Classification techniques used in remote sensing satellite imageries: A survey
Pabitha Muthu, Dr. S Siva Ranjani
Satellite image classification method involves grouping the image component values into meaty classes. Several satellite image classification strategies and techniques available. Satellite image classification strategies will be broadly classified into three classes 1. Automatic 2. Manual 3. Hybrid All three strategies have their own benefit sand disadvantages. Majority of the satellite image classification strategies fall into initial class. Satellite image classification wants choice of acceptable classification method supported the necessities. This analysis work is a study on satellite image classification strategies and techniques. The analysis work conjointly compares numerous researchers’ comparative results on satellite image classification strategies. This work compares performance of conventionally recent classification techniques on satellite information. Additionally, there are many problems requiring thought in respect of the classification of remotely perceived data: 1) The way to choose the right size of coaching samples 2) The way to create the classifier parameters 3) The way to mix classifiers in an economical way. This review suggests that effective use of multiple options of remotely perceived information and also the choice of an acceptable classification methodology are crucial for rising classification accuracy. Additional analysis, however, is required to spot and cut back uncertainties within the image-processing to enhance classification accuracy.