CN117788448A - Lithium battery enhancement method based on Laplacian algorithm - Google Patents
Lithium battery enhancement method based on Laplacian algorithm Download PDFInfo
- Publication number
- CN117788448A CN117788448A CN202311864904.4A CN202311864904A CN117788448A CN 117788448 A CN117788448 A CN 117788448A CN 202311864904 A CN202311864904 A CN 202311864904A CN 117788448 A CN117788448 A CN 117788448A
- Authority
- CN
- China
- Prior art keywords
- image
- filter
- lithium battery
- laplace
- algorithm
- 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.)
- Pending
Links
- WHXSMMKQMYFTQS-UHFFFAOYSA-N Lithium Chemical compound [Li] WHXSMMKQMYFTQS-UHFFFAOYSA-N 0.000 title claims abstract description 77
- 229910052744 lithium Inorganic materials 0.000 title claims abstract description 77
- 238000000034 method Methods 0.000 title claims abstract description 64
- 230000004044 response Effects 0.000 claims abstract description 37
- 238000004364 calculation method Methods 0.000 claims abstract description 32
- 238000007781 pre-processing Methods 0.000 claims abstract description 10
- 230000002708 enhancing effect Effects 0.000 claims abstract description 8
- 238000012545 processing Methods 0.000 claims description 37
- 238000003708 edge detection Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 7
- 238000009499 grossing Methods 0.000 claims description 7
- 239000011159 matrix material Substances 0.000 claims description 6
- 230000008569 process Effects 0.000 claims description 4
- 238000001514 detection method Methods 0.000 abstract description 8
- 230000007547 defect Effects 0.000 abstract description 7
- 238000001914 filtration Methods 0.000 description 7
- 230000009286 beneficial effect Effects 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 229920000642 polymer Polymers 0.000 description 3
- 230000001629 suppression Effects 0.000 description 3
- 238000013461 design Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 230000003044 adaptive effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012216 screening Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses a lithium battery enhancement method based on a Laplacian algorithm, which comprises the following steps: s10, preprocessing an image to obtain a preprocessed image A; s20, designing a fast local Laplace filter, wherein the fast local Laplace filter comprises the following steps: s21, determining the size of a filter; s22, determining weight distribution of the filter; s23, generating a filter template; s30, calculating a quick local Laplace response, which comprises the following steps: s31, edge filling is carried out on the preprocessed image A, and a filled image B is obtained; s32, performing first convolution calculation on the filled image B to obtain a quick local Laplace response value R and a quick local Laplace response chart C; s40, enhancing the image, namely performing second convolution calculation on the filled image B and the quick local Laplace response graph C to obtain an enhanced image D. The invention solves the problems of high error rate, high complexity, low response speed and the like of the existing image recognition method for carrying out defect detection on the lithium battery by adopting the Laplace algorithm.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a lithium battery enhancement method based on a Laplacian algorithm.
Background
Various means are used in the lithium battery production process to monitor and control the quality of the battery. The image recognition method is a method for monitoring the production quality of the lithium battery, and the method analyzes images of the positions of the battery core, the housing, the electrode lugs and the like to judge whether the product has defects or flaws, so that a strong auxiliary effect is achieved on the quality inspection of the battery. In order to detect defects of a lithium battery, an image of the lithium battery needs to be first identified.
For example, a machine vision-based polymer soft package lithium battery base angle automatic detection device disclosed in chinese patent document with application publication number CN113744239a comprises: the image acquisition module is used for acquiring a first base angle image of the polymer soft package lithium battery; the image processing module is connected with the image acquisition module and is used for preprocessing the first base angle image to obtain a second base angle image; the detection module is connected with the image processing module and is used for detecting the bottom angle damage of the second bottom angle image to obtain a detection result; and the rejecting module is connected with the detecting module and is used for rejecting the polymer soft package lithium battery with the detection result of the damage of the base angle.
