[Retracted] An Efficient Multispectral Image Classification and Optimization Using Remote Sensing Data
Algorithm 2
Proposed Inception v3 and VGG 16 algorithm.
Source image labeled:
Target image unlabeled:
Class labels,
Assign parameters:
Epoch value =100;
Min-batch size: Bs =32;
Learning rate
1st and 2nd decay rate exponential
With and
Epsilon:
Initially use VGG16 and inception v3 model trained with dataset and pre-training with CNN to initialize feature generation G and classified images alone
Shuffle the random source and target image data set and organize it into
Groups each size
Select mini-batches:
Need to train the by optimizing (1) to (10) using Adam same is required to be done for C1 to C4 (11)
End for
End for
Classify the different class target domain using source and target
