New energy lithium battery surface defect detection method based on adaptive deep learning

Abstract

The invention discloses a new energy lithium battery surface defect detection method based on adaptive deep learning. Including: nonlinear mapping of the lithium battery surface grayscale image; transforming the decoupled illumination component and reflection component into the frequency domain; filtering, inverse Fourier transform, and exponential transformation of the frequency domain data to obtain the reconstructed lithium battery image ; Based on morphological processing and background difference, enhance the grayscale response at the defect; perform image segmentation, connected domain analysis and screening processing, and use the results as annotated images; design operators to simulate illumination details and perform sample enhancement on lithium battery surface grayscale images Operation; train a deep convolutional neural network based on the enhanced sample image set and annotated images; implement lithium battery surface defect detection based on the trained network. The present invention can improve the detection efficiency and reduce the false detection rate in the lithium battery surface defect detection scenario.

Description

technical field [0001] The invention relates to the technical fields of computer vision and deep learning, in particular to a method for detecting surface defects of new energy lithium batteries based on adaptive deep learning. Background technique [0002] Lithium batteries are widely used in mobile devices, new energy vehicles, home appliances and other fields, bringing great convenience to our lives. Lithium batteries have the characteristics of high energy density, long service life and low self-discharge rate. They are generally batteries that use lithium alloys as negative electrode materials and non-aqueous electrolyte solutions. For flexible packaging batteries, polymer shells are usually put on the business solution, and the structure is packaged with aluminum-plastic film. It is precisely because of the characteristics of flexible packaging that often due to poor sealing, unevenness, or Bumps in the battery will lead to convex hull defects on the surface of the ba...

AI Technical Summary

Problems solved by technology

Due to the complexity of the light control system and the technical limitations of the frame rate of the camera, the nominal detection time of this method is about a quarter of a second, but the system obviously cannot call all light sources, and cannot directly obtain the result through image acquisition of a single light source

The parameters of this kind of system are too complicated, and the work of parameter adjustment and calibration during production and maintenance is time-consuming and laborious, and the theoretical performance cannot be optimal, resulting in a high false alarm rate although the missed detection rate is low

[0007] Therefore, the existing lithium battery surface defect detection technology has the problems of low detection efficiency and high false detection rate.

Images