CN103925893B - Quality detection method of battery cells - Google Patents
Quality detection method of battery cells Download PDFInfo
- Publication number
- CN103925893B CN103925893B CN201410153942.3A CN201410153942A CN103925893B CN 103925893 B CN103925893 B CN 103925893B CN 201410153942 A CN201410153942 A CN 201410153942A CN 103925893 B CN103925893 B CN 103925893B
- Authority
- CN
- China
- Prior art keywords
- image
- gray
- electrode
- edge
- value
- 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
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a quality detection method of battery cells. The method includes the steps that firstly, all electrodes of one battery cell are imaged through X-rays, and an image is collected; secondly, smoothing is carried out on the X-ray image to filter out noise points, and gray stretching is also carried out to improve the contrast ratio; thirdly, the portion, uneven in gray level, of the image is adjusted according to a curve about the change of the horizontal direction of a calibration point and the change of the grey level of the calibration point in the X-ray image to improve the quality of the image; fourthly, the portion, at the edges of the electrodes, of the image is divided to extract an ROI image of each electrode; fifthly, the gray average and variance of the ROI images of the electrodes are adjusted to be of the same gray level; sixthly, the edge line of the ROI image of each electrode is detected to work out the positions of the cathode and the anode and the distance between the cathode and the anode; seventhly, whether the distance between the cathode and the anode is up to standard or not is judged, and a detection result signal is sent out; eighthly, detection is carried out on the next battery cell. By means of the method, the distance between the two electrodes of each battery cell can be accurately detected, and then the quality of the battery cell can be effectively judged.
Description
Technical field
The present invention relates to a kind of winding battery detection technique, particularly relate to a kind of by winding battery
The quality determining method of X-ray picture.
Background technology
The battery formed with the battery core of winding method combination forming, referred to as winding battery.Winding battery
Also referred to as battery core, battery insider is referred to as core.
In the production technology of winding battery, the distance of each layer negative and positive two electrode determines the quality of battery.
Under the conditions of undamaged to battery, the distance accurately and effectively detecting winding battery the two poles of the earth is battery
One of key link in technique.
Summary of the invention
It is an object of the invention to overcome the deficiencies in the prior art, it is provided that the quality inspection of a kind of winding battery
Survey method, can detect winding battery the two poles of the earth accurately and effectively by the electrode x-ray image of winding battery
Distance, thus effectively judge the quality of battery.
In order to achieve the above object, the technical solution used in the present invention is:
The quality determining method of a kind of winding battery, comprises the following steps:
Step 1. passes through X-ray electrode all to winding battery imaging, and gathers image;
X-ray image is smoothed by step 2., filters noise spot, and uses gray scale to stretch, carries
High-contrast;
Step 3. is according to the song of x-ray image acceptance of the bid position, point level direction with fixed point gray-value variation
Line, adjusts the image that gray scale is uneven, improves picture quality;
Step 4. splits electrode edge image, and extracts electrode edge ROI image;
The gray average of each electrode edge ROI image and variance are adjusted to same gray level by step 5.;
The edge of step 6. detecting electrode ROI image, calculates position and the distance at negative and positive the two poles of the earth;
According to technological requirement, step 7. judges that the distance at negative and positive the two poles of the earth is the most qualified, provide testing result letter
Number;
Step 8. enters the detection process of next battery.
It is preferred that the noise spot that filters described in step 2 specifically includes:
Step S001. uses Gauss operator to x-ray image smoothing processing, the Gauss being smoothed
Operator formula is:
G (x)=A*exp ((-(x-(k-1)/2)2)/(2*s2)), wherein, A is constant (general value 1), x
For the grey scale pixel value of image, k is the size of wave filter, s smothing filtering coefficient;
Step S002. uses gray scale stretching to remove the pixel exceeding gray scale values in image,
And improve the contrast of image, gray scale stretching operator is as follows:
F (x)=(((x-xL)/(xH-xL))gamma) * (yH-yL)+yL,
Wherein, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after conversion, and [xL, xH] is for becoming
Changing the gray-value variation scope in the horizontal direction of rear gray value, [yL, yH] is the ash on vertical direction
Angle value excursion, gamma is gray value smoothing factor.
