DEFECT DETECTION METHOD, DEVICE AND SYSTEM

1 . A method for detecting defects, comprising:

obtaining an average grayscale value of an image of an object to be detected;

constructing a mapping table, wherein elements of the mapping table comprise a mapped value corresponding to each grayscale value within a grayscale value range of the image, a mapped value corresponding to a grayscale value greater than or equal to a reference value is a first value, a mapped value corresponding to a grayscale value less than the reference value is a second value, and the reference value is an absolute value of a difference between the average grayscale value and a preset grayscale value;

searching for a mapped value corresponding to a grayscale value of each pixel in the image from the mapping table;

segmenting the image to obtain at least one suspicious defect sub-image based on the mapped value corresponding to the grayscale value of each pixel, wherein a mapped value corresponding to a grayscale value of each pixel in each suspicious defect sub-image is the first value; and

inputting the at least one suspicious defect sub-image into a machine learning model to obtain a defect detection result.

2 . The method according to claim 1 , wherein obtaining the average grayscale value of the image of the object to be detected comprises:

obtaining an original grayscale value range of the image; and

performing contrast stretching on the image to expand the original grayscale value range to the grayscale value range,

wherein the average grayscale value is an average grayscale value of the image after being subjected to the contrast stretching.

3 . The method according to claim 2 , wherein performing the contrast stretching on the image comprises:

converting the original grayscale value I1(x,y) of each pixel in the image into a grayscale value I2(x,y) according to the following formula:

I

⁢

2

⁢

(

x

,

y

)

=

{

c

,

I

⁢

1

⁢

(

x

,

y

)

<

a

d

,

I

⁢

1

⁢

(

x

,

y

)

>

b

I

⁢

1

⁢

(

x

,

y

)

-

a

b

-

a

·

(

d

-

c

)

,

a

≤

I

⁢

1

⁢

(

x

,

y

)

≤

b

wherein a is a lower limit of the original grayscale value range, b is an upper limit of the original grayscale value range, c is a lower limit of the grayscale value range, and d is an upper limit of the grayscale value range.

4 . The method according to claim 3 , wherein the first value is d, and the second value is c.

5 . The method according to claim 3 , wherein c=0, and d=255.

6 . The method according to claim 1 , wherein the machine learning model comprises a residual neural network model, and a total number of convolutional layers and fully connected layers in the residual neural network model is 14.

7 . The method according to claim 6 , wherein a difference between a maximum original grayscale value and a minimum original grayscale value of a non-defect region, other than the defect, in the image ranges from 35 to 50.

8 . The method according to claim 7 , wherein the difference between the maximum original grayscale value and the minimum original grayscale value is 40.

9 . The method according to claim 8 , wherein the maximum original grayscale value is 105, and the minimum original grayscale value is 75.

10 . The method according to claim 1 , wherein segmenting the image to obtain at least one suspicious defect sub-image based on the mapped value corresponding to the grayscale value of each pixel comprises:

segmenting the image to obtain a plurality of connected regions based on the mapped value corresponding to the grayscale value of each pixel, wherein a mapped value corresponding to a grayscale value of each pixel in each connected region is the first value;

in a case where two adjacent connected regions satisfy a preset condition, combining the two connected regions into a suspicious defect sub-image, wherein areas of the two connected regions are respectively a first area and a second area less than or equal to the first area, an area of an overlapped region of the two connected regions is a third area, and the preset condition comprises a ratio of the third area to the first area being greater than a preset ratio; and

in a case where the two connected regions do not satisfy the preset condition, determining the two connected regions as two suspicious defect sub-images.

11 . The method according to claim 10 , wherein the preset ratio is greater than 0.5 and less than 1.

12 . The method according to claim 11 , wherein the preset ratio is 0.8.

13 . The method according to claim 1 , wherein a data type of an element in the mapping table is an unsigned char type.

14 . The method according to claim 1 , wherein the defect detection result comprises a defect type.

15 . The method according to claim 1 , wherein the object to be detected comprises an electrode plate of a battery.

16 . The method according to claim 15 , wherein the battery comprises a lithium battery.

17 . A defect detection apparatus, comprising:

an obtaining module, configured to obtain an average grayscale value of an image of an object to be detected;

a construction module, configured to construct a mapping table, wherein elements of the mapping table comprise a mapped value corresponding to each grayscale value within a grayscale value range of the image, a mapped value corresponding to a grayscale value greater than or equal to a reference value is a first value, a mapped value corresponding to a grayscale value less than the reference value is a second value, and the reference value is an absolute value of a difference between the average grayscale value and a preset grayscale value;

a search module, configured to search for a mapped value corresponding to a grayscale value of each pixel in the image from the mapping table;

a segmentation module, configured to segment the image to obtain at least one suspicious defect sub-image based on the mapped value corresponding to the grayscale value of each pixel, wherein a mapped value corresponding to a grayscale value of each pixel in each suspicious defect sub-image is the first value; and

an input module, configured to input the at least one suspicious defect sub-image into a machine learning model to obtain a defect detection result.

18 . A defect detection apparatus, comprising:

a memory; and

a processor coupled to the memory and configured to execute, based on instructions stored in the memory, the method according to claim 1 .

19 . A defect detection system, comprising:

the defect detection apparatus according to claim 17 ; and

an imaging apparatus, configured to scan the object to be detected, to obtain the image.

20 . A non-transitory computer-readable storage medium storing computer program instructions, wherein when executed by a processor, the computer program causes the processor to perform a defect detection method that comprises:

obtaining an average grayscale value of an image of an object to be detected;

constructing a mapping table, wherein elements of the mapping table comprise a mapped value corresponding to each grayscale value within a grayscale value range of the image, a mapped value corresponding to a grayscale value greater than or equal to a reference value is a first value, a mapped value corresponding to a grayscale value less than the reference value is a second value, and the reference value is an absolute value of a difference between the average grayscale value and a preset grayscale value;

searching for a mapped value corresponding to a grayscale value of each pixel in the image from the mapping table;

segmenting the image to obtain at least one suspicious defect sub-image based on the mapped value corresponding to the grayscale value of each pixel, wherein a mapped value corresponding to a grayscale value of each pixel in each suspicious defect sub-image is the first value; and

inputting the at least one suspicious defect sub-image into a machine learning model to obtain a defect detection result.