Depth Analysis With Stereo Cameras

Slides:



Advertisements
Similar presentations
Autonomous Tracking Unit John Berglund Randy Cuaycong Wesley Day Andrew Fikes Kamran Shah Professor: Dr. Rabi Mahapatra CPSC Spring 1999 Autonomous.
Advertisements

ELECTRONIC SYSTEMS NESI NanoRacks Embedded System Integration Board HARDWARE Functional Design.
Sumitha Ajith Saicharan Bandarupalli Mahesh Borgaonkar.
Night Vision James Stacy Brian Herre Maurio Grando Eric Faller Chris Bawiec James Team Bender.
Super Fast Camera System Performed by: Tokman Niv Levenbroun Guy Supervised by: Leonid Boudniak.
Preliminary Design Review Jan 31, 2006 Brianna Bethel Robert Havlik Jessica Lowry Alex Silva.
1 Color Discriminating Tracking System Lloyd Rochester Sam Duncan Ben Schulz Fernando Valentiner.
Department of Electrical & Computer Engineering Team Hollot By: Cory Brett Jonathan Katon Thomas Pavlu Haiyan Xu NavFocus Midway Design Review.
Ahmed Abdel-Fattah Jerry Chang Derrick Culver Matt Zenthoefer.
DE2-115 Control Panel - Part II
Critical Design Review 27 February 2007 Black Box Car System (BBCS) ctrl + z: Benjamin Baker, Lisa Furnish, Chris Klepac, Benjamin Mauser, Zachary Miers.
USB host for web camera connection
Eye Detector Project Midterm Review John Robertson Roy Nguyen.
Lab 2: Capturing and Displaying Digital Image
© Copyright Xilinx 2004 All Rights Reserved 9 November, 2004 XUP Virtex-II Pro Development System.
1.  Project Goals.  Project System Overview.  System Architecture.  Data Flow.  System Inputs.  System Outputs.  Rates.  Real Time Performance.
Diffuse Optical Tomography Optimization and Miniaturization ECE 4902-Spring 2014 Thomas Capuano (EE&BME), Donald McMenemy (EE), David Miller (EE), Dhinakaran.
Adding the TSE component to BANSMOM system and Software Development m Yumiko Kimezawa October 4, 20121RPS.
Matt Waldersen T.J. Strzelecki Rick Schuman Krishna Jharjaria.
Facial Recognition Alex Newcomb, Tom Stefanyk. Group Members Alex Newcomb In charge of web server, image compression and facial recognition database The.
ECE532 Final Project Demo Disparity Map Generation on a FPGA Using Stereoscopic Cameras ECE532 Final Project Demo Team 3 – Alim, Muhammad, Yu Ting.
3. ISP Hardware Design & Verification
MTA-SZTAKI Roboteyepair an ultra-high speed stereo vision system based on CNN technology.
FPGA-based Platform for Real-Time Stereo Vision Sergiy Zhelnakov, Pil Woo (Peter) Chun, Valeri Kirischian Supervisor: Dr. Lev Kirischian Reconfigurable.
Intruder Alert System By: Jordan Tymburski Rachita Bhatia.
Universal Measurement System with Web Interface Maciej Lipiński Ph.D. Krzysztof Poźniak, MSc Grzegorz Kasprowicz Wilga r.
MULTIMEDIA INPUT / OUTPUT TECHNOLOGIES
UAV IMAGING G6: Shen, Yubing, Yushi. PANDABOARD Dual-Core 1.2 GHz ARM Cortex-A9 CPU 1 GB DDR2 SDRAM 5V Power Supply.
ECE 477 Final Presentation Team 1  Spring 2013 Zelun Tie Xin Jin Ranmin Chen Hang Xie.
Final Year Project(EPT4046) Development of an internet controlled Surveillance Mobile Robot By Mimi Madihah Bt Mohd Idris Id: BACHELOR OF ENGINEERING.
IMAGE COMPRESSION AND DECOMPRESSION USING ARM9 SYSTEM Under the guidance of Mrs. B. SATYA SRI DEVI,M.Tech, Assistant Professor By V. KUSUMA KUMARI Roll.
Identify internal hardware devices (e. g
Lab 4 HW/SW Compression and Decompression of Captured Image
BASIC PROGRAMMING C SCP1103 (02)
Gesture Control interface
Lab 0: Familiarization with Equipment and Software
Computer Architecture & Operations I
Voice Manipulator Department of Electrical & Computer Engineering
<Add team picture or relevant project picture here>
Depth Analysis With Stereo Cameras
DE2-115 Control Panel - Part II
BASIC PROGRAMMING C SCP1103 (02)
2D GPU Platform with Hardware-Accelerated Features
Computer Hardware By Millie Hay.
SUBMITTED BY EDGEFX TEAM
Nicole Stodola, Chris Pederson and Gerry Finlay
Accelerometer-Based Character Recognition Pen
Course Name: Computer Application Topic: Central Processing Unit (CPU)
Vision for Robotic Applications
Real-time Image Processing System
This chapter provides a series of applications.
Computer Basics.
Serial Data Hub (Proj Dec13-13).
AT91RM9200 Boot strategies This training module describes the boot strategies on the AT91RM9200 including the internal Boot ROM and the U-Boot program.
Neurochip3.
DVR LH-Series 1 Key Features H.264 8ch/16ch Stand Alone DVR
ChipScope Pro Software
Low cost FPGA implimentation of tracking system from USB to VGA
ECE 477 Final Presentation Team 2 Spring 2012
Depth Analysis With Stereo Camera
ChipScope Pro Software
Nicole Stodola, Chris Pederson and Gerry Finlay
Nicole Stodola, Chris Pederson and Gerry Finlay
Nicole Stodola, Chris Pederson and Gerry Finlay
Accelerometer-Based Character Recognition Pen
Depth Analysis With Stereo Camera
ECE 477 Final Presentation Team 13  Spring 2011
Nicole Stodola, Chris Pederson and Gerry Finlay
PRELIMINARY DESIGN REVIEW
Embedded Image Processing: Edge Detection on FPGAs
Presentation transcript:

