Sie sind auf Seite 1von 7

Edge Detection of Textured Image

ABSTRACT been carried out and a comparison of


Edge detection is one of the most existing techniques is shown.
important steps leading to the
analysis of processed image data. Keywords
Image texture segmentation is an
important problem and occurs Texture, Edge Detection, wavelet
frequently in many image-processing Pyramid, Histogram Filter, Pixel
applications. The image can be Difference.
segmented by detecting there
boundaries. This paper presents 1. INTRODUCTION
some algorithms that are required for
Edge detection is the process of
Edge detection of textured image.
detecting the edges of the image.
Algorithms for Edge detection are
This task becomes particularly
presented that works directly for
difficult in the case of textured
non-textured image. These algorithm
images. Texture is a term that refers
can be carried out for edge detection
to properties that represent the
of textured images also by
surface of an object. We might
preprocessing the image through
define texture as something
discrete Filters. Experimental results
consisting of mutually related
on images containing various
elements therefore we consider a
synthetic and natural textures have
group of pixels [1]. Image
segmentation is the process of
partitioning an image into
homogenous regions. The existing
segmentation methods are commonly
classified according to the texture
description. The algorithm for
texture segmentation deals with first
extracting textural features of image
by Gabor transform or Discrete
Wavelet Transform. The statistical
Features can be derived from these
images. In case of Wavelet transform
Wavelet Pyramid or Wavelet packet
Filters are used. In case of proposed
method the image is first
smoothened and passed through
Wavelet pyramid filters, the image is
than passed through Histogram filter.
Edge detection algorithms are then
applied on these images so as to get
the boundary of different textures.
For non-textured image boundary
can be detected by smoothing and
then directly applying edge detection
algorithm [2].
2. Basic Concepts
2.1 Statistical analysis
• Co-occurrence matrix: This for output statistics computation as
method of texture description is features.
based on the repeated occurrence
of some Grey level configuration
in the texture; this configuration 2.2.1 Gabor Filter
varies rapidly with distance Using a Gabor filter bank, an image
slowly in coarse textures. An can be decomposed into orientational
occurrence of some grey level components lying in a specified
configuration may be described frequency range. This
by a matrix of relative decomposition simplifies higher
frequencies Pφ ,d (a, b) describing level image processing like
how frequently two pixels with extraction of contours or pattern
the grey level a, b appear in the recognition.Here, the input textural
window separated by a distance d image is filtered through a filter
in the direction φ . [1] bank having Gabor Filters and
Gaussian smoothing filters in
cascade. The feature vector is
2.2 Textural Feature Extraction constructed from the output statistics
The filtering approaches to texture of the images obtained from each
classification generally compute the branch of the filter bank. Figure 2
[3] illustrates the block diagram for
the procedure of textural feature
Figure:1 Textural Feature
vector construction [3].
Extraction

filter output statistics as features.


Figure 1 describes a typical scheme
of textural feature extraction. The Gabor Wavelet Transform
textured image is filtered through a
bank of filters tuned to different
frequencies. The filtered image
undergoes a nonlinear transform
followed by a smoothing operation
one quarter of the original size,
which makes up the original image.

• Wavelet Pyramid
Decomposition:
The pyramid wavelet transform
recursively decomposes sub-signals
Figure: 2 Gabor filter Bank. in the low frequency channel. This
has the effect of concentrating the
energy of the image towards the low
2.2.2 Discrete Wavelet Transform end of the frequency spectrum,
Wavelet transform is capable of emitting the high frequency
providing the time and frequency information and approaching an
information simultaneously, hence approximation of the image [5].
giving a time-frequency
representation of the signal. A
multi-resolution approach is • Wavelet Packet (Tree)
suggested to give a robust Decomposition.
segmentation process. This is when Pyramid wavelet transform may not
an image is decomposed and be suitable for quasi-periodic
represented at different scales. The signals, whose dominant frequency
discrete wavelet transform is channels are located in the middle
implemented with a 2-channel frequency region [5].
analysis filter bank [5]shown in
Figure 3. First the low pass filter
(intensity) and high pass filter
(texture) are applied to the rows of
the image. After this stage the
columns are down sampled by a
factor of 2 (the odd numbered
columns are discarded). After this
stage the same technique is applied
to the 2 resultant images, however Figure 3: DWT 2 channel Analysis
this time the filters are applied to the Filter bank
columns of the image and the rows
are then down sampled. The
resultant is 4 frequency bands, each 3. Outline Of Propose Method
1) From difference of
The algorithm for textured and non- Pixel
textured images is given. for non- 2) Read the input Pixel.
textured images only smoothing and 3) Read another input
edge detecting algorithm is required, pixel.
while for detecting edge of textured
image the image need to be 4) Take the difference of
preprocessed by passing it through the RGB values of two
wavelet pyramid filter as shown in pixels.
Figure 4. 5) Display the difference
of Pixels.

