4.3 Classification Accuracy Assessment
The measurement of accuracy level from landuse classification result is using a reference where the data truth is 100 assured. Other kind of reference
that is commonly used, i.e. first, by aggregating reference data directly from the field; second, by using aerial photograph or other image that has same resolution
with aerial photograph. This research occupied vector data acquired from Ikonos Multispectral data
2001 with 4 bands red, green, blue and near IR. Acquisition date of Ikonos image as reference is the same with acquisition date of Landsat7 ETM + as
classified data. “Accuracy assessment reference data should be collected as close as possible to the date of the collection of the remotely sensed data used to make
the map “Congalton, 1987. In order to minimize the error in creating vector on Ikonos data, then the geometry process of Ikonos will use the same Ground
Control Point with Landsat Image. There are five common sampling schemes that have been applied for
collecting reference data Congalton, 1987: 1 Simple random sampling
2 Systematic sampling 3 Stratified random sampling
4 Cluster sampling 5 Stratified systematic unaligned sampling
59
In this study, stratified random sampling was used for collecting reference data. Stratified Random Sampling is similar to simple random sampling;
however, some prior knowledge about the study area is used to divide the area into groups or strata, and then each of strata of is randomly sampled. In
Stratified random sample each sample units in the study area has an equal chance of being selected. In the case of accuracy assessment, the map has been
stratified into land cover or vegetation types i. e., land cover types, no matter how small, will be included in the sample. In Stratified also pick random x,y
coordinates to go and collect samples. The main advantage of Stratified Random Sampling is the good statistical properties that result from the random selection
of samples. The aggregation of reference data in form of vector can be seen in Figure 4.14.
Tea Garden Settlement
a
60
Paddy Field Grass
b
Forest Farm
c
61
Bush Water Body
d Figure 4.14. Vector Data as Reference
From the vector result as figured above, then the result of classification obtained from likelihood and back propagation neural network method will
be cut by using that vector data. The result is shown in this Figure 4.15
62
Tea Garden Settlement
Paddy Field
Grass Forest Farm
Bush Water
Figure 4.15. Vector Cutting of Classification Result Using Back Propagation Neural Network
63
Tea Garden Settlement Paddy
Field
Grass Forest Farm
Bush Water
Figure 4.16. Vector Cutting of Classification Result Using Maximum Likelihood
64
65
The cutting from vector data above used to calculate accuracy level from classification of both methods back propagation neural network and maximum
likelihood. On the calculation method of confusion matrix, there are 3 types of accuracy that can be quantified, i.e. overall accuracy, producer’s accuracy and
user’s accuracy. Producer’s accuracy is calculated by dividing the number of correctly classified samples by the column total or the class in the confusion
matrix. User’s accuracy is calculated by dividing the number of correctly classified samples by the row total for the category in the confusion matrix.
Overall calculation can be seen in Table 4.6 and 4.7 In Table 4.5 below, it can be observed that producer’s accuracy and user’s
accuracy of both methods. Table 4.5. The Producer’s Accuracy and User’s Accuracy for Back Propagation
and Maximum Likelihood Classification Methods
Back Propagation Maximum Likelihood
Class Prod. Acc
User. Acc Prod. Acc
User. Acc
Tea Garden 61.6
87.9 76.9
70 Settlement 95.9
97.4 94.5 95