabbed cannot be accurately positioned and grabbed by the mechanical arm. Therefore, according to the embodiment of the invention, each frame of gray level image is analyzed, the area where the medicine to be grabbed is located in each frame of gray level image is obtained as a target area, the motion vector of each pixel point in the target area is obtained according to the position change of the same pixel point in the target areas of two adjacent frames of gray level images, the fuzzy area with motion fuzzy defects is screened out according to the change of the motion vector in each target area, the fuzzy area is analyzed, the gray level of each pixel point in the fuzzy area is adjusted, the fuzzy area is enhanced, the medicine to be grabbed is accurately positioned, and the mechanical arm is enabled to accurately and efficiently grab the medicine to be grabbed.
Referring to fig. 1, a flowchart of a positioning and grabbing method for a mechanical arm based on image processing according to an embodiment of the invention is shown, and the method includes the following steps:
Step S1: and acquiring gray images of the target object of the continuous preset number of frames.
Specifically, the embodiment of the invention takes the medicine on the mechanical arm grabbing conveyor belt as an example for analysis. Wherein each drug on the conveyor belt is identical. The mechanical arm is arranged at one side of the end of the conveyor belt, the visual sensor is arranged at the tail end of the mechanical arm, the visual sensor faces the conveyor belt at a 45-degree depression angle, and medicines in the moving direction of the conveyor belt are shot. The mechanical arm grabs the medicine closest to the mechanical arm on the conveyor belt, so that the target object in the embodiment of the invention is the medicine closest to the mechanical arm on the conveyor belt. In the process of positioning and identifying the medicines on the conveyor belt by the mechanical arm, acquiring videos acquired by shooting by a visual sensor at the tail end of the mechanical arm, wherein the acquisition frame rate of the visual sensor is 30 frames per second, and in order to improve the grabbing efficiency of the mechanical arm, the embodiment of the invention acquires the last 10 continuous frames of images containing the medicines to be grabbed in the last second before the mechanical arm grabs, and an implementer can set continuous preset number of frames of images according to actual conditions without limitation. Wherein the number of pixels in each frame of image is the same.
Step S2: acquiring a target area in each frame of gray level image according to the size of the connected domain in the gray level image; obtaining a motion vector of each pixel point in a target area according to the position change of the same pixel point in the target area of two adjacent frames of gray images; and obtaining the enhancement value of each target area according to the change condition of the length of the motion vector in each target area, and screening out the fuzzy area.
Specifically, it is known that the background is black in each frame of gray level image, and the region where the medicine is located is a region with a larger gray level value, so that the connected domain in each frame of gray level image is acquired through the connected domain algorithm, and one connected domain defaults to one medicine. The connected domain algorithm is not described in detail in the prior art. The acquired gray level images of each frame may have a plurality of medicines, namely a plurality of connected domains, because the mechanical arm grabs the medicines with the closest distance, according to the visual principle of near-far and far-far, the area of each connected domain in each gray level image of each frame is acquired, and the connected domain with the largest area in each gray level image of each frame is taken as the target area in each gray level image of each frame, wherein the target area in each gray level image of each frame is the area where the target object, namely the medicine to be grabbed, is located, and therefore, the number of pixels in each target area is the same and only one target area exists in each gray level image of each frame. The method for obtaining the area of the connected domain is the prior art, and will not be described in detail. So far, determining the region to be grabbed of the medicine in each frame of gray level image, namely a target region. Because the medicine to be grabbed moves on the conveyor belt, motion blur defects can be generated in target areas in each frame of gray level image, and in order to analyze whether the motion blur defects exist in each target area, the embodiment of the invention acquires the motion vector of each pixel point in the target area according to the position change of the same pixel point in the target areas of two adjacent frames of gray level images, wherein the same pixel points in different target areas refer to the pixel points representing the same area of the medicine to be grabbed. And then, according to the change condition of the length of the motion vector in each target area, a target area with motion blur defect, namely a blur area, is screened out, and the blur area is enhanced subsequently, so that the medicine to be grabbed is accurately positioned and identified. The specific method for acquiring the fuzzy area is as follows:
(1) A motion vector is acquired.
Preferably, the method for acquiring the motion vector comprises the following steps: and (3) obtaining a motion vector from each pixel point in the target area of the (i-1) th frame gray image to the same pixel point in the target area of the i th frame gray image by an optical flow method, and taking the motion vector as a motion vector of a corresponding pixel point in the target area of the i th frame gray image. The optical flow method is a known technique and will not be described in detail. Since the first frame gray-scale image does not have an adjacent previous frame gray-scale image, the target region of the first frame gray-scale image is not analyzed.
(2) A blurred region is acquired.
Since the motion blur defect is a ghost caused by the motion of the medicine to be grabbed in the time interval between two adjacent frames of gray images, when the lengths of the motion vectors in the target area are more similar, the clearer the target area is, and the motion blur defect is not existed; when the motion vector lengths in the target area are not uniform, the more likely the target area has motion blur defects. Therefore, the enhancement value of each target area is acquired according to the change of the length of the motion vector in each target area, and the fuzzy area is screened out.
