aula_01 Flashcards
First photograph:
First permanent photograph of a camera image was made in 1825 by Joseph
Nicéphore Niépce using a sliding wooden box camera
Pinhole Camera principle
Light enters a dark box through a small hole and creates an inverted image
on the wall opposite the hole
How is an image formed through a lens
All the rays of light that came from an object in direction
to the lens converge, on the other side, in another point
at a certain distance from the lens. This distance is
called focal distance.
All the points that verify this fact are denoted the focal
plane.
Basic camera geometry
- Far objects appear smaller.
- Lines project to lines.
- Lines in 3D project to lines in 2D.
- Distances and angles are not preserved.
Digital Cameras: several alternatives
- Several interfaces (Firewire, GigE, CameraLink,
USB, . . . ). - Scientific usage (high resolution, long exposure
time, . . . ). - High speed (ex. 1000 fps).
- Linear (ex. 10000 lines per second).
- 3D
- Infrared (ex. 8 to 14 µm).
- High dynamic range (ex. using a prism and two
sensors). - Multispectral
Image as a function
-A digital image is represented by a rectangular
matrix of scalars or vectors.
Each one represented by f(i,j) which are named pixels
- An image can be represented by a two-dimensional
function, f(x; y), where x and y are spatial coordinates.
- Meaning of f in a given point in space, (x; y), depends on
the source (visible light, x-rays, ultrasound, radar, . . . ).
- Spatial coordinates and the function values are
continuous quantities.
-To convert f(x; y) into a digital image, it is necessary to
perform spatial sampling and amplitude quantization.
Digitalization: Sampling and quantization
- Continuous light distribution is spatially sampled, then there is a
still image created by time sampling that
discrete distribution - Resulting values are quantized to a finite set of
numerical values - Sampling process means digitalizing the coordinate values
*Quantization means digitalizing the amplitude values
Types of digital images
*black and white(binary) - (0,1)
*grayscale image
*color-indexed images
*color images - RGB
why are used color-indexed images?
Color indexed images (palettized imaged) are images that use a limited number of colors to represent the full range of colors in an image. These images are often used to reduce the amount of data required to represent an image (usuallu use 1/3 of original size)
They have different advantages:
-Size: Color indexed images are typically smaller in size than full-color images, because they use a limited number of colors to represent the image.
-Processing speed: Color indexed images can be processed faster than full-color images, because there are fewer colors to deal with.
-Compatibility: Color indexed images are often used in applications that require compatibility with older systems or software that do not support full-color images.
-Efficient use of color: In some cases, an image may not require the full range of colors that a full-color image can represent.
Number of pixels and range of grayscale digital images
Grayscale (Intensity images):
1 bit/pix : range [0,1] binary image
8 bit/pix : range [0,255] Universal photo
12 bit/pix : range [0,4095] high quality
14 bit/pix : range [0,16383] professional quality
16 bit/pix : range [0,65535] highest quality
Number of pixels and range of color digital images
3 channel 24 bit/pix : range [0,255]^3 RGB universal
3 channel 36 bit/pix : range [0,4095]^3 RGB high quality
3 channel 42 bit/pix : range [0,16383]^3 RGB professional
4 channel 32 bit/pix : range [0,255]^4 CMYK
Other types of images?
Infra red (Gobi camera)
Depth image (Kinect)
Point Cloud ( swissranger)
Digital Videos, what caractherizes them?
- A video signal can be represented by a 3D-function
v(x; y; t), where x and y are spatial coordinates and t denotes time. - The process of converting analog video into digital video
requires spatial and temporal sampling, besides
amplitude quantization. - A digital video is a temporal sequence of digital images
which we represent by v(i; j; k), with k=t/T - T indicates the time period between two consecutive
images (frames). Therefore, the frame rate 1/T (Hz) is
inversely proportional to T.
How are objects perceived? (Color models)
Objects are perceived as having a color depending on
the spectrum of the reflected light (or emitted)
Different spectra induces similar color sensations?
It may not.
Then It is important to be able to describe color objectively