It is preferred that the fixed point horizontal direction position described in step 3 and the song of fixed point gray-value variation
Line, fixed point conic section formula is f (x)=A*x*x+B*x+C, and wherein A, B, C are respectively
The multinomial coefficient of conic section, x is the position of the pixel horizontal direction of image, and f (x) is pixel
Compensating for gray-scale value.
It is preferred that the extraction electrode edge ROI image described in step 4, specifically include:
Step S101., in battery ROI image, calculates average intensity change in vertical direction adjacent column
Maximum position, is set to first electrode edge;
Step S102. uses Hough transform detection of straight lines method, calculates first electrode edge direction;
Hough operator formula is: ρ=x*cos (θ)+y*sin (θ);In formula, (ρ, θ) is rectangular coordinate
It is that (x, y) corresponding to the coordinate points in polar coordinate system at midpoint;
Step S103., according to electrode deflection direction, corrects to vertical direction;
Step S104. is alternately distributed feature according to anode and cathode in image, extracts the ROI figure of each layer electrode
Picture.
It is preferred that the gray average by each electrode ROI image described in step 5 is adjusted to same with variance
Gray level uses normalization operator to be adjusted in same gray level by the image of different grey-scale, and normalization is calculated
Son is as follows:
If x > mean
F (x)=mean0+sqrt (std0* (x-mean) * (x-mean)/std);
others
F (x)=mean0-sqrt (std0* (x-mean) * (x-mean)/std);
In formula, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after being adjusted, mean0
Being respectively desired gray average and variance with std0, the gray scale that mean with std is respectively image is equal
Value and variance.
It is preferred that the detection at the electrode ROI image edge described in step 6 includes sharp-edged edge
Image and edge blurry and interrupted edge image;Edge image clearly, the projection of its electrode direction
In curve, the Gray Projection value of edge changes greatly, and exceedes default gray value, and fuzzy image,
The drop shadow curve of its electrode direction is shallower, and the Gray Projection value of edge is less than presetting gray value;
Specifically include:
Step S301. obtains the half-tone information of electrode local ROI image, uses gray scale stretching to remove super
Go out the pixel of default intensity value ranges, and improve the contrast of image;
Step S302. uses the gradient operator of horizontal direction, detecting electrode ROI image edge;
The gradient operator of horizontal direction is as follows: H=(-1,0,1 ,-1,0,1 ,-1,0,1);
Step S303. uses morphology to open operation, removes the noise spot of edge image;
Step S304. passes through Hough transform detection of straight lines method, extracts the original position of cathode edge;
Step S305. filters the background area of more than negative electrode, the Gray Projection figure on calculated level direction
Picture;
In step S306. detection Gray Projection image, greatest gradient value position, i.e. anode edge is initial
Position;
Step S307., according to the gray distribution features of anode and cathode ROI image, calculates negative and positive two electrode district
The maximum classification line of territory pixel grey scale distribution, the true demarcation line of i.e. two electrodes;
Step S308. thus calculates position and the distance of negative and positive two electrode of each layer of battery.
It is preferred that the anode initiating terminal that the technological requirement described in step 7 includes cell top end falls at negative electrode
On, the anode covered cathode of bottom, it is qualified that the distance at electrode negative and positive the two poles of the earth is judged in preset range,
Otherwise, for exception, and provide detection abnormal signal.
Compared with prior art, the invention has the beneficial effects as follows: by electrode layer is split, gone
Make an uproar, edge extracting etc. operates, and obtains electrode edge line, thus obtains the distance at negative and positive the two poles of the earth, enters one
Step judges whether battery meets the requirements, and can detect the distance at winding battery the two poles of the earth accurately and effectively,
Thus effectively judge the quality of battery.
Accompanying drawing explanation
Fig. 1 is the method flow diagram of the present invention;
Fig. 2 is the specific embodiment flow chart of the present invention.
