CN115100608B - Electric drill shell glass fiber exposure identification method and system - Google Patents
Electric drill shell glass fiber exposure identification method and system Download PDFInfo
- Publication number
- CN115100608B CN115100608B CN202211022236.6A CN202211022236A CN115100608B CN 115100608 B CN115100608 B CN 115100608B CN 202211022236 A CN202211022236 A CN 202211022236A CN 115100608 B CN115100608 B CN 115100608B
- Authority
- CN
- China
- Prior art keywords
- run
- jump
- gray
- gray level
- electric drill
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/50—Extraction of image or video features by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The invention relates to the field of identification methods by using electronic equipment, in particular to an electric drill shell glass fiber exposure identification method and system, which comprises the following steps: acquiring the gray level of a pixel in a gray level image of the electric drill shell; acquiring the gray level jump characteristic of each pixel in the gray level image by utilizing the gray level of each pixel and the circular neighborhood edge pixels of each pixel; acquiring the run length characteristics of each pixel in the gray-scale image according to the run length of each pixel in different directions; constructing a gray level jump run co-occurrence matrix by utilizing the gray level jump characteristics and the run characteristics; obtaining a high-jump long-run advantage and a low-jump short-run advantage in a gray-scale image by using a gray-scale jump run co-occurrence matrix; acquiring a high-jump long-run dominant limit value and a low-jump short-run dominant limit value of a gray scale image; and identifying the electric drill shell according to the high-jump long-run advantage, the low-jump short-run advantage and the corresponding limit values of the high-jump long-run advantage and the low-jump short-run advantage in the gray scale image. The method is used for identifying the exposure of the glass fibers of the electric drill shell, and can improve the accuracy and efficiency of identification.
Description
Technical Field
The invention relates to the field of identification methods by using electronic equipment, in particular to an electric drill shell glass fiber exposure identification method and system.
Background
The electric drill shell is mostly made of glass fiber reinforced nylon, and compared with pure nylon, the glass fiber reinforced nylon has the advantages that the mechanical strength, rigidity, heat resistance, creep resistance and fatigue resistance are greatly improved. The method is widely used for manufacturing plastic parts with heat-resistant and stress-resistant structures. The electric drill shell is usually formed by injection molding, but due to improper process parameter setting in the injection molding process, glass fibers are exposed on the electric drill shell. The glass fiber is exposed because the flowability of the glass fiber relative to the nylon material is much poorer during injection molding. The glass fiber is exposed, so that the appearance of the electric drill shell is influenced, and the glass fiber is unevenly distributed in the electric drill shell, so that the uniformity of the performance of each part of the electric drill shell is influenced, the quality of the electric drill is further influenced, and more serious production accidents can be caused.
At present, the glass fiber exposure is identified by adopting a manual inspection or image processing mode: the manual inspection mode mainly comprises the steps of identifying and judging the injection-molded electric drill shell by means of the experience of a detector; and the image processing mode is to identify the electric drill shell after the injection molding is finished after the image processing.
However, manual inspection is inefficient. And the influence of particle sense and illumination on the surface of the electric drill shell cannot be eliminated by a common image processing mode, so that the electric drill shell with the exposed glass fibers is difficult to identify. Therefore, the invention provides the method and the system for identifying the glass fiber exposure of the electric drill shell, which are used for identifying the glass fiber exposure of the electric drill shell by using electronic equipment and can effectively improve the identification precision and efficiency.
Disclosure of Invention
The invention provides an electric drill shell glass fiber exposure identification method and system, which comprises the following steps: acquiring the gray level of a pixel in a gray level image of the electric drill shell; acquiring the gray level jump characteristic of each pixel in the gray level image by utilizing the gray level of each pixel and the circular neighborhood edge pixels of each pixel; acquiring the run length characteristic of each pixel in the gray-scale image according to the run lengths of the pixels in different directions; constructing a gray level jump run co-occurrence matrix by utilizing the gray level jump characteristics and the run characteristics; obtaining the high-jump long-run advantage and the low-jump short-run advantage in the gray-scale map by utilizing the gray-scale jump run co-occurrence matrix; acquiring a high-jump long-run dominant limit value and a low-jump short-run dominant limit value of a gray scale image; compared with the prior art, the method has the advantages that the gray scale jump characteristic and the run characteristic are obtained by utilizing the gray scale value of the pixels in the image of the electric drill shell, the co-occurrence matrix is further constructed according to the two characteristics, the high-jump long run advantage and the low-jump short run advantage in the gray scale image are calculated according to the co-occurrence matrix, and finally the glass fiber exposure of the electric drill shell is identified by utilizing the high-jump long run advantage and the low-jump short run advantage. The invention applies the electronic equipment to identify the glass fiber exposure of the electric drill shell, and can effectively improve the identification precision and efficiency.
In order to achieve the purpose, the invention adopts the following technical scheme that the method for identifying the exposure of the glass fiber of the electric drill shell comprises the following steps:
and acquiring a gray scale image of the electric drill shell to be identified.
