ENG 7854 / 9804 Industrial Machine Vision. Midterm Exam March 1, 2010.

Similar documents
Practice Exam Sample Solutions

ECE 172A: Introduction to Intelligent Systems: Machine Vision, Fall Midterm Examination

APS Sixth Grade Math District Benchmark Assessment NM Math Standards Alignment

Understanding Tracking and StroMotion of Soccer Ball

A Nondestructive Bump Inspection in Flip Chip Component using Fuzzy Filtering and Image Processing

Image Processing: Final Exam November 10, :30 10:30

Islamic University of Gaza Faculty of Engineering Computer Engineering Department

Computer Vision I - Filtering and Feature detection

EECS 556 Image Processing W 09. Image enhancement. Smoothing and noise removal Sharpening filters

Lecture 2 Image Processing and Filtering

Image Processing. Bilkent University. CS554 Computer Vision Pinar Duygulu

1. (10 pts) Order the following three images by how much memory they occupy:

Coarse-to-fine image registration

CHAPTER 3 IMAGE ENHANCEMENT IN THE SPATIAL DOMAIN

Perception. Autonomous Mobile Robots. Sensors Vision Uncertainties, Line extraction from laser scans. Autonomous Systems Lab. Zürich.

Detecting and Identifying Moving Objects in Real-Time

Computer Vision I - Basics of Image Processing Part 2

Filtering and Enhancing Images

Time Stamp Detection and Recognition in Video Frames

Motion and Tracking. Andrea Torsello DAIS Università Ca Foscari via Torino 155, Mestre (VE)

EXAM SOLUTIONS. Image Processing and Computer Vision Course 2D1421 Monday, 13 th of March 2006,

Anno accademico 2006/2007. Davide Migliore

Filtering Images. Contents

Babu Madhav Institute of Information Technology Years Integrated M.Sc.(IT)(Semester - 7)

SECTION 5 IMAGE PROCESSING 2

Computer Vision 2. SS 18 Dr. Benjamin Guthier Professur für Bildverarbeitung. Computer Vision 2 Dr. Benjamin Guthier

EE795: Computer Vision and Intelligent Systems

Problem definition Image acquisition Image segmentation Connected component analysis. Machine vision systems - 1

Digital Image Processing COSC 6380/4393

Counting Particles or Cells Using IMAQ Vision

NAME :... Signature :... Desk no. :... Question Answer

Image Processing. Traitement d images. Yuliya Tarabalka Tel.

Texture. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors. Frequency Descriptors

CS4733 Class Notes, Computer Vision

FPGA IMPLEMENTATION FOR REAL TIME SOBEL EDGE DETECTOR BLOCK USING 3-LINE BUFFERS

C E N T E R A T H O U S T O N S C H O O L of H E A L T H I N F O R M A T I O N S C I E N C E S. Image Operations II

BSc (Hons) Computer Science. with Network Security. Examinations for / Semester 2

Lesson 6: Contours. 1. Introduction. 2. Image filtering: Convolution. 3. Edge Detection. 4. Contour segmentation

Detection of Edges Using Mathematical Morphological Operators

(Sample) Final Exam with brief answers

Last update: May 4, Vision. CMSC 421: Chapter 24. CMSC 421: Chapter 24 1

Robot vision review. Martin Jagersand

COMPUTER AND ROBOT VISION

Segmentation. (template) matching

Image Processing using LabVIEW. By, Sandip Nair sandipnair.hpage.com

JNTUWORLD. 4. Prove that the average value of laplacian of the equation 2 h = ((r2 σ 2 )/σ 4 ))exp( r 2 /2σ 2 ) is zero. [16]

Digital Image Processing

Digital Image Fundamentals

CS443: Digital Imaging and Multimedia Binary Image Analysis. Spring 2008 Ahmed Elgammal Dept. of Computer Science Rutgers University

Laboratory of Applied Robotics

CS 559 Computer Graphics Midterm Exam March 22, :30-3:45 pm

Objective 1 : The student will demonstrate an understanding of numbers, operations, and quantitative reasoning.

Mathematics - LV 6 Correlation of the ALEKS course Mathematics MS/LV 6 to the State of Texas Assessments of Academic Readiness (STAAR) for Grade 6

Lecture 7: Most Common Edge Detectors

ECG782: Multidimensional Digital Signal Processing

Outline 7/2/201011/6/

Digital Image Processing. Prof. P. K. Biswas. Department of Electronic & Electrical Communication Engineering

Announcements. Edges. Last Lecture. Gradients: Numerical Derivatives f(x) Edge Detection, Lines. Intro Computer Vision. CSE 152 Lecture 10

Dietrich Paulus Joachim Hornegger. Pattern Recognition of Images and Speech in C++

Object Shape Recognition in Image for Machine Vision Application

Hybrid filters for medical image reconstruction

Light-weight Contour Tracking in Wireless Sensor Networks

Motion. 1 Introduction. 2 Optical Flow. Sohaib A Khan. 2.1 Brightness Constancy Equation

Training-Free, Generic Object Detection Using Locally Adaptive Regression Kernels

EECS150 - Digital Design Lecture 14 FIFO 2 and SIFT. Recap and Outline

Level-set MCMC Curve Sampling and Geometric Conditional Simulation

Finally: Motion and tracking. Motion 4/20/2011. CS 376 Lecture 24 Motion 1. Video. Uses of motion. Motion parallax. Motion field

CS4442/9542b Artificial Intelligence II prof. Olga Veksler

Examination in Image Processing

Multimedia Information Retrieval

CS4442/9542b Artificial Intelligence II prof. Olga Veksler

Image processing. Reading. What is an image? Brian Curless CSE 457 Spring 2017

A New Technique of Extraction of Edge Detection Using Digital Image Processing

Homework 3: Shading, Raytracing, and Image Processing

From Pixels to Blobs

Image Processing. Filtering. Slide 1

What will we learn? Neighborhood processing. Convolution and correlation. Neighborhood processing. Chapter 10 Neighborhood Processing

Chapter 3: Intensity Transformations and Spatial Filtering

Midterm Exam Solutions

Estimating the wavelength composition of scene illumination from image data is an

Connected components - 1

ME132 February 3, 2011

Filtering of impulse noise in digital signals using logical transform

ECG782: Multidimensional Digital Signal Processing

UNIT-2 IMAGE REPRESENTATION IMAGE REPRESENTATION IMAGE SENSORS IMAGE SENSORS- FLEX CIRCUIT ASSEMBLY

Computer Vision I. Announcements. Fourier Tansform. Efficient Implementation. Edge and Corner Detection. CSE252A Lecture 13.

Tracking Under Low-light Conditions Using Background Subtraction

CSAP Achievement Levels Mathematics Grade 7 March, 2006

COMPUTER VISION > OPTICAL FLOW UTRECHT UNIVERSITY RONALD POPPE

Image restoration. Restoration: Enhancement:

The SIFT (Scale Invariant Feature

CS6670: Computer Vision

Why is computer vision difficult?

CS223b Midterm Exam, Computer Vision. Monday February 25th, Winter 2008, Prof. Jana Kosecka

Image Filtering with MapReduce in Pseudo-Distribution Mode

Announcements. Edge Detection. An Isotropic Gaussian. Filters are templates. Assignment 2 on tracking due this Friday Midterm: Tuesday, May 3.

APS Seventh Grade Math District Benchmark Assessment NM Math Standards Alignment

Connecticut Alternate Assessment: Individual Student Report Performance Literals Mathematics

Short Survey on Static Hand Gesture Recognition

Boundary descriptors. Representation REPRESENTATION & DESCRIPTION. Descriptors. Moore boundary tracking

Transcription:

ENG 7854 / 9804 Industrial Machine Vision Midterm Exam March 1, 2010. Instructions: a) The duration of this exam is 50 minutes (10 minutes per question). b) Answer all five questions in the space provided. c) Exam is closed book, closed notes. d) Exam will be graded out of 50 points as indicated; 10 points for Problem 1 10 points for Problem 2 10 points for Problem 3 10 points for Problem 4 10 points for Problem 5 Student Name: Student ID: ENG 7854/9804 Industrial Machine Vision Page 1 of 10