Detailed description of the invention
Idea of the invention is that and overcome the deficiencies in the prior art, it is provided that the quality inspection of a kind of winding battery
Survey method, winding battery imaging in X-ray is fuzzyyer, and contrast is relatively low, and noise image or discrete
Image is more.Therefore need image is carried out denoising, then detect negative and positive the two poles of the earth according to gray feature
Distance.The detection major way of winding battery is by the X-ray all electrode imagings to winding battery,
The original position detecting each layer negative and positive two electrode edge calculates the distance of two electrodes.Wind according to power
The technological requirement of battery, the anode initiating terminal of cell top end must fall on negative electrode, and the anode of bottom must
Palpus covered cathode, therefore, calculates both distances by negative and positive the two poles of the earth gray feature in the picture,
Thus judge battery quality.
Be described in detail referring to the drawings below in conjunction with embodiment, in order to the technical characteristic of the present invention and
Advantage is interpretated more in-depth.
The method flow diagram of the present invention, as it is shown in figure 1, the quality determining method of a kind of winding battery, wraps
Include following steps:
Step 1. passes through X-ray electrode all to winding battery imaging, and gathers image;
Step 2. uses Gauss operator to x-ray image smoothing processing, filtering image noise point;Gray scale is drawn
Stretch and remove the pixel exceeding default intensity value ranges, and improve the contrast of image;
Step 3. is bent with fixed point gray-value variation according to x-ray image acceptance of the bid position, point level direction
Line, compensates the pixel that in image, the upper grey scale change of each row is bigger;
Step 4. splits electrode edge image, and extracts electrode edge ROI image;
The gray average of each layer electrode ROI image and variance are adjusted to same gray level by step 5.;
The edge of step 6. detecting electrode ROI image, calculates position and the distance at negative and positive the two poles of the earth;
According to technological requirement, step 7. judges that the distance at negative and positive the two poles of the earth is the most qualified, provide testing result letter
Number;
Step 8. enters the detection process of next battery.
It is preferred that the noise spot that filters described in step 2 specifically includes:
Step S001. uses Gauss operator to x-ray image smoothing processing, the Gauss being smoothed
Operator formula is:
G (x)=A*exp ((-(x-(k-1)/2)2)/(2*s2)), wherein, A is constant, and x is the pixel of image
Gray value, k is the size of wave filter, s smothing filtering coefficient;
Setup parameter value k=3, s=1.0, original image is I=
0.7,0,0.2,0.5,0.3,0.1,0.7,0.5,0.7}, then after Gaussian smoothing algorithm, image
Data are as follows: g (I)=and 0.227,0.162,0.148,0.418,0.333,0.287,0.590,
0.525,0.511}。
Step S002. uses the pixel in gray scale stretching removal image beyond default intensity value ranges,
And improve the contrast of image, gray scale stretching operator is as follows:
F (x)=(((x-xL)/(xH-xL))gamma)*(yH-yL)+yL,
Wherein, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after conversion, and [xL, xH] is for becoming
Changing the gray-value variation scope in the horizontal direction of rear gray value, [yL, yH] is the ash on vertical direction
Angle value excursion, gamma is gray value smoothing factor.
Setup parameter value xL=0, xH=0.5, yL=0.2, yH=0.7, gamma=1.0, original image is
I={112,227,36,66,54,161,252,200,193}, then after gray scale stretches, view data
As follows: f (x)={ 163,255,87,117,105,212,255,251,244}
It is preferred that the fixed point horizontal direction position described in step 3 and the song of fixed point gray-value variation
Line, the i.e. position of image pixel horizontal direction and the conic section relation of its template grey scale pixel value, mend
Repay the gray value of the uneven image of gray scale, improve the quality of image.
It is preferred that the extraction electrode edge ROI image described in step 4, specifically include:
Step S101., in battery ROI image, calculates average intensity change in vertical direction adjacent column
Maximum position, is set to first electrode edge;
Step S102. uses Hough transform detection of straight lines method, calculates first electrode edge direction;
Straight line in rectangular coordinate system, after Hough transform, in polar coordinate system, corresponds to one
Point, therefore can accurately detect in rectangular coordinate system all according to the gray value of the pixel in polar coordinate system
The straight line in direction;
Step S103., according to electrode deflection direction, corrects to vertical direction;
Step S104. is alternately distributed feature according to anode and cathode in image, extracts the ROI figure of each layer electrode
Picture.
