SlideShare a Scribd company logo
1 of 45
1
AOIEA
深度學習於表面瑕疵檢測:
Strategies for imbalanced data training
蔡篤銘 教授
元智大學
工業工程與管理學系
2018 Taiwan AOI Forum and show
物體表面特徵(surface texture)
2
High-resolution
Resolution
Uniform Low contrast Patterned Homogeneous
texture
Repeated
patterns
Heterogeneous
texture
•Web material
•Paper
•Steel surface
•Mura
•Backlight panel
•Assembled PCB
•IC die
•TFT-LCD
•IC wafer
•TFT-LCD
•Color filter
•Solar cell/wafer
Uniform Low contrast Patterned Homogeneous
texture
Repeated
patterns
Heterogeneous
texture
Inspection in manufacturing
Product & process properties in manufacturing:
• Small , subtle local defect in
− Size
− Shape
− Deformation
− Gray scale/color unevenness
• Imbalanced data
– Many positive (defect-free) samples
– Only a few or No negative (defect) samples
Requirements:
• High precision/accuracy
– Location
– Detection rate
• Computationally fast
(Real-time computation to meet the cycle time)
3
Implement the deep learning models for
• CNN regression for image positioning
• GAN-based techniques to generate synthesized defects for
the data-imbalance problem
• Autoencoding models
– Convolutional Autoencoders (CAE) with one-class SVM classification
for unsupervised anomaly detection
– Variational Autoencoders (VAE) image reconstruction for unsupervised
defect detection
4
5
Deep learning for PCB positioning
6PCB assembly PCB inspection
Deep learning for PCB positioning
A precise positioning system can be used for
 Automated assembly
 Automated visual defect detection
Bare PCB Assembled PCB
 Problem
Predict Angle (θ)、Horizontal displacement (x) and Vertical displacement (y)
of a PCB w.r.t. the template.
7
(θ, x, y)=(-20°, -20, -20) (θ, x, y)=(4°, 0, 12) (θ, x, y)=(20°, 20, 20)
(θ, x, y)=(0°, 0, 0)PCB board
Deep learning for PCB positioning
Template
8
Traditional machine learning for regression
Deep learning for regression
Deep learning regression v.s. traditional ML
regression
9
CNN for regression:
CNN as feature extractor
for SVR regression:
DNN for regression:
CNN + SVR for regression:
Using Convolutional Neural Network (CNN)
and Support Vector Regression (SVR) models
to predict θ, x and y.
10
Linear SVR
Nonlinear SVR (with Kernel transformation)
Polynomial:
Radial Basis Function(RBF):
Support Vector Regression (SVR)
x = input feature(s)
y = output response(s)
11
• Input PCB image into DNN model to predict (θ, x, y)
• Number of hidden layers: 3
Deep Neural Network (DNN) model
12
• Input PCB image into CNN model to predict (θ, x, y)
Convolutional Neural Network (CNN) model
13
Note:
1. Kernel function “Radial Basis Function” is used
2. CNN model is the same as the previous one,
the number of features to SVR is 3.
3. Training time: 2.2 hours (for 9,261 samples)
CNN+SVR model for regression
14
CNN as feature extractor for regression
Note:
1. Kernel function “Radial Basis Function” is used
2. CNN model is the same as the previous one; the last layer of CNN
contains 128 nodes (i.e., number of features to SVR is 128)
3. Training time: 2.2 hours (for 9,261 samples)
15
The good thing about Deep learning for Regression:
The user needs only provide ONE SINGLE template image
16
It creates the input image (x) and the corresponding output (θ, x, y) for DL training
Reference PCB image
provided by users:
Augmented image:
Corresponding output: (θ, x, y)=(-20°, -20, -20) (θ, x, y)=(0°, 0, 0) (θ, x, y)=(20°, 20, 20)
180 pixel
180 pixel
The training samples are automatically generated
by Image Augmentation from the template
17
Range of
Training samples
• Image size: 180 × 180 (pixels)
• Angles: 0°、±2°、±4°、 …、± 20° (every two angles for θ)
• Horizontal displacement: 0、±2、±4、…、 ± 20 (pixel) (every two pixels for x)
• Vertical displacement : 0、±2、±4、…、 ± 20 (pixel) (every two pixels for y)
• Number of training samples: 9,261 (21 × 21 × 21)
Testing samples
• Image size :180 × 180 (pixels)
• Angles : 0°、±1°、±2°、…、± 20°
• Horizontal displacement : 0、±1、±2、…、±20 (pixel)
• Vertical displacement : 0、±1、±2、…、±20 (pixel)
• Number of testing samples :68,921 (41 × 41 × 41)
Note: The images with odd θ, x and y are unseen to the training model
18
Evaluate positioning accuracy by mean error, variance and maximum error.
Angle error (degree) Horizontal error (pixel) Vertical error (pixel)
Mean Variance Max Mean Variance Max Mean Variance Max
DNN 0.111 0.008 0.689 0.119 0.009 1.009 0.120 0.008 0.876
CNN 0.124 0.008 0.634 0.162 0.008 0.657 0.133 0.007 0.563
CNN+SVR 0.049 0.001 0.464 0.049 0.001 0.344 0.055 0.002 0.506
CNN as
feature
extractor
0.068 0.003 0.499 0.066 0.003 0.484 0.069 0.003 0.515
Positioning accuracy
19
Model Time (seconds)
DNN 0.00123
CNN 0.00166
CNN+SVR 0.00230
CNN as feature extractor 0.00541
Note: Equipment
1. CPU: Intel® Core™ i7-6700K CPU @ 4.00GHz × 8
2. GPU: GeForce GTX 1080 Ti
Computation time of each model
It achieves 2-milliseconds efficiency.
20
Defect inspection by image subtraction
Template fT ,
(a) Normal (b) Scratch (c) Extrusion
Test image
,
Image
subtraction
from ,
Result
∆ ,
(d) Intrusion
∆ , fT , ,
Template Aligned
Saw-mark defect detection
in heterogeneous solar wafer images using
- GAN-based training samples generation
- CNN classification
21
Multicrystalline solar wafer inspection
• Multicrystalline silicon wafers
 A multicrystalline solar wafer presents random shapes, sizes and directions of
crystal grains in the surface and results in a heterogeneous texture.
 A saw-mark defect is a severe flaw of wafers when cutting the silicon ingot into
thin wafers using the multi-wire saws.
Defect-free solar wafer image
White saw-mark defect Saw-mark defect caused
by impurity
Solar wafer image
with a black saw-mark defect
22
The proposed deep learning scheme is
composed of two phases:
Defect samples generation using the CycleGAN (Cycle-consistent Generative Adversarial Networks), and then
Defect detection using the CNN (Convolutional Neural Networks) based on
the true defect-free samples and the synthesized defective samples.
• The CycleGAN model combines both the adversarial loss (i.e. GAN) and the cycle consistency loss .
• GAN measures the adversarial lose between the generated images and the target image.
• The consistency lose prevents the learned forward and backward mappings from contradiction.
• It uses unpaired datasets (not specific paired samples in GAN) for the training, and is suited for our
application.
CycleGAN model used for defect patches generation
23
Real solar wafer surfaces
For training the CycleGAN:
• Use a small set of true defect patches (60 for black, 90 for white defects) as the target dataset,
and then randomly collect a small set of defect-free patches (60 & 90) as the input dataset to
the CycleGANs.
Real defect-free samples
Real black saw-mark samples
Real white saw-mark samples
24
Using the CycleGAN model to generate
the defective samples
Synthesized defects:
• Whenever we change the input set with different defect-free patches to the trained
CycleGAN, a new defective set is created.
Real defect-free samples input to the trained CycleGAN
Generated black saw-mark patches
Generated white saw-mark patches
25
The CNN model for classification
• A simple CNN with 3 convolutional layers is used for the training.
• A lean CNN model gives better computational efficiency in the inspection process.
• Training information:
– For the CycleGAN models, 150 (60 & 90 Black and White sawmarks) real defective
patches and 150 (60&90) real defect-free patches are used as the training samples.
– For the CNN model, a total of 4000 real defect-free patches and 4000 synthesized
sawmark patches are used as the training samples.
– Patch size 50 50 pixels
– Training time : 3 hours for CycleGAN , and 1 hour for CNN
CNN model used for defect detection
26
Postprocessing with conventional machine
vision techniques
• The saw-mark in a small windowed patch contains only subtle changes and, thus,
the entire saw-mark region may not be completely detected in the full-sized solar
wafer image.
• Apply the horizontal projection line by line in the resulting binary image B , to
intensify the horizontal saw-mark in the image . That is
P ∑ , , ∀
• The maximum projection value is used as the discriminant measure for saw-mark
detection, i.e. P ∗
P , ∀ . If the horizontal projection P ∗
is large
enough, a saw-mark at line ∗
is declared.
Note: The mean computation time is 0.004 seconds for an image patch of size 50×50
pixels on a PC with an Intel Core 2, 3.6GHz CPU and an NVIDIA GTX 1070 GPU .
27
Detection results on sawmarks
• Detection results of defect-free solar wafer images
Test images Detection result projection 28
Detection results on sawmarks
)( yP
y
)( yP
y
)( yP
y
• Detection results of defective solar wafer images
Test images Detection result projection 29
Detection results on stains / foreign particles
30
Real stains defect samples
Synthesized stain defect samples
Detection results on stains / foreign particles
• Detection results of defective solar wafer images
(a) Test images
(b) Detection result
31
Additional test: Using the CycleGAN model
to generate bump defects
• Bumps defect
32
Real defect samples
Real defect-free samples
Generated defect samples
Autoencoders for defect detection
-Autoencoders for image reconstruction
- Autoencoders for feature extraction
33
Unsupervised Autoencoders for
image reconstruction
, ,
Trained model
Autoencoding Model
Self-reference comparison:
∆ , , ,
	∆ , 	 ∆ 	, , 	 	 	
Encoder Decoder
Image Image
Testing image Reconstructed image
34
Defect detection in TFT-LCD
• Thin Film Transistor-Liquid Crystal Display (TFT-LCD) comprises vertical
data lines and horizontal gate lines .
• The main types of defects are pinholes, particles and scratches defect.
defect-free
LCD image
LCD image with
Particle defects
LCD image with
Pinhole defects
LCD image with
Scratch defects 35
VAE (Variational AutoEncoder) model for
image reconstraction
- The Model
• Structure of the VAE model
			 Zp Z p Z 		
Defect-free Defect-free
Latent variables
36
VAE model for image reconstruction
- Detection results by image subtraction
• Defect-free image
• Defect image (true defects)
Original image Restored image Image subtraction Binarization
, , ∆ ,
Original image Restored image Image subtraction Binarization
, , ∆ ,
37
Use AE for feature extraction for
anomaly detection (with one-class SVM)
• Testing image is reconstructed from the trained AE model.
• The features are extracted from the last layer of the Encoder , and used as
the input data of the one-class SVM to identify the anomalies.
ZEncoder DecoderEncoder Decoder
Test
image
Anomaly detection
Feature maps
One-class
SVM
Trained AE model
38
Use AE for feature extraction for
anomaly detection (one-class SVM)
• One-class SVM (Support Vector Machine)
Ni
NiR
CR
i
ii
N
i
i
,,2,1,0
,,2,1,
s.t.
Min
22
1
2




 



ax
Ra
Positive samples
Support vector
Outlier
Center point (a)
Radius (R)
	 a 	 	, 	 	 	
39
Use AE for feature extraction for anomaly
detection (one-class SVM) : The Model
• Structure of the AE model:
ZEncoder DecoderEncoder Decoder
Feature maps
Use AE for feature extraction for anomaly
detection (one-class SVM) : Training samples (LCD)
 Train only positive (defect-free) samples:
− Original 256 256 LCD images are rotated between 	0°
and 35°
with	5°
increment.
− 256 random image patches of size 28×28 are used for training.
− Using the AE model to extract the features for the one-class SVM model.
 Testing data:
− 160 positive samples
− 33 negative (true defect) samples
Computation time : training 15 minutes , testing 0.00008 seconds
41
Use AE for feature extraction for anomaly
detection (one-class SVM) : Detection results (LCD)
• Feature size : 490 (10 feature maps of size 7*7)
• Testing data : 160 positive samples 、 33 negative (true defect) samples
Prediction
Actual Normal Outlier
Normal 89% 11%
Outlier 0% 100%
Type I error 11%
Type II error 0%
Overall recognition rate 90%
42
Use AE for feature extraction for anomaly
detection (one-class SVM) : Detection results (LCD)
• Feature size : 49 (use only the best feature map with size 7*7)
Testing data : 160 positive samples , 33 negative (true defect) samples
Prediction
Actual Normal Outlier
Normal 92% 8%
Outlier 0% 100%
Type I error 8%
Type II error 0%
Overall recognition rate 92%
43
A thought on Deep Learning:
• Can Deep Learning replace Machine Vision?
• MV as preprocessing , and DL as postprocessing
– MV for defect detection, and DL for defect classification
– Or, vice versa
• MV in DL models?
– “Convolution” and “pooling” are parts of image processing operations
– Human knowledge in DL models
(e.g. embed the known defect features to the DL model)
44
45
Thank you
AOIEA

More Related Content

What's hot

演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824
演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824
演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824Beckett Hsieh
 
Diabetic Retinopathy.pptx
Diabetic Retinopathy.pptxDiabetic Retinopathy.pptx
Diabetic Retinopathy.pptxNGOKUL3
 
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...Joonhyung Lee
 
Data Science - Part XVII - Deep Learning & Image Processing
Data Science - Part XVII - Deep Learning & Image ProcessingData Science - Part XVII - Deep Learning & Image Processing
Data Science - Part XVII - Deep Learning & Image ProcessingDerek Kane
 
Prospects of Deep Learning in Medical Imaging
Prospects of Deep Learning in Medical ImagingProspects of Deep Learning in Medical Imaging
Prospects of Deep Learning in Medical ImagingGodswll Egegwu
 
Number plate recognition system using matlab.
Number plate recognition system using matlab.Number plate recognition system using matlab.
Number plate recognition system using matlab.Namra Afzal
 
AOI應用於生醫及生物檢測之實例探討
AOI應用於生醫及生物檢測之實例探討AOI應用於生醫及生物檢測之實例探討
AOI應用於生醫及生物檢測之實例探討CHENHuiMei
 
Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)Vivek reddy
 
ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)
ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)
ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)Masakazu Iwamura
 
CNN and its applications by ketaki
CNN and its applications by ketakiCNN and its applications by ketaki
CNN and its applications by ketakiKetaki Patwari
 
Face recognition technology
Face recognition technologyFace recognition technology
Face recognition technologyranjit banshpal
 
Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)
Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)
Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)YuMaruyama
 
Metal Surface Defect Inspection Through Deep Neural Network
Metal Surface Defect Inspection Through Deep Neural NetworkMetal Surface Defect Inspection Through Deep Neural Network
Metal Surface Defect Inspection Through Deep Neural NetworkMd Fantacher Islam
 
Viola-Jones Object Detection
Viola-Jones Object DetectionViola-Jones Object Detection
Viola-Jones Object DetectionVenugopal Boddu
 
Minor project report on logo detection on vehicles
Minor project report on logo detection on vehiclesMinor project report on logo detection on vehicles
Minor project report on logo detection on vehiclesAvinash Singh Yadav
 
Lecture 9&10 computer vision segmentation-no_task
Lecture 9&10 computer vision segmentation-no_taskLecture 9&10 computer vision segmentation-no_task
Lecture 9&10 computer vision segmentation-no_taskcairo university
 
Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...
Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...
Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...Universitat Politècnica de Catalunya
 
Convolutional neural network
Convolutional neural network Convolutional neural network
Convolutional neural network Yan Xu
 
Deep Learningと画像認識   ~歴史・理論・実践~
Deep Learningと画像認識 ~歴史・理論・実践~Deep Learningと画像認識 ~歴史・理論・実践~
Deep Learningと画像認識   ~歴史・理論・実践~nlab_utokyo
 

What's hot (20)

演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824
演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824
演講-臨床醫學統計方法攻略-三星統計威廉顧問-20130824
 
Diabetic Retinopathy.pptx
Diabetic Retinopathy.pptxDiabetic Retinopathy.pptx
Diabetic Retinopathy.pptx
 
Mask R-CNN
Mask R-CNNMask R-CNN
Mask R-CNN
 
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
CutMix: Regularization Strategy to Train Strong Classifiers with Localizable ...
 
Data Science - Part XVII - Deep Learning & Image Processing
Data Science - Part XVII - Deep Learning & Image ProcessingData Science - Part XVII - Deep Learning & Image Processing
Data Science - Part XVII - Deep Learning & Image Processing
 
Prospects of Deep Learning in Medical Imaging
Prospects of Deep Learning in Medical ImagingProspects of Deep Learning in Medical Imaging
Prospects of Deep Learning in Medical Imaging
 
Number plate recognition system using matlab.
Number plate recognition system using matlab.Number plate recognition system using matlab.
Number plate recognition system using matlab.
 
AOI應用於生醫及生物檢測之實例探討
AOI應用於生醫及生物檢測之實例探討AOI應用於生醫及生物檢測之實例探討
AOI應用於生醫及生物檢測之實例探討
 
Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)Brain tumor detection by scanning MRI images (using filtering techniques)
Brain tumor detection by scanning MRI images (using filtering techniques)
 
ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)
ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)
ディープラーニングを用いた物体認識とその周辺 ~現状と課題~ (Revised on 18 July, 2018)
 
CNN and its applications by ketaki
CNN and its applications by ketakiCNN and its applications by ketaki
CNN and its applications by ketaki
 
Face recognition technology
Face recognition technologyFace recognition technology
Face recognition technology
 
Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)
Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)
Efficient Lifelong Learning with A-GEM ( ICLR 2019 読み会 in 京都 20190602)
 
Metal Surface Defect Inspection Through Deep Neural Network
Metal Surface Defect Inspection Through Deep Neural NetworkMetal Surface Defect Inspection Through Deep Neural Network
Metal Surface Defect Inspection Through Deep Neural Network
 
Viola-Jones Object Detection
Viola-Jones Object DetectionViola-Jones Object Detection
Viola-Jones Object Detection
 
Minor project report on logo detection on vehicles
Minor project report on logo detection on vehiclesMinor project report on logo detection on vehicles
Minor project report on logo detection on vehicles
 
Lecture 9&10 computer vision segmentation-no_task
Lecture 9&10 computer vision segmentation-no_taskLecture 9&10 computer vision segmentation-no_task
Lecture 9&10 computer vision segmentation-no_task
 
Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...
Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...
Skin Lesion Detection from Dermoscopic Images using Convolutional Neural Netw...
 
Convolutional neural network
Convolutional neural network Convolutional neural network
Convolutional neural network
 
Deep Learningと画像認識   ~歴史・理論・実践~
Deep Learningと画像認識 ~歴史・理論・実践~Deep Learningと画像認識 ~歴史・理論・実践~
Deep Learningと画像認識   ~歴史・理論・実践~
 

Similar to Deep Learning for Imbalanced Data Defect Detection

Convolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detectionConvolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detectionDarian Frajberg
 
Weakly Supervised Whole Slide Image Analysis Using Cloud Computing
Weakly Supervised Whole Slide Image Analysis Using Cloud ComputingWeakly Supervised Whole Slide Image Analysis Using Cloud Computing
Weakly Supervised Whole Slide Image Analysis Using Cloud ComputingSean Yu
 
Optimizing Terascale Machine Learning Pipelines with Keystone ML
Optimizing Terascale Machine Learning Pipelines with Keystone MLOptimizing Terascale Machine Learning Pipelines with Keystone ML
Optimizing Terascale Machine Learning Pipelines with Keystone MLSpark Summit
 
05 contours seg_matching
05 contours seg_matching05 contours seg_matching
05 contours seg_matchingankit_ppt
 
Anjali (qa qc on cr dr)
Anjali (qa qc on cr dr)Anjali (qa qc on cr dr)
Anjali (qa qc on cr dr)globalsoin
 
Computer Graphics
Computer GraphicsComputer Graphics
Computer GraphicsAdri Jovin
 
Real time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithmReal time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithmajayrampelli
 
74 real time-image-processing-applied-to-traffic-queue-d
74 real time-image-processing-applied-to-traffic-queue-d74 real time-image-processing-applied-to-traffic-queue-d
74 real time-image-processing-applied-to-traffic-queue-dravi247272
 
5707_10_auto-encoder.pptx
5707_10_auto-encoder.pptx5707_10_auto-encoder.pptx
5707_10_auto-encoder.pptxSidoriOne
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用CHENHuiMei
 
presentation
presentationpresentation
presentationlei zhang
 
Computer graphics - Nitish Nagar
Computer graphics - Nitish NagarComputer graphics - Nitish Nagar
Computer graphics - Nitish NagarNitish Nagar
 
SPIE 10059-36(Reheman Baikejiang)
SPIE 10059-36(Reheman Baikejiang)SPIE 10059-36(Reheman Baikejiang)
SPIE 10059-36(Reheman Baikejiang)Reheman Baikejiang
 
Build Your Own 3D Scanner: 3D Scanning with Structured Lighting
Build Your Own 3D Scanner: 3D Scanning with Structured LightingBuild Your Own 3D Scanner: 3D Scanning with Structured Lighting
Build Your Own 3D Scanner: 3D Scanning with Structured LightingDouglas Lanman
 
Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...
Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...
Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...Lionel Briand
 
CyberSec_JPEGcompressionForensics.pdf
CyberSec_JPEGcompressionForensics.pdfCyberSec_JPEGcompressionForensics.pdf
CyberSec_JPEGcompressionForensics.pdfMohammadAzreeYahaya
 

Similar to Deep Learning for Imbalanced Data Defect Detection (20)

Convolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detectionConvolutional Neural Network for pixel-wise skyline detection
Convolutional Neural Network for pixel-wise skyline detection
 
Weakly Supervised Whole Slide Image Analysis Using Cloud Computing
Weakly Supervised Whole Slide Image Analysis Using Cloud ComputingWeakly Supervised Whole Slide Image Analysis Using Cloud Computing
Weakly Supervised Whole Slide Image Analysis Using Cloud Computing
 
Optimizing Terascale Machine Learning Pipelines with Keystone ML
Optimizing Terascale Machine Learning Pipelines with Keystone MLOptimizing Terascale Machine Learning Pipelines with Keystone ML
Optimizing Terascale Machine Learning Pipelines with Keystone ML
 
05 contours seg_matching
05 contours seg_matching05 contours seg_matching
05 contours seg_matching
 
Anjali (qa qc on cr dr)
Anjali (qa qc on cr dr)Anjali (qa qc on cr dr)
Anjali (qa qc on cr dr)
 
Temporal Segment Network
Temporal Segment NetworkTemporal Segment Network
Temporal Segment Network
 
Computer Graphics
Computer GraphicsComputer Graphics
Computer Graphics
 
PillQC-Deep-Learning
PillQC-Deep-LearningPillQC-Deep-Learning
PillQC-Deep-Learning
 
Real time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithmReal time-image-processing-applied-to-traffic-queue-detection-algorithm
Real time-image-processing-applied-to-traffic-queue-detection-algorithm
 
74 real time-image-processing-applied-to-traffic-queue-d
74 real time-image-processing-applied-to-traffic-queue-d74 real time-image-processing-applied-to-traffic-queue-d
74 real time-image-processing-applied-to-traffic-queue-d
 
5707_10_auto-encoder.pptx
5707_10_auto-encoder.pptx5707_10_auto-encoder.pptx
5707_10_auto-encoder.pptx
 
深度學習在AOI的應用
深度學習在AOI的應用深度學習在AOI的應用
深度學習在AOI的應用
 
presentation
presentationpresentation
presentation
 
Computer graphics - Nitish Nagar
Computer graphics - Nitish NagarComputer graphics - Nitish Nagar
Computer graphics - Nitish Nagar
 
SPIE 10059-36(Reheman Baikejiang)
SPIE 10059-36(Reheman Baikejiang)SPIE 10059-36(Reheman Baikejiang)
SPIE 10059-36(Reheman Baikejiang)
 
Build Your Own 3D Scanner: 3D Scanning with Structured Lighting
Build Your Own 3D Scanner: 3D Scanning with Structured LightingBuild Your Own 3D Scanner: 3D Scanning with Structured Lighting
Build Your Own 3D Scanner: 3D Scanning with Structured Lighting
 
OBDPC 2022
OBDPC 2022OBDPC 2022
OBDPC 2022
 
Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...
Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...
Testing Dynamic Behavior in Executable Software Models - Making Cyber-physica...
 
IMAGE PROCESSING
IMAGE PROCESSINGIMAGE PROCESSING
IMAGE PROCESSING
 
CyberSec_JPEGcompressionForensics.pdf
CyberSec_JPEGcompressionForensics.pdfCyberSec_JPEGcompressionForensics.pdf
CyberSec_JPEGcompressionForensics.pdf
 

More from CHENHuiMei

QIF對AOI設備業之衝擊與機會
QIF對AOI設備業之衝擊與機會QIF對AOI設備業之衝擊與機會
QIF對AOI設備業之衝擊與機會CHENHuiMei
 
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉CHENHuiMei
 
基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統CHENHuiMei
 
使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善CHENHuiMei
 
IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君CHENHuiMei
 
精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照CHENHuiMei
 
When AOI meets AI
When AOI meets AIWhen AOI meets AI
When AOI meets AICHENHuiMei
 
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠CHENHuiMei
 
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士CHENHuiMei
 
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民CHENHuiMei
 
2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益CHENHuiMei
 
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢CHENHuiMei
 
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章CHENHuiMei
 
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏CHENHuiMei
 
200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略CHENHuiMei
 
2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案CHENHuiMei
 
2008 year Computer Vision
2008 year Computer Vision2008 year Computer Vision
2008 year Computer VisionCHENHuiMei
 
2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術CHENHuiMei
 

More from CHENHuiMei (20)

QIF對AOI設備業之衝擊與機會
QIF對AOI設備業之衝擊與機會QIF對AOI設備業之衝擊與機會
QIF對AOI設備業之衝擊與機會
 
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
產研融合推手-台大AOI設備研發聯盟_台大陳亮嘉
 
基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統基於少樣本深度學習之橡膠墊片檢測系統
基於少樣本深度學習之橡膠墊片檢測系統
 
使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善使用人工智慧檢測三維錫球瑕疵_台大傅楸善
使用人工智慧檢測三維錫球瑕疵_台大傅楸善
 
IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君IIoT發展趨勢及設備業者因應之_微軟葉怡君
IIoT發展趨勢及設備業者因應之_微軟葉怡君
 
精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照精密機械的空間軌跡精度光學檢測法_台大范光照
精密機械的空間軌跡精度光學檢測法_台大范光照
 
Report
ReportReport
Report
 
Deep learning
Deep learningDeep learning
Deep learning
 
When AOI meets AI
When AOI meets AIWhen AOI meets AI
When AOI meets AI
 
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
2018AOI論壇_基於生成對抗網路之非監督式AOI技術_工研院蔡雅惠
 
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
2018AOIEA論壇Keynote_眺望趨勢 量測設備未來10年發展重點_致茂曾一士
 
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
2018AOI論壇Keynote_AI入魂製造領域現況與趨勢_工研院熊治民
 
2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益2018AOI論壇_AOI and IoT產線應用_工研院周森益
2018AOI論壇_AOI and IoT產線應用_工研院周森益
 
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
2018AOI論壇_AOI參與整廠協作之實務建議_達明機器人黃鐘賢
 
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
2018AOI論壇_深度學習在電腦視覺應用上的疑問_中央大學曾定章
 
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
2018AOI論壇_時機已到 AOI導入邊緣運算_SAS林育宏
 
200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略200704 Dr. Schenk 產品策略
200704 Dr. Schenk 產品策略
 
2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案2007 TFT LCD-AOI教學檔案
2007 TFT LCD-AOI教學檔案
 
2008 year Computer Vision
2008 year Computer Vision2008 year Computer Vision
2008 year Computer Vision
 
2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術2008 tech 高分子微光學應用技術
2008 tech 高分子微光學應用技術
 

Recently uploaded

SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024Scott Keck-Warren
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...Fwdays
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions
 
Science&tech:THE INFORMATION AGE STS.pdf
Science&tech:THE INFORMATION AGE STS.pdfScience&tech:THE INFORMATION AGE STS.pdf
Science&tech:THE INFORMATION AGE STS.pdfjimielynbastida
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptxLBM Solutions
 
Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksSoftradix Technologies
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticscarlostorres15106
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024BookNet Canada
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubKalema Edgar
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...shyamraj55
 
New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024BookNet Canada
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersThousandEyes
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 3652toLead Limited
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Enterprise Knowledge
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machinePadma Pradeep
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDGMarianaLemus7
 
Unlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power SystemsUnlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power SystemsPrecisely
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsMemoori
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupFlorian Wilhelm
 

Recently uploaded (20)

SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024SQL Database Design For Developers at php[tek] 2024
SQL Database Design For Developers at php[tek] 2024
 
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks..."LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
"LLMs for Python Engineers: Advanced Data Analysis and Semantic Kernel",Oleks...
 
Pigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping ElbowsPigging Solutions Piggable Sweeping Elbows
Pigging Solutions Piggable Sweeping Elbows
 
Science&tech:THE INFORMATION AGE STS.pdf
Science&tech:THE INFORMATION AGE STS.pdfScience&tech:THE INFORMATION AGE STS.pdf
Science&tech:THE INFORMATION AGE STS.pdf
 
Key Features Of Token Development (1).pptx
Key  Features Of Token  Development (1).pptxKey  Features Of Token  Development (1).pptx
Key Features Of Token Development (1).pptx
 
Benefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other FrameworksBenefits Of Flutter Compared To Other Frameworks
Benefits Of Flutter Compared To Other Frameworks
 
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmaticsKotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
Kotlin Multiplatform & Compose Multiplatform - Starter kit for pragmatics
 
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
#StandardsGoals for 2024: What’s new for BISAC - Tech Forum 2024
 
Unleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding ClubUnleash Your Potential - Namagunga Girls Coding Club
Unleash Your Potential - Namagunga Girls Coding Club
 
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptxVulnerability_Management_GRC_by Sohang Sengupta.pptx
Vulnerability_Management_GRC_by Sohang Sengupta.pptx
 
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
Automating Business Process via MuleSoft Composer | Bangalore MuleSoft Meetup...
 
New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
New from BookNet Canada for 2024: BNC BiblioShare - Tech Forum 2024
 
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for PartnersEnhancing Worker Digital Experience: A Hands-on Workshop for Partners
Enhancing Worker Digital Experience: A Hands-on Workshop for Partners
 
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
Tech-Forward - Achieving Business Readiness For Copilot in Microsoft 365
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024
 
Install Stable Diffusion in windows machine
Install Stable Diffusion in windows machineInstall Stable Diffusion in windows machine
Install Stable Diffusion in windows machine
 
APIForce Zurich 5 April Automation LPDG
APIForce Zurich 5 April  Automation LPDGAPIForce Zurich 5 April  Automation LPDG
APIForce Zurich 5 April Automation LPDG
 
Unlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power SystemsUnlocking the Potential of the Cloud for IBM Power Systems
Unlocking the Potential of the Cloud for IBM Power Systems
 
AI as an Interface for Commercial Buildings
AI as an Interface for Commercial BuildingsAI as an Interface for Commercial Buildings
AI as an Interface for Commercial Buildings
 
Streamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project SetupStreamlining Python Development: A Guide to a Modern Project Setup
Streamlining Python Development: A Guide to a Modern Project Setup
 

Deep Learning for Imbalanced Data Defect Detection

  • 1. 1 AOIEA 深度學習於表面瑕疵檢測: Strategies for imbalanced data training 蔡篤銘 教授 元智大學 工業工程與管理學系 2018 Taiwan AOI Forum and show
  • 2. 物體表面特徵(surface texture) 2 High-resolution Resolution Uniform Low contrast Patterned Homogeneous texture Repeated patterns Heterogeneous texture •Web material •Paper •Steel surface •Mura •Backlight panel •Assembled PCB •IC die •TFT-LCD •IC wafer •TFT-LCD •Color filter •Solar cell/wafer Uniform Low contrast Patterned Homogeneous texture Repeated patterns Heterogeneous texture
  • 3. Inspection in manufacturing Product & process properties in manufacturing: • Small , subtle local defect in − Size − Shape − Deformation − Gray scale/color unevenness • Imbalanced data – Many positive (defect-free) samples – Only a few or No negative (defect) samples Requirements: • High precision/accuracy – Location – Detection rate • Computationally fast (Real-time computation to meet the cycle time) 3
  • 4. Implement the deep learning models for • CNN regression for image positioning • GAN-based techniques to generate synthesized defects for the data-imbalance problem • Autoencoding models – Convolutional Autoencoders (CAE) with one-class SVM classification for unsupervised anomaly detection – Variational Autoencoders (VAE) image reconstruction for unsupervised defect detection 4
  • 5. 5 Deep learning for PCB positioning
  • 6. 6PCB assembly PCB inspection Deep learning for PCB positioning A precise positioning system can be used for  Automated assembly  Automated visual defect detection Bare PCB Assembled PCB
  • 7.  Problem Predict Angle (θ)、Horizontal displacement (x) and Vertical displacement (y) of a PCB w.r.t. the template. 7 (θ, x, y)=(-20°, -20, -20) (θ, x, y)=(4°, 0, 12) (θ, x, y)=(20°, 20, 20) (θ, x, y)=(0°, 0, 0)PCB board Deep learning for PCB positioning Template
  • 8. 8 Traditional machine learning for regression Deep learning for regression Deep learning regression v.s. traditional ML regression
  • 9. 9 CNN for regression: CNN as feature extractor for SVR regression: DNN for regression: CNN + SVR for regression: Using Convolutional Neural Network (CNN) and Support Vector Regression (SVR) models to predict θ, x and y.
  • 10. 10 Linear SVR Nonlinear SVR (with Kernel transformation) Polynomial: Radial Basis Function(RBF): Support Vector Regression (SVR) x = input feature(s) y = output response(s)
  • 11. 11 • Input PCB image into DNN model to predict (θ, x, y) • Number of hidden layers: 3 Deep Neural Network (DNN) model
  • 12. 12 • Input PCB image into CNN model to predict (θ, x, y) Convolutional Neural Network (CNN) model
  • 13. 13 Note: 1. Kernel function “Radial Basis Function” is used 2. CNN model is the same as the previous one, the number of features to SVR is 3. 3. Training time: 2.2 hours (for 9,261 samples) CNN+SVR model for regression
  • 14. 14 CNN as feature extractor for regression Note: 1. Kernel function “Radial Basis Function” is used 2. CNN model is the same as the previous one; the last layer of CNN contains 128 nodes (i.e., number of features to SVR is 128) 3. Training time: 2.2 hours (for 9,261 samples)
  • 15. 15 The good thing about Deep learning for Regression: The user needs only provide ONE SINGLE template image
  • 16. 16 It creates the input image (x) and the corresponding output (θ, x, y) for DL training Reference PCB image provided by users: Augmented image: Corresponding output: (θ, x, y)=(-20°, -20, -20) (θ, x, y)=(0°, 0, 0) (θ, x, y)=(20°, 20, 20) 180 pixel 180 pixel The training samples are automatically generated by Image Augmentation from the template
  • 17. 17 Range of Training samples • Image size: 180 × 180 (pixels) • Angles: 0°、±2°、±4°、 …、± 20° (every two angles for θ) • Horizontal displacement: 0、±2、±4、…、 ± 20 (pixel) (every two pixels for x) • Vertical displacement : 0、±2、±4、…、 ± 20 (pixel) (every two pixels for y) • Number of training samples: 9,261 (21 × 21 × 21) Testing samples • Image size :180 × 180 (pixels) • Angles : 0°、±1°、±2°、…、± 20° • Horizontal displacement : 0、±1、±2、…、±20 (pixel) • Vertical displacement : 0、±1、±2、…、±20 (pixel) • Number of testing samples :68,921 (41 × 41 × 41) Note: The images with odd θ, x and y are unseen to the training model
  • 18. 18 Evaluate positioning accuracy by mean error, variance and maximum error. Angle error (degree) Horizontal error (pixel) Vertical error (pixel) Mean Variance Max Mean Variance Max Mean Variance Max DNN 0.111 0.008 0.689 0.119 0.009 1.009 0.120 0.008 0.876 CNN 0.124 0.008 0.634 0.162 0.008 0.657 0.133 0.007 0.563 CNN+SVR 0.049 0.001 0.464 0.049 0.001 0.344 0.055 0.002 0.506 CNN as feature extractor 0.068 0.003 0.499 0.066 0.003 0.484 0.069 0.003 0.515 Positioning accuracy
  • 19. 19 Model Time (seconds) DNN 0.00123 CNN 0.00166 CNN+SVR 0.00230 CNN as feature extractor 0.00541 Note: Equipment 1. CPU: Intel® Core™ i7-6700K CPU @ 4.00GHz × 8 2. GPU: GeForce GTX 1080 Ti Computation time of each model It achieves 2-milliseconds efficiency.
  • 20. 20 Defect inspection by image subtraction Template fT , (a) Normal (b) Scratch (c) Extrusion Test image , Image subtraction from , Result ∆ , (d) Intrusion ∆ , fT , , Template Aligned
  • 21. Saw-mark defect detection in heterogeneous solar wafer images using - GAN-based training samples generation - CNN classification 21
  • 22. Multicrystalline solar wafer inspection • Multicrystalline silicon wafers  A multicrystalline solar wafer presents random shapes, sizes and directions of crystal grains in the surface and results in a heterogeneous texture.  A saw-mark defect is a severe flaw of wafers when cutting the silicon ingot into thin wafers using the multi-wire saws. Defect-free solar wafer image White saw-mark defect Saw-mark defect caused by impurity Solar wafer image with a black saw-mark defect 22
  • 23. The proposed deep learning scheme is composed of two phases: Defect samples generation using the CycleGAN (Cycle-consistent Generative Adversarial Networks), and then Defect detection using the CNN (Convolutional Neural Networks) based on the true defect-free samples and the synthesized defective samples. • The CycleGAN model combines both the adversarial loss (i.e. GAN) and the cycle consistency loss . • GAN measures the adversarial lose between the generated images and the target image. • The consistency lose prevents the learned forward and backward mappings from contradiction. • It uses unpaired datasets (not specific paired samples in GAN) for the training, and is suited for our application. CycleGAN model used for defect patches generation 23
  • 24. Real solar wafer surfaces For training the CycleGAN: • Use a small set of true defect patches (60 for black, 90 for white defects) as the target dataset, and then randomly collect a small set of defect-free patches (60 & 90) as the input dataset to the CycleGANs. Real defect-free samples Real black saw-mark samples Real white saw-mark samples 24
  • 25. Using the CycleGAN model to generate the defective samples Synthesized defects: • Whenever we change the input set with different defect-free patches to the trained CycleGAN, a new defective set is created. Real defect-free samples input to the trained CycleGAN Generated black saw-mark patches Generated white saw-mark patches 25
  • 26. The CNN model for classification • A simple CNN with 3 convolutional layers is used for the training. • A lean CNN model gives better computational efficiency in the inspection process. • Training information: – For the CycleGAN models, 150 (60 & 90 Black and White sawmarks) real defective patches and 150 (60&90) real defect-free patches are used as the training samples. – For the CNN model, a total of 4000 real defect-free patches and 4000 synthesized sawmark patches are used as the training samples. – Patch size 50 50 pixels – Training time : 3 hours for CycleGAN , and 1 hour for CNN CNN model used for defect detection 26
  • 27. Postprocessing with conventional machine vision techniques • The saw-mark in a small windowed patch contains only subtle changes and, thus, the entire saw-mark region may not be completely detected in the full-sized solar wafer image. • Apply the horizontal projection line by line in the resulting binary image B , to intensify the horizontal saw-mark in the image . That is P ∑ , , ∀ • The maximum projection value is used as the discriminant measure for saw-mark detection, i.e. P ∗ P , ∀ . If the horizontal projection P ∗ is large enough, a saw-mark at line ∗ is declared. Note: The mean computation time is 0.004 seconds for an image patch of size 50×50 pixels on a PC with an Intel Core 2, 3.6GHz CPU and an NVIDIA GTX 1070 GPU . 27
  • 28. Detection results on sawmarks • Detection results of defect-free solar wafer images Test images Detection result projection 28
  • 29. Detection results on sawmarks )( yP y )( yP y )( yP y • Detection results of defective solar wafer images Test images Detection result projection 29
  • 30. Detection results on stains / foreign particles 30 Real stains defect samples Synthesized stain defect samples
  • 31. Detection results on stains / foreign particles • Detection results of defective solar wafer images (a) Test images (b) Detection result 31
  • 32. Additional test: Using the CycleGAN model to generate bump defects • Bumps defect 32 Real defect samples Real defect-free samples Generated defect samples
  • 33. Autoencoders for defect detection -Autoencoders for image reconstruction - Autoencoders for feature extraction 33
  • 34. Unsupervised Autoencoders for image reconstruction , , Trained model Autoencoding Model Self-reference comparison: ∆ , , , ∆ , ∆ , , Encoder Decoder Image Image Testing image Reconstructed image 34
  • 35. Defect detection in TFT-LCD • Thin Film Transistor-Liquid Crystal Display (TFT-LCD) comprises vertical data lines and horizontal gate lines . • The main types of defects are pinholes, particles and scratches defect. defect-free LCD image LCD image with Particle defects LCD image with Pinhole defects LCD image with Scratch defects 35
  • 36. VAE (Variational AutoEncoder) model for image reconstraction - The Model • Structure of the VAE model Zp Z p Z Defect-free Defect-free Latent variables 36
  • 37. VAE model for image reconstruction - Detection results by image subtraction • Defect-free image • Defect image (true defects) Original image Restored image Image subtraction Binarization , , ∆ , Original image Restored image Image subtraction Binarization , , ∆ , 37
  • 38. Use AE for feature extraction for anomaly detection (with one-class SVM) • Testing image is reconstructed from the trained AE model. • The features are extracted from the last layer of the Encoder , and used as the input data of the one-class SVM to identify the anomalies. ZEncoder DecoderEncoder Decoder Test image Anomaly detection Feature maps One-class SVM Trained AE model 38
  • 39. Use AE for feature extraction for anomaly detection (one-class SVM) • One-class SVM (Support Vector Machine) Ni NiR CR i ii N i i ,,2,1,0 ,,2,1, s.t. Min 22 1 2          ax Ra Positive samples Support vector Outlier Center point (a) Radius (R) a , 39
  • 40. Use AE for feature extraction for anomaly detection (one-class SVM) : The Model • Structure of the AE model: ZEncoder DecoderEncoder Decoder Feature maps
  • 41. Use AE for feature extraction for anomaly detection (one-class SVM) : Training samples (LCD)  Train only positive (defect-free) samples: − Original 256 256 LCD images are rotated between 0° and 35° with 5° increment. − 256 random image patches of size 28×28 are used for training. − Using the AE model to extract the features for the one-class SVM model.  Testing data: − 160 positive samples − 33 negative (true defect) samples Computation time : training 15 minutes , testing 0.00008 seconds 41
  • 42. Use AE for feature extraction for anomaly detection (one-class SVM) : Detection results (LCD) • Feature size : 490 (10 feature maps of size 7*7) • Testing data : 160 positive samples 、 33 negative (true defect) samples Prediction Actual Normal Outlier Normal 89% 11% Outlier 0% 100% Type I error 11% Type II error 0% Overall recognition rate 90% 42
  • 43. Use AE for feature extraction for anomaly detection (one-class SVM) : Detection results (LCD) • Feature size : 49 (use only the best feature map with size 7*7) Testing data : 160 positive samples , 33 negative (true defect) samples Prediction Actual Normal Outlier Normal 92% 8% Outlier 0% 100% Type I error 8% Type II error 0% Overall recognition rate 92% 43
  • 44. A thought on Deep Learning: • Can Deep Learning replace Machine Vision? • MV as preprocessing , and DL as postprocessing – MV for defect detection, and DL for defect classification – Or, vice versa • MV in DL models? – “Convolution” and “pooling” are parts of image processing operations – Human knowledge in DL models (e.g. embed the known defect features to the DL model) 44