A Proficient Roi Segmentation with Denoising and Resolution Enhancement

Abstract

Region-of-interest (ROI) detection technology, which has been introduced into the remote sensing image analysis field for segmentation. In existing method, Frequency domain analysis and salient region detection (FDA-SRD) method are used, which has drawbacks: inaccurate, missing detection and  low image resolution. Efficient, denoisy, enhanced and accurate ROI images based on Switching Bilateral Filter(SBF) and wavelet (DWT and SWT)transforms are proposed. First, Denoising the remote sensing image using switching bilateral filter. Second, Extract the desired output by using frequency domain analysis and salient region detection. Finally, Image resolution and contrast are enhanced by discrete wavelet transform and stationary wavelet transform. Compared with existing models, the proposed method is more efficient and provides more visually accurate detection results.

General Terms: Color segmentation, Median filter, ROI segmentation, SBF, Wavelet transforms.