And grading all gray values in the gray image of the electric drill shell to obtain the gray level of each pixel point in the gray image.
And acquiring the gray level jump characteristic of each pixel point in the gray level image by utilizing the gray level of each pixel point in the gray level image and the gray level of the pixel point at the edge of the circular neighborhood.
And acquiring the run length characteristics of each pixel point in the gray-scale image according to the run lengths of each pixel point in the gray-scale image in different directions.
And constructing a gray level jump and run co-occurrence matrix by utilizing the gray level jump characteristics and the run characteristics of all pixel points in the gray level image.
And obtaining the high-jump long-run advantage and the low-jump short-run advantage in the gray-scale map by utilizing the number of the gray-scale jump features, the number of the run features and the co-occurrence frequency of the gray-scale jump features and the run features in the gray-scale jump run co-occurrence matrix.
And acquiring a high-jump long-run dominant limit value and a low-jump short-run dominant limit value of the gray scale image without glass fiber exposure.
And identifying whether the glass fiber is exposed on the shell of the electric drill to be identified according to the ratio of the high-jump long run advantage and the low-jump short run advantage in the gray scale image to the corresponding limit value.
Further, in the electric drill shell glass fiber exposure identification method, the gray level of each pixel point in the gray scale image is obtained according to the following mode:
and obtaining a gray level histogram according to all gray level values in the gray level graph of the electric drill shell.
And performing Gaussian smoothing on the gray level histogram to obtain a local minimum value in the gray level histogram.
And taking each local minimum as a segmentation point, and segmenting all gray values into different levels to obtain the gray level of each pixel point in the gray map.
Further, according to the electric drill shell glass fiber exposure identification method, the gray level jump characteristic of each pixel point in the gray level graph is obtained according to the following mode:
and taking each pixel point in the gray-scale image as a center, and taking R as a radius to make a circle, and obtaining a circular neighborhood of each pixel point in the gray-scale image and a neighborhood edge pixel point.
Judging the gray level of each neighborhood edge pixel point and the gray level of the center pixel point: if the gray level of the neighborhood edge pixel point is greater than that of the central pixel point, the position of the neighborhood edge pixel point is marked as 1, and if the gray level of the neighborhood edge pixel point is less than or equal to that of the central pixel point, the position of the neighborhood edge pixel point is marked as 0.
And taking adjacent neighborhood edge pixel points from 1 to 0 or from 0 to 1 as one jump, and counting the jump times of the neighborhood edge pixel points along the same direction.
And taking the hopping times as the hopping characteristics of the central pixel point to obtain the gray hopping characteristics of each pixel point in the gray map.
Further, according to the electric drill shell glass fiber exposure identification method, the run length characteristics of each pixel point in the gray level graph are obtained according to the following modes:
and counting the continuous same number of gray levels of all pixel points in the gray-scale image in different directions to obtain the run length of each pixel point in the gray-scale image in each direction.
And counting the maximum value of the run of each pixel point in each direction in the gray-scale image, taking the maximum value as the run characteristic of each pixel point, and obtaining the run characteristic of each pixel point in the gray-scale image.
Further, in the method for identifying the exposure of the glass fiber on the electric drill shell, the calculation expression of the high jump length run advantage in the gray scale map is as follows:
in the formula (I), the compound is shown in the specification,for the high transition long run advantage in the grayscale,is shown asA plurality of gray-level jump features, wherein,is shown asOne run bitThe steps of (1) performing the sign,representing gray level jump characteristicsAnd run characteristicsThe number of times of co-occurrence,is the number of run-length features,the number of the gray scale jump features,is aboutAs a function of (c).
Further, in the electric drill housing glass fiber exposure identification method, the calculation expression of the low-jump short-run advantage in the gray scale map is as follows:
in the formula (I), the compound is shown in the specification,for the low jump short run advantage in the grey scale,is shown asA plurality of gray-level jump features, wherein,denotes the firstThe characteristics of the run length are such that,representing gray level jump characteristicsAnd run length characteristicsThe number of times of co-occurrence,is the number of run-length features,the number of the gray scale jump features,to relate toAs a function of (c).
Further, in the electric drill shell glass fiber exposure identification method, the process of identifying whether the electric drill shell to be identified has glass fiber exposure is as follows:
and acquiring a high-jump long-run advantage limit value and a low-jump short-run advantage limit value of the gray-scale image without glass fiber exposure by utilizing the number of the gray-scale jump characteristics, the number of the run characteristics and the co-occurrence frequency of the gray-scale jump characteristics and the run characteristics in the gray-scale jump run co-occurrence matrix.
And acquiring the ratio of the high jump long-run advantage to the corresponding limit value and the ratio of the low jump short-run advantage to the corresponding limit value of the gray scale image.
And judging the two ratios: when the two ratios are less than 1, the electric drill shell to be identified has no glass fiber exposure, and when the ratio in the two ratios is more than or equal to 1, the electric drill shell to be identified has glass fiber exposure.