The existing device for identifying and detecting the image of the lithium battery can save labor cost and realize the purposes of identifying the image characteristics and automatically detecting the defects of the lithium battery. In addition, as described in the foregoing patent document, in the conventional scheme, a laplace algorithm is generally adopted to identify the image features of the lithium battery, and then whether the definition of the image reaches a preset standard is determined, so that the defect of the battery is detected; however, when the above method is applied to the field of lithium battery edge detection, there are the following problems:
firstly, the error rate is too high; due to the complexity of the detection environment, when the battery is subjected to edge detection by the Laplacian algorithm, the algorithm is subjected to larger noise interference, so that an edge detection result is inaccurate, and an excessively high error rate is generated;
secondly, because the laplace algorithm needs to carry out complex parameter setting and adjustment in practical application, and the appearance difference of batteries of different batches and models is large, when the batteries of different batches and models are detected, the parameter setting and adjustment are often very complicated, and difficulties are often encountered, so that the production efficiency is reduced, and the operability and the applicability of the laplace algorithm are relatively poor;
thirdly, when processing large-scale image data, the Gaussian Laplace algorithm involves multiple convolution operations and pixel operation on the image, and has long processing time and high requirements on hardware equipment, so that the response speed is low, and the application scene of the Gaussian Laplace algorithm in the aspect of real-time image processing is limited;
fourth, the laplace algorithm generally has higher accuracy for images with clear boundaries and simple structures, and when extracting complex image structures or details, the algorithm cannot fully extract and highlight key features, so that the processing result is not accurate and effective enough.
In view of this, there is a need for improvements in existing methods of lithium battery image recognition based on the laplace algorithm.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a lithium battery enhancement method based on a Laplace algorithm, which aims to solve the problems of high error rate, high complexity, low response speed and poor complex image processing capability of the existing image recognition method for detecting defects of a lithium battery by adopting the Laplace algorithm, thereby improving the efficiency and the accuracy of image recognition of the lithium battery.
The invention discloses a lithium battery enhancement method based on a Laplacian algorithm, which comprises the following steps:
s10, preprocessing an image to obtain a preprocessed image A;
s20, designing a fast local Laplace filter, wherein the fast local Laplace filter comprises the following steps: s21, determining the size of a filter; s22, determining weight distribution of the filter; s23, generating a filter template;
s30, calculating a quick local Laplace response, which comprises the following steps: s31, edge filling is carried out on the preprocessed image A, and a filled image B is obtained; s32, performing first convolution calculation on the filled image B to obtain a quick local Laplace response value R and a quick local Laplace response chart C;
s40, enhancing the image, namely performing second convolution calculation on the filled image B and the quick local Laplace response graph C to obtain an enhanced image D.
Preferably, the image preprocessing method includes: and obtaining an original image, carrying out graying treatment on the original image to obtain a gray image, and carrying out noise removal and smoothing treatment on the gray image to obtain the preprocessed image A.
Preferably, the size of the filter is m×m, m is greater than or equal to the size of the edge feature of the lithium battery, the edge feature is a set of edge pixel points of the lithium battery, and the size of the edge feature is an average value of the width of the set of edge feature pixel points in the extending direction of the edge feature pixel points.
Preferably, the weight distribution of the filter is gaussian weighted distribution, the filter template is an m×m matrix, and the weight values of each point of the filter are taken as elements of the matrix.
Preferably, the method for edge filling the preprocessed image a is to fill pixels with a width of m-1 outside the edge of the preprocessed image a, and the pixel values of the filled pixels are all 0.
Preferably, the method for performing the first convolution calculation on the filled image B includes: and using a window with the size of m multiplied by m as a filter window, sequentially sliding the filter window on the filled image B, sliding one pixel point each time, and performing calculation once each time, wherein the calculation is to calculate convolution of the filter template and pixel values of each point in the filter window, and traversing the filled image B.
Preferably, the second convolution calculation method for the filled image B and the fast local laplace response graph C uses a window with a size of m×m as a filter window, slides the filter window on the filled image B sequentially, slides a pixel point each time, and performs calculation once each time, where the calculation is to calculate, for each pixel point of the filled image B in the filter window, a convolution of a fast local laplace response value R and pixel values of the points, and traverses the filled image B.
Preferably, the binarization processing is performed after the image enhancement: and converting the pixel value in the enhanced image D into a binary value through the set threshold parameter to obtain a binary image E.
Preferably, the binarization processing is performed by directly converting a certain point pixel value of the enhanced image D into 1 if the point pixel value is higher than the threshold parameter, and converting the point pixel value into 0 if the point pixel value is lower than the threshold parameter.
Preferably, after the binarization processing, a Canny algorithm is adopted to perform edge detection and feature extraction on the binary image E.
