Processing

PImage img;

void setup(){
	size(345, 567);
	// The image file must be in the data  folder of the current sketch
	// to load successfully
	img = loadImage("buzz3.jpg"); // Load the image into the program
}

void draw(){
	// display the image at its actual size at point(0,0)
	image(img, 0, 0);
	// Displays the image at point(0, height/2) at half of its size
	// image(img, 0, height/2, img.width/2, img.height/2);
}

openCV
here you can download
http://opencv.org/releases.html
http://opencv.jp/

in python-openCV

import cv2
img = cv2.imread('input.png')
cv2.imwrite('output.png', img)
img = imread('input.png')
imwrite('output.png', img)

imshow(img)
imageinfo('input.png')
help images

Digital Image

Typically, the functional operation requires discrete values
-sample the two-dimensional(2D) space on a regular grid
-Quantize each sample(rounded to “nearest integer”)
Matrix of integer values(Range:0-255)

image, height map, zoomed in
Pixel count
Image Histogram, intensity bins

color, red channel, green channel, blue channel

Raster image formats store a series of colored dot “pixels”
Number of bits for each pixel represents the depth of color
– 1 bit-per-pixel: 2colors (black or white, binary)
– 4 bit-per-pixel: 16colors
– 8 bit-per-pixel: 256 different colors 2^8 = 256

24 bit per pixel usually means 8 bits per color
at the two highest levels, the pixels themselves can carry up to 16777,216 different colors

Common raster image format
-gif, jpg, ppm, tif, bmp, etc

Exercise to do on own
-openCV/Python
(opencv.org, python.prg)
-matlab/octave
(mathworks.com, www.gnu.org/software/octave)
-Processing
(processing.org)

import cv2

# Read images
img = cv2.imread("mona-lisa.png")
print "Read image from file; size: {}x{}".format(img.shape[1], img.shape[0])
cv2.imshow("Mona Lisa", img)

# and run different operations on them
img_smooth = cv2.GaussianBlur(img, (7, 7), 0) #smooth image using a Gaussian Blur
cv2.imshow("Smoothed", img_smooth)

img_gray = cv2.cvtColor(img_smooth, cv2.COLOR_BGR2GRAY)
cv2.imshow("Grayscale", img_gray)
buzz=imread('buzz3.jpg')
imshow(buzz) 

Pixel

A “picture element” that contains the light intensity at some location (i,j) in the image
i, j
I(i,j) = some numeric value

width * height
= 1280 * (1280 /2)
= 0.8192mp image

Characteristics of a digital image
Original Image, Zoomed In, Values
Piots of values at a slice

A two-dimensional array of pixels and respective intensities
image can be represented as a matrix
intensity value range from 0 = black to 255 = white
two dimension array

8 bits allow to represent 2^8 = 256 different values

Digital image is a Function
x or i, y or j

Digital image

Rays of light -> Picture
Illumination, optics, sensor, processing, display

make an image a “computable” object
A digital image

digital image – pixel and image resolution
discrete (matrix) and continuous (function) representations
grayscale and color images
digital image formats

A digital image(W X H)
column and rows
width, height
square image -> 512 x 512 pixels = 262 = 0.26mp image
mp: mega pixel image

Numeric representation in 2-D (x and y)
Referred to as I(x,y) in continuous function from I(i,j) in discrete
Image resolution expressed in terms of width and height of the image

Extend FP / DP

Compared to FP/DP, CP has better specification and support for
Dynamic range
Vary focus point-by-point
Field of view vs. resolution
Exposure time and frame rate
Bursts

Images in News
Kennedy Assassination
Rodney King Beatings in LA
9/11 images
7/7 London bombing
Virginia Tech
Michael Richards
Russian meteor
Beast with a Billion eyes(literally)

Participatory Data
Handheld, citizen, etc.
Institutional Imagery
Satellite, Airborne, Recon, UAV
etc.
Incidental
security cameras, ATMs, etc

Computer Vision and Computer Graphics
images(2D) -> geometry(3D), Photometry(Appearance)

Utimate Camera
object, light rays, lens, film(retina)

Emerging Field of computational photography
what will a camera look like in 10 years? 20 years?
what novel images can we get? what are their uses?

SLR(film)

-Smart phone camera comes into the market, several billion devices
-Number of photos taken each year

DSLR advantages
– more light
– depth of field
– shutter lag
– control field of view
– better glass
– other (flash, man, modes, …)

phone advantages
– computation
– data
– programmers

Film and digital cameras have roughly the same features and controls
– zoom and focus
– aperture and exposure
– shutter release and advance
– one shutter press = one snapshot

For FP/DP we can use, but cp allows us to change
optics, illumination, sensor, movement
exploit wavelength, speed, depth, polarization, etc.

Probes, actuators, network

Why study comp. photography

1.Pervasiveness of photography
2.Computational photography as it relates to other disciplines
3.Computational photography vs. photography

Traditional Film/Digital Camera processes
Novel Camera
Sensor/Detector, lens
generalized optics(ray bender)

Almost everone has a camera
e.g. smaller, ubiquitous
Significant Improvements in option
Field of applied optics has studied
every aspect of the lens

Camera phones
widest selling electronic platform
Google earth, youtube, flickr
text, speech, music, images, video, 3D…
key element for art, research, products, special-computing …

Panorama

3D Scene -> illumination -> Optics -> Sensor -> Processing -> Display -> User

7 pictures,/ 3012 * 2304 (7imp)
11262 * 2691 /(31mp)/ FOV = 15172 * 2446

Taking pictures
These days, robotic cameras
match and merge together pictures
some overlap can thus be merged to create a panorama

detection and matching
basically some features are matched in two pictures
warping

Fade, blend, or cut

Five steps to make a panorama
1.capture images
2.detection and matching
3.warping
4.blending, fading, cutting
5.cropping

Dual Photography

The concept of “dual photography”
Computational Photography(Rays to Pixels)

Novel Illumination
Photocell capture the light

Projector(controllable light source), modulator(controllable aperture)

Novel Camera
modulator(controllable aperture)

Reflective properties of ray of light
light source, light sensor
Reflection of light depends on the kind of surface: spectacular(mirror), diffuse(matte)

projector pattern – camera image

Enables Imaging

Unbounded Dynamic Range
Variable
-Focus
-Depth of Field
-Resolution
-Lighting
-Reflectance

Supports and enhance the medium of photography

3D scene

(1)Illumination
(2)Optics
(3)Sensor
(4)Processing
(5)Display
(6)User
Computation can be embedded in all aspects of these elements to support photography

Rays to Pixels

Aperture
Generalized optics

Novel Camera
Processing, Sensor, Aperture, Generalized Optics

pixel, display