Preferably, the method for acquiring the blurred region comprises the following steps: for any target area, acquiring the length of a motion vector of each pixel point in the target area; and obtaining the addition result of the maximum moving vector length and the minimum moving vector length in the target area as the changing weight of the moving vector length in the target area. The larger the change weight, the larger the length of the motion vector in the target area, and the more likely the motion blur defect exists in the target area. In order to further determine the motion blur defect condition of the target area, the difference value between the maximum motion vector length and the minimum motion vector length in the target area is adjusted through the change weight, and the overall change value of the target area is obtained. The larger the overall change value, the more likely the motion blur defect is present in the target region. And further, obtaining the enhancement value of the target area according to the overall change value of the target area and the variance of the length of the motion vector in the target area. The larger the enhancement value, the more likely the target region is that motion blur defects are present, and the more enhancement is required for the target region. And when the normalized enhancement value is larger than a preset enhancement value threshold, taking the corresponding target area as a fuzzy area. The blurred region is the target region that needs to be enhanced.
As an example, taking the q-th target area as an example, a calculation formula for obtaining the enhancement value of the q-th target area is:
In the method, in the process of the invention, Enhancement value for the q-th target region; /(I)The maximum motion vector length in the q-th target area; The minimum motion vector length in the q-th target area; /(I) A change weight for the q-th target region; is the variance of the length of the motion vector in the q-th target region.
It should be noted that the number of the substrates,The larger the motion vector length in the q-th target area, the more the motion blur defect is more likely to exist in the q-th target area, and the enhancement is needed, namely the greater the change degree of the motion vector length in the q-th target area isThe larger; at the position ofThe greater the basis, the more weight/>The larger the q-th target area, the more the change degree of the length of the motion vector of each pixel point in the q-th target area is, the more motion blur defect is likely to exist in the q-th target area,/>The larger; The larger the motion vector length in the q-th target area is, the more chaotic the motion vector length in the q-th target area is, the more motion blur defect is likely to exist in the q-th target area,/> The larger; thus,/>The larger the q-th target region, the more likely the motion blur defect is present, and the more enhancement is required.
And acquiring the enhancement value of each target area according to the method for acquiring the enhancement value of the q-th target area. In the embodiment of the invention, the preset enhancement value threshold is set to be 0.7, and the magnitude of the preset enhancement value threshold can be set by an implementer according to actual conditions, so that the method is not limited. And when the normalized enhancement value is larger than a preset enhancement value threshold, taking the corresponding target area as a fuzzy area. The blurred region is the target region that needs to be enhanced. So far, the fuzzy area needing to be enhanced is determined, and the efficiency of positioning and identifying the medicine to be grabbed is improved.
Step S3: taking the gray level image of the fuzzy area and all gray level images before the gray level image as a reference image of the fuzzy area; and according to the change speed of the moving vector angle of each pixel point in the target area of each frame of reference image, the angle between the moving vector of each pixel point in the target area of each frame of reference image and the moving vector of the same pixel point in the previous frame of reference image is changed, the enhancement value of the target area of each frame of reference image and the fluctuation of the moving vector module length difference between each pixel point in the target area of each frame of reference image and surrounding pixel points are obtained, and the fuzzy value of each pixel point in the target area of each frame of reference image is obtained.
Specifically, as the medicine to be grabbed gradually approaches the vision sensor, the motion blur defect of each pixel point in the target area of the corresponding captured gray image is more obvious, and in order to accurately enhance the fuzzy area, the embodiment of the invention analyzes each pixel point in the fuzzy area. In order to accurately and efficiently adjust the gray value of each pixel point in the fuzzy area, the embodiment of the invention takes the gray image of the fuzzy area and all gray images before the gray image as the reference image of the fuzzy area, only analyzes the reference image of each fuzzy area, and improves the efficiency of enhancing each fuzzy area. And then, analyzing the motion vector of each pixel point in the target area of each frame of reference image and the enhancement value of the target area of each frame of reference image, obtaining the fuzzy value of each pixel point in the target area of each frame of reference image, improving the efficiency and accuracy of the actual fuzzy value of each pixel point in the fuzzy area obtained later, further accurately adjusting the gray value of each pixel point in the fuzzy area, accurately enhancing the fuzzy area, and accurately positioning and identifying the medicine to be grabbed.