Further, according to the electric drill shell glass fiber exposure identification method, the gray level image of the electric drill shell to be identified is obtained according to the following mode:
and acquiring an electric drill shell image to be identified after injection molding.
And performing semantic segmentation on the injection-molded electric drill shell image to be identified to obtain the electric drill shell connected domain image to be identified.
And carrying out gray processing on the communicated domain image of the electric drill shell to be identified to obtain a gray image of the electric drill shell to be identified.
The invention also provides an electric drill shell glass fiber exposure identification system, which comprises an acquisition unit, a processing unit, a calculation unit and a control unit:
the acquisition unit is used for acquiring the image of the electric drill shell to be identified, which is subjected to injection molding on the conveying belt, by arranging the camera on the track arranged above the conveying belt.
The processing unit and the data master controller perform semantic segmentation and gray level grading on the image acquired by the acquisition unit to acquire the gray level of each pixel point in the electric drill shell gray level image.
The data master controller of the computing unit computes the gray level jump characteristics and the run characteristics of all pixel points in the gray level image according to the gray level of each pixel point in the gray level image of the electric drill shell, which is obtained by the processing unit, further constructs a co-occurrence matrix according to the two characteristics, and computes the high-jump long-run advantages and the low-jump short-run advantages of the gray level image by utilizing the co-occurrence matrix.
The control unit and the data master controller identify the exposure of the glass fibers in the electric drill shell by utilizing the high-jump long-run advantage and the low-jump short-run advantage of the gray scale image obtained by the calculation unit, and the injection molding controller adjusts the injection molding process parameters according to the identification result.
The invention has the beneficial effects that:
the method comprises the steps of obtaining gray level jump characteristics and run characteristics by utilizing gray level values of pixels in an electric drill shell image, further constructing a co-occurrence matrix according to the two characteristics, calculating high jump long run advantages and low jump short run advantages in a gray level image according to the co-occurrence matrix, and finally identifying glass fiber exposure of the electric drill shell by utilizing the high jump long run advantages and the low jump short run advantages. The invention applies the electronic equipment to identify the glass fiber exposure of the electric drill shell, and can effectively improve the identification precision and efficiency.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.
Fig. 1 is a schematic flow chart of a method for identifying glass fiber exposure of an electric drill housing according to embodiment 1 of the present invention;
fig. 2 is a schematic flow chart of a method for identifying the exposure of glass fibers on an electric drill housing according to embodiment 2 of the present invention;
fig. 3 is a schematic diagram of a circular neighborhood provided in embodiment 2 of the present invention;
fig. 4 is a schematic diagram of a gray level jump run-length co-occurrence matrix according to embodiment 2 of the present invention;
fig. 5 is a block diagram of a system for identifying exposure of glass fibers to an electric drill casing according to embodiment 3 of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
The embodiment provides an electric drill shell glass fiber exposure identification method, which is used for identifying glass fiber exposure of an electric drill shell by using electronic equipment, and can effectively improve identification precision and efficiency.
The embodiment of the invention provides a method for identifying glass fiber exposure of an electric drill shell, which comprises the following steps of:
s101, obtaining a gray scale image of the electric drill shell to be identified.
The method comprises the following steps of carrying out semantic segmentation and graying processing on an electric drill shell image to obtain a grayscale image.
And S102, grading all gray values in the gray image of the electric drill shell to obtain the gray level of each pixel point in the gray image.
Wherein each gray level is obtained by analyzing the gray histogram.
S103, obtaining the gray level jump characteristic of each pixel point in the gray-scale image by utilizing the gray levels of each pixel point in the gray-scale image and the pixel points at the edge of the circular neighborhood.
Wherein, the circular neighborhood takes each pixel point as a central point and the radius asThe circular area of (a).
And S104, acquiring the run length characteristics of each pixel point in the gray-scale image according to the run lengths of each pixel point in the gray-scale image in different directions.
The run is the number of the pixels with continuous and consistent gray levels in a certain direction.
And S105, constructing a gray level jump and run co-occurrence matrix by utilizing the gray level jump characteristics and the run characteristics of all pixel points in the gray level image.
Wherein the values in the matrix represent the number of times that the corresponding gray scale jump characteristic and the run characteristic co-occur.
And S106, obtaining a high-jump long-run advantage and a low-jump short-run advantage in the gray-scale image by utilizing the number of the gray-scale jump features, the number of the run features and the co-occurrence frequency of the gray-scale jump features and the run features in the gray-scale jump run co-occurrence matrix.
When the number of the pixel points of the high-jump long-run is large, the advantage of the high-jump long-run is large.
S107, acquiring a high-jump long-run advantage limit value and a low-jump short-run advantage limit value of the gray-scale image without exposed glass fibers.
Wherein the limit value is used for the subsequent identification of the exposed glass fiber.
And S108, identifying whether the glass fiber exposure exists in the electric drill shell to be identified according to the ratio of the high-jump long run advantage and the low-jump short run advantage in the gray scale image to the corresponding limit value.
When the ratio of the two ratios is greater than or equal to 1, the electric drill shell is exposed by the glass fibers.
