IP Library Granted Patent US 12663385
Granted Patent B2
US 12663385 · App. 18/448,909 · Granted Jun 23, 2026

Inspection method and inspection apparatus for wound cell

Inventors: Fenglin Zhang (Ningde, CN); Jianlin Liu (Ningde, CN); Shaoteng Ren (Ningde, CN); Yingqin Su (Ningde, CN)
Assignee: CONTEMPORARY AMPEREX TECHNOLOGY (HONG KONG) LIMITED
G01N23/04H01M10/4285G01N2223/1016
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 12663385
App. No.
18/448,909
Granted
Jun 23, 2026
Kind
B2
Abstract

An inspection method includes obtaining an image of a corner region of the wound cell that includes an image of a cathode electrode plate at N layers of the corner region of the wound cell and an anode electrode plate at the N layers, and determining, based on the image of the corner region, amounts of misalignment between the cathode electrode plate and the anode electrode plate that are adjacent in the wound cell. N is a positive integer.

Claims (37)

1 . An inspection method for a wound cell, comprising:

obtaining an image of a corner region of the wound cell, wherein in the image of the corner region, a cathode electrode plate and an anode electrode plate are visible at each of N layers of the corner region of the wound cell, N being a positive integer;

determining, from the image of the corner region of the wound cell, first endpoints on a first plane for the cathode electrode plate at each of the N layers and second endpoints on the first plane for the anode electrode plate at each of the N layers, the first plane being a plane on which a winding axis of the wound cell is located, the first plane being perpendicular to a direction of X-rays used to obtain the image, the first endpoints and the second endpoints being output as coordinate values by a neural network model obtained by training using a plurality of marked wound cell images that include information about marked endpoints on the first plane for the cathode electrode plate and for the anode electrode plate; and

determining, based on the first endpoints and the second endpoints, amounts of misalignment between the cathode electrode plate and the anode electrode plate that are adjacent in the wound cell, the determining the amounts of misalignment comprises determining, for each of the N layers and based on the coordinate values, a distance between the first endpoint and a corresponding second endpoint.

2 . The inspection method according to claim 1 , wherein N is a total number of layers of the cathode electrode plate and anode electrode plate in the corner region of the wound cell.

3 . The inspection method according to claim 1 , wherein the image of the corner region is obtained by using a micro-focus X-ray source to emit X-ray that penetrate the corner region for imaging, wherein a direction of the X-ray is perpendicular to a direction of the winding axis of the wound cell.

4 . The inspection method according to claim 1 , further comprising, before determining, based on the image of corner region, the amounts of misalignment between the cathode electrode plate and the anode electrode plate that are adjacent in the wound cell:

performing image enhancement on the image of the corner region.

5 . The inspection method according to claim 4 , wherein performing the image enhancement on the image of the corner region comprises:

increasing contrast of the image of the corner region to highlight pixels on the first plane for the cathode electrode plate at each of the N layers and pixels on the first plane for the anode electrode plate at each of the N layers.

6 . The inspection method according to claim 4 , wherein performing the image enhancement on the image of the corner region further comprises:

using a first grayscale value to replace a grayscale value for a first region and using a second grayscale value to replace grayscale values for other regions in the image of the corner region except the first region, wherein the first region comprises a region in which pixels on the first plane for the anode electrode plate at each of the N layers and pixels on the first plane for the cathode electrode plate at each of the N layers are located.

7 . The inspection method according to claim 1 , further comprising, before obtaining the image of the corner region of the wound cell:

obtaining a plurality of initial image frames of the corner region continuously acquired; and

performing average denoising on the plurality of initial image frames of the corner region to obtain the image of the corner region.

8 . The inspection method according to claim 1 , wherein:

the image of the corner region is obtained by using a micro-focus X-ray source to emit an X-ray that penetrates the corner region for imaging;

a direction of the X-ray is perpendicular to a direction of the winding axis of the wound cell; and

a control power of the micro-focus X-ray source is positively related to a thickness of the corner region.

9 . The inspection method according to claim 1 , wherein the corner region comprises four corner regions of the wound cell.

10 . The inspection method according to claim 1 , further comprising:

determining a minimum amount of misalignment between the cathode electrode plate and the anode electrode plate that are adjacent in the wound cell;

determining that the wound cell passes an acceptance criterion in response to the minimum amount of misalignment being within a preset range; and

determining that the wound cell fails the acceptance criterion in response to the minimum amount of misalignment being beyond the preset range.

11 . An inspection apparatus for a wound cell, comprising:

an X-ray source and a detector, the X-ray source and the detector being configured to obtain an image of a corner region of the wound cell, wherein in the image of the corner region, a cathode electrode plate and an anode electrode plate are visible at each of N layers of the corner region of the wound cell, N being a positive integer; and

a processor and a memory storing instructions that, when executed by the processor, cause the processor to:

determine, from the image of the corner region of the wound cell, first endpoints on a first plane for the cathode electrode plate at each of the N layers and second endpoints on the first plane for the anode electrode plate at each of the N layers, the first plane being a plane on which a winding axis of the wound cell is located, the first plane being perpendicular to a direction of X-rays used to obtain the image, the first endpoints and the second endpoints being output as coordinate values by a neural network model obtained by training using a plurality of marked wound cell images that include information about marked endpoints on the first plane for the cathode electrode plate and for the anode electrode plate; and

determine, based on the first endpoints and the second endpoints, amounts of misalignment between the cathode electrode plate and the anode electrode plate that are adjacent in the wound cell, the determining the amounts of misalignment comprises determining, for each of the N layers and based on the coordinate values, a distance between the first endpoint and a corresponding second endpoint.

12 . The inspection apparatus according to claim 11 , wherein N is a total number of layers of the cathode electrode plate and anode electrode plate in the corner region of the wound cell.

13 . The inspection apparatus according to claim 11 , wherein the image of the corner region is obtained by using a micro-focus X-ray source to emit an X-ray that penetrates the corner region for imaging, wherein a direction of the X-ray is perpendicular to a direction of the winding axis of the wound cell.

14 . The inspection apparatus according to claim 11 , wherein the processor is further configured to:

perform image enhancement on the image of the corner region.

15 . The inspection apparatus according to claim 14 , wherein the processor is further configured to:

increase contrast of the image of the corner region to highlight pixels on the first plane for the cathode electrode plate at each of the N layers and pixels on the first plane for the anode electrode plate at each of the N layers.

16 . The inspection apparatus according to claim 14 , wherein the processor is further configured to:

use a first grayscale value to replace a grayscale value for a first region and use a second grayscale value to replace grayscale values for other regions in the image of the corner region except the first region, wherein the first region comprises a region in which pixels on the first plane for the anode electrode plate at each of the N layers and pixels on the first plane for the cathode electrode plate at each of the N layers are located.