Optimization of Hyperspectral Images and Performance Evