IP Library Granted Patent US 12,664,642
Granted Patent B2
US 12,664,642 · App. 18/503,170 · Granted Jun 23, 2026

Appearance detection method and apparatus for bare cell, computer device, and storage medium

Inventors: Yinhang Tu (Ningde, CN); Xiaofeng Liu (Ningde, CN); Jinping Zhu (Ningde, CN)
Assignee: CONTEMPORARY AMPEREX TECHNOLOGY (HONG KONG) LIMITED
G06T7/001G06T7/13G06T2207/30108
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,664,642
App. No.
18/503,170
Filed
Nov 6, 2023
Granted
Jun 23, 2026
Kind
B2
Art Unit
2662
USPC
382/141
Abstract

Provided are an appearance detection method and apparatus for a bare cell, a computer device, a computer-readable storage medium, a computer program product, and an appearance detection system for a bare cell, where the method includes: obtaining a cell picture of the bare cell by shooting, the cell picture being obtained by shooting in alignment with positions of tabs on an edge of the bare cell; determining a detection object in the cell picture; and performing appearance detection based on the detection object to obtain detection result information. The cell picture is obtained by shooting in alignment with the positions of the tabs on the edge of the bare cell, such that imaging of cell edges is parallel light reflection imaging.

Claims (86)

1 . An appearance detection method for a bare cell, comprising:

obtaining a cell picture of the bare cell by shooting, the cell picture being obtained by shooting in alignment with positions of tabs on an edge of the bare cell, so that a tab region and an edge of a cell separator adjacent to the tab region are included in the cell picture, wherein the cell picture comprises a first cell picture, an exposure of the first cell picture is higher than a preset exposure threshold;

determining a detection object in the cell picture, the detection object comprising an edge of a cell separator, wherein determining the detection object in the cell picture further comprises:

determining a reference position in the first cell picture based on a preset cell model, wherein determining the reference position comprises matching the preset cell model from the first cell picture to obtain center position coordinates of the cell separator and establishing a spatial coordinate system based on the center position coordinates,

setting edge finding regions free from interference of tab regions in a space following mode, wherein for an edge where the tabs are arranged, a plurality of different edge finding regions are set to skip over tab regions, and the edge of the cell separator is obtained by fitting based on points found in real time in the plurality of different edge finding regions, and

finding the edge of the cell separator based on the reference position; and

performing appearance detection based on the detection object to obtain detection result information, the detection result information comprising pin pulling detection result, wherein the performing appearance detection further comprises:

re-establishing a spatial coordinate system by using an intersection point of edges of the cell separator as an origin of the coordinate system,

determining a pin pulling detection region based on the edge of the cell separator and positions of the tabs; and

performing detection on the pin pulling detection region to obtain the pin pulling detection result,

wherein the performing detection on the pin pulling detection region further comprises:

obtaining distance values between edge points of the cell separator in the pin pulling detection region and a corresponding edge of the cell separator to obtain a plurality of distance values; and

obtaining the pin pulling detection result by analysis based on the plurality of distance values.

2 . The method according to claim 1 , wherein the detection result information comprises size data; and the performing appearance detection based on the detection object to obtain detection result information comprises:

calculating distances between all points on an edge of the cell separator and an opposite edge to obtain a plurality of pieces of edge distance data; and

obtaining the size data based on the plurality of pieces of edge distance data.

3 . The method according to claim 2 , wherein the edge distance data comprises length data and width data; and the obtaining the size data based on the plurality of pieces of edge distance data comprises:

performing averaging on a plurality of pieces of length data to obtain an average length value;

performing averaging on a plurality of pieces of width data to obtain an average width value; and

obtaining the size data based on the average length value and the average width value.

4 . The method according to claim 1 , wherein the performing detection on the pin pulling detection region to obtain the pin pulling detection result comprises:

looking for and obtaining spot information within the pin pulling detection region, and obtaining the pin pulling detection result by analysis based on the spot information.

5 . The method according to claim 4 , wherein the spot information comprises at least one of an area of a single spot, a size of a spot, and a sum of spot areas.

6 . The method according to claim 1 , wherein the obtaining the pin pulling detection result by analysis based on the plurality of distance values comprises:

sorting the plurality of distance values to obtain sorted distance values;

selecting a maximum value of the sorted distance values; and

obtaining the pin pulling detection result by analysis based on the selected maximum value.

7 . The method according to claim 1 , wherein after the performing appearance detection based on the detection object to obtain detection result information, the method further comprises: storing the detection result information by binding the information with identification information of the bare cell.

8 . The method according to claim 1 , wherein the cell picture is shot by using two cameras respectively in alignment with the positions of the tabs on the edge of the bare cell.

9 . An appearance detection system for a bare cell, the appearance detection system comprising: an image obtaining apparatus and a host computer, wherein the image obtaining apparatus is configured to obtain a cell picture by shooting in alignment with positions of tabs on an edge of the bare cell, and send the cell picture to the host computer; and the host computer is configured to perform appearance detection on the bare cell by using a method according to claim 1 .

10 . The system according to claim 9 , wherein the image obtaining apparatus comprises two cameras.

11 . An appearance detection apparatus for a bare cell, the appearance detection apparatus comprising:

a picture obtaining module configured to obtain a cell picture of the bare cell by shooting, the cell picture being obtained by shooting in alignment with positions of tabs on an edge of the bare cell, so that a tab region and an edge of a cell separator adjacent to the tab region are included in the cell picture, wherein the cell picture comprises a first cell picture, an exposure of the first cell picture is higher than a preset exposure threshold;

a picture processing module configured to determine a detection object in the cell picture, the detection object comprising an edge of a cell separator, wherein the picture processing module is further configured to:

determine a reference position in the first cell picture based on a preset cell model, wherein determining the reference position comprises matching the preset cell model from the first cell picture to obtain center position coordinates of the cell separator and establishing a spatial coordinate system based on the center position coordinates,

setting edge finding regions free from interference of tab regions in a space following mode, wherein for an edge where the tabs are arranged, a plurality of different edge finding regions are set to skip over tab regions, and the edge of the cell separator is obtained by fitting based on points found in real time in the plurality of different edge finding regions, and

find the edge of the cell separator based on the reference position; and

a picture detection module configured to perform appearance detection based on the detection object to obtain detection result information, the detection result information comprising pin pulling detection result,

wherein the picture detection module is further configured to:

re-establishing a spatial coordinate system by using an intersection point of edges of the cell separator as an origin of the coordinate system,

determine a pin pulling detection region based on the edge of the cell separator and positions of the tabs; and

perform detection on the pin pulling detection region to obtain the pin pulling detection result, wherein the picture detection module is further configured to:

obtain distance values between edge points of the cell separator in the pin pulling detection region and a corresponding edge of the cell separator to obtain a plurality of distance values; and

obtain the pin pulling detection result by analysis based on the plurality of distance values.

12 . The appearance detection apparatus according to claim 11 , wherein the detection result information comprises size data; and the picture processing module is further configured to:

calculate distances between all points on an edge of the cell separator and an opposite edge to obtain a plurality of pieces of edge distance data; and

obtain the size data based on the plurality of pieces of edge distance data.

13 . The appearance detection apparatus according to claim 12 , wherein the edge distance data comprises length data and width data; and the picture processing module is further configured to:

perform averaging on a plurality of pieces of length data to obtain an average length value;

perform averaging on a plurality of pieces of width data to obtain an average width value; and

obtain the size data based on the average length value and the average width value.

14 . The appearance detection apparatus according to claim 11 , wherein the picture processing module is further configured to:

look for and obtaining spot information within the pin pulling detection region, and obtaining the pin pulling detection result by analysis based on the spot information.

15 . The appearance detection apparatus according to claim 11 , wherein the picture processing module is further configured to:

sort the plurality of distance values to obtain sorted distance values;

select a maximum value of the sorted distance values; and

obtain the pin pulling detection result by analysis based on the selected maximum value.

16 . A computer device, comprising:

a memory; and

one or more processors,

wherein the memory stores computer-readable instructions that, when executed by the one or more processors, cause the one or more processors to:

obtain a cell picture of a bare cell by shooting, the cell picture being obtained by shooting in alignment with positions of tabs on an edge of the bare cell, so that a tab region and an edge of a cell separator adjacent to the tab region are included in the cell picture, wherein the cell picture comprises a first cell picture, an exposure of the first cell picture is higher than a preset exposure threshold,

determine a detection object in the cell picture, the detection object comprising an edge of a cell separator,

determine a reference position in the first cell picture based on a preset cell model,

match the preset cell model from the first cell picture to obtain center position coordinates of the cell separator and establish a spatial coordinate system based on the center position coordinates,

set edge finding regions free from interference of tab regions in a space following mode, wherein for an edge where the tabs are arranged, a plurality of different edge finding regions are set to skip over tab regions, and the edge of the cell separator is obtained by fitting based on points found in real time in the plurality of different edge finding regions,

find the edge of the cell separator based on the reference position, perform appearance detection based on the detection object to obtain detection result information,

re-establish a spatial coordinate system by using an intersection point of edges of the cell separator as an origin of the coordinate system,

determine a pin pulling detection region based on the edge of the cell separator,

perform detection on the pin pulling detection region to obtain pin pulling detection result, wherein the pin pulling detection result is included in the detection result information,

obtain distance values between edge points of the cell separator in the pin pulling detection region and a corresponding edge of the cell separator to obtain a plurality of distance values, and

obtain the pin pulling detection result by analysis based on the plurality of distance values.

17 . The computer device according to claim 16 , wherein the computer-readable instructions when executed by the one or more processors further cause the one or more processors to:

calculate distances between all points on an edge of the cell separator and an opposite edge to obtain a plurality of pieces of edge distance data; and

obtain size data based on the plurality of pieces of edge distance data,

wherein the size data is included in the detection result information.

18 . The computer device according to claim 17 , wherein the computer-readable instructions when executed by the one or more processors further cause the one or more processors to:

perform averaging on a plurality of pieces of length data of the edge distance data to obtain an average length value;

perform averaging on a plurality of pieces of width data the edge distance data to obtain an average width value; and

obtain the size data based on the average length value and the average width value.

19 . The computer device according to claim 16 , wherein the computer-readable instructions when executed by the one or more processors further cause the one or more processors to:

look for and obtaining spot information within the pin pulling detection region, and obtaining the pin pulling detection result by analysis based on the spot information.

20 . The computer device according to claim 16 , wherein the computer-readable instructions when executed by the one or more processors further cause the one or more processors to:

sort the plurality of distance values to obtain sorted distance values;

select a maximum value of the sorted distance values; and

obtain the pin pulling detection result by analysis based on the selected maximum value.

Assignments (2)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Aug 6, 2024
From: CONTEMPORARY AMPEREX TECHNOLOGY CO., LIMITED
To: CONTEMPORARY AMPEREX TECHNOLOGY (HONG KONG) LIMITED
Reel/Frame 068338/0402 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 6, 2023
From: TU, YINHANG; LIU, XIAOFENG; ZHU, JINPING
To: CONTEMPORARY AMPEREX TECHNOLOGY CO., LIMITED
Reel/Frame 065476/0052 →
Priority Claims (1)
CN 202210957577.6 · Aug 10, 2022 · national
Continuity (2)
Continuation PCTCN2023084032 · Mar 27, 2023
Related Publication 20240070852A1 · Feb 29, 2024
References Cited (33)
US 20140362248A1 · Ishida · 2014 [cited by examiner]
US 20210209739A1 · Wen et al. · 2021 [cited by applicant]
CN 107941805A · 2018 [cited by applicant]
CN 109269419A · 2019 [cited by applicant]
CN 109358064A · 2019 [cited by applicant]
CN 109443418A · 2019 [cited by examiner]
CN 111192246A · 2020 [cited by applicant]
CN 111403800A · 2020 [cited by examiner]
CN 111584923A · 2020 [cited by examiner]
CN 112557390A · 2021 [cited by applicant]
CN 112577421A · 2021 [cited by applicant]
CN 113097570A · 2021 [cited by examiner]
CN 113409296A · 2021 [cited by examiner]
CN 214957024U · 2021 [cited by applicant]
CN 113989232A · 2022 [cited by applicant]
CN 114740001A · 2022 [cited by examiner]
CN 114813745A · 2022 [cited by applicant]
CN 115829913A · 2023 [cited by applicant]
DE 202022101348U1 · 2022 [cited by applicant]
JP H08101133A · 1996 [cited by examiner]
WO WO2021169335A1 · 2021 [cited by examiner]
The extended European search report received in the counterpart EP application 23797624.6, dated May 7, 2025, 7 pages. [cited by applicant]
International Search Report received in the corresponding International Application PCT/CN2023/084032, mailed Jun. 8, 2023. [cited by applicant]
Written Opinion received in the corresponding International Application PCT/CN2023/084032, mailed Jun. 8, 2023. [cited by applicant]
Linghu Rong, “Defect Detection of Rubber Damping Pad Based on VisionPro”, published on Shangxi Electronic Technology vol. 1 on Dec. 31, 2016, p. 3-4. [cited by applicant]
Lushen Wu et al., “Design of Solar Cell Panel Calibration System Based on Machine Vision”, published on Modular Machine Tool Automatic Manufacturing Technique vol. 5 on May 31, 2018, p. 23-29. [cited by applicant]
Fanrong Kong et al., “Industrial Intelligent Transformation Methods and Practices”, published by Shanghai Science and Technology Press on Jul. 31, 2020, p. 193-197. [cited by applicant]
Yong Huang, “Power Battery and Energy Management Technology”, published by Chongqing University Press on Sep. 30, 2021, p. 42-46. [cited by applicant]
Xuehong Sun, “Machine Vision Technology and Applications”, published by Mechanical Engineering Industry Press on Apr. 30, 2021, p. 64-71. [cited by applicant]
First Office Action received in the corresponding Chinese Application 20221095577.6, mailed Jun. 6, 2023. [cited by applicant]
Second Office Action received in the corresponding Chinese Application 20221095577.6, mailed Aug. 22, 2023. [cited by applicant]
Notification to Grant Patent Right for Invention received in the corresponding Chinese Application 202210957577.6, mailed Mar. 11, 2024. [cited by applicant]
Office Action, mailed Nov. 27, 2025, for corresponding European Patent Application Serial No. 23797624.6. [cited by applicant]