Input wavelet Histogram Edge Output


Image Low pass pyramid filter Detection Image

Fig:4 Textured Image Edge


detection

• Algorithm For Textured Image


Edge Detection
• From Image Difference.
1) Preprocess the image
by passing through Low Pass 1) Read the image.
filter. 2) Smooth the image by
2) Pass the image applying smoothing Filter.
through Wavelet Pyramid 3) Take the difference of
Filter. these images.
3) Segment the image by
passing through Histogram • Histogram filter Algorithm
filter.
1) Read image pixels.
4) Detect the image by
2) Convert RGB into
Edge Detection algorithm.
HSV Hue saturation value.
3) Find intensity part of
• Edge Detection algorithm HSB.
4) Create Histogram for famous Lena image, the output has
255 binary level. taken less calculations. Fig 6 (c)
5) if histogram value lies shows output from Typical textured
between 10 and 50 make it image used image processing,, the
constant less devide it by 4. result of detecting edges of this from
algorithm mentioned for textured
6) display output image image edge detection is also found
good
• Wavelet Pyramid [5]
1) Read Image.
2) Filter the image
through Low pass and high
pass Filter.
3) Downsample output
of (b) by 2
4) Repeat step (a) & (b) (a) original image (b)Edge
such that low passed image is Detected
passed as input of (b). image
5) Above step is Figure 5: Edge Detection
repeated till a certain level of
pyramid.
(a)
4. RESULTS
The method has been tested on a
large number of various images
including synthetic and natural
textures. The method of finding edge
requires less time as it is simply the (b)
difference, rather than applying filter
mask like as used in laplacian and
(c)
sobel operator. Although, this
method showed good results it
appeared that is was not robust (this Figure 6: Textured image Edge
is due to the random nature of detection
genetic algorithm). Fig. 5 shows an (a) original image
example of Edge Detection of
[2] Alberto Martin and Sabri
Tosunoglu, "Image Processing
Technique For Machine Vision",
IEEE Transactions on Pattern
Analysis and Machine Intelligence,
Vol. 5, February 2000.
[3] Dennis Dunn William E. Higgins
(b) after histogram filter and Joseph Wakely, “Texture
(c) after edge detection algorithm Segmentation using 2-D Gabor
Elementary Functions”, IEEE
Transactions on Pattern Analysis and
5. CONCLUSION
Machine Intelligence, Vol. 16,
This paper has presented a novel February 1994.
algorithm for Edge detection of
[4] Milan Sonka, Vaclav Hlavac and
textured images. The main
Roger Boyle, “Image Processing,
contribution is on applying a
Analysis And Machine Vision”,
combination of Wavelet pyramid,
Chapman & Hall 1995.
histogram filter and Edge Detection
Filter. The method presented has [5] S.C. Liew, H. Lim, L.K. Kwoh,
shown good behavior for both and G.K. Tay, “Texture Analysis Of
textured and non-textured images. SAR Images”, IGARSS '95,
The algorithm for Edge Detection International Volume 2 1995.
requires less computation than edge
detection through masking since the
computation required for masking
requires convolution. Algorithm for
textured image may fail if the
textures of images have very low
intensity difference. A combination
of Gabor Filter along with
Histogram filter also can be used to
segment multi-textured image.
6.REFERENCES
[1] Anil K. Jain, “Fundamentals of
Digital Image Processing”, Prentice
Hall, 1989

Das könnte Ihnen auch gefallen