Preferably, the method for acquiring the blur value of each pixel point in the target area of each frame of reference image comprises the following steps: acquiring the size of a time interval between two adjacent frames of reference images as a first duration; wherein, the time interval between every two adjacent frames of reference images is equal; and acquiring the ratio of the moving vector angle of each pixel point in the target area of the reference image of the k frame to the first duration, and taking the ratio as the angular speed of the moving vector of the corresponding pixel point in the target area of the reference image of the k frame. Because the motion blur defect is formed by the movement of the medicine to be grabbed, when the motion blur defect of a certain pixel point in the target area of the reference image of the k frame is more serious, the distribution of the motion vectors of the pixel point tends to be concentrated to be discrete, so that when the angular speed of the motion vector of the pixel point is higher, the degree of the discrete motion vector of the pixel point is larger, and the motion blur defect of the pixel point is more serious. And acquiring the angular acceleration of the motion vector of each pixel point in the target area of the k-th frame reference image according to the difference of the motion vector angle between each pixel point in the target area of the k-th frame reference image and the same pixel point in the (k-1) -th frame reference image, the first duration and the angular velocity of the motion vector of each pixel point in the target area of the (k-1) -th frame reference image. When the angular acceleration is larger, the discrete trend of the motion vector of the corresponding pixel point is more obvious, and the motion blur defect of the corresponding pixel point is more serious. When the enhancement value of the target area of the kth frame reference image is larger, the pixel point in the target area of the kth frame reference image is more likely to have motion blur defect. And acquiring fluctuation, namely variance, of the mode length difference of the motion vector between each pixel point and the pixel points in the four adjacent domains in the target area of the reference image of the k frame, wherein when the variance is larger, the difference between the motion vector of the corresponding pixel point and the motion vector of other surrounding pixel points is larger, so that the possibility of motion blur defects of the corresponding pixel point is larger. In the embodiment of the invention, the surrounding pixel points of each pixel point are set to be four adjacent areas of each pixel point, and an operator can set the surrounding pixel points of each pixel point according to actual conditions, so that the method is not limited. If a pixel point is a boundary pixel point, and a pixel point in a four-neighborhood is not present, then the neighborhood pixel point present in the four-neighborhood of the pixel point is analyzed. Therefore, the fuzzy value of each pixel point in the target area of the k-th frame reference image is obtained according to the angular speed and the angular acceleration of the movement vector of each pixel point in the target area of the k-th frame reference image, the enhancement value of the target area of the k-th frame reference image and the fluctuation of the movement vector mode length difference between each pixel point in the target area of the k-th frame reference image and the pixel points in the four adjacent domains.
As an example, taking the ith pixel point in the target area of the kth frame reference image as an example, the size of the displacement vector angle of the ith pixel point in the target area of the kth frame reference image isThe first time period is t, so that the angular velocity of the motion vector of the ith pixel point in the target area of the kth frame reference image is/>. According to the method for acquiring the angular velocity of the motion vector of the ith pixel point in the target area of the kth frame of reference image, the angular velocity of the motion vector of each pixel point in the target area of each frame of reference image is acquired. Wherein the ith pixel point in the (k-1) th frame reference image and the ith pixel point in the target area of the (k-1) th frame reference image are pixel points corresponding to the same area in the medicine to be grabbed, so that the displacement vector angle of the ith pixel point in the (k-1) th frame reference image is/>Sum angular velocity/>And the size of the moving vector angle of the ith pixel point in the target area of the first duration and the kth frame reference image is/>Acquiring angular acceleration/>, of a motion vector of an ith pixel point in a target area of a kth frame reference image. The embodiment of the invention is based on the existing formulaTo push over the acquisition angle acceleration, where/>Representing an initial displacement; /(I)Representing the displacement after the change; Representing the displacement variation; /(I) Representing an initial speed; t represents a change time; a represents acceleration. Thus, will/>Instead of x,/>, in the formulaReplace/>, in the formulaAnd/>Replace/>, in the formulaI.e. to obtain the formula/>, In embodiments of the invention、/>And t are scalar quantities, thereby obtaining/>Also scalar. Thus, the angular acceleration of the motion vector of the ith pixel point in the target area of the kth frame reference image is acquiredThe calculation formula of (2) is as follows:
In the method, in the process of the invention, Angular acceleration of a motion vector of an ith pixel point in a target area of a kth frame reference image; the moving vector angle of the ith pixel point in the target area of the kth frame reference image; /(I) The size of the displacement vector angle of the ith pixel point in the target area of the (k-1) th frame reference image; /(I)An angular velocity of a motion vector for an i-th pixel point within a target region of a (k-1) -th frame reference image; t is a first duration.
It should be noted that the number of the substrates,The larger the difference between the movement vector angle of the ith pixel point in the (k-1) th frame reference image and the movement vector angle of the ith pixel point in the target area of the kth frame reference image is, the more the possibility that the motion blur defect exists in the ith pixel point in the target area of the kth frame reference image is indirectly indicated as the larger the differenceThe larger; thus,/>The larger the pixel point is, the greater the possibility that the motion blur defect exists in the ith pixel point in the target area of the kth frame reference image is.
The calculation formula for acquiring the fuzzy value of the ith pixel point in the target area of the kth frame reference image is as follows:
In the method, in the process of the invention, The fuzzy value of the ith pixel point in the target area of the kth frame reference image is obtained; /(I)Enhancement values for a target region of a kth frame reference image; /(I)The moving vector angle of the ith pixel point in the target area of the kth frame reference image; t is a first duration; /(I)The angular velocity of the motion vector of the ith pixel point in the target area of the kth frame reference image; /(I)Angular acceleration of a motion vector of an ith pixel point in a target area of a kth frame reference image; n is the total number of neighbor pixel points of the ith pixel point in the target area of the kth frame reference image, and the embodiment of the invention is 4; /(I)The motion vector module length of the ith pixel point in the target area of the kth frame reference image is the motion vector module length; /(I)The motion vector module length of the n neighborhood pixel point of the i pixel point in the target area of the k frame reference image is the motion vector module length of the n neighborhood pixel point; /(I)And the average value of the motion vector modular length difference value of the ith pixel point and each neighborhood pixel point in the target area of the kth frame reference image.