The beneficial effect of this embodiment lies in:
according to the method, the gray level jump characteristic and the run characteristic are obtained by utilizing the gray level value of the pixels in the electric drill shell image, the co-occurrence matrix is further constructed according to the two characteristics, the high-jump long run advantage and the low-jump short run advantage in the gray level image are calculated according to the co-occurrence matrix, and finally the glass fiber exposure of the electric drill shell is identified by utilizing the high-jump long run advantage and the low-jump short run advantage. The electronic equipment is applied to the embodiment to identify the exposed glass fibers of the electric drill shell, so that the identification precision and efficiency can be effectively improved.
Example 2
The main purposes of this embodiment are: and processing the acquired electric drill shell image by using computer vision, analyzing the characteristics of the electric drill shell image, identifying the exposure of the glass fibers, and adjusting the production process of the electric drill shell. The embodiment provides an electric drill shell glass fiber exposure identification method, which is used for identifying glass fiber exposure of an electric drill shell by using electronic equipment, and can effectively improve identification precision and efficiency.
In the injection molding production process of the electric drill shell, the glass fibers on the surface of the electric drill shell are exposed probably due to improper process parameter setting in the injection molding processing process. The exposed glass fibers need to be detected and identified, and the production process is adjusted according to the identification result so as to improve the production qualified rate.
The embodiment designs an electric drill shell glass fiber exposure identification method and system. The system comprises a camera, an embedded system and the like. The embedded system comprises a data preprocessing module, a glass fiber exposure identification module and an injection molding process parameter adjusting module.
The embodiment of the invention provides an electric drill shell glass fiber exposure identification method, as shown in fig. 2, comprising the following steps:
the method comprises the following steps: and shooting an electric drill shell image, and performing semantic segmentation processing on the electric drill shell image.
And erecting a camera above the conveying belt after the injection molding process of the electric drill shell, and shooting the injection molded electric drill shell image. The image includes the drill housing and the background. And inputting the shot image to an image preprocessing module.
The present embodiment employs a DNN semantic segmentation approach to identify objects in a segmented image.
The relevant content of the DNN network is as follows:
a. the data set used is a drill casing image data set acquired from a top view.
b. The pixels to be segmented are divided into 2 types, that is, the labeling process of the labels corresponding to the training set is as follows: and in the single-channel semantic label, the pixel at the corresponding position belongs to the background class and is marked as 0, and the pixel belonging to the electric drill shell is marked as 1.
c. The role of the network is classification, so the loss function used is the cross entropy loss function.
Therefore, the electric drill shell image is processed through the DNN, and the electric drill shell connected domain information in the image is obtained. For ease of analysis, the drill housing image was converted to a grey scale image. And transmitting the gray-scale image to the glass fiber exposed identification module.
Step two: and (4) analyzing the gray image characteristics of the electric drill shell to obtain the texture characteristics of the electric drill shell.
And (3) analyzing the gray level image of the electric drill shell, wherein the gray level image of the electric drill shell is partially brighter and partially darker under the influence of illumination. The electric drill shell is made of glass fiber reinforced nylon materials, the surface of the electric drill shell presents granular sensation under illumination, the granular sensation is obvious in a bright area, and the granular sensation is not obvious in a dark area.
The glass fiber is exposed to be in a short line shape with brighter color. However, under the interference of illumination and granular sensation, the glass fiber exposure is difficult to be accurately identified through threshold segmentation or edge detection.
In the embodiment, gray level grading is performed by combining a gray level histogram, the gray level jump characteristic and the run characteristic of each pixel point are obtained according to the gray level, and a gray level jump and run co-occurrence matrix is established by combining the gray level jump characteristic and the run characteristic of all the pixel points. And combining the matrix to obtain the high-jump long-run advantage and the low-jump short-run advantage texture characteristics of the electric drill shell image, so that the glass fiber exposure can be identified according to the texture characteristics in the following process.
a. Gray level grading:
in order to facilitate the acquisition of the subsequent gray level jump characteristic and the run characteristic, all gray values in the image need to be graded.
And drawing a gray level histogram, and analyzing the gray level histogram, wherein each peak in the histogram is distributed in the peak by the gray level which represents the existence of an image characteristic in the image. Therefore, the gray level histogram is subjected to Gaussian smoothing, local minimum values in the gray level histogram are obtained, all gray levels are divided into a plurality of levels by taking the local minimum values as dividing points, and different image features are divided into different gray levels.
b. Acquiring gray level jump characteristics and run characteristics of pixel points:
1. in order to measure the local change characteristics of the pixel points, the gray level jump characteristic of each pixel point in the image is obtained.
Obtaining the radius of each pixel point asCircular neighborhood of (the present embodiment)). Obtain neighborhood edge pixels, e.g., gray pixels in FIG. 3, with radius ofHas a circular neighborhood ofAnd each edge pixel point. Judging whether the gray level of each pixel point on the neighborhood edge is greater than that of the central pixel point, if so, marking the position of the neighborhood edge pixel point asOtherwise, it is marked as. Thus obtainingA binary number of bits.