Depth Analysis With Stereo Cameras Timo Hohn and Leo Nickerson Group 5

Functionality Motivation Take stereo images of a scene and calculate the distance objects are away from the two cameras. This distance will be represented on an output image as a color gradient that is displayed on an external VGA monitor. Motivation 3D depth analysis for remote robot imagery, where more information of the scene is necessary. Object detection and avoidance applications for robot navigation where IR systems are not available.

Hardware Block Diagram

Software Flow Diagram Start (User Initiates Analysis) Collect JPEGS from Both Cameras Trim Images To Overlap Pixel Recognition Depth Analysis of Matched Pixels Update output image Modify Video Buffer to Include Output Image as Overlay

Design Challenges Interfacing the Cameras` NTSC video in and RS232 communication to the DE2 board Adapting the camera communication library to work well with the DE2 board Using the onboard TV decoder to collect frame buffers from one of the two cameras Sending manipulated frames to VGA out. Application Notes are available from previous projects Pixel Recognition Open source libraries from the OpenCV project will be used for matching pixels May perform image subtraction to reduce number of pixels required to analyze

Included FPGA and External Components Cyclone II Processor (currently economy version) VGA and TV decoder accessed as Avalon-MM slaves Use of onboard SRAM modules and the possibility of larger SDRAM storage depending on memory use GPIO pins (Tx, Rx, and GND pins for each camera, which use 3.3V logic) Additional +5V power supply for camera modules (either USB wall plug or battery packs) A single onboard push button for initiating image capture

Example C++ Module For Finding Distances #include "DistFind.h" #include <cmath> DistFind::DistFind(int x_res, int fov_degree, double width){ x_res = x_res; fov_rad = (double)fov_degree*PI/180; rad_per_pixel = fov_rad/x_res; width = width; } double DistFind::find(int x_camera1, int x_camera2){ double offset1 = rad_per_pixel/2 + abs(x_camera1-320)*rad_per_pixel; double offset2 = rad_per_pixel/2 + abs(x_camera2-320)*rad_per_pixel; … … (x_camera1>320) ? offset1*=-1 : ; (x_camera2<320) ? offset2*=-1 : ; double phi1 = PI/2 + offset1; double phi2 = PI/2 + offset2; double angle_c = PI - phi1 - phi2; double sine_ratio = width/sin(angle_c); double side1 = sine_ratio*sin(phi2); double side2 = sine_ratio*sin(phi1); return (phi1<phi2) ? sin(phi1)*side1 : sin(phi2)*side2; }

Unit Testing Integration Testing Using stored images to test depth and pixel recognition tasks Sending an image out the VGA port to test that interface Collect raw data from NTSC video and observe its accuracy Take raw video input from one camera and route it directly to the VGA port Starting with a basic, high contrast environment with a single object. Work our way up to more complex environments involving overlapping objects.

Optional Features Additional Features Secondary Features Performing depth calculations on real time video (additional hardware required) Automation and optimization of frequent image captures (eliminate the need for user initialization) Secondary Features Removal of the NTSC video overlay and only display still images Simplification of analysis environment if pixel recognition algorithms deem inaccurate

Questions?