+ All Categories
Home > Documents > An Introduction to Real-time Machine Vision in Mechatronics Dr. Onur TOKER.

An Introduction to Real-time Machine Vision in Mechatronics Dr. Onur TOKER.

Date post: 21-Dec-2015
Category:
View: 236 times
Download: 1 times
Share this document with a friend
34
An Introduction to Real-time Machine Vision in Mechatronics Dr. Onur TOKER
Transcript

An Introduction to Real-time Machine Vision in

Mechatronics

Dr. Onur TOKER

Dr. Onur TOKER 2

Outline

• RT Machine Vision ? Mechatronics ?

• Review of previous experiments

• Image sensors (CMOS versus CCD)

• CMUCam, and cwCAM

• Interfacing a CCD camera to an 8-bit uC

• Difficulties in real-time machine vision

• Conclusion

Dr. Onur TOKER 3

RT Machine Vision ? Mechatronics ?

Machine vision is the ability of a computation machine to "see."

•Visual object tracking

•Object recognition

•Automated inspection, sorting

•Pattern recognition, etc.

RT: There is no strict real-time system. There are systems with very short event response latency times.

Dr. Onur TOKER 4

Experiment #1

• 1-D tracking system

• Analog video camera & PCI grabber

• VB 6 & VFW based

• Simple algorithm

• PID control

• Pentium 2/350MHz

Dr. Onur TOKER 5

Experiment #2

• Line following

• Wireless video camera and ToyCar

• Processing on a remote PC

• VC++ & DirectX based

• Simple algorithm

• Pentium 3/1GHz

Dr. Onur TOKER 6

Experiment #3

• Intel 8051

• Very primitive machine vision

• Rapid prototyping board

• LDR sensor

• MOSFET driver

Dr. Onur TOKER 7

Experiment #3

Dr. Onur TOKER 8

BOE-BOT kit

Simple kit

PBASIC

Not very flexible

Very small RAM

IR LEDs &

photo transistors

Dr. Onur TOKER 9

BOE-BOT demo

Dr. Onur TOKER 10

Other demos

CMUCam demo

(Color tracking) WAM demo (MIT 1995)

(Tracking by stereo machine vision)

Dr. Onur TOKER 11

CMOS image sensors

A CMOS sensor (OV7620)

CUMCam uses such a sensor 2nd PCB has a Scenix

uC

Digital output

Easy to interface

DALSA CMOS Sensor

Dr. Onur TOKER 12

CCD image sensors

DALSA CCD Sensor

Analog output

Difficult to interface

Require several support chips

Dr. Onur TOKER 13

CMOS versus CCD

CMOS sensor

640x480 mode

CCD sensor

640x480 (NTSC output)

Under same lightning,

same distance,

comparable budget,

CCD image is better.

Dr. Onur TOKER 14

CMUCam architecture

CMOS sensor

SX28

uCuC/DSP

•“User device” issues high level commands

•SX28 does the processing (Limited built-in functions)

•SX28 replies

Serial I/O

CMUCam

Dr. Onur TOKER 15

What is wrong with CMUCam ?

• Serial I/O (Low bandwidth)• Low frame rate (Max. 17fps)• CMOS sensor• Processing done by SX28• Limited to built in functions• Not much flexibility• Instead of FPGA, uses SX28• Very compact design

Dr. Onur TOKER 16

Proposed cwCam architecture

CCD camera

Video ADC

FPGA

uC/DSP

uC/DSP

uC/DSP

•Co-operating windowing approach (Discussed later)

•Parallel processors

•Parallel application specific digital architectures in the FPGA

•ASIC CPU cores in FPGA

cwCam

Dr. Onur TOKER 17

Digitized video signal

VSYNC

One field

One frame

Dr. Onur TOKER 18

A single field

Several HSYNCs

VSYNC

Dr. Onur TOKER 19

Video ADC speed ?

VSYNC

HSYNC ???

Conclusion:Use 10MHz

ADC

Dr. Onur TOKER 20

Machine Vision with an Analog Industrial camera

• NTSC/30fps or PAL/25fps

• Even/odd field interlacing: 60fips/50fips rate

• 31ms VSYNC, 4.7us HSYNC for NTSC

• Needs a high speed ADC (AD9048 is 35 MHz)

• Most 8-bit uCs are too slow for this task

• Scenix SX28AC/DP 13.3 ns instruction cycle

• FPGA for accurate and high resolution capture

Dr. Onur TOKER 21

Scenix SX28AC/DP

• 13.3 ns instruction cycle (75MHz clock)

• 10MHz video sampling = 100 ns loop time

• 1 Branch=3 cycles

• 4 instruction loop OK, but int. RAM too small

• 8051 too slow !

• PIC16F877 too slow !

• USE AN FPGA !

Dr. Onur TOKER 22

Our FPGAs (Prototyping boards)

Spartan II FPGA 50 Kgate

8MB RAM

8051

Dr. Onur TOKER 23

Our ADC (AD9048)

•35MSPS, 8-bit Flash ADC, Bipolar, 550mW, DIP 28 available

•AD9203, 40MSPS, 10-bit,CMOS, 74mW, No DIP available

Actual photo of AD9048 used in our video digitizer

Dr. Onur TOKER 24

Cortex-I approach

•Bederson, 1992

•Logarithmic structured space variant pixel geometry

•Based on human vision system

•For real-time machine vision, reduce data to < 1500 pixels

Dr. Onur TOKER 25

Co-operating windowing (1)

•Nassif & Capson, 1997

•2 Watch windows (200x20)

•1 Peripheral window (40x40 … 200x200)

•1 Foveal window (20x20)

•Object tracking at 113Hz

Dr. Onur TOKER 26

Co-operating windowing (2)

Dr. Onur TOKER 27

Where we are at cwCAM ?

•AD9048 Video ADC board design completed (PCB layout !)

•AD9048 interfaced to 8051 prototyping board and tested

•Logic design is being done by Xilinx ISE software

•Mixed VHDL and graphical logic designs

•Tedious and long taskCCD camera

Video ADC

FPGA

cwCam

Dr. Onur TOKER 28

Human Vision ?

HMD and Dual monitor support

PUMA robot arm and dual camera set

Dr. Onur TOKER 29

Conclusion

•Real time machine vision requires innovative use of software and hardware techniques.

•Cortex-I (Human Eye), Co-operating windowing, etc.

•Innovative use of FPGAs and uC/DSPs.

•High frame rate CCD sensors.

•Optimum designs likely to be an application specific one.

•cwCAM is based on co-operating windowing approach and innovative hardware/software techniques.

Dr. Onur TOKER 30

QUESTIONS ?

Dr. Onur TOKER 31

Other slides

Dr. Onur TOKER 32

Prototyping / Final product

Prototyping board Serial download, EEPROM based, 9V battery

Final design EPROM based minimum size PCB

Dr. Onur TOKER 33

A Student ProjectLine following

robot

Phototransistor based sensors

Dr. Onur TOKER 34

Image sensor types

1. Charged coupled devices (CCD)

2. Charge injection devices (CID)

3. CMOS Active Pixel Sensors (CMOS)

•They all convert incident light (photons) into electronic charge (electrons) by a photo-conversion process.

•Color sensors can be made by coating each individual pixel with a filter color (e.g. red, green, and blue).

•Beyond that point, everything is different.


Recommended