Will be selected fromToOr fromToCalled one-hop, statisticsThe jump times of the binary digit number are taken as the jump characteristics of the central pixel point and are recorded as. The hopping characteristic can reflect the gray level change complexity around the central pixel point to a certain extent.
2. In order to measure the continuous change characteristics of the pixel points, the run-length characteristics of each pixel point in the image are obtained.
The run is the number of continuous and consistent gray levels of the pixel points in a certain direction. Obtaining the run length of a pixel point in each direction, and countingThe maximum value is taken as the run-length characteristic of the pixel point and is recorded as. The run-length characteristics can reflect the continuity of the pixel points to a certain extent.
c. Acquiring a gray level jump run co-occurrence matrix:
step b, the gray level jump characteristic of each pixel point is obtainedAnd run characteristicsThen, a gray level jump run co-occurrence matrix is constructed according to all the gray level jump characteristics and all the run characteristics, and the value in the matrix represents the co-occurrence times of the corresponding gray level jump characteristics and the run characteristics, such as the first matrixGo to the firstValue of columnRepresenting a characteristic of gray level jump in an image asThe run length is characterized byPixel point of has appearedNext, the process is repeated. The gray level jump run co-occurrence matrix is shown in fig. 4.
d. Acquiring the advantages of high jump and long run and the advantages of low jump and short run of the electric drill shell:
the fiber is exposed to be in a brighter short-line shape, the run length of the fiber is longer than that of other surrounding pixel points in a region with larger brightness, and the run length of the fiber is shorter than that of other surrounding pixel points in a region with smaller brightness. In the area with high brightness, the granular sensation is obvious, the gray level change around the pixel point is complex, and the gray level jump of the pixel point is large at the moment. In the area with small brightness, the granular sensation is not obvious, the gray level change around the pixel point is smooth, and the gray level jump of the pixel point is small at the moment.
WhereinDenotes the firstA plurality of gray scale jump features;denotes the firstA run length feature;representing gray level jump characteristicsAnd run length characteristicsCo-occurrence ofThe number of times;the number of run characteristics;the number of the gray level jump characteristics;to relate toIs a function of (a) a function of (b),will be provided withThe normalization is carried out, and the normalization is carried out,will be provided withThe normalization is carried out, and the normalization is carried out,denotes the firstA plurality of gray scale jump features;denotes the firstA run length feature; when in useThe smaller the difference value after the normalization is,the larger;representation is paired according to jump characteristic and run characteristicEnhancing, when the jump characteristic is larger and the run characteristic is larger, the enhancing degree is larger, when the jump characteristic is smaller or the run characteristic is smaller, the enhancing degree is smaller,avoidOne of them being too largeBecomes larger. Giving greater weight to high jump long runs by boosting; when the number of the pixel points of the high-jump long-run is more, the advantage of the high-jump long-runIs relatively large.
Similarly, the low-jump short-run advantage of the image is obtained according to the gray-level jump run matrix:
WhereinRepresenting acquisition hopping featuresAndin order to avoid denominator of;Representing pairs of features according to jump and runAnd weakening is carried out, the weakening degree is larger if the jump characteristic is larger or the run characteristic is larger, and the weakening degree is smaller when the jump characteristic is smaller and the run characteristic is smaller. By weakening the lower jump short runs to give more weight; when the number of low-jump short-run pixels is large, the low-jump short-run advantageIs relatively large.
Therefore, texture characteristics of the high-jump long-run advantage and the low-jump short-run advantage of the electric drill shell image are obtained.
Step three: and identifying the exposure of the glass fibers by combining the texture characteristics of the electric drill, and adjusting the production process of the electric drill shell.
And step two, texture characteristics of the high-jump long-run advantage and the low-jump short-run advantage of the electric drill shell image are obtained.
Wherein the content of the first and second substances,the number of run characteristics;the number of the gray level jump characteristics; if no glass fiber is exposed, the granular sensation of a bright area of the electric drill shell is obvious, and the value in the gray level jump run-length co-occurrence matrix is concentrated at the lower left corner of the matrix.The high jump long run advantage when the left lower corner of the matrix is concentrated for all values in the gray jump run matrix is assumed; if no glass fiber is exposed, the granular sensation of a darker area of the electric drill shell is not obvious, and the value in the gray level jump run-length co-occurrence matrix is concentrated at the upper right corner of the matrix.The advantage of high jump long-run is assumed when all values in the gray jump run matrix are concentrated in the upper right corner of the matrix.Is composed ofOf (2) is calculated.Denotes the firstA gray level jump feature;denotes the firstA run length feature;representing gray level jump characteristicsAnd run length characteristicsThe number of co-occurrences;the number of run characteristics;the number of the gray level jump characteristics;is aboutAs a function of (a) or (b),is shown asA gray level jump feature;denotes the firstA run length feature;to relate toAs a function of (c).
Similarly, the limit value of the advantage of low-jump short-run length without glass fiber exposure is obtained:
In the formula (I), the compound is shown in the specification,to assume the low jump short run advantage when all values in the gray jump run matrix are centered in the lower left corner of the matrix,the advantage of low jump short run when all values in the gray jump run matrix are concentrated in the upper right corner of the matrix is assumed.
Combining high-jump long-run advantagesAnd low jump short run advantageRecognizing the exposure of the glass fiber:
when the temperature is higher than the set temperatureOrWhen the electric drill is used, the glass fiber is exposed on the shell of the electric drill;
when in useWhen the electric drill is used, the glass fiber is not exposed on the shell of the electric drill.
And finally, completing the identification of the exposed glass fibers, and if the glass fibers are identified to be exposed, transmitting the result to the injection molding process parameter adjusting module. The glass fiber is exposed because the flowability of the glass fiber is much lower than that of the nylon material, so that the glass fiber stays on the surface of the shell of the electric drill. When the injection molding process parameter adjusting module receives the glass fiber exposure prompt, the control system increases the injection speed and simultaneously increases the temperature of the mold so as to achieve the purpose of reducing the flow speed difference between the glass fibers and the nylon material. Thereby reducing the exposure of glass fibers of subsequent products and improving the qualification rate of electric drill shell products.
The beneficial effect of this embodiment lies in:
according to the embodiment, gray level jump characteristics and run characteristics are obtained by utilizing gray level values of pixels in an electric drill shell image, a co-occurrence matrix is further constructed according to the two characteristics, high-jump long-run advantages and low-jump short-run advantages in the gray level image are calculated according to the co-occurrence matrix, and finally, glass fiber exposure of the electric drill shell is identified by utilizing the high-jump long-run advantages and the low-jump short-run advantages. The embodiment applies the electronic equipment to identify the exposed glass fibers of the electric drill shell, and can effectively improve the identification precision and efficiency.
Example 3
The embodiment provides an electric drill shell glass fiber exposure identification system, and electronic equipment is applied to identify glass fiber exposure of an electric drill shell, so that the identification precision and efficiency can be effectively improved.
The embodiment of the invention provides an electric drill shell glass fiber exposure identification system, which comprises an acquisition unit, a processing unit, a calculation unit and a control unit, as shown in fig. 5:
the acquisition unit is used for acquiring an electric drill shell image to be identified, which is injected and molded on the conveyor belt, by arranging the camera on a track arranged above the conveyor belt;
the processing unit inputs the image acquired by the acquisition unit into the data master controller, performs semantic segmentation on the image by using the data master controller to obtain an electric drill shell connected domain image, performs graying processing on the connected domain image to obtain an electric drill shell gray image, performs gray level grading on the gray image to obtain the gray level of each pixel point in the electric drill shell gray image;
the computing unit is used for computing a gray level jump characteristic by the data master controller according to the gray levels of each pixel point and the adjacent edge pixel points in the gray level image acquired by the processing unit, computing a run characteristic by the aid of runs of each pixel point in the gray level image in different directions, further constructing a co-occurrence matrix according to the two characteristics, and computing a high-jump long-run advantage, a low-jump short-run advantage and a corresponding limit value of the gray level image by the aid of the co-occurrence matrix;
the control unit judges the ratio of the high-jump long-run advantage to the high-jump long-run advantage limit value, the low-jump short-run advantage to the low-jump short-run advantage limit value of the gray scale image obtained by the calculation unit by using the data main controller: when the ratio of the two ratios is more than or equal to 1, the glass fiber is exposed in the electric drill shell; and the injection molding controller adjusts the injection molding process parameters according to the identification result.
The beneficial effect of this embodiment lies in:
according to the embodiment, gray level jump characteristics and run characteristics are obtained by utilizing gray level values of pixels in an electric drill shell image, a co-occurrence matrix is further constructed according to the two characteristics, high-jump long-run advantages and low-jump short-run advantages in the gray level image are calculated according to the co-occurrence matrix, and finally, glass fiber exposure of the electric drill shell is identified by utilizing the high-jump long-run advantages and the low-jump short-run advantages. The electronic equipment is applied to the embodiment to identify the exposed glass fibers of the electric drill shell, so that the identification precision and efficiency can be effectively improved.
The above description is only for the purpose of illustrating the preferred embodiments of the present invention and is not to be construed as limiting the invention, and any modifications, equivalents, improvements and the like that fall within the spirit and principle of the present invention are intended to be included therein.
