betway必威(中国)官方网站

Solution Based on AI-CV to Inspecting Defects of Breathable Membranes

Solution brief

The waterproof and breathable membrane inspection/mounting plan currently covers structural components and functional components of mobile phone, such as die cutting, stamping, CNC, softgood, etc.

It may effectively solve three major difficulties in the field:

1. The difference in the occurrence probability of different types of defects is large, and imbalance of defect samples.

2. Defects are small target defects.

3. Defect types have different semantic levels.

Currently, the system covers more than 40 kinds of defect types, such as: Edge whitening, external glue overflow, glue imprinting, cover displacement, bad fool-proofing, outer frame burr and filament, cover clamping waste, fool-proofing hole blocking, foreign object in breathable membrane, foreign object on glue, breathable membrane imprinting, cover impenetrable, breathable membrane scratch, breathable membrane damage, missing drainage hole blocking, breathable membrane black/white lines, burr and filament in hole, deviation, imprinting, glue overflow in hole, damage, bubbles, clamping waste in hole, dislocation, cover scratch, membrane corrosion, lack of product, etc.

Solution function

Cell phone waterproof and breathable membrane detection and installation solution based on Dihuge TimesAI deep learning development platform. We can provide one-stop services for online real-time detection of defects in precision parts, defect elimination, and data analysis and statistics. The system's performance indicators are excellent and can be continuously updated iteratively. It can replace and outperform manual work, improve yield, and reduce personnel to increase efficiency.

Bright spot

This system achieves online real-time full detection of die-cutting products in precision parts of cell phones, effectively solves the three major difficulties in defect detection in the field (Unbalanced defect samples, small defects as targets, and different semantic levels of defect types), and provides one-stop services for defect elimination and data analysis statistics; The system performance indicators are excellent, which can replace and outperform manual labor, improve yield, and reduce personnel to increase efficiency.

Application scenario

Online real-time full inspection of die-cutting products in mobile phones, pads, and other electronic devices

Product case

The inspection system for cell phone precision parts (die-cut waterproof and breathable membrane) of the top three companies in the global industry is developed based on Dihuge TimesAI platform, adapting to various line structures and sorting logic of users. It detects nearly forty types of defects, such as: White edge, external glue overflow, glue imprinting, cover displacement, bad anti-dulling, outer frame hairy wire, cover clamping waste, anti-dulling blocked hole, breathable membrane foreign object, foreign object on glue, breathable membrane imprinting, cover incomplete cutting, scratches on the breathable membrane, breathable membrane breakage, leaky waste blocked hole, black/white lines on the breathable membrane, hairy wire in hole, repeat deviation, mark, glue overflow in hole, breakage, air bubble, clamping waste in the hole, offset position, cover scratches, membrane corrosion, lack of material, etc. Defect detection performance index: Escape＜0.5%, overkill＜3%.

Scenario
Scenario
Scenario