Torque marker inspection for threaded connections

Industry

Manufacturing

Description

Checks if a torque seal is broken or straight. Detects a threaded connection (nut or bolt) head with a torque seal. If the nut slips or starts to loosen, the painted line will be broken and classified as DEFECT. If the line is straight - the model returns OK.

Architecture / Pipeline
Pipeline:

Pipeline: Input -> Detector -> Tracker -> Selector -> Pre-processing -> Classifier -> Output


Input: still images from GigE camera


Detector: YOLOX Tiny@288x480, patches of unrestricted size containing objects.


Tracker: SORT


Selector: allows to filter out partially visible objects on the frame borders. 


Pre-processing: object box is extended to given % in height/width from the center and cropped according to frame borders


Classifier: ResNet34, predicting probability of two classes, resolution 64x64, RGB, aspect ratio preserved


Output: softmax(preds)[:, 1] (Anomaly score)

Classes

Objects: "NUT", Classification: "OK", "DEFECT"

Thresholds

The following thresholds were used for test: XXX, XXX. Thresholds can be defined in a config file.

Code optimization

ONNX, TensorRT

Use case keywords

assembly, bolt, nut, marker, threaded connection, torque seal

Current State

Ready to deploy

Version

1.20, build 15.12.2022

Intended uses

Inspection of torque seals (markers) on threaded connections

How to use

1. Get a fresh GPU
2. Fill it with well-pruned models
3. Stir

Input

Still images



Output

Object name ("Nut"), Object class ("OK", "Defect"), bounding box location and size

Limitations and bias

The module was tested with input image sizes from 640x480 to 1920x1080 on Siemens IPC 520A industrial PC.


Training data

Real-life images: 212; synthetic images: 3,110 (generated in NVIDIA Omniverse with domain, light, optics, textures, size, shape randomization)

Training procedure

6 epochs, trained on DGX2 server (NVIDIA V100 GPU)

Evaluation results

Accuracy, performance, ROC/AUC
Precision: 96.2%, Recall: 97.4% with confidence threshold 0.43

Target hardware

GPU: NVIDIA Jetson Xavier NX, CPU: ARM64; RTX4000/amd64 version is under development

Benchmark

Frame rate: 20 FPS on Jetson Xavier NX

Size

3.4 GB

Profile Image

<path to hi-res image at google drive>

Dataset samples

<path to hi-res images at google drive>



Start working with Data Monsters

Submit your application, and our team will get in touch with you soon.

Partnerships
There are many ways to partner with Data Monsters. Find the right fit for you.
Partner with us