Claims (9)
1. An electric drill shell glass fiber exposure identification method is characterized by comprising the following steps:
acquiring a gray scale image of an electric drill shell to be identified;
grading all gray values in a gray image of the electric drill shell to obtain the gray level of each pixel point in the gray image;
acquiring the gray level hopping characteristic of each pixel point in the gray level map by utilizing the gray level of each pixel point in the gray level map and the gray level of the pixel points at the edge of the circular neighborhood;
acquiring the run length characteristics of each pixel point in the gray-scale image according to the run lengths of each pixel point in the gray-scale image in different directions;
constructing a gray level jump and run co-occurrence matrix by utilizing the gray level jump characteristics and the run characteristics of all pixel points in the gray level map; the gray level jump run-length co-occurrence matrix is constructed in the following way:
the characteristic of gray level jump among all pixel points of the statistical gray level map isAnd run is characterized byThe number of the pixel points;
the characteristic of gray jump among all pixel points of the gray map is taken asAnd run is characterized byThe number of the pixel points is an element value, and a gray level jump run co-occurrence matrix is constructed;
obtaining a high-jump long-run advantage and a low-jump short-run advantage in a gray level map by utilizing the number of gray level jump features, the number of run features and the co-occurrence frequency of the gray level jump features and the run features in the gray level jump run co-occurrence matrix;
acquiring a high-jump long-run dominant limit value and a low-jump short-run dominant limit value of a gray scale map without glass fiber exposure;
and identifying whether the glass fiber is exposed on the shell of the electric drill to be identified according to the ratio of the high-jump long run advantage and the low-jump short run advantage in the gray scale image to the corresponding limit value.
2. The method for identifying the exposure of the glass fibers of the electric drill casing as claimed in claim 1, wherein the gray scale of each pixel point in the gray scale map is obtained as follows:
obtaining a gray level histogram according to all gray levels in the gray level image of the electric drill shell;
performing Gaussian smoothing on the gray level histogram to obtain a local minimum value in the gray level histogram;
and taking each local minimum value as a segmentation point, and segmenting all gray values into different levels to obtain the gray level of each pixel point in the gray map.
3. The method for identifying the exposure of the glass fibers of the electric drill shell as claimed in claim 1, wherein the gray jump characteristic of each pixel point in the gray map is obtained as follows:
taking each pixel point in the gray-scale image as a center, taking R as a radius to make a circle, and obtaining a circular neighborhood of each pixel point in the gray-scale image and neighborhood edge pixel points;
judging the gray level of each neighborhood edge pixel point and the gray level of the center pixel point: if the gray level of the neighborhood edge pixel point is greater than that of the central pixel point, the position of the neighborhood edge pixel point is marked as 1, and if the gray level of the neighborhood edge pixel point is less than or equal to that of the central pixel point, the position of the neighborhood edge pixel point is marked as 0;
taking adjacent neighborhood edge pixel points from 1 to 0 or from 0 to 1 as one-time hopping, and counting the hopping times of the neighborhood edge pixel points along the same direction;
and taking the hopping times as the hopping characteristics of the central pixel point to obtain the gray level hopping characteristics of each pixel point in the gray level image.
4. The method for identifying the exposure of the glass fibers of the electric drill shell as claimed in claim 1, wherein the run length characteristics of each pixel point in the gray scale map are obtained as follows:
counting the continuous same number of gray levels of all pixel points in the gray level image in different directions to obtain the run length of each pixel point in the gray level image in all directions;
and counting the maximum value of the run of each pixel point in each direction in the gray-scale image, taking the maximum value as the run characteristic of each pixel point, and obtaining the run characteristic of each pixel point in the gray-scale image.
5. The method for identifying the exposure of the glass fibers of the electric drill shell as claimed in claim 1, wherein the calculation expression of the high jump length run advantage in the gray scale map is as follows:
in the formula (I), the compound is shown in the specification,for the high transition long run advantage in the grayscale,is shown asA plurality of gray-level jump features, wherein,denotes the firstThe characteristics of a run are such that,representing gray level jump characteristicsAnd run characteristicsThe number of co-occurrences is such that,is the number of run-length features,the number of the characteristics of the gray level jump,is aboutAs a function of (c).
6. The electric drill shell glass fiber exposure identification method as claimed in claim 1, wherein the calculation expression of the low-jump short-run advantage in the gray scale map is as follows:
in the formula (I), the compound is shown in the specification,for the low jump short run advantage in the grey scale,denotes the firstA number of grey-scale jump features that,denotes the firstThe characteristics of a run are such that,representing gray level jump characteristicsAnd run characteristicsThe number of times of co-occurrence,is the number of run-length features,the number of the gray scale jump features,is aboutAs a function of (c).
7. The method for identifying the exposure of the glass fibers of the electric drill casing as claimed in claim 1, wherein the process for identifying whether the glass fibers of the electric drill casing to be identified are exposed is as follows:
acquiring a high-jump long-run dominant limit value and a low-jump short-run dominant limit value of a gray scale image without glass fiber exposure by using the number of gray jump features, the number of run features, and the co-occurrence times of the gray jump features and the run features in the gray jump run co-occurrence matrix;
acquiring the ratio of the high-jump long-run advantage to the corresponding limit value and the ratio of the low-jump short-run advantage to the corresponding limit value of the gray scale map;
and judging the two ratios: when the two ratios are less than 1, the electric drill shell to be identified has no glass fiber exposure, and when the ratio in the two ratios is more than or equal to 1, the electric drill shell to be identified has glass fiber exposure.
