Machine vision inspection has been widely applied in the key quality control links of motor rotor production. It can efficiently detect rotor defects, such as welds on the outer ring, bending and missed installation of insulating sheets. These defects can directly cause serious faults or failures of the motor. The visual system can accurately detect anomalies through reliable, high-contrast images and algorithms, ensuring the consistency and reliability of the products.

1.Inspection background
The rotor of a motor is the core component of the motor, and its assembly accuracy directly determines the operational stability and service life of the motor. The traditional manual detection method has problems such as low efficiency, high missed detection rate and inconsistent judgment standards. By adopting machine vision inspection technology, it is possible to achieve automated and high-precision screening of rotor defects, meeting the quality control requirements of large-scale production.

2.Object
The test for the motor rotor, two core test content includes: 1. The rotor outer ring weld any decision; 2. The rotor insulation slices bend, packing defect identification.

3.Solution
3.1 welding line inspection
Test content: Check if there are any welds on the outer ring of the rotor, or if the insulating sheet is bent or missing.
300x200Light architecture diagram
300x200Effect picture   
Algorithmic Logic
A: Filter and reduce noise on the collected images to enhance the edge features of the weld seam. 
B: set the weld width, the length of the threshold range, beyond or without corresponding characteristics are judged to be abnormal. 
C: automatic marking anomaly area, synchronous signal output test result.

3.2 Inspection of missed installation and bending of insulating sheets
300x200Light architecture diagram
300x200Effect picture   
Algorithmic Logic
A: Extract the contour features of the insulating sheet and compare them with the standard template. 
B: to determine if the outline is missing packing, if contour offset than threshold judgement for bending. 
C: support defect types of classified statistics to facilitate the production end back problem.

4.Advantages
4.1 High-precision detection: The detection accuracy rate of weld seam and insulating sheet defects can reach over 99.8%, far exceeding the level of manual detection. 
4.2 High efficiency operation: single rotor testing takes 2 seconds or less, can match the production line of high-speed production rhythm.
4.3 High stability security: to test the result is not affected by personnel status, support 7 * 24 hours of continuous work, reduce the quality cost.