Scribd
Upload
28 views1 page

ML Algorithm

Uploaded by

apurba.sarker.as
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
28 views1 page

ML Algorithm

Uploaded by

apurba.sarker.as
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 1

ML Algorithm Application Data Type

Melt pool images


Support Vector Classification of melt pool images to locate lack of
(thermal or visible
Machine (SVM) fusion (LoF) pores
spectrum)
Melt pool images
Distinguishing pores from normal printing states
Decision Tree (DT) (thermal or visible
using melt pool images
spectrum)
Melt pool images
K-Nearest Identifying pores in melt pool images by comparing
(thermal or visible
Neighbor (KNN) to known examples
spectrum)
Quadratic Melt pool images
Differentiating between defect and non-defect states
Discriminant (thermal or visible
in melt pool images
Analysis (QDA) spectrum)
High-speed camera
Deep Neural Predicting laser power levels from high-speed camera
images of the melt
Network (DNN) images of the melt pool
pool
Deep High-speed digital
Correlating spatters and temperature from melt pool
Convolutional images and thermal
images to process parameters like scanning speed and
Neural Network images of the melt
laser power
(DCNN) pool
- Classifying porosity occurrences during the DED
High-speed digital
process<br> - In-situ defect identification using
Convolutional images, coaxial
coaxial melt pool images<br> - Predicting humping,
Neural Network camera images, and
spatter, or normal fabrication states<br> - Extracting
(CNN) thermal images of
quality indicators from raw images for dilution
the melt pool
quantification and defect prediction
Self-Organizing Predicting porosity locations in melt pool thermal Thermal images of
Map (SOM) images using unsupervised learning the melt pool
Scale Invariant High-speed digital
Extracting melt pool signatures for unsupervised
Feature Transform images of the melt
learning to identify defects
(SIFT) pool
Histogram of High-speed digital
Extracting features from melt pool images to identify
Oriented Gradients images of the melt
defects using unsupervised learning
(HOG) pool
Functional High-speed digital
Principal Selecting essential features from melt pool images for images and thermal
Component training traditional ML models images of the melt
Analysis (FPCA) pool
Ensemble Deep
Real-time defect detection with low latency for Time-series melt
Learning
identifying various welding defects on-the-fly pool infrared images
Architecture
Generative
Enhancing image contrast ratio of thermal images for Thermal images of
Adversarial
better melt pool segmentation the melt pool
Network (GAN)
Gravitational
Reducing dimensionality of infrared images and Infrared images of
Super-Pixels
performing real-time melt pool segmentation the melt pool
Algorithm

You might also like