It is preferred that the gray average by each electrode ROI image described in step 5 is adjusted to same with variance
Gray level uses normalization operator to be adjusted in same gray level by the image of different grey-scale, and normalization is calculated
Son is as follows:
If x > mean
F (x)=mean0+sqrt (std0* (x-mean) * (x-mean)/std);
others
F (x)=mean0-sqrt (std0* (x-mean) * (x-mean)/std);
In formula, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after being adjusted, mean0
Being respectively desired gray average and variance with std0, the gray scale that mean with std is respectively image is equal
Value and variance.
Setting parameter value mean0=120, std0=120, original image
I={174,32,124,66,142,67,11,233,217}, then after normalization, view data is:
F (x)={ 190,10,127,53,149,55,0,255,244}.
It is preferred that the detection at the electrode ROI image edge described in step 6 includes sharp-edged edge
Image and edge blurry and interrupted edge image;Edge image clearly, its gray average relatively big and
Grey scale change is obvious, and fuzzy image then gray average is less than normal, and grey scale change is less, and edge is not
Continuously.
Step 6 specifically includes:
Step S301. obtains the half-tone information of electrode local ROI image, uses gray scale stretching algorithm,
Remove the pixel exceeding default intensity value ranges in image, and improve the contrast of image;
Step S302. uses the gradient operator of horizontal direction, detecting electrode ROI image edge;
The gradient operator of horizontal direction is as follows: H=(-1,0,1 ,-1,0,1 ,-1,0,1);
Step S303. uses morphology to open operation, removes the noise spot of edge image;
Step S304. passes through Hough transform detection of straight lines method, extracts the original position of cathode edge;
Use Hough transform, exactly the straight line in rectangular coordinate system can be expressed as polar coordinate system
In point, thus can detect directive straight line accurately;
Step S305. filters the background area of more than negative electrode, the Gray Projection figure on calculated level direction
Picture;
In step S306. detection Gray Projection image, greatest gradient value position, i.e. anode edge is initial
Position;
Step S307., according to the gray distribution features of anode and cathode ROI image, calculates negative and positive two electrode district
The maximum classification line of territory pixel grey scale distribution, the true demarcation line of i.e. two electrodes;
Step S308. thus calculates position and the distance of negative and positive two electrode of each layer of battery.
It is preferred that the anode initiating terminal that the technological requirement described in step 7 includes cell top end falls at negative electrode
On, the anode covered cathode of bottom, it is qualified that the distance at electrode negative and positive the two poles of the earth is judged in preset range,
Otherwise, for exception, and provide detection abnormal signal.
As in figure 2 it is shown, winding battery imaging in X-ray is fuzzyyer in concrete operations, contrast is relatively
Low, and noise image or discrete picture more.Therefore need to process, image denoising then according to gray scale
The distance at feature detection negative and positive the two poles of the earth.