It should be noted that the number of the substrates,The larger the pixel points in the target area of the reference image of the kth frame are, the greater the possibility of motion blur defects is, namely/>The larger; /(I)The larger, i.e./>The larger the motion blur defect exists in the ith pixel point in the target area of the kth frame reference image, the more serious the motion blur defect exists in the ith pixel point in the target area of the kth frame reference image, namely the greater the degree of dispersion of the motion vector of the ith pixel point in the target area of the kth frame reference imageThe larger; /(I)The larger the motion blur defect is, the more obvious the discrete trend of the motion vector of the ith pixel point in the target area of the kth frame reference image is, and the more serious the motion blur defect of the ith pixel point in the target area of the kth frame reference image is indirectly illustrated as followsThe larger; /(I)The larger the difference between the movement vector module length of the ith pixel point in the target area of the kth frame reference image and the movement vector module length of other surrounding pixel points is larger, the more discrete the movement vector of the ith pixel point in the target area of the kth frame reference image is, namely/>The larger; thus,/>The larger the motion blur defect existing in the ith pixel point in the target area of the kth frame reference image is, the more serious the motion blur defect is.
And acquiring the fuzzy value of each pixel point in the target area of each frame of reference image according to the method for acquiring the fuzzy value of the ith pixel point in the target area of the kth frame of reference image. It should be noted that the first frame of reference image in the reference image of each blurred region does not have an adjacent previous frame of reference image, and thus the target region in the first frame of reference image in the reference image of each blurred region is not analyzed.
Step S4: and correcting the fuzzy value of each pixel point in the fuzzy area according to the change trend of the fuzzy value of the same pixel point in the corresponding reference image of each pixel point in the fuzzy area, and obtaining the actual fuzzy value of each pixel point in the fuzzy area.
Specifically, the fuzzy value is only analyzed between the reference image of the current frame and the reference image of the previous frame, so that in order to accurately determine the actual fuzzy value of each pixel point in the fuzzy area and further accurately enhance the fuzzy area, the embodiment of the invention analyzes the change trend of the fuzzy value of the same pixel point in the corresponding reference image of each pixel point in the fuzzy area. When a certain pixel point in the blurring area has a motion blurring defect, the blurring value of the pixel point which is the same as the pixel point in the reference image changes according with the increasing trend, and the increasing degree is larger. Therefore, according to the change trend of the fuzzy value of each pixel point in the fuzzy area in the corresponding reference image, the fuzzy value of each pixel point in the fuzzy area is corrected, and the actual fuzzy value of each pixel point in the fuzzy area is obtained.
Preferably, the method for acquiring the actual fuzzy value is as follows: for any fuzzy region, taking the j-th pixel point in the fuzzy region as a target pixel point; sequencing the fuzzy values of the pixel points which are the same as the target pixel point in the reference image of the fuzzy region according to the time sequence to obtain a fuzzy value sequence; subtracting the adjacent previous fuzzy value from the last fuzzy value of the fuzzy value sequence, and taking the obtained difference value as a target difference value; stopping obtaining the target difference value when the target difference value is smaller than a preset target difference value threshold; the result of accumulating the obtained target difference is used as a first result; taking the quantity of fuzzy values participating in acquiring a first result as a first quantity; taking the product of the first result and the first number as an adjustment value; taking the normalized regulating value as a first weight of a j-th pixel point in the fuzzy area; and taking the product of the first weight of the jth pixel point in the fuzzy area and the fuzzy value as the actual fuzzy value of the jth pixel point in the fuzzy area.
As an example, taking the jth pixel point in the mth fuzzy area as an example, the jth pixel point in the mth fuzzy area is the target pixel point. If the m-th blurred region is in the fifth frame gray level image, the 5 gray level images of the first frame to the fifth frame are all reference images of the m-th blurred region. And obtaining fuzzy values of pixel points of the same region as the to-be-grabbed medicine represented by the target pixel point in the reference image of the mth fuzzy region, wherein the total number of the fuzzy values is 5, and the 5 fuzzy values are sequenced according to the time sequence to obtain a fuzzy value sequence. The last fuzzy value in the fuzzy value sequence is the fuzzy value of the target pixel point. The difference obtained by subtracting the 4 th fuzzy value from the 5 th fuzzy value in the fuzzy value sequence is the first target difference, and the preset target difference threshold is set to 0 according to the embodiment of the invention, and the operator can set the preset target difference threshold according to the actual situation without limitation. And stopping obtaining the target difference value when the first target difference value is smaller than a preset target difference value threshold, wherein the first target difference value is a first result, and the first number is 2, namely the 5 th fuzzy value and the 4 th fuzzy value respectively. When the first target difference value is larger than or equal to a preset target difference value threshold, the difference value between the 4 th fuzzy value and the 3 rd fuzzy value is obtained and used as a second target difference value, when the second target difference value is larger than or equal to the preset target difference value threshold, the difference value between the 3 rd fuzzy value and the 2 nd fuzzy value is obtained and used as a third target difference value, and the like is performed until the target difference value is smaller than the preset target difference value threshold, and the obtaining of the target difference value is stopped. And accumulating all the obtained target differences to obtain a first result, wherein the number of fuzzy values participating in obtaining the first result is the first number. The product of the first result and the first quantity is the regulating value; the normalized adjustment value is the first weight of the jth pixel point in the mth fuzzy area. According to the first weight and the fuzzy value of the jth pixel point in the mth fuzzy area, the calculation formula for obtaining the actual fuzzy value of the jth pixel point in the mth fuzzy area is as follows:
In the method, in the process of the invention, The actual fuzzy value of the jth pixel point in the mth fuzzy region; /(I)The fuzzy value of the jth pixel point in the mth fuzzy region; /(I)The adjustment value of the jth pixel point in the mth fuzzy area; norm is a normalization function; /(I)Is a first weight.
