08-16-2017, 10:29 PM
Digitization includes sampling of image and quantization of sampled values. After converting the image intobinary, processing is performed which maybe:
-Image enhancement
- Image reconstruction
- Image compression
Image compression deals with minimizing the no of bits required to represent an image.
An image in the "real world" is considered to be a function of two real variables, for example, F(x,y) with F as the amplitude (e.g. brightness) of the image at the position (x,y). An image contains regions-of-interests to which specific image processing operations are performed. For digitization, the given Image is sampled on a discrete grid and each sample or pixel is quantized using a finite number of bits. images be available in digitized form, that is, arrays of finite length binary words for image processing. The digitized image is processed by a computer and to display it, it is converted to analog form.
Video compression
There are loseless and lossy compression standards.
Lossless Compression: Here compressed video is numerically identical to the original content on a pixel-by-pixel basis and Motion Compensation is not used
Bit Plane Coding:A video frame with NxN pixels is taken and each pixel is encoded by K bits. Then Converts this frame into K x (NxN) binary frames and each binary frame is encoded independently.
Predictive Coding: The difference between the actual pixel value, and its most likely prediction is called the differential or the error signal and its value is entropy encoded.
Applications of Lossless Compression are in Medical Imaging, Contribution video applications etc
Lossy Compression:
Transform Coding (MPEG-X)
Its desirable characters are Content decorrelation: packing the most amount of energy in the fewest number of coefficients, Fast implementation, Content-Independent basis functions
used transformations are:
-Karhunen-Loeve Transform (KLT)
-Discrete Fourier Transform (DFT/FFT)
-Discrete Cosine Transform (DCT)
-Walsh-Hadamard Transform (WHT)
Vector Quantization (VQ)
Based on Shannon s Theory that Purely digital signals could be compressed by assigning shorter code-words to more probable signals and that the maximum achievable compression could be determined
from a statistical description of the signal and that Coding vectors or groups of symbols (speech samples or pixels), rather than individual symbols or samples.
Subband Coding (Wavelets)
Each frame is filtered to create a set of smaller frames (subbands) and then Each band is encoded separately (different bit rates, encoder, etc )
Fractals
This is based on mandelbrot set and concept of Self similarity.
Model-Based Coding
models Designed for different objects in natural scenes encoder and the decoder are given copies of the models.
Detailed seminar report download:
[attachment=707]