Operating process is as follows:
(1) Gaussian smoothing operator g (x)=A*exp ((-(x-(k-1)/2) is used2)/(2*s2))), filter little making an uproar
Sound point.In formula, A is constant (general value 1), and x is the grey scale pixel value of image, and k is wave filter
Size, s smothing filtering coefficient;
Use gray scale stretching to remove in image and exceed the pixel presetting intensity value ranges, and improve image
Contrast, gray scale stretching operator formula as follows:
F (x)=(((x-xL)/(xH-xL))gamma)*(yH-yL)+yL,
Wherein, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after conversion, and [xL, xH] is for becoming
Changing the gray-value variation scope in the horizontal direction of rear gray value, [yL, yH] is the ash on vertical direction
Angle value excursion, gamma is gray value smoothing factor;
(2) according to the conic section relation of the horizontal direction position of fixed point in image Yu its gray value,
Improve the picture quality that gray scale is uneven;
(3) use morphology to open operation (first corrode and expand afterwards), remove noise image less in image;
(4) detect first electrode edge position of battery, use Hough transform detection of straight lines method
Calculate edge direction, and correcting electrode is on vertical direction;
(5) split the local ROI image of each electrode layer, and use normalization operator will there is certain ash
The topography of degree difference, adjusts to same grey scale change, is consequently adapted to certain grey scale change
Electrode image detection;
(6) for sharp-edged electrode image, according to the half-tone information of electrode topography, use
Horizontal direction gradient operator (H=(-1,0,1 ,-1,0,1 ,-1,0,1)) detection image edge pixels, and
Open operation with morphology, remove part noise spot in edge pixel, then detected by Hough transform
Linear method extracts the position of electrode edge;
(7) for edge blurry and interrupted electrode image, according to the grey scale change of electrode topography
Gray feature in feature and electrode edge, the Gray Projection on calculated level direction, detection gray scale is thrown
The original position of greatest gradient value position, i.e. electrode in shadow curve;
(8) demarcation line, negative and positive the two poles of the earth produces relatively fuzzy watershed area because electrode roll necessarily fluctuates around existence
Territory, therefore, need to by calculating at the adjacent area grey scale change maximum of two electrode local area image,
The i.e. maximum classification line of negative and positive two electrode zone pixel distribution, it is determined as the true demarcation line of two electrodes;
(9) by the initiating terminal position at the negative and positive the two poles of the earth calculated in (7) and (8), i.e. obtain two electrodes away from
From, thus judge whether electrode meets technique and prescription.
Above content is to combine concrete optimal way further description made for the present invention, no
Should assert the present invention be embodied as be confined to described above.For those skilled in the art
For, without departing from the inventive concept of the premise, it is also possible to make some simple deduction or replace,
Within the protection domain that the claim being regarded as being submitted to by the present invention determines.
Claims (6)
1. a quality determining method for winding battery, comprises the following steps:
Step 1. passes through X-ray electrode all to winding battery imaging, and gathers image;
X-ray image is smoothed by step 2., filters noise spot, and uses gray scale to stretch, carries
High-contrast;
Step 3. is according to the song of x-ray image acceptance of the bid position, point level direction with fixed point gray-value variation
Line, adjusts the image that gray scale is uneven, improves picture quality;
Step 4. splits electrode edge image, and extracts electrode edge ROI image;
The gray average of each electrode edge ROI image and variance are adjusted to same gray level by step 5.;
The edge of step 6. detecting electrode ROI image, calculates position and the distance at negative and positive the two poles of the earth;
According to technological requirement, step 7. judges that the distance at negative and positive the two poles of the earth is the most qualified, provide testing result letter
Number;
Step 8. enters the detection process of next battery;
The edge of the detecting electrode ROI image described in step 6 includes sharp-edged edge image and limit
The edge image that edge is fuzzy and interrupted;Edge image clearly, in the drop shadow curve of its electrode direction,
The Gray Projection value of edge changes greatly, and exceedes default gray value, and fuzzy image, its electrode
The drop shadow curve in direction is shallower, and the Gray Projection value of edge is less than presetting gray value;
Specifically include:
Step S301. obtains the half-tone information of electrode local ROI image, uses gray scale stretching to remove super
Go out the pixel of default intensity value ranges, and improve the contrast of image;
Step S302. uses the gradient operator of horizontal direction, detecting electrode ROI image edge;
The gradient operator of horizontal direction is as follows: H=(-1,0,1 ,-1,0,1 ,-1,0,1);
Step S303. uses morphology to open operation, removes the noise spot of edge image;
Step S304. passes through Hough transform detection of straight lines method, extracts the original position of cathode edge;
Step S305. filters the background area of more than negative electrode, the Gray Projection figure on calculated level direction
Picture;
In step S306. detection Gray Projection image, greatest gradient value position, i.e. anode edge is initial
Position;
Step S307., according to the gray distribution features of anode and cathode ROI image, calculates negative and positive two electrode district
The maximum classification line of territory pixel grey scale distribution, the true demarcation line of i.e. two electrodes;
Step S308. thus calculates position and the distance of negative and positive two electrode of each layer of battery.