It should be noted that the number of the substrates,The larger the first weight/>The larger the blur value of the jth pixel point in the mth blur area is, the more accords with the change rule of the blur value of the pixel point with the motion blur defect, and the greater the possibility that the jth pixel point in the mth blur area has the motion blur defect is, thus/>Participation acquisition/>The greater the extent of (2); thus/>The larger the pixel point in the mth fuzzy area is, the more the possibility that the j pixel point in the mth fuzzy area has motion fuzzy defects is, and the greater the gray value of the j pixel point in the mth fuzzy area is adjusted.
According to the method for acquiring the actual fuzzy value of the jth pixel point in the mth fuzzy region, acquiring the actual fuzzy value of each pixel point in each fuzzy region.
Step S5: and adjusting the gray value of each pixel point in the fuzzy area according to the actual fuzzy value to obtain an enhanced fuzzy area, and carrying out mechanical arm positioning grabbing.
Specifically, in order to accurately enhance the fuzzy area, the gray value of each pixel point in the fuzzy area is adjusted according to the actual fuzzy value of each pixel point in the fuzzy area. In order to accurately adjust the gray value of each pixel point in the fuzzy area, the pixel points in the fuzzy area are firstly divided according to the actual fuzzy value of each pixel point in the fuzzy area, the motion fuzzy defective pixel points and the non-motion fuzzy defective pixel points in the fuzzy area are determined, the gray value of the motion fuzzy defective pixel points is weakened, the gray value of the non-motion fuzzy defective pixel points is enhanced, the influence caused by the motion fuzzy defect is reduced, and the fuzzy area is enhanced.
Preferably, the method for adjusting the gray value of each pixel point in the fuzzy area according to the actual fuzzy value is as follows: normalizing the actual fuzzy value of each pixel point in the fuzzy area to obtain a fuzzy characteristic value of the corresponding pixel point; the result of the negative correlation of the fuzzy characteristic value is used as a non-fuzzy characteristic value of the corresponding pixel point; when the fuzzy characteristic value is larger than a preset fuzzy characteristic value threshold value, the corresponding pixel point is a motion fuzzy defect pixel point, and the product of the gray value of the corresponding pixel point and the non-fuzzy characteristic value is used as an adjustment gray value of the corresponding pixel point. When the fuzzy demarcation value is smaller than or equal to a preset demarcation value threshold value, the corresponding pixel point is a non-motion fuzzy defect pixel point, and the product of the gray value of the corresponding pixel point and the non-fuzzy characteristic value is obtained and used as a gray adjustment value of the corresponding pixel point; and taking the addition result of the gray value of the corresponding pixel point and the gray adjustment value as the adjustment gray value of the corresponding pixel point. In the embodiment of the invention, the preset fuzzy characteristic value threshold is set to be 0.6, and the magnitude of the preset fuzzy characteristic value threshold can be set by an implementer according to actual conditions, so that the method is not limited.
Taking the jth pixel point in the mth fuzzy area in the step S4 as an example, the calculation formula for obtaining the adjustment gray value of the jth pixel point in the mth fuzzy area is as follows:
In the method, in the process of the invention, The gray value is adjusted for the j pixel point in the m-th fuzzy area; /(I)The gray value of the jth pixel point in the mth fuzzy area; /(I)The actual fuzzy value of the jth pixel point in the mth fuzzy region; norm is a normalization function; /(I)The fuzzy characteristic value of the jth pixel point in the mth fuzzy region is obtained; /(I)Is the non-fuzzy characteristic value of the j pixel point in the m-th fuzzy area.
When the following is performedIn this case, it is explained that the jth pixel point in the mth blurred region is a non-motion blurred defective pixel point, and the gray value of the jth pixel point in the mth blurred region is enhanced, so that the pixel value is expressed as/>Increase/>, based on; Wherein, when/>When the maximum gray value overflows, the maximum gray value is taken as/>. When (when)In this case, the j pixel point in the m-th blurring region is described as a motion blurring defect pixel point, the gray value of the j pixel point in the m-th blurring region is weakened, and the influence of the motion blurring defect is reduced, so thatAcquisition/>。
According to the method for acquiring the adjustment gray value of the jth pixel point in the mth fuzzy region, the adjustment gray value of each pixel point in each fuzzy region is acquired, the enhancement of each fuzzy region is realized, and the influence caused by motion fuzzy defects is reduced.
According to the obtained enhanced fuzzy region, the medicine to be grabbed in each frame of gray level image is accurately positioned, the mechanical arm accurately calculates the grabbing path, the influence of motion fuzzy defects in each frame of gray level image is eliminated, and the accuracy of grabbing and positioning of the mechanical arm is improved. And after the grabbing path of the mechanical arm is planned and determined, controlling the mechanical arm to move along the planned path in real time. The embodiment of the invention uses a PID (pro-dynamic INTEGRAL DIFFERENTIAL) algorithm to adjust the joint angle and ensure the accurate movement of the mechanical arm. Therefore, the optimization of the positioning grabbing system of the mechanical arm is realized. The PID algorithm is a well-known algorithm, and will not be described in detail.
The present invention has been completed.
In summary, the embodiment of the invention acquires the gray level images of the preset number of frames; obtaining a motion vector according to the position change of pixel points in a target area of two adjacent frames of gray images, and screening out a fuzzy area; taking the gray level image of the fuzzy area and all gray level images before the gray level image as a reference image; acquiring a fuzzy value of a pixel point according to the motion vectors in the reference image and the reference image of the previous frame; according to the change of the fuzzy value of the same pixel point in the reference image, acquiring an actual fuzzy value, adjusting a gray value, acquiring an enhanced fuzzy region, and carrying out mechanical arm positioning grabbing. According to the invention, the actual fuzzy value of the pixel points in the fuzzy area is obtained, the gray value of each pixel point in the fuzzy area is adjusted, the fuzzy area is enhanced, the target object is accurately positioned and identified, and the accuracy of positioning and grabbing the target object by the mechanical arm is improved.
Based on the same inventive concept as the method embodiment, the embodiment of the invention also provides a mechanical arm positioning and grabbing system based on image processing, which comprises the following steps: the system comprises a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor executes the computer program to realize the steps in the embodiment of the mechanical arm positioning grabbing method based on image processing, such as the steps shown in fig. 1. The mechanical arm positioning and grabbing method based on image processing is described in detail in the above embodiments, and will not be described again.
It should be noted that: the sequence of the embodiments of the present invention is only for description, and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the particular order shown, or sequential order, to achieve desirable results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
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.
Claims (8)
1. The mechanical arm positioning and grabbing method based on image processing is characterized by comprising the following steps of:
Acquiring gray images of a target object of a continuous preset number of frames;
Acquiring a target area in each frame of gray level image according to the size of the connected domain in the gray level image; obtaining a motion vector of each pixel point in a target area according to the position change of the same pixel point in the target area of two adjacent frames of gray images; obtaining an enhancement value of each target area according to the change condition of the length of the motion vector in each target area, and screening out a fuzzy area;
Taking the gray level image of the fuzzy area and all gray level images before the gray level image as a reference image of the fuzzy area; according to the change speed of the moving vector angle of each pixel point in the target area of each frame of reference image, the angle between the moving vector of each pixel point in the target area of each frame of reference image and the moving vector of the same pixel point in the previous frame of reference image is changed, the enhancement value of the target area of each frame of reference image and the fluctuation of the moving vector module length difference between each pixel point in the target area of each frame of reference image and surrounding pixel points are obtained, and the fuzzy value of each pixel point in the target area of each frame of reference image is obtained;
Correcting the fuzzy value of each pixel point in the fuzzy area according to the change trend of the fuzzy value of the same pixel point in the corresponding reference image of each pixel point in the fuzzy area, and obtaining the actual fuzzy value of each pixel point in the fuzzy area;
According to the actual fuzzy value, the gray value of each pixel point in the fuzzy area is adjusted to obtain an enhanced fuzzy area, and the mechanical arm positioning and grabbing are carried out;
According to the change speed of the moving vector angle of each pixel point in the target area of each frame of reference image, the angle between the moving vector of each pixel point in the target area of each frame of reference image and the moving vector of the same pixel point in the previous frame of reference image is changed, the enhancement value of the target area of each frame of reference image and the fluctuation of the moving vector module length difference between each pixel point in the target area of each frame of reference image and surrounding pixel points are adopted, the method for acquiring the fuzzy value of each pixel point in the target area of each frame of reference image is as follows:
Acquiring the size of a time interval between two adjacent frames of reference images as a first duration; wherein, the time interval between every two adjacent frames of reference images is equal;
Acquiring the ratio of the moving vector angle of each pixel point in the target area of the reference image of the k frame to the first duration, and taking the ratio as the angular speed of the moving vector of the corresponding pixel point in the target area of the reference image of the k frame;
Acquiring the angular acceleration of the motion vector of each pixel point in the target area of the kth frame reference image according to the difference of the motion vector angle between each pixel point in the target area of the kth frame reference image and the same pixel point in the (k-1) frame reference image, the first duration and the angular velocity of the motion vector of each pixel point in the target area of the (k-1) frame reference image;
Acquiring a fuzzy value of each pixel point in a target area of a kth frame reference image according to the angular speed and the angular acceleration of the motion vector of each pixel point in the target area of the kth frame reference image, the enhancement value of the target area of the kth frame reference image and the fluctuation of the motion vector mode length difference between each pixel point in the target area of the kth frame reference image and the pixel points in the four adjacent domains;
the calculation formula of the fuzzy value is as follows:
In the method, in the process of the invention, The fuzzy value of the ith pixel point in the target area of the kth frame reference image is obtained; /(I)Enhancement values for a target region of a kth frame reference image; /(I)The moving vector angle of the ith pixel point in the target area of the kth frame reference image; t is a first duration; /(I)The angular velocity of the motion vector of the ith pixel point in the target area of the kth frame reference image; Angular acceleration of a motion vector of an ith pixel point in a target area of a kth frame reference image; n is the total number of the neighborhood pixel points of the ith pixel point in the target area of the kth frame reference image; /(I) The motion vector module length of the ith pixel point in the target area of the kth frame reference image is the motion vector module length; /(I)The motion vector module length of the n neighborhood pixel point of the i pixel point in the target area of the k frame reference image is the motion vector module length of the n neighborhood pixel point; /(I)And the average value of the motion vector modular length difference value of the ith pixel point and each neighborhood pixel point in the target area of the kth frame reference image.
