Loading presentation...

Present Remotely

Send the link below via email or IM

Copy

Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.

DeleteCancel

Make your likes visible on Facebook?

Connect your Facebook account to Prezi and let your likes appear on your timeline.
You can change this under Settings & Account at any time.

No, thanks

Butterfly Identification

Final project presentation
by

Mariyana Koleva

on 26 June 2013

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of Butterfly Identification

The goal of the project is to develop a
butterfly classifier
which can be incorporated in a butterfly identification phone app or used on its own.

Main focus:
Develop an automatic classifier which identifies correctly the name of a butterfly species based on a segmented image of the butterfly taken in an uncontrolled environment.

Main challenges:
Overview
Automated Identification of Butterfly Images
Project Presentation
Contents:
Project Overview
Methodology
Evaluation
Results and Conclusion

Project Description
The most significant parts of the system are the feature selection and representation and the classifiers.

The bag-of-words method was used for feature representation and selection. The features investigated are :
colour
shape/pattern
combinations of those

The classifiers compared are:
k-Nearest-Neighbour
One-against-all Support Vector Machine
BAG-OF-WORDS APPROACH
Colour feature
Shape/Pattern feature
k-Nearest-Neighbour -
4 steps required to classify an image:
1. Feature extraction and representation
2. Codebook construction
3. Bag-of-words representation of an image
4. Training of the classifier
Similar approach is employed as with the colour feature. However, the visual words and their descriptors are different.
More sophisticated classifiers,
used to achieve the optimal results.
Features
Methodology
Idea: represent images using a fixed set of visual words (vocabulary, codebook); spatially invariant representation.
Feature extraction and representation
Codebook construction
Image representation
A codebook, or a vocabulary, is constructed by clustering all pixels from the dataset into a fixed number of clusters.
R-G-B values
{10,70,150}
Codebook – set of cluster centroids (visual words)
Feature vector
Histogram
Each pixel in an image is then represented as its closest centroid (visual word) . The whole image is represented as a frequency histogram of the number of codewords occurring in it normalized over #pixels in the image, i.e. an m-dimensional feature vector, where m is the size of the vocabulary.
0.4
0.003
0

0.021
Originally images are matrices of pixels, with each pixel laying in a 3D space - either the RGB (Red-Green-Blue) or HSV (Hue-Saturation-Value) colour spaces.
R-G-B values
{150,78,43}
As a result, images are represented as vectors using their colour feature and the bag-of-words approach.

Colour-based classification: unreliable, sensitive to illumination changes, not discriminative enough.
Bag-of-words
Classifiers
Pixels, described with RGB values
Circular patches, represented with SIFT descriptors
Shape codewords
Colour codewords
. . .
. . .
Feature vector
Concatenated Histogram
0.4
0.003
0

0.021
Combined Features
colourSIFT - Extract descriptors from each channel and concatenate them together
histogram concatenation - concatenate colour and shape histograms
simpler classifier, no training phase, # neighbours optimized with cross-validation.
multi-class SVM -
Occlusion
Shape deformations
Intra-class variance
Side view
Illumination, low quality images, difficult to gather data etc.
Conclusion and application
Features
Evaluation
Classifiers
Datasets
Other work
Results are averaged over multiple runs of 10-fold cross validation.
colour

shape

combined
For each feature - size of vocabulary and other classification parameters are tuned in. Overall and per-class performance are observed.
Colour vs shape feature
The shape/pattern feature (above)
outperforms the colour feature (below) on 9/10 classes.
Combined Features
colourSIFT improves accuracy of SVM & KNN;
concatenated histograms - only SVM
Best method: concatenated histograms of colour+colourSIFT
KNN - k and distance measure tuned-in
KNN>SVM with colour feature
13-class dataset (1012 images)
Results on the extended dataset
Although the performance generally deteriorates, combining features makes the accuracy more stable.
Main contributions:
data
feature engineering
classification

Possible development directions:
Conservation
Collection of scientific data
Education

Questions?
Mariyana Koleva
SVM>KNN with shape/combined
SVM error<<KNN's error
Several projects on Butterfly, Bee or Moth identification achieved comparable or better results, but in a much more controlled environment - supervision, specialist equipment, limited classes, family level classification.
The most notable project is a Flower Identification system:
10 species (830 images)
Full transcript