The quality determining method of winding battery the most according to claim 1, it is characterised in that step
The noise spot that filters described in rapid 2 specifically includes:
Step S001. uses Gauss operator to x-ray image smoothing processing, the Gauss being smoothed
Operator formula is:
G (x)=A*exp ((-(x-(k-1)/2)2)/(2*s2)), wherein, A is constant, and x is the pixel of image
Gray value, k is the size of wave filter, s smothing filtering coefficient;
Step S002. uses gray scale stretching to remove the pixel exceeding gray scale values in image,
And improve the contrast of image, gray scale stretching operator is as follows:
F (x)=(((x-xL)/(xH-xL))gamma) * (yH-yL)+yL,
Wherein, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after conversion, and [xL, xH] is for becoming
Changing the gray-value variation scope in the horizontal direction of rear gray value, [yL, yH] is the ash on vertical direction
Angle value excursion, gamma is gray value smoothing factor.
The quality determining method of winding battery the most according to claim 2, it is characterised in that:
Fixed point horizontal direction position described in step 3 and the curve of fixed point gray-value variation, fixed point two
Secondary curve equation is f (x)=A*x*x+B*x+C, and wherein to be respectively conic section many for A, B, C
Binomial coefficient, x is the position of the pixel horizontal direction of image, and f (x) is the compensating for gray-scale value of pixel.
The quality determining method of winding battery the most according to claim 3, it is characterised in that
Extraction electrode edge ROI image described in step 4, specifically includes:
Step S101., in battery ROI image, calculates average intensity change in vertical direction adjacent column
Maximum position, is set to first electrode edge;
Step S102. uses Hough transform detection of straight lines method, calculates first electrode edge direction;
Hough operator formula is: ρ=x*cos (θ)+y*sin (θ);In formula, (ρ, θ) is rectangular coordinate
It is that (x, y) corresponding to the coordinate points in polar coordinate system at midpoint;
Step S103., according to electrode deflection direction, corrects to vertical direction;
Step S104. is alternately distributed feature according to anode and cathode in image, extracts the ROI figure of each layer electrode
Picture.
The quality determining method of winding battery the most according to claim 4, it is characterised in that
The gray average of each electrode ROI image be adjusted to same gray level with variance adopt described in step 5
Being adjusted in same gray level by the image of different grey-scale with normalization operator, normalization operator is as follows:
If x > mean
F (x)=mean0+sqrt (std0* (x-mean) * (x-mean)/std);
others
F (x)=mean0-sqrt (std0* (x-mean)*(x-mean)/std);
In formula, x is image slices vegetarian refreshments gray value, and f (x) is the gray value after being adjusted, mean0
Being respectively desired gray average and variance with std0, the gray scale that mean with std is respectively image is equal
Value and variance.
The quality determining method of winding battery the most according to claim 5, it is characterised in that:
Technological requirement described in step 7 includes that the anode initiating terminal of cell top end falls on negative electrode, the sun of bottom
Pole covered cathode, it is qualified that the distance at electrode negative and positive the two poles of the earth is judged in preset range, otherwise, for different
Often, and provide detection abnormal signal.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410153942.3A CN103925893B (en) | 2014-04-17 | 2014-04-17 | Quality detection method of battery cells |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410153942.3A CN103925893B (en) | 2014-04-17 | 2014-04-17 | Quality detection method of battery cells |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103925893A CN103925893A (en) | 2014-07-16 |
CN103925893B true CN103925893B (en) | 2017-01-11 |
Family
ID=51144194
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410153942.3A Active CN103925893B (en) | 2014-04-17 | 2014-04-17 | Quality detection method of battery cells |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103925893B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106127793B (en) * | 2016-07-29 | 2019-07-09 | 广东正业科技股份有限公司 | A kind of extracting method of battery anodic-cathodic profile |
