Basic Principles Photogrammetry V: Image Convolution & Moving Window:

Slides:



Advertisements
Similar presentations
3-D Computer Vision CSc83020 / Ioannis Stamos  Revisit filtering (Gaussian and Median)  Introduction to edge detection 3-D Computater Vision CSc
Advertisements

Linear Filtering – Part I Selim Aksoy Department of Computer Engineering Bilkent University
Spatial Filtering (Chapter 3)
Sliding Window Filters and Edge Detection Longin Jan Latecki Computer Graphics and Image Processing CIS 601 – Fall 2004.
1Ellen L. Walker ImageJ Java image processing tool from NIH Reads / writes a large variety of images Many image processing operations.
Lecture 2: Filtering CS4670/5670: Computer Vision Kavita Bala.
Image Filtering CS485/685 Computer Vision Prof. George Bebis.
7. Neighbourhood operations A single pixel considered in isolation conveys information on the intensity and colour at a single location in an image, but.
CS 376b Introduction to Computer Vision 02 / 27 / 2008 Instructor: Michael Eckmann.
Lecture 1: Images and image filtering
1 Image Filtering Readings: Ch 5: 5.4, 5.5, 5.6,5.7.3, 5.8 (This lecture does not follow the book.) Images by Pawan SinhaPawan Sinha formal terminology.
CS443: Digital Imaging and Multimedia Filters Spring 2008 Ahmed Elgammal Dept. of Computer Science Rutgers University Spring 2008 Ahmed Elgammal Dept.
1 Image filtering Images by Pawan SinhaPawan Sinha.
1 Image filtering
1 Images and Transformations Images by Pawan SinhaPawan Sinha.
Image Analysis Preprocessing Arithmetic and Logic Operations Spatial Filters Image Quantization.
CSCE 441: Computer Graphics Image Filtering Jinxiang Chai.
Spatial Filtering: Basics
Basis beeldverwerking (8D040) dr. Andrea Fuster Prof.dr. Bart ter Haar Romeny dr. Anna Vilanova Prof.dr.ir. Marcel Breeuwer Convolution.
Digital Image Processing Lecture12: Basics of Spatial Filtering.
Multimedia Data Introduction to Image Processing Dr Sandra I. Woolley Electronic, Electrical.
Linear Filtering – Part I Selim Aksoy Department of Computer Engineering Bilkent University
Lecture 03 Area Based Image Processing Lecture 03 Area Based Image Processing Mata kuliah: T Computer Vision Tahun: 2010.
Convolution and Filtering
Chapter 5: Neighborhood Processing
CSC508 Convolution Operators. CSC508 Convolution Arguably the most fundamental operation of computer vision It’s a neighborhood operator –Similar to the.
October 7, 2014Computer Vision Lecture 9: Edge Detection II 1 Laplacian Filters Idea: Smooth the image, Smooth the image, compute the second derivative.
Digital Image Processing Lecture 16: Segmentation: Detection of Discontinuities Prof. Charlene Tsai.
CS654: Digital Image Analysis
Sejong Univ. CH3. Area Processes Convolutions Blurring Sharpening Averaging vs. Median Filtering.
Machine Vision Edge Detection Techniques ENT 273 Lecture 6 Hema C.R.
CSE 185 Introduction to Computer Vision Image Filtering: Spatial Domain.
Sliding Window Filters Longin Jan Latecki October 9, 2002.
Lecture 1: Images and image filtering CS4670/5670: Intro to Computer Vision Noah Snavely Hybrid Images, Oliva et al.,
Filters– Chapter 6. Filter Difference between a Filter and a Point Operation is that a Filter utilizes a neighborhood of pixels from the input image to.
Leaves Recognition By Zakir Mohammed Indiana State University Computer Science.
Spatial Filtering (Chapter 3) CS474/674 - Prof. Bebis.
Digital Image Processing, Assoc. Prof. Dr. Setyawan Widyarto 1-1 Convolution.
Miguel Tavares Coimbra
Spatial Image Enhancement
CS262: Computer Vision Lect 06: Face Detection
Photogrammetry V: Dr. Gamal Hassan Seedahmed Dept. of Surveying Eng.
Basis beeldverwerking (8D040) dr. Andrea Fuster dr. Anna Vilanova Prof
Digital Image Processing Lecture 16: Segmentation: Detection of Discontinuities Prof. Charlene Tsai.
Image Pre-Processing in the Spatial and Frequent Domain
A Gentle Introduction to Bilateral Filtering and its Applications
Filtering – Part I Gokberk Cinbis Department of Computer Engineering
CS-565 Computer Vision Nazar Khan Lecture 4.
Convolutional Networks
Other Algorithms Follow Up
Histogram Histogram is a graph that shows frequency of anything. Histograms usually have bars that represent frequency of occuring of data. Histogram has.
Computer Vision Lecture 9: Edge Detection II
Image filtering Images by Pawan Sinha.
Image Processing, Lecture #8
Digital Image Processing
Image filtering Images by Pawan Sinha.
Image Processing, Lecture #8
Spatial operations and transformations
Lecture: Image manipulations (2) Filters and kernels
Digital Image Processing Week IV
Non-local Means Filtering
Edge Detection in Computer Vision
CSC 578 Neural Networks and Deep Learning
Image Filtering Readings: Ch 5: 5. 4, 5. 5, 5. 6, , 5
Lecture 7 Spatial filtering.
All about convolution.
Recursive Implementation of Anisotropic Filtering
Fundamentals of Spatial Filtering
Spatial operations and transformations
FREQUENTLY USED 3x3 CONVOLUTION KERNELS
Presentation transcript:

Basic Principles Photogrammetry V: Image Convolution & Moving Window: Dr. Gamal Hassan Seedahmed Dept. of Surveying Eng. Faculty of Eng. University of Khartoum Gamal.Seedahmed@gmail.com

Background Convolution is the process of combining the input signal with the effect of the response function (Filter). Input signal here refer to a digital image in terms of pixel values.

Convolution in Digital Images Input image is the signal Response function is a small array called a kernel or a filter Process is termed a Moving Window Product of kernel and image

Kernels Generally, sum of kernel elements equals one or zero Common size is 3x3, but 5x5, 7x7, etc. are also used. (very rare even dimensions)

The 3 Elements of Image Convolution You need an input image. You need a convolution kernel. Then you end up with an output image, which is the result of convolving the kernel with image.

The 3 Elements of Image Convolution (cont’d) Convolution kernel (K) Input image (I) Task: convolve the kernel to the left with the image to the right

The 3 Elements of Image Convolution (cont’d) The practical implementation of convolution requires the centering of the convolution kernel (K) on each image location and compute the sum of the multiplication of the kernel with local image neighborhood. Convolution kernel Centered at image Location (1,1). Center of the kernel

The 3 Elements of Image Convolution (cont’d) 1 2 3 4 5 6 7 8 9 Convolution kernel Centered at image Location (1,1). Center of the kernel Centering the kernel at the boundary of the image creates a problem. In the example above five of the elements of kernel are sitting outside the image {1,2,3,4,7}.

The 3 Elements of Image Convolution (cont’d) 1 2 3 4 5 6 7 8 9 One of the techniques of dealing with the boundary problem is to ignore the boundary. Then the starting image location for the Convolution is (2,2) (indicated above by a blue square).

The 3 Elements of Image Convolution: How to compute the convolution of an image with a kernel Input image Output image Now let us compute the result of the convolution at output image on Location (2,2):

The 3 Elements of Image Convolution: How to compute the convolution of an image with a kernel Output image Observe that we put zeros around the boundary of the image. This is known as padding by zeros to avoid the boundary problem.

The 3 Elements of Image Convolution: How to compute the convolution of an image with a kernel Input image Output image Now let us compute the result of the convolution at output image on Location (2,3):

The 3 Elements of Image Convolution: How to compute the convolution of an image with a kernel Input image Output image Now let us compute the result of the convolution at output image on Location (2,4):

The 3 Elements of Image Convolution: How to compute the convolution of an image with a kernel Input image Output image Then we need to repeat the process for the rest of the image. The practical implementation of convolution is very simple..

Average Kernel (Mean Filter) The kernel above is known as the average kernel of the Mean Filter. This is a 3 x 3 mean filter.

Gaussian Kernel Please very this. Question: Where is the number 16 come from? It came from the summation of the kernel elements. Please very this.

Vertical Edge Filter

Horizontal Edge Filter