8. The method for identifying the exposure of the glass fibers of the electric drill casing as claimed in claim 1, wherein the gray scale map of the electric drill casing to be identified is obtained as follows:
collecting an electric drill shell image to be identified after injection molding;
performing semantic segmentation on the injection-molded electric drill shell image to be identified to obtain an electric drill shell connected domain image to be identified;
and carrying out gray processing on the communicated domain image of the electric drill shell to be identified to obtain a gray image of the electric drill shell to be identified.
9. The utility model provides a fine identification system that exposes of electric drill shell glass, its characterized in that, includes acquisition unit, processing unit, computational element and the control unit:
the acquisition unit is used for acquiring an electric drill shell image to be identified, which is injected and molded on the conveyor belt, by arranging the camera on a track arranged above the conveyor belt;
the processing unit and the data master controller perform semantic segmentation and gray scale grading on the image acquired by the acquisition unit to acquire the gray scale of each pixel point in the gray scale image of the electric drill shell;
the data master controller of the computing unit computes gray level jump characteristics and run characteristics of all pixel points in the gray level image according to the gray level of each pixel point in the gray level image of the electric drill shell, which is obtained by the processing unit, further constructs a co-occurrence matrix according to the two characteristics, and computes high-jump long-run advantages and low-jump short-run advantages of the gray level image by utilizing the co-occurrence matrix;
the control unit and the data master controller identify the exposure of the glass fibers in the electric drill shell by utilizing the high-jump long-run advantage and the low-jump short-run advantage of the gray scale image obtained by the calculation unit, and the injection molding controller adjusts the injection molding process parameters according to the identification result.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211022236.6A CN115100608B (en) | 2022-08-25 | 2022-08-25 | Electric drill shell glass fiber exposure identification method and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211022236.6A CN115100608B (en) | 2022-08-25 | 2022-08-25 | Electric drill shell glass fiber exposure identification method and system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN115100608A CN115100608A (en) | 2022-09-23 |
CN115100608B true CN115100608B (en) | 2022-11-11 |
Family
ID=83300131
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202211022236.6A Active CN115100608B (en) | 2022-08-25 | 2022-08-25 | Electric drill shell glass fiber exposure identification method and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN115100608B (en) |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112069852B (en) * | 2020-09-07 | 2023-12-12 | 凌云光技术股份有限公司 | Low-quality two-dimensional code information extraction method and device based on run-length coding |
CN114049489B (en) * | 2022-01-13 | 2022-04-12 | 山东鹰联光电科技股份有限公司 | Early warning method for breakage of optical fiber cable sheath |
CN114841940A (en) * | 2022-04-24 | 2022-08-02 | 南通金丝楠膜材料有限公司 | Injection molding silver thread defect detection method and system based on gray level run matrix |
CN114757949B (en) * | 2022-06-15 | 2022-09-27 | 济宁市海富电子科技有限公司 | Wire and cable defect detection method and system based on computer vision |
-
2022
- 2022-08-25 CN CN202211022236.6A patent/CN115100608B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN115100608A (en) | 2022-09-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN115082683B (en) | Injection molding defect detection method based on image processing | |
CN108562589B (en) | Method for detecting surface defects of magnetic circuit material | |
CN104680519B (en) | Seven-piece puzzle recognition methods based on profile and color | |
CN113160192A (en) | Visual sense-based snow pressing vehicle appearance defect detection method and device under complex background | |
CN114359190B (en) | Plastic product molding control method based on image processing | |
CN114219805B (en) | Intelligent detection method for glass defects | |
CN111815564B (en) | Method and device for detecting silk ingots and silk ingot sorting system | |
CN114187289B (en) | Plastic product shrinkage pit detection method and system based on computer vision | |
CN114372955A (en) | Casting defect X-ray diagram automatic identification method based on improved neural network | |
CN115294410B (en) | Plastic product molding control method based on pattern recognition | |
CN115578389A (en) | Defect detection method of groove MOS device | |
CN115880699B (en) | Food packaging bag detection method and system | |
CN117173158B (en) | Intelligent detection method and system for quality of precise connector | |
CN106372624A (en) | Human face recognition method and human face recognition system | |
CN116309577B (en) | Intelligent detection method and system for high-strength conveyor belt materials | |
CN113256624A (en) | Continuous casting round billet defect detection method and device, electronic equipment and readable storage medium | |
CN106874942A (en) | A kind of object module fast construction method semantic based on regular expression | |
CN114882026A (en) | Sensor shell defect detection method based on artificial intelligence | |
CN113570628A (en) | Leukocyte segmentation method based on active contour model | |
CN106997590A (en) | A kind of image procossing and detecting system based on detection product performance | |
CN206897873U (en) | A kind of image procossing and detecting system based on detection product performance | |
Wu et al. | Automatic gear sorting system based on monocular vision | |
CN115100608B (en) | Electric drill shell glass fiber exposure identification method and system | |
CN109815957A (en) | A kind of character recognition method based on color image under complex background | |
CN116167970A (en) | Sand mold detection system and detection method |
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 |