2. The method for positioning and grabbing a mechanical arm based on image processing as claimed in claim 1, wherein the method for acquiring the target area is as follows:
Acquiring a connected domain in each frame of gray level image through a connected domain algorithm;
Taking the connected domain with the largest area in each frame of gray level image as a target area in each frame of gray level image; the target areas in each frame of gray level image are the areas where the target objects are located, and the number of pixel points in each target area is the same.
3. The method for positioning and grabbing a mechanical arm based on image processing as claimed in claim 1, wherein the method for obtaining the enhancement value of each target area and screening out the fuzzy area according to the change condition of the length of the motion vector in each target area is as follows:
For any target area, acquiring the length of a motion vector of each pixel point in the target area;
obtaining the addition result of the maximum movement vector length and the minimum movement vector length in the target area as the change weight of the movement vector length in the target area;
Adjusting the difference value between the maximum motion vector length and the minimum motion vector length in the target area through the change weight to obtain the overall change value of the target area;
acquiring an enhancement value of the target area according to the overall change value of the target area and the variance of the length of the motion vector in the target area;
and when the normalized enhancement value is larger than a preset enhancement value threshold, taking the corresponding target area as a fuzzy area.
4. The method for positioning and grabbing a mechanical arm based on image processing as claimed in claim 3, wherein the calculation formula of the enhancement value is:
In the method, in the process of the invention, Enhancement value for the q-th target region; /(I)The maximum motion vector length in the q-th target area; The minimum motion vector length in the q-th target area; /(I) A change weight for the q-th target region; /(I)Is the variance of the length of the motion vector in the q-th target region.
5. The method for positioning and grabbing a mechanical arm based on image processing as claimed in claim 1, wherein the calculation formula of the angular acceleration is:
In the method, in the process of the invention, Angular acceleration of a motion vector of an ith pixel point in a target area of a kth frame reference image; /(I)The moving vector angle of the ith pixel point in the target area of the kth frame reference image; /(I)The size of the displacement vector angle of the ith pixel point in the target area of the (k-1) th frame reference image; /(I)An angular velocity of a motion vector for an i-th pixel point within a target region of a (k-1) -th frame reference image; t is a first duration.
6. The mechanical arm positioning and grabbing method based on image processing as claimed in claim 1, wherein the actual fuzzy value obtaining method is as follows:
For any fuzzy region, taking the j-th pixel point in the fuzzy region as a target pixel point;
Sequencing the fuzzy values of the pixel points which are the same as the target pixel point in the reference image of the fuzzy region according to the time sequence to obtain a fuzzy value sequence;
Subtracting the adjacent previous fuzzy value from the last fuzzy value of the fuzzy value sequence, and taking the obtained difference value as a target difference value;
stopping obtaining the target difference value when the target difference value is smaller than a preset target difference value threshold;
the result of accumulating the obtained target difference is used as a first result;
taking the quantity of fuzzy values participating in acquiring a first result as a first quantity;
Taking the product of the first result and the first number as an adjustment value;
taking the normalized regulating value as a first weight of a j-th pixel point in the fuzzy area;
and taking the product of the first weight of the jth pixel point in the fuzzy area and the fuzzy value as the actual fuzzy value of the jth pixel point in the fuzzy area.
7. The method for positioning and grabbing a mechanical arm based on image processing as claimed in claim 1, wherein the method for adjusting the gray value of each pixel point in the fuzzy area according to the actual fuzzy value is as follows:
Normalizing the actual fuzzy value of each pixel point in the fuzzy area to obtain a fuzzy characteristic value of the corresponding pixel point;
the result of the negative correlation of the fuzzy characteristic value is used as a non-fuzzy characteristic value of the corresponding pixel point;
when the fuzzy characteristic value is larger than a preset fuzzy characteristic value threshold value, taking the product of the gray value of the corresponding pixel point and the non-fuzzy characteristic value as an adjustment gray value of the corresponding pixel point;
When the fuzzy demarcation value is smaller than or equal to a preset demarcation value threshold value, obtaining the product of the gray value of the corresponding pixel point and the non-fuzzy characteristic value as a gray adjustment value of the corresponding pixel point; and taking the addition result of the gray value of the corresponding pixel point and the gray adjustment value as the adjustment gray value of the corresponding pixel point.
