Research

The Industrial Big Data Intelligent Analysis Project developed the world's first intelligent robot for detecting surface defects in nuclear power plant reactor containment structures, effectively ensuring the safety of critical facilities in my country. The crack width recognition error is 0.05mm, the defect location error is less than 10cm, and the defect detection rate is greater than 90%.

Key technologies include:

1. A SLAM algorithm that fuses point and line features, extracting corner points from continuous frames, using KLT optical flow to track these corner points, and employing an LSD line segment detector to extract line features.

2. A multi-branch detection model based on a rotating frame crack detection perspective, using a geometric distance-aware loss function to extract differential depth features.

This technology as a whole has reached an internationally advanced level, and the algorithm for identifying micro-cracks and defects has reached an internationally leading level.