The invention has the beneficial effects that: the lithium battery image enhancement method based on the Laplace filter algorithm has remarkable beneficial effects in the aspects of improving image quality, suppressing noise, accelerating processing speed, improving working efficiency and the like. The method has the advantages that the image processing of the lithium battery is more accurate and reliable, powerful support is provided for analysis, diagnosis and evaluation of the lithium battery, the method has good application universality, the problems of high error rate, high complexity, low response speed and poor complex image processing capability of the existing image recognition method for detecting the lithium battery by adopting the Laplace algorithm are solved, and the efficiency and the accuracy of the image recognition of the lithium battery are improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a block flow diagram of a Laplacian algorithm-based lithium battery enhancement method of the present invention;
FIG. 2 is a flow chart of a method of designing a fast local Laplace filter in step S20;
FIG. 3 is a flowchart of a method for computing the fast local Laplace at step S30;
FIG. 4 is a flow chart of the method of edge detection and feature extraction in step S60;
FIG. 5 is a schematic illustration of a pre-processed image of a lithium battery;
fig. 6 is a schematic diagram of a lithium battery image after being processed by a fast local laplace filter.
Detailed Description
For a further understanding of the invention, its features and advantages, reference is now made to the following examples, which are illustrated in the accompanying drawings in which:
referring to fig. 1-4, the method for enhancing the lithium battery based on the laplace algorithm disclosed by the invention comprises the following steps:
s10, preprocessing an image;
s20, designing a rapid local Laplace filter;
s30, calculating a quick local Laplace response;
s40, enhancing the image;
s50, binarization processing;
s60, edge detection and feature extraction.
The image preprocessing method comprises the steps of obtaining an original image of a lithium battery, carrying out graying treatment on the original image through a computer, converting the original image into a gray image, and then carrying out noise removal and smoothing treatment on the gray image so as to reduce interference and noise points in the image; after the image preprocessing, a preprocessed image of the lithium battery is obtained, as shown in fig. 5. Specifically, the original image of the lithium battery may be obtained by a photographing device, such as a CCD camera, a line scan camera or an infrared camera, and the original image data is transmitted to a computer in a wired or wireless transmission manner, and the methods of graying, noise removal and smoothing are all existing methods, which are not described herein.
The filter in the laplace algorithm is an m×m square pixel area, the side length of the square pixel area can be set freely, and the filter is used for filtering an image. The size of the filter determines the local extent of the process, and too large a filter may result in loss of image detail and too small a filter may not be effective in enhancing the image. In step S20, the method for designing the fast local laplacian filter includes the following steps:
s21, determining the size of a filter, namely determining the value of m; in general, the size of the filter can be set manually, and three factors need to be considered in the setting process: first, the size of the edge feature in the image, if the lithium battery edge feature in the image has a wider size, such as occupies 5 pixels, the value of m should not be smaller than the size of the edge feature, where the edge feature refers to the collection of essential pixel points in the image that constitute the lithium battery edge; second, the larger the noise level, the larger the size of the filter, the larger the value of m, so that the noise can be smoothed; third, if the computational effort is low, a smaller filter should be used.
In lithium battery image recognition, since noise level and computer power are controllable and can be assumed to be environment-noiseless and computer power is infinite, the size of the filter can be set to only consider the size of the edge feature of the detected lithium battery in the image, if the set of pixels forming the lithium battery shell edge extends in the image with an average width of 3 pixels, the size of the edge feature is 3 pixel points; in this embodiment, the m value of the filter is not smaller than the dimension of the edge feature, and the dimension of the edge feature can be estimated by the existing algorithm, which is not described herein.
S22, determining weight distribution of the filter; the weight of the filter refers to the contribution degree of different pixel points to the central pixel, and the sum of the weights of all the pixel points is 1; in this embodiment, the weighting distribution mode of the filter adopts gaussian weighting distribution, that is, the pixel points closer to the center pixel have larger weights, and the weights of the pixel points farther from the center pixel gradually decrease. The designed weight distribution can enable the filter to carry out smoothing processing on the image in a local range, and the details of the image are reserved while the edge information is enhanced.
S23, generating a filter template: and generating a template of the local Laplace filter according to the determined filter size m and the weight distribution. The template is an m x m matrix in which each element represents the weight of the corresponding pixel point. The filter template may be expressed in the form:
wherein wij represents the weight of the ith row and the jth column of pixel points. Filling each element in the filter template according to the weight distribution manner determined in step S22.
