| |
| |
|
| |
About
Vision IPKS |
|
| |
Vision
IPKS (Image Processing Kernel System) is an extensive library
of about 300 "CO / FORTRAN callable low-level Image Processing
(IP) functions. These routines are useful in building the
state-of.the-art IP applied. cations. Visionlabs shall continue
to add more routines
to the library thus creating a comprehensive IP kernel a bonanza
for IP system designers. |
|
| |
Who would use
Vision IPKS ? |
|
| |
Vision IPKS is a software
toolkit designed for a wide range of users. Students who are
learning IP will find it an excellent tool. Research organizations
planning to embark on IF activity and organizations involved
in town planning, agriculture, medical imaging, material verification,
quality control, remote sensing, geological and geographic survey
etc, will find IPKS an ideal platorm to work with. |
|
| |
What applications
can be developed using Vision IPKS ? |
|
| |
Vision IPKS can be used
as a base to develop a wide range of applications in various
fields. Some of the possible applications are analysis of retinal
imagery, structural and micro structural analysis of materials
(metallic, nonmetallic, inorganic, ceramic, plastics., rocks,
minerals, ores), microscopial and macroscopial, analysis, diagnosis
of cancer and heriditary diseases, analysis of chromosomes,
analysis of DNA fingure prints, etc,. |
|
| |
What
hardware do you need for Vision IPKS ? |
|
| |
Vision IPKS will run
on any PC.286/386/486 based computer or compatible with enough
disk capacity. Since image files occupy large diskspace, a hard
disk with a capacity of 80MB to 200MB hard disk is highly recommended.
A high resolution colour graphics display adapter card (VGA/SVGA)
and colour monitor is required for display of images. Vision
IPKS can also be ported to other workstations such as SUN, SP
ARC. APOLLO and VAX etc,. |
|
| |
What operating systems
does Vision IPKS support? |
|
| |
Vision IPKS runs on DOS and
UNIX platforms. It can also be easily ported to any other operating
systems of your choice. |
|
| |
Why should you use
Vision IPKS ? |
|
| |
After graduating from the
basics, what you really need is flexibility and freedom to experiment
with IP technology.And that is exactly what IPKS is meant for
I You can select various routines available in IPKS and program
it for your specific application. Using Vision ~, you can enhance
your knowledge and skills in IP to enable better decision ~making
regarding large investments in IP systems in future. |
|
| |
How to use Vision
IPKS ? |
|
| |
To use Vision IPKS you should
be familiar with "C" or FORTRAN programming. To develop
a program for any specific application, you simply call IPKS
subroutines available as tiny modules in your "CO or FORTRAN
programs. |
|
| |
What file formats
does Vision IPKS support ? |
|
| |
Vision IPKS is independent
of any image file format. You can write your own routines' for
file handling for any of the existing image file formatS such
as !MG, TIFF, BMP, etc,. |
|
| |
Is Vision IPKS compatible
to any existing IP system? |
|
| |
Vision IPKS contains standard
library routines independent. dent of any customis.ed IP system.
The output can be displayed on any existing display hardware
with suitable routines. Since Vision IPKS handles pure image
rues for input and output, it can easily be interfaced to any
existing IP systems with minor modifications. |
|
| |
|
|
| |
Routines Available
In VISION IPKS |
|
| |
Orthogonal Transform
- Fit – radix
- Walsh – handmard transform
- Haar transform
- Discrete cosine transform
- Slant transform
- Fourier spectrum computation
- Wht spectrum computation
- Convolution (direct & fit)
- Correlation (direct & fit)
- Distribution computation in
- Power spectrum domain
- Different filter
|
|
| |
Image Registration
- Correlation (coarse and fine search)
- SSDA (constant, monotonic threshold, auto or manual threshold
selection, coarse search or fine search)
|
|
| |
Image Geometric
Correction
- Affine transform – bilinear, quadratic,minimum & maximum
value
- Parameter computation – (rotation)angle, reference points)
- General second order transform – bilinear,quadratic, minimum
& maximum value.
|
|
| |
Image Enhancement
and Smoothing
- Histogram equalization
- Histogram hyperbolization
- General histogram transform
- Iterative edge and line weights
- Iterative contrast sensitive weights
- e – filter
- Fast mean, median and mode filters
- Edge preserving smoothing
- Standard hysteresis smoothing
- Symmetric hysteresis smoothing
|
|
| |
Edge and Line detection
- Laplacian operator
- Sobel operator
- Roberts operator
- Perwitt operator
- Kirsch operator
- Heuckel operator
- Robinson operator
- Hough transform
- Feri & chen method
- Fast hueckel operator
- K asavand iterative method
|
|
| |
Restoration
- Inverse filter
- Wiener filter
- Constrained least-square method
- psedo - inverse filter
|
|
| |
Application of
Relaxation Labelling
- Line and curve enhancement
- Edge enhancement
- Noise removal
|
|
| |
Texture Analysis
- Co-ocurance matrix
- Difference statistics
- Local exterma
- Texture edge detection
- Texture edge preserving smoothing
- Auto – regressive model
- Auto – correclation
- Fourier features
|
|
| |
Region Segmentation
- k-s test
- Heuristic method
- Iterative merging
- Iterative thresholding
- Split & merge method
|
|
| |
Image Classification
- Maximum likelihood
- Parallelopiped
- Minimum distance
- Clustering by iso-data algorithm
- k – I transform
|
|
| |
Geometric Feature
Measurement
- Shape features – centroid, circumscribed quadrilateral,
area, perimeter, elongatedness,moments
- Thinking – hilditech, stefanelli & rosenfield, pattern
adaptive thinning
- Boundary detection
- Boundary description – chaincode, slope, curvature, fourier
descripter
- Expansion, contraction, shrinking & projection
|
|
| |
General Utilities
- Fundamental statistics
- Histogram computation
- Thresholding
- Linear filters
- Gray scale translation
- Requantization
- Arithmetic operations
- Transposition
|
|
| |
FFT
of input with Filters |
|
| |
|
|
| |
Hiss Pass Filter
|
Low Pass Filter |
|
|
|
|
| |
Edge Detection
|
|
| |
Edge Detection |
|
| |
|
|
| |
Image Enhancement |
|
| |
 |
 |
Input Image |
Enhanced Image |
|
|
| |
|
|