8. An image processing-based mechanical arm positioning and grabbing system, comprising a memory, a processor and a computer program stored in the memory and capable of running on the processor, wherein the processor, when executing the computer program, realizes the steps of an image processing-based mechanical arm positioning and grabbing method as claimed in any one of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410145369.5A CN117671014B (en) | 2024-02-02 | 2024-02-02 | Mechanical arm positioning grabbing method and system based on image processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410145369.5A CN117671014B (en) | 2024-02-02 | 2024-02-02 | Mechanical arm positioning grabbing method and system based on image processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117671014A CN117671014A (en) | 2024-03-08 |
| CN117671014B true CN117671014B (en) | 2024-04-19 |
Family
ID=90086686
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410145369.5A Active CN117671014B (en) | 2024-02-02 | 2024-02-02 | Mechanical arm positioning grabbing method and system based on image processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117671014B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009066783A1 (en) * | 2007-11-22 | 2009-05-28 | Semiconductor Energy Laboratory Co., Ltd. | Image processing method, image display system, and computer program |
| CN114419073A (en) * | 2022-03-09 | 2022-04-29 | 荣耀终端有限公司 | Motion blur generation method and device and terminal equipment |
| CN114842323A (en) * | 2022-07-04 | 2022-08-02 | 山东西曼克技术有限公司 | Intelligent robot sorting optimization method based on classification recognition |
| WO2023066072A1 (en) * | 2021-10-20 | 2023-04-27 | 上海微创微航机器人有限公司 | Catheter positioning method, interventional surgery system, electronic device and storage medium |
| CN116128769A (en) * | 2023-04-18 | 2023-05-16 | 聊城市金邦机械设备有限公司 | Track vision recording system of swinging motion mechanism |
| WO2023113143A1 (en) * | 2021-12-16 | 2023-06-22 | Samsung Electronics Co., Ltd. | Method and electronic device for enhancing image quality |
| WO2023134793A2 (en) * | 2022-12-13 | 2023-07-20 | 苏州迈创信息技术有限公司 | Machine vision-based machine tool part online inspection method |
| CN117422627A (en) * | 2023-12-18 | 2024-01-19 | 卓世科技(海南)有限公司 | AI simulation teaching method and system based on image processing |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108055452B (en) * | 2017-11-01 | 2020-09-18 | Oppo广东移动通信有限公司 | Image processing method, device and equipment |
-
2024
- 2024-02-02 CN CN202410145369.5A patent/CN117671014B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009066783A1 (en) * | 2007-11-22 | 2009-05-28 | Semiconductor Energy Laboratory Co., Ltd. | Image processing method, image display system, and computer program |
| WO2023066072A1 (en) * | 2021-10-20 | 2023-04-27 | 上海微创微航机器人有限公司 | Catheter positioning method, interventional surgery system, electronic device and storage medium |
| WO2023113143A1 (en) * | 2021-12-16 | 2023-06-22 | Samsung Electronics Co., Ltd. | Method and electronic device for enhancing image quality |
| CN114419073A (en) * | 2022-03-09 | 2022-04-29 | 荣耀终端有限公司 | Motion blur generation method and device and terminal equipment |
| CN114842323A (en) * | 2022-07-04 | 2022-08-02 | 山东西曼克技术有限公司 | Intelligent robot sorting optimization method based on classification recognition |
| WO2023134793A2 (en) * | 2022-12-13 | 2023-07-20 | 苏州迈创信息技术有限公司 | Machine vision-based machine tool part online inspection method |
| CN116128769A (en) * | 2023-04-18 | 2023-05-16 | 聊城市金邦机械设备有限公司 | Track vision recording system of swinging motion mechanism |
| CN117422627A (en) * | 2023-12-18 | 2024-01-19 | 卓世科技(海南)有限公司 | AI simulation teaching method and system based on image processing |
Non-Patent Citations (2)
| Title |
|---|
| 一种视觉 SLAM 中运动模糊应对方法;郭科，方俊永，王 潇 ，张晓红，刘学;哈尔滨工业大学学报;20190809;第51卷(第8期);全文 * |
| 基于图像序列的模糊聚类嘴唇分割算法;武利娟;王士林;傅剑文;;信息技术;20161225(第12期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117671014A (en) | 2024-03-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN115633259B (en) | Automatic regulation and control method and system for intelligent camera based on artificial intelligence | |
| CN110225226B (en) | Visual tracking system and method | |
| EP2916537B1 (en) | Image processing device | |
| CN105374049B (en) | Multi-corner point tracking method and device based on sparse optical flow method | |
| CN113744315B (en) | Semi-direct vision odometer based on binocular vision | |
| CN111212241A (en) | High-speed automatic driving automatic exposure control method based on image gradient and entropy fusion | |
| CN109978908B (en) | Single-target rapid tracking and positioning method suitable for large-scale deformation | |
| CN117671014B (en) | Mechanical arm positioning grabbing method and system based on image processing | |
| CN112270691B (en) | Monocular video structure and motion prediction method based on dynamic filter network | |
| CN113052794A (en) | Image definition recognition method based on edge features | |
| CN116665097A (en) | Self-adaptive target tracking method combining context awareness | |
| CN109064497B (en) | Video tracking method based on color clustering supplementary learning | |
| CN116489516A (en) | Specific object tracking shooting method and system | |
| CN114923629A (en) | Method for detecting vibration amplitude of spinning high-speed spindle during rotation | |
| CN114598849B (en) | Building construction safety monitoring system based on thing networking | |
| CN111260686A (en) | Target tracking method and system for anti-shielding multi-feature fusion of self-adaptive cosine window | |
| CN111145216A (en) | Tracking method of video image target | |
| CN114972084A (en) | Image focusing accuracy evaluation method and system | |
| CN108734139B (en) | Correlation filtering tracking method based on feature fusion and SVD self-adaptive model updating | |
| CN117557626B (en) | Auxiliary positioning method for spray head installation of aerosol sprayer | |
| CN113177901B (en) | Multi-frame moving image fusion method and system for robot vision | |
| CN116205914B (en) | Waterproof coating production intelligent monitoring system | |
| CN115620286B (en) | Automatic data labeling system and method based on big data | |
| CN111507916B (en) | Method for acquiring stable image of object of article on assembly line | |
| CN113536934B (en) | Active hiding method for patrol robot during tracking task execution |
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 |