Generally, the preprocessed image obtained after the image preprocessing in S10 is a two-dimensional matrix a, and the size is n×s; in step S30, the method for calculating the fast local laplace response includes the following steps:
s31, edge filling is carried out on the preprocessed image A: filling pixel points with the width of m-1 outside the edge of the preprocessed image A, so that each pixel point can be positioned at the central pixel position of a filter window in certain calculation during subsequent convolution calculation; after edge filling, the preprocessed image a is converted into a filled image B, the size of the filled image B is (n+m-1) × (s+m-1), and the filling mode can adopt zero filling or copying of edge pixel values, in this embodiment, zero filling is adopted, that is, the values of the pixel points filled at the edge of a are all 0.
S32, performing first convolution calculation on the filled image B; the specific method comprises the following steps:
a window with the size of m multiplied by m is used as a filter window, the filter window is sequentially slid on the filled image B, one pixel point is slid each time, and one calculation is performed each time, the calculation method is that,
and (3) calculating convolution of the filter template and pixel values of each point in the filter window to obtain a quick local Laplacian response value of the pixel point. The calculation formula is as follows:
response(i,j)＝ΣΣ(w(i,j)*I(i+k,j+l))；k＝1，l＝1；
the response (i, j) represents a fast local Laplacian response value of the pixel point in the ith row and the jth column in the filled image B, and the subsequent brief description is R (i, j); w (I, j) represents the weight of the ith row and jth column in the filter template, and I (i+k, j+l) represents the pixel value of the kth row and jth column pixel point in the filter window;
traversing all pixel points of the filled image B, taking a response value when each point in the preprocessed image A is taken as a central point as a new pixel value of the point, and thus obtaining a filtered image, namely a quick local Laplace response graph C, as shown in fig. 6; the fast local laplacian response map C has the same size as the preprocessed image a, i.e., n×s.
The principle of the steps S31 and S32 is that the filter template can achieve the purpose of edge enhancement by calculating the difference between the pixel point in the image and the neighboring pixel point. Specifically, the filter template performs convolution operation on the image, multiplies the weight in the filter by the pixel value of the corresponding pixel point, and accumulates the result to obtain the enhanced pixel value, while the weight distribution in the local laplace filter template can enable the filter to have higher response to the edge region in the image, so that the edge information is enhanced, and meanwhile, the smoothing processing of the filter is also beneficial to reducing the influence of noise and improving the image quality.
Through the above steps S10 to S30, the local laplace response of the filter can be quickly calculated. The rapid algorithm can effectively improve the calculation efficiency of the filter, particularly when processing large-size images, the calculation time is remarkably reduced, and the response chart C of the rapid local Laplacian filter is used for subsequent image enhancement and feature extraction steps, so that rapid processing and enhancement of lithium battery images are realized.
In step S40, the image enhancement method is to perform a second convolution calculation on the image B filled in step S31 and the fast local laplace response chart C, so as to obtain an enhanced image D. Specifically, the method comprises the following steps:
a window with the size of m multiplied by m is used as a filter window, the filter window is sequentially slid on the filled image B, one pixel point is slid each time, and one calculation is performed each time, the calculation method is that,
for each pixel point of the image B within the filter window, a convolution of the fast local laplace response value R with the pixel values of these points is calculated, resulting in an enhanced pixel value. The specific calculation formula is as follows:
enhanced_pixel(i,j)＝ΣΣ(R(i,j)*I(i+k,j+l))；k＝1，l＝1；
enhanced_pixel (i, j) represents an enhanced pixel value of a j-th pixel point of an i-th row in the enhanced image, and is abbreviated as EP (i, j) in the following; r (i, j) represents the value of the pixel point in the ith row and the jth column in the quick local Laplace response graph C, namely the quick local Laplace response value; i (i+k, j+l) represents the pixel value of the kth row and the kth column of pixel points of the image B in the filter window;
and traversing all pixel points of the filled image B, and replacing the original pixel points with the enhanced pixel values EP (i, j) to obtain an enhanced image D.