CN106767574B (en) * | 2016-12-12 | 2019-01-29 | 上海电气钠硫储能技术有限公司 | A kind of sodium-sulphur battery detection x-ray detection device |
CN107941805A (en) * | 2017-12-01 | 2018-04-20 | 无锡先导智能装备股份有限公司 | Battery core quality determining method |
CN110222679B (en) * | 2019-05-10 | 2023-07-11 | 惠州市德赛电池有限公司 | General battery polarity automatic detection method based on deep learning |
CN113313677B (en) * | 2021-05-17 | 2023-04-18 | 武汉工程大学 | Quality detection method for X-ray image of wound lithium battery |
CN113409294B (en) * | 2021-06-30 | 2023-03-24 | 广东利元亨智能装备股份有限公司 | Core-pulling detection method of roll core, electronic equipment and storage medium |
CN113674197B (en) * | 2021-07-02 | 2022-10-04 | 华南理工大学 | Method for dividing back electrode of solar cell |
CN115839684A (en) * | 2022-10-20 | 2023-03-24 | 宁德时代新能源科技股份有限公司 | Gap detection device and gap detection method |
CN118096765A (en) * | 2024-04-28 | 2024-05-28 | 江苏时代新能源科技有限公司 | Battery cell detection method and device, electronic equipment and storage medium |
Family Cites Families (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001283935A (en) * | 2000-03-30 | 2001-10-12 | Sony Corp | Inspection method of battery sealing structured and battery-sealing structure inspection device |
JP2002198097A (en) * | 2000-12-25 | 2002-07-12 | Sony Corp | Method for inspecting cell |
CN101539533B (en) * | 2009-03-11 | 2011-06-22 | 华南理工大学 | Automatic detecting device and method for battery internal defect |
CN102737395B (en) * | 2011-04-15 | 2017-07-11 | 深圳迈瑞生物医疗电子股份有限公司 | Image processing method and device in a kind of medical X-ray system |
CN103196373B (en) * | 2013-03-11 | 2015-09-30 | 江南大学 | A kind of automatic testing method of battery plus-negative plate terminal lug length |
-
2014
- 2014-04-17 CN CN201410153942.3A patent/CN103925893B/en active Active
Also Published As
Publication number | Publication date |
---|---|
CN103925893A (en) | 2014-07-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN103925893B (en) | Quality detection method of battery cells | |
CN109410230A (en) | One kind can antimierophonic improvement Canny method for detecting image edge | |
CN104063866B (en) | A kind of particle size detection method in ore transmit process | |
CN107742291B (en) | Defect detection method and device for photovoltaic glass | |
CN102855617B (en) | Method and system for processing adaptive images | |
CN104574337B (en) | Based on the image enchancing method that bilateral gamma correction and multi-scale image merge | |
CN105067638A (en) | Tire fetal-membrane surface character defect detection method based on machine vision | |
CN105913396A (en) | Noise estimation-based image edge preservation mixed de-noising method | |
WO2013148566A1 (en) | Image blur detection | |
CN105096269A (en) | Radial image distortion rectifying method and system based on distorted linear structure detection | |
CN110648330B (en) | Defect detection method for camera glass | |
CN105046683A (en) | Object detection method based on adaptive-parameter-adjustment Gaussian mixture model | |
CN108171674B (en) | Vision correction method for projector image with any visual angle | |
CN104537650A (en) | Polarity direction judging method and device for aluminum electrolytic capacitor | |
CN103413311A (en) | Edge-based fuzzy detection method | |
Hou et al. | Welding image edge detection and identification research based on canny operator | |
WO2010050412A1 (en) | Calibration index determination device, calibration device, calibration performance evaluation device, system, method, and program | |
CN104866847A (en) | Rapid and robust answering card identifying method | |
CN103489167A (en) | Automatic image sharpening method | |
CN102693535B (en) | Method for detecting light bundling device area in DR image | |
CN115908410B (en) | Pressure vessel laser welding control method based on machine vision | |
CN110889807A (en) | Image processing method for channel type X-ray security inspection equipment | |
CN115060754B (en) | Stainless steel product surface quality detection method | |
CN106709952A (en) | Automatic calibration method of display screen | |
CN114140416A (en) | Glass edge detection method and system based on machine vision |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |