MATLAB 影像處理及擴展應用研討會 醫學影像、機器學習、物聯網(IoT)

MATLAB 影像處理及擴展應用研討會 醫學影像、機器學習、物聯網(IoT)

Agenda

Time Topics

Welcome

13:30 - 14:30 MATLAB 於影像處理介紹

14:30 - 15:20 應用機器學習技法於影像處理領域

15:20 - 15:30 Coffee Break

15:30 - 16:30 MATLAB與物聯網: 連結至穿戴式裝置硬體之訊號處理

Q&A and Wrap UP

Applications: Image and Video Processing

▪ Medical imaging

▪ Surveillance

▪ Robotics

▪ Automotive safety

▪ Consumer electronics

▪ Geospatial computing

▪ Machine vision

▪ and more…

Different Aspects of Image Processing

▪ Importing, visualizing and investigating

▪ Enhancing images

▪ Aligning multiple images

▪ Segmenting images

▪ Detecting image features

▪ Classifying images

Technical Computing Workflow

Share Explore & Discover

Access

Image acquisition Image data in files

Reporting and Documentation

Outputs for Design

Deployment Visualization

tools

Application Development

Data/Image Analysis and

Modeling

Image Processing

Consider this image from the Centers for Disease Control:

 Our goal:

To develop an algorithm to detect and quantify infection.

How many cells are in the image, and how many are infected?

Quantifying infection across multiple images…

…Despite widely varying image quality

Identify key challenges, consider strategies:

•

Challenges:

•

Differences in color

•

Differences in illumination

•

Contiguity of cells

•

Low resolution/poor quality

•

Strategies:

•

Using apps to explore images

•

Pre-processing

•

Watershed segmentation

•

Morphological segmentation

DEMO

In this session…

…we quantified rates of infection in heterogeneous images

What if we wanted to classify the type of infection,

differentiating several species of parasites?

Agenda

Time Topics

Welcome

13:30 - 14:30 MATLAB 於影像處理介紹

14:30 - 15:20 應用機器學習技法於影像處理領域

15:20 - 15:30 Coffee Break

15:30 - 16:30 MATLAB與物聯網: 連結至穿戴式裝置硬體之訊號處理

Q&A and Wrap UP

Machine Learning

A machine learning algorithm takes examples of inputs and outputs associated with a task and produces a program that can automatically differentiate them.

If brightness > 0.5 then ‘hat’

If edge_density < 4 and major_axis > 5

then “boat”

…

‘boats’

‘mugs’

‘hats’

Hand Written Program

Machine Learning

𝑚𝑜𝑑𝑒𝑙 = 𝑓𝑖𝑡𝑐svm(image_features, label)

‘boats’

‘mugs’

‘hats’

Computer Vision

Machine Learning Workflow Using Images

Training Data Feature Extraction, Encoding

Machine Learning

Classifier ‘babesiosis’

Input Image Feature Extraction,

Encoding Classification

‘babesiosis’ ‘plasmodium’ ‘chagas’

Bag of Words

Performimage processing, analysis, and algorithm development Image Processing Toolbox™ provides a comprehensive set of reference- standard algorithms, functions, and apps for image processing, analysis, visualization, and algorithm development. You can perform

image analysis^,imagesegmentation,image enhancement^{, noise}

reduction, geometric transformations, andimageregistration. Many toolbox functions support multicore processors, GPUs, and C-code generation.

Image Processing Toolbox supports a diverse set of

image

processing, displaying, and navigating large images.

Image Processing Toolbox

Bag: image processing, analysis , image, pixels, enhancement

Class / Label

Training Data

Vocabulary / Bag of Words

Bag of “Visual Words” ( features)

‘babesiosis

’

Class / Label

Training Data

Vocabulary / Bag of Features

What is a Classifier ?

Training Data

Features Classifier

Classification ‘babesiosis’

‘plasmodium’

‘chagas’

Class Membership

Machine Learning

Encoded images

So let’s give it a try…

Using Machine Learning for Computer Vision

▪ Image Processing Toolbox

• Provides 100s of validated functions

• Indispensable for image processing applications

▪ Computer Vision System Toolbox

• Provides tools to generate image features for training classifiers

• See doc for full list of provided image features

▪ Statistics and Machine Learning Toolbox

• Provides learning algorithms to train classifiers

Agenda

Time Topics

Welcome

13:30 - 14:30 MATLAB 於影像處理介紹

14:30 - 15:20 應用機器學習技法於影像處理領域

15:20 - 15:30 Coffee Break

15:30 - 16:30 MATLAB與物聯網: 連結至穿戴式裝置硬體之訊號處理

Q&A and Wrap UP

The Challenge of IoT

?

Devices Insight

What is the Internet of Things?

Smart Connected

Devices Exploratory Analysis

Data Aggregator

(Server or Cloud) Deploy analytics to cloud

Deploy algorithms to smart devices

Business Systems Integration

Look at the Pieces: Devices

▪ Sensors and Human Interaction

▪ May have strict energy budget

▪ Required embedded programming skills

▪ Bandwidth is expensive in power and dollars

▪ Goal for device is to do as much data reduction locally as possible

Smart, Connected Devices

• Communication

• Embedded Sensor Analytics

• Data Reduction

Data Aggregator

• Storage

• On-Line analytics

• Visualization & reporting

Look at the Pieces: Data Collection & Online Analytics

▪ Server or Cloud-based

▪ Log and analyze information across collection of devices

▪ Real-time information for situational awareness

▪ Requires cloud/web development skills and operations support

▪ Data Intake must be scalable and reliable

▪ True “Big Data”

Look at the Pieces: Exploratory Analysis

Exploratory Analysis

• Historical analytics

• Algorithm development

▪ Desktop-based

▪ Access historical device information

▪ Analyze past performance for predictive modeling and deep insight

▪ Requires data analysis / data science skills

▪ Heavy use of statistics and signal processing

techniques

▪ Deploy complex algorithms to both cloud and edge devices

IoT – Design challenges

• Embedded development is challenging

• Increasing algorithmic complexity

• Need connectivity to cloud resources

• Streaming data management and storage

• Online and real-time analytics

• Ability to visualize results and make decisions

• Advanced analysis algorithms

Business Systems

• Integrating

various software components

• Acquiring data from external sources

MathWorks Capabilities for IoT

Leveraging MATLAB Analytics and Model-Based Design

Overview

Smart Connected

Devices Exploratory Analysis

Data Aggregator

(Server or Cloud) Deploy analytics to cloud

Deploy algorithms to smart devices

Business Systems Integration

• Data storage

• Online analytics

• Visualization and reporting

• Embedded algorithms for sensing, data reduction, and control

• Models and simulation of system

• Historical analytics

• Algorithm development Monitor and control nodes/devices

Publish data to Cloud

Model Device Communication

Physical Component Modeling

• Electronic

• Mechanical

• Hydraulic, etc.

Communications Protocol Modeling

• LTE, Zigbee, 802.11, etc.

Automatic Code Generation

• Programmable chips (MCU, DSP, etc.)

• FPGAs

Verification/Validation and Process Support

• Model- and Code proving

• Lifecycle management tools

MATLAB & Simulink Capabilities for IoT

Deployment

• .NET, COM components

• Java components

• Multicore and GPU systems

• Spreadsheet plug-ins

• Database plug-ins

• Hadoop

• Cloud services (AWS)

• ThingSpeak Apps

• Smartphone/tablet integration

Analysis, Modeling, Design

• Data visualization

• Statistics

• Regression

• Machine learning (supervised &unsupervised)

• Neural networks

• Optimization (gradient-based & stochastic)

• Symbolic computing

• Image analysis

• Financial analysis

• Geospatial computing

• Object recognition

• Speech recognition Data Clean-up

• Filtering

• Image processing

• Signal processing

• Telemetry

• RF sampling

Real-Time Sources

• Sensors

• GPS

• Instrumentation

• Cameras

• Communication systems

• Machines:

• embedded systems

• fieldbus

• Financial datafeeds File I/O

• Text

• Spreadsheet

• XML

• CDF/HDF

• Image

• Audio

• Video

• Geospatial

• Web content

Repositories

• Databases (SQL)

• NoSQL

• Hadoop

Communication Protocols

• CAN

• DDS

• OPC

• XCP

Business and Transactional

Data Engineering,

Scientific, and Field Data

Analytics Design and Development

Deploy and integrate in Target to

embedded

MATLAB and Simulink are well positioned for:

A. Analytics that increasingly require both business and engineering data

B. Developing embedded systems which have increasing analytic content

C. Deploying the increasing number of analytic-rich applications that run on both traditional IT and embedded platforms

D. Enabling domain experts to do data science

Why MATLAB?

C D

A

B

Customer Case Study

Customer Study: iSonea

Cloud and Embedded Analytics

Opportunity

• Develop an acoustic respiratory monitoring system for wheeze detection and asthma management

Analytics in cloud and embedded

• Captures 30 seconds of windpipe sound and processes the data locally to clean up and reduce ambient noise

• Invokes spectral processing and pattern-detection analytics for wheeze detection on iSonea server in the cloud

• Provides feedback to the patient on their smartphone Benefit

• Eliminates error-prone self-reporting and visits to the doctor

Sensor Analytics for IoT

Signal analysis for classification Application examples

▪ Mobile sensing

▪ Structural health monitoring (SHM)

▪ Fault and event detection

▪ Automated trading

▪ Radar post-processing

▪ Advanced surveillance

▪ ...

Case Study: Human Activity Analysis and Classification

Classificatio n

Feature Extraction

Dataset courtesy of:

Davide Anguita, Alessandro Ghio, Luca Oneto, Xavier Parra and Jorge L. Reyes-Ortiz.

Human Activity Recognition on Smartphones using a Multiclass Hardware-Friendly Support Vector Machine.

International Workshop of Ambient Assisted Living (IWAAL 2012). Vitoria-Gasteiz, Spain. Dec 2012 http://archive.ics.uci.edu/ml/datasets/Human+Activity+Recognition+Using+Smartphones

Machine Learning Signal

Processing

Sensor Data Analytics Workflow – the bigger picture

• Domain knowledge

• Open-ended problem

• Long discovery cycles

• Built-in algorithms

Machine Learning Connect and

Acquire

Signal Processing

Sensor Data Analytics Workflow – the bigger picture

• Different tools and environments

• Disconnect between hardware and analysis

• Inefficiencies in data sharing

• MATLAB Connects to DAQ interfaces and sensors directly. E.g.

• Android Sensor Support

• iPhone and iPad Sensor Support

Machine Learning Connect and

Acquire

Signal Processing

Embedded Implementation

Sensor Data Analytics Workflow – the bigger picture

• Signal analysis vs. on-line DSP

• From Machine Learning theory to pre-trained, low-footprint classifiers

• MATLAB vs. C/C++

• Streaming algorithms, data sources and

visualization for System modelling and simulation

Leverage Built-in Algorithms, Apps, and Technologies

▪ Signal Processing Toolbox™

Built-in algorithms and Apps to process and

analyse signals ^_ ^cheby2_filter

 rms

 pwelch

 periodogra m

 xcov

 findpeaks

 …

Leverage Built-in Algorithms, Apps, and Technologies

▪ Signal Processing Toolbox™

▪ Parallel Computing Toolbox™

Accelerate computationally and data- intensive problems using multicore

processors, GPUs and computer clusters

 parfor

Leverage Built-in Algorithms, Apps, and Technologies

▪ Signal Processing Toolbox™

▪ Parallel Computing Toolbox™

▪ Statistics and Machine Learning Toolbox™

Functions and apps to describe, analyze, and model data.

Regression, clustering and classification algorithms to draw inferences from data and build predictive models

>> classificationLearner

Leverage Built-in Algorithms, Apps and Technologies

▪ Signal Processing Toolbox™

▪ Parallel Computing Toolbox™

▪ Statistics and Machine Learning Toolbox™

▪ Neural Network Toolbox™

Functions and apps to design, train, visualize, and simulate neural

networks

 >> nprtool

 patternnet

Leverage Built-in Algorithms, Apps and Technologies

▪ Signal Processing Toolbox™

▪ Parallel Computing Toolbox™

▪ Statistics Toolbox™

▪ Neural Network Toolbox™

• DSP System Toolbox™

Streaming algorithms, data sources and visualization for system

modelling and simulation

 BiquadFilter

 MatFileReader

 Autocorrelator

 SpectrumEstimator

 TimeScope

Leverage Built-in Algorithms, Apps and Technologies

▪ Signal Processing Toolbox™

▪ Parallel Computing Toolbox™

▪ Statistics Toolbox™

▪ Neural Network Toolbox™

▪ DSP System Toolbox™

▪ MATLAB Coder™

>> codegen

Signal Processing and Machine Learning Techniques for Sensor Data Analytics

Summary

▪ Extensive set of de-facto standard functions for signal processing and machine learning

▪ Environment accelerates insight and

automation: visualisation, apps, language, documentation

▪ Path to embedded products, from on-line simulation to automatic code generation

Any Questions?