In step S50, binarization processing is used to highlight the edges and details of the lithium battery, specifically, the method is that the pixel value in the enhanced image D is converted into two values, namely 0 or 1, through the set threshold parameter, so that the enhanced image D is converted into a two-value image E with only two colors of black and white; the threshold value parameter is in the range of 0 to 255, which represents the pixel gray value of the image, when in processing, if the pixel value of a certain point of the image is higher than the threshold value parameter, the pixel value of the point is directly converted into 1, and if the pixel value of the point is lower than the threshold value parameter, the pixel value of the point is converted into 0.
The method for setting the threshold parameter is determined according to the characteristics and the processing requirements of the image, and specifically comprises the following steps:
fixed threshold: a fixed threshold parameter is directly set, and the pixel value of the image is set to be 1 which is larger than the threshold parameter and 0 which is smaller than the threshold parameter. The method is suitable for the condition that the brightness distribution of the image is relatively uniform in the lithium battery image detection. In this embodiment, a fixed threshold method is used.
Adaptive threshold: the threshold parameters are set according to the pixel gray values of the local areas of the image, and different threshold parameters are used for different areas. The method is suitable for the conditions of uneven image brightness distribution and large illumination variation.
Histogram-based threshold selection: the selection of suitable threshold parameters by analyzing the pixel histogram of the image enables the edges and features of the image to be highlighted. The method can automatically select the threshold parameters according to the gray level distribution of the image, and is suitable for different types of images.
In step S60, the specific method of edge detection and feature extraction is to perform an edge detection algorithm on the binary image E in S50 to extract edge information of the lithium battery, where the edge detection algorithm in this embodiment is a Canny algorithm, and specifically includes:
s61, gaussian filtering: the binary image E is first gaussian filtered to eliminate noise in the image. Gaussian filtering is an existing smoothing method that reduces the effect of noise by weighted averaging of pixels in an image.
S62, calculating gradient amplitude and direction: after Gaussian filtering, the gradient amplitude and direction of each pixel point in the image are calculated. Gradients, which represent the directions in the image in which the pixel values change most rapidly, are often used to detect edges.
S63, non-maximum suppression: and carrying out non-maximum suppression on the gradient amplitude image, screening out a pixel point with the maximum gradient amplitude, and reserving a local maximum value in the edge direction.
S64, double threshold processing: two thresholds, a high threshold and a low threshold, are set. Classifying pixels in the image according to gradient amplitude values: pixels above the high threshold are considered strong edges, pixels below the low threshold are considered weak edges, and pixels located between the two are considered possible edges.
S65, edge connection: and finally, forming a complete edge line by connecting the weak edge pixel points according to the strong edge pixel points.
The principle of the Canny algorithm is to detect edges in an image by gradient computation and thresholding. In particular, gaussian filtering is used to smooth the image and reduce noise, making edge detection more stable and reliable. By calculating the gradient magnitude and direction, the direction in the image in which the pixel value changes most rapidly can be found, which directions may be the location of the edge.
In the non-maximum value inhibition stage, the Canny algorithm selects a local maximum value as a possible edge point by comparing the gradient amplitude of the pixel point with the gradient amplitudes of the surrounding pixel points. Thus, the detection result can be more accurate, and the blurred edge is prevented from occurring.
In the double-threshold processing, the Canny algorithm divides the pixel points into strong and weak edges by setting high and low thresholds. A strong edge is a pixel that is believed to be an edge, while a weak edge is a pixel that is likely to be an edge. Finally, strong edges and surrounding weak edge pixel points are connected together through edge connection to form continuous edge lines.
After the image of the lithium battery is processed by adopting the steps S10-S60, key characteristic parameters in the image, such as size, position, shape and the like, can be conveniently extracted based on the shape and geometric characteristics of the lithium battery, and the lithium battery is classified, identified or quality evaluated according to the extracted characteristic parameters; in addition, the computer can output the result of the processing of the image of the lithium battery in the steps, and the method can include marking the edge of the lithium battery, displaying characteristic parameters or generating a processing report and the like, so that the efficiency of identifying the image of the lithium battery is greatly improved.
Through the steps S10 to S60, the technical scheme provided by the invention can realize rapid processing and enhancement of the lithium battery image, and more accurately extract the edge and characteristic information of the lithium battery, thereby achieving the purposes of image analysis and application. The solution combines the design and application of a fast local Laplace filter and subsequent edge detection and feature extraction steps, and can provide an efficient, accurate and reliable image processing solution in the lithium battery related field.
Specifically, the lithium battery enhancement method based on the Laplace algorithm has the following beneficial effects:
firstly, the image definition and contrast are improved: the invention can enhance the edge and detail information of the image and improve the definition and contrast of the image by carrying out local Laplace filtering processing on the lithium battery image. The key details in the lithium battery image are more clearly visible, and accurate analysis and identification are facilitated;
secondly, image noise and interference are suppressed: the algorithm of the invention can effectively inhibit noise and interference in the lithium battery image through filtering operation. Noise and interference are one of the main factors influencing the image quality and the visual effect, and the quality and the visual effect of the lithium battery image are improved by reducing the influence of the noise and the interference;
thirdly, fast running and high resource utilization: the invention optimizes the Laplace filter algorithm, so that the Laplace filter algorithm has the characteristics of quick operation and high resource utilization rate, and the multi-core characteristic of the modern CPU can be fully utilized, the calculation efficiency of the algorithm is improved, and the processing time is saved through parallel calculation, image blocking processing and quick algorithm design;
fourth, improve the image processing efficiency of the lithium battery: the technical scheme of the invention can improve the efficiency of image processing of the lithium battery. By rapid image enhancement and noise suppression, the time and workload of subsequent analysis and recognition are reduced, which is very beneficial to the situation that a large number of images need to be processed in the production, test and research processes of lithium batteries, and the working efficiency and accuracy can be improved.
In conclusion, the lithium battery image enhancement method based on the Laplace filter algorithm has remarkable beneficial effects in the aspects of improving image quality, suppressing noise, accelerating processing speed, improving working efficiency and the like. The method has the advantages that the image processing of the lithium battery is more accurate and reliable, powerful support is provided for analysis, diagnosis and evaluation of the lithium battery, the method has good application universality, the problems of high error rate, high complexity, low response speed and poor complex image processing capability of the existing image recognition method for detecting the lithium battery by adopting the Laplace algorithm are solved, and the efficiency and the accuracy of the image recognition of the lithium battery are improved.
The above is merely an embodiment of the present invention, and is not intended to limit the present invention. Various modifications and variations of the present invention will be apparent to those skilled in the art. Any modification, equivalent replacement, improvement, or the like, which is within the spirit and principles of the present invention, should be included in the scope of the claims of the present invention.
Claims (10)
1. The lithium battery enhancement method based on the Laplace algorithm is characterized by comprising the following steps of:
s10, preprocessing an image to obtain a preprocessed image A;
s20, designing a fast local Laplace filter, wherein the fast local Laplace filter comprises the following steps: s21, determining the size of a filter; s22, determining weight distribution of the filter; s23, generating a filter template;
s30, calculating a quick local Laplace response, which comprises the following steps: s31, edge filling is carried out on the preprocessed image A, and a filled image B is obtained; s32, performing first convolution calculation on the filled image B to obtain a quick local Laplace response value R and a quick local Laplace response chart C;
s40, enhancing the image, namely performing second convolution calculation on the filled image B and the quick local Laplace response graph C to obtain an enhanced image D.
2. The laplace algorithm-based lithium battery enhancement method as claimed in claim 1, wherein the image preprocessing method comprises: and obtaining an original image, carrying out graying treatment on the original image to obtain a gray image, and carrying out noise removal and smoothing treatment on the gray image to obtain the preprocessed image A.
3. The laplace algorithm-based lithium battery enhancement method as claimed in claim 1, wherein the size of the filter is m×m, m is greater than or equal to the size of the edge feature of the lithium battery, the edge feature is a set of edge pixel points of the lithium battery, and the size of the edge feature is an average value of widths of the set of edge feature pixel points in the extending direction of the set of edge feature pixel points.
4. The laplace algorithm-based lithium battery enhancement method as claimed in claim 3, wherein the weight distribution of the filter is gaussian weight distribution, the filter template is an m x m matrix, and the weight values of each point of the filter are taken as elements of the matrix.
5. The method for enhancing a lithium battery based on the laplace algorithm as claimed in claim 3, wherein the method for edge filling the preprocessed image a is to fill pixels with a width of m-1 outside the edge of the preprocessed image a, and pixel values of the filled pixels are all 0.