Problem 1: (10 points: 5 + 5) Figure 1 represents several white (i.e., 1 ) binary blobs on a black (i.e., 0) background. a) Perform the first pass of a basic connected component labeling algorithm assuming 8-neighbor connectivity. Label the pixels that make up the white binary blobs beginning with the label 2 and construct the equivalence table in the space provided. b) Manually resolve the equivalence table and indicate the results of the second pass of connected component labeling in Figure 2. Figure 1 ENG 7854/9804 Industrial Machine Vision Page 2 of 10

Equivalence Table: Figure 2 ENG 7854/9804 Industrial Machine Vision Page 3 of 10

Problem 2: (10 points: 4 + 4 + 2) The 5 X 5 image shown below is subject to random ( Salt-and-Pepper ) noise: a) Reduce the effects of the noise by applying a local averaging filter over a 3 X 3 kernel. Specify the kernel used. b) Apply a median filter also over a 3 X 3 kernel. c) Discuss the relative advantages and disadvantages of both techniques. Note: The values in parentheses are included for determining values along the perimeter of the image. (0) (0) (0) (0) (0) (0) (0) (0) 0 135 0 135 0 (0) (0) 0 0 0 0 0 (0) (180) 180 180 180 0 135 (0) (180) 180 180 180 0 0 (0) (180) 180 180 180 0 135 (0) (180) (180) (180) (180) (0) (0) (0) Results of local averaging filter ENG 7854/9804 Industrial Machine Vision Page 4 of 10

Results of Median filter ENG 7854/9804 Industrial Machine Vision Page 5 of 10

Problem 3: (10 points: 5 + 5) a) Write pseudo code that implements a convolution over a 3X3 kernel. Assume an image, f(y,x), with k gray levels and a size of n_rows and m_col (y corresponds to the rows and x corresponds to the columns). Assume the first pixel in the image is given by f(1,1) and make sure the algorithm handles the pixels along the perimeter of the image in an appropriate manner. ENG 7854/9804 Industrial Machine Vision Page 6 of 10

ENG 7854/9804 Industrial Machine Vision Page 7 of 10

Problem 4: (10 points) The crack code shown below represents a white (i.e., 1 ) binary blob on a black (i.e., 0 ) background. Crack Code: 1, 1, 2, 1, 1, 0, 1, 2, 2, 2, 1, 2, 3, 0, 3, 3, 3, 2, 3, 0, 3, 0, 0, 0 a) Sketch the contour of the blob in the space provided. Note that the reference frame (0, 0) is in the upper left hand corner of the image and that the coordinates of the first pixel encountered during contour tracking are given by x=3, y=4 (beginning in the upper left-hand corner). b) Provide the chain code for the same blob. c) Calculate the perimeter of the region using the chain code representation. Based on the complexity, is the blob closer to a circle or a square? Justify your answer. ENG 7854/9804 Industrial Machine Vision Page 8 of 10

Problem 5: (10 points: 2 + 2 + 2 + 2 + 2) a) A histogram equalization algorithm is applied to an 8-bit gray level image of size 640 by 480. Ideally how many pixels will be at a gray level of 128 after the original image is processed? Explain. b) In order to reduce the noise in an image, 16 successive images of the same scene are averaged on a pixel-by-pixel basis. What is the theoretical improvement in signal to noise ratio? In what way is the image quality adversely affected by this form of averaging? Explain. c) Sketch and/or describe an 8 bit gray level image that is composed of only one spatial frequency in the X direction and no spatial frequencies in the Y direction. ENG 7854/9804 Industrial Machine Vision Page 9 of 10

d) An image of a checkerboard pattern is acquired with a camera that has a spatial resolution of 64 by 64 pixels. If the checkerboard pattern consists of 1156 squares (i.e., 34 by 34) do you foresee any issues in the resulting image? From a theoretical point of view, what is the maximum number of checkerboard squares that can be represented in an image that has a spatial resolution of 64 by 64 pixels? e) We wish to determine the best fit circle corresponding to a set of experimental image points. What are the parameters that must be determined and what is the minimum number of data points required for and an overdetermined system of equations (i.e., more equations that unknowns)? How many data points would result in an overdetermined system of equations for fitting an ellipse? ENG 7854/9804 Industrial Machine Vision Page 10 of 10