%0 Journal Article %1 noauthororeditor %A S.Nivetha1, %A Dr.R.Jensi2, %D 2019 %J Advances in Engineering: an International Journal (ADEIJ) %K Classification. Classifier Multi Prediction Signatures temporal %N 3 %P 01-13 %T CLASSIFICATION AND COMPARISION OF REMOTE SENSING IMAGE USING SUPPORT VECTOR MACHINE AND K-NEAREST NEIGHBOUR ALGORITHMS %U https://airccse.com/adeij/papers/2319adeij03.pdf %V 2 %X Remote sensing is collecting information about an object without any direct physical contact with the particular object. It is widely used in many fields such as oceanography, geology, ecology. Remote sensing uses the Satellite to detect and classify the particular object or area. They also classify the object on the earth surfaces which includes Vegetation, Building, Soil, Forest and Water. The approach uses the classifiers of previous images to decrease the required number of training samples for the classifier training of an incoming image. For each incoming image, a rough classifier is predicted first based on the temporal trend of a set of previous classifiers. The predicted classifier is then fine-tuned into a more accurate manner with current training samples. This approach can be further applied as sequential image data, with only a small number of training samples, which are being required from each image. This method uses LANSAT 8 images for Training and Testing processes. First, using the Classifier Prediction technique the Signatures are being generated for the input images. The generated Signatures are used for the Training purposes. SVM Classification is used for classifying the images. The final results describes that the leverage of a priori information from previous images will provide advantageous improvement for future images in multi temporal image classification.

@article{noauthororeditor, abstract = {Remote sensing is collecting information about an object without any direct physical contact with the particular object. It is widely used in many fields such as oceanography, geology, ecology. Remote sensing uses the Satellite to detect and classify the particular object or area. They also classify the object on the earth surfaces which includes Vegetation, Building, Soil, Forest and Water. The approach uses the classifiers of previous images to decrease the required number of training samples for the classifier training of an incoming image. For each incoming image, a rough classifier is predicted first based on the temporal trend of a set of previous classifiers. The predicted classifier is then fine-tuned into a more accurate manner with current training samples. This approach can be further applied as sequential image data, with only a small number of training samples, which are being required from each image. This method uses LANSAT 8 images for Training and Testing processes. First, using the Classifier Prediction technique the Signatures are being generated for the input images. The generated Signatures are used for the Training purposes. SVM Classification is used for classifying the images. The final results describes that the leverage of a priori information from previous images will provide advantageous improvement for future images in multi temporal image classification.}, added-at = {2019-06-03T09:18:22.000+0200}, author = {S.Nivetha1 and Dr.R.Jensi2}, biburl = {https://www.bibsonomy.org/bibtex/293481357335cee1bbb27e85ecdbfd735/adeij_journal}, interhash = {67a37cc26752e5e9f588a50bc16e511e}, intrahash = {93481357335cee1bbb27e85ecdbfd735}, journal = {Advances in Engineering: an International Journal (ADEIJ)}, keywords = {Classification. Classifier Multi Prediction Signatures temporal}, month = may, number = 3, pages = {01-13}, timestamp = {2019-06-03T09:18:22.000+0200}, title = {CLASSIFICATION AND COMPARISION OF REMOTE SENSING IMAGE USING SUPPORT VECTOR MACHINE AND K-NEAREST NEIGHBOUR ALGORITHMS}, url = {https://airccse.com/adeij/papers/2319adeij03.pdf}, volume = 2, year = 2019 }