6. The laplace algorithm-based lithium battery enhancement method as set forth in any one of claims 3 to 5, wherein the method for performing the first convolution calculation on the filled image B includes: and using a window with the size of m multiplied by m as a filter window, sequentially sliding the filter window on the filled image B, sliding one pixel point each time, and performing calculation once each time, wherein the calculation is to calculate convolution of the filter template and pixel values of each point in the filter window, and traversing the filled image B.
7. The method of claim 6, wherein the second convolution calculation of the filled image B and the fast local laplace response map C is performed by sequentially sliding a filter window over the filled image B, one pixel at a time, and one calculation per sliding, using a window of size m x m as a filter window, the calculation being a convolution of a fast local laplace response value R and pixel values of the points for each pixel of the filled image B within the filter window, and traversing the filled image B.
8. The method for enhancing a lithium battery based on the laplace algorithm according to any one of claims 1 to 5, wherein the binarization processing is performed after the image enhancement: and converting the pixel value in the enhanced image D into a binary value through the set threshold parameter to obtain a binary image E.
9. The laplace algorithm-based lithium battery enhancement method as set forth in claim 8, wherein the binarization processing is performed by directly converting a certain pixel value of the enhanced image D into 1 if the certain pixel value is higher than the threshold parameter, and converting the certain pixel value into 0 if the certain pixel value is lower than the threshold parameter.
10. The laplace algorithm-based lithium battery enhancement method as set forth in claim 8, wherein after the binarization process, edge detection and feature extraction are performed on the binary image E using a Canny algorithm.
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311370727 | 2023-10-20 | ||
| CN2023113707274 | 2023-10-20 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117788448A true CN117788448A (en) | 2024-03-29 |
Family
ID=90381671
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311864904.4A Pending CN117788448A (en) | 2023-10-20 | 2023-12-29 | Lithium battery enhancement method based on Laplacian algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117788448A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118297839A (en) * | 2024-06-04 | 2024-07-05 | 山东华尚电气有限公司 | Transformer winding state detection method based on vision |
-
2023
- 2023-12-29 CN CN202311864904.4A patent/CN117788448A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118297839A (en) * | 2024-06-04 | 2024-07-05 | 山东华尚电气有限公司 | Transformer winding state detection method based on vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109410230B (en) | Improved Canny image edge detection method capable of resisting noise | |
| AU2016101504A4 (en) | Method for edge detection | |
| CN106856002B (en) | Unmanned aerial vehicle shooting image quality evaluation method | |
| CN118392891A (en) | Chip packaging defect detection method and system based on machine vision | |
| CN112070717B (en) | Power transmission line icing thickness detection method based on image processing | |
| CN118505693B (en) | Holographic printing quality detection method and system based on computer vision | |
| CN109472788B (en) | A method for detecting flaws on the surface of aircraft rivets | |
| CN118799796A (en) | Digital energy storage power station safety early warning system based on cloud computing | |
| CN115131359A (en) | Method for detecting pitting defects on surface of metal workpiece | |
| CN120070418B (en) | Structural facility deformation monitoring method and system based on machine vision | |
| CN112529853A (en) | Method and device for detecting damage of netting of underwater aquaculture net cage | |
| CN120635015A (en) | Facade crack detection method, device, equipment and medium using multi-scale feature fusion | |
| CN116934787A (en) | Image processing method based on edge detection | |
| CN117091530A (en) | A device and method for detecting roundness error of riveting holes based on machine vision | |
| CN120147975A (en) | Real-time monitoring method and system for liquid level in chemical storage tanks based on image recognition | |
| CN117745552A (en) | Self-adaptive image enhancement method and device and electronic equipment | |
| CN116935496A (en) | Electronic cigarette smoke visual detection method | |
| CN119850619A (en) | Method and device for detecting surface defects of LED special-shaped screen based on image analysis | |
| CN117788448A (en) | Lithium battery enhancement method based on Laplacian algorithm | |
| CN120374564B (en) | A detection and judgment method and device for a simple cow assembly machine | |
| CN117115174B (en) | Automatic detection method and system for appearance of pliers | |
| CN119540326A (en) | Wire rope diameter measurement method, system, medium and terminal based on machine vision | |
| CN108288267A (en) | A kind of scanning electron microscope (SEM) photograph image sharpness based on dark is without with reference to evaluation method | |
| CN119251220B (en) | Spacer vibration test result detection method based on machine vision | |
| CN118710671B (en) | A method, system and device for edge detection of multi-phase material images |
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 |