Skip to main content

Review Edge and Texture Fusion for Plant Leaf Classification

By


 

1. summary

Plants are an important resource in life, which can be developed in the fields of industry, medicine and food. classification plants with automated digital technique, can be done simply by extraction features of the leaves. In this paper, the classification of plants is based forms and texture. Based on the theory of plant taxonomy, plants can be classified based on the shape of the leaves and flowers. However, in general, the classification is based on leaf analysis. Classification based on the shape of the flower is difficult due to the structure of interest is 3D and can only be obtained when the flowers bloom.

Feature extraction used in this paper is the Sobel edge detection and Gabor filters. In edge detection, there are three stages: filtering, enhancement, and detection. Sobel edge detection find the approximate absolute gradient magnitude at each point to detect edges. In Gabor filters have the ability to do the multi-resolution decomposition, and are ideal for segmenting texture in the spatial and spatial-frequency domains. Then the method of feature extraction are tested on classification and regression tree (CART) classifier and the Radial Basis Function (RBF) classifier. Variation of the RBF NN is better in terms of interpolation, by taking input non-linear and linear output. Cost function is usually square error is assumed to be used to evaluate the suitability of parameters. Activation function using a Gaussian function by the equation:
Where X = the input feature vector, L = number of hidden units, uj = average, and Σj = convariance matrix of the Gaussian function.

This paper shows the degree of accuracy RBF classification higher than in CART. The level of accuracy of classification in RBF squared error of 85.93 with 56.2. While the CART rate of accuracy classification of premises 79.26 67.38 squared error. So that RBF has a level squared error is smaller than CART.

2) Excess

  • Gabor filters have the ability to perform multi-resolution decomposition
  • Gabor filters are ideal for texture segmentation because it requires the simultaneous measurement the spatial and spatial-frequency domains.
  • Better in terms of RBF interpolation, taking a non-linear input and provide linear output, thus helping pemetakan complex models.
  • Sobel edge detector find the approximate absolute gradient magnitude at each point to detect edges.

3) Weakness

  • RBF output is limited to the interval (0,1)

4) Suggestions

  • Improving the classification accuracy of the proposed feature reduction technique.


Labels:

Comments

Post a Comment

Popular posts from this blog

Water Found on Mars, Google Changed Logo

By
The discovery of water on the surface of Mars has been confirmed by NASA. With this, speculation that withdrawal of life on Earth have occurred on Mars, it could re-debated. Director of Planetary Science Division of NASA James Green has not been able to confirm it. According to James, NASA will investigate more deeply to find the answer. Topics related to water flowing on Mars and all the lively debate on various social media. Start of Twitter and Facebook. Google did not want to miss. The search engine giant directly change the logo on its search engine page by issuing Doodle titled "Evidence of water found on Mars" to celebrate the discovery of NASA. Doodle created by Nate Swinehart shows an illustration of the red planet Mars. The Mars is cool to suck their drinking water from a glass. He also spinning like a planet that is being rotated. The Mars is exactly in the middle of the word "Google", replaces the letter "o". Purple background color selected...
Post a Comment
Read more

An Supervised Artificial Neural Network Method for Sattelite Image Segmentation

By
Image segmentation   is   an important   step   in image processing   ( image processing).   The main purpose   of   segmentation   is   to   simplify   and   or   to   change the   representation   of   an   image   into a form that   is easier   to   analyze.   Already there are   several   methods of   image segmentation   are   found ,   but   most of   these methods are not   suitable   for   satellite imagery   and   -Method   method   requires   a   knowledge of   the initial   ( a priori   knowledge) .   To   overcome these problems ,   a   satellite   image segmentation   method   is developed   using   an   artificial neural network method   without   learning   ( unsupervised )   called ...
Post a Comment
Read more

How to choose between the Canon 700D or Nikon D5300

By
Canon EOS   700D   and Nikon   D5300   is a   digital SLR camera   aimed at   novice photographers .   Nevertheless ,   there are   features   on   both cameras are   much the same with   a more   sophisticated   camera   /   semi - pro,   so it   is good enough to   be used   in various   scene conditions .   From the shape   and size ,   at a second glance   something like the   camera .   But   when viewed   more closely ,   overall   more   compact   Nikon   D5300   a few   millimeters   and   about 100   grams lighter .   The new lens   Nikon   18-55mm   VR   II also   shorter   in   the off condition .   This can be achieved   because   the plastic   material   of the   camera   thinner . In   ...
Post a Comment
Read more