FORESTRY CLASSIFICATION OF BEIJING USING SUPERVISED &
UNSUPERVISED CLASSIFICATIONSubject: Urban Remote SensingName:
Muhammed Usman FaridStudent Number: 201429180003Dated:
2015/06/23INTRODUCTIONFor the completion of the course, I have
selected the site Beijing and its forestry land use classification.
As from the first glace of visual ground trooping we can say that,
Beijing is a green city. Although for the representation of the
forestry in the area, I have used the ENVI 4.7 Software. I have
used the TM band Width 2 for to represent the site of forestry. I
used the Supervised and Unsupervised classification of site by
using ENVI 4.7 software.SITE AREABeijing,
formerlyRomanizedasPeking. Beijing is the second largest Chinese
city byurban populationafterShanghaiand is the
nation'spolitical,cultural, and educational center. The Ariel map
of the city is as under;
Figure 1: Map of Beijing
SUPERVISED CLASSIFICATION OF BEIJINGDevelopment of a
classification scheme by selecting representative areas using
reference sources such as higher resolution. Determine the number
of cluster centers and initialize the cluster centers. Maximum
likelihood supervised classification. MLC is performed according to
the following steps. Display the three-band overlay composite
image. Using box-cursor to choose representative training samples
for each of the desired classes from the color composite image.
Color-encode and show the classified image. Estimate the number of
pixels and area for each class and show the statistics for each
class.Supervised Classification Maximum Likelihood Map
Figure 2: MaxlikelihoodUNSUPERVISED CLASSIFICATION OF
BEIJINGAnother broad of classification is unsupervised
classification. It doesnt require human to have the foreknowledge
of the classes, and mainly using some clustering algorithm to
classify an image data. The computer uses techniques to determine
which pixels are related and groups them into classes.The two most
frequently used algorithms are the K-mean and the ISODATA
clustering algorithm.In general, both of them assign first an
arbitrary initial cluster vector.The ISODATA algorithm is similar
to the k-means algorithm with the distinct difference that the
ISODATA algorithm allows for different number of clusters while the
k-means assumes that the number of clusters is known a
priority.UNSUPERVISED CLASSIFICATION OF BEIJING USING ISO-DATA AND
K-MEANS
Figure 2: ISO-DATA
Figure 3: K-MeansRESULTSThe selection of suitable band
combination is essential for SPOT 3 image for visual
interpretation. The images of Beijing Forest Reserve and its
surrounding area appear much better after performed enhancement
technique. In this study, the band combination of 4-3-2
(Red-Green-Blue) was selected as the best combination and later
used for digital classification. Meanwhile Forest appear more
expanded compared to the unsupervised technique. Maximum Likelihood
Classifier (MLC) and the original mechanism of ISODATA classifier
in the unsupervised technique. The vegetation mapping accuracy in
both techniques referred to producers and users
accuracies.CONCLUSIONThe study on evaluation of accuracy from the
supervised and unsupervised techniques had produced a baseline for
the Beijing Forestry. Accuracy assessment showed that the overall
accuracies were less than 80% and 71% for unsupervised and
supervised classifications, respectively. The advantage of
classification is obvious. We can get the physical meaningful
reflectance and their multivariate spreads. We can know the
estimate the area coverage by forestry.
ADVANTAGES AND DISADVANTAGES OF SUPERVISED AND UNSUPERVISED
CLASSIFICATIONSupervised classification Disadvantages The
disadvantage of this approach is its computational cost, since
performing wrapper FSS is slow.Unsupervised Classification
Disadvantages As the analyst has little control over the groupings
determined in unsupervised classification, assigning those
groupings to reset classes can be more difficult and complicated.
It should be noted, however, that in other situations where the
final set of land cover classes is more open to adjustment this
disadvantage may not be an issue in the classification. A fiction
than the unsupervised method for similar reasons. The inherent
disadvantages of the unsupervised method are advantages of the
supervised method, and vice versa.ADVANTAGES AND DISADVANTAGES OF
SUPERVISED AND UNSUPERVISED CLASSIFICATIONSupervised classification
Advantages This image give us a visible picture by comparing both
ground trooping and computer generation.Unsupervised Classification
Advantages It might give us the image and classification of the
site without inputting any ground trooping data, It will show the
band representation with the three coordinator red, green and blue.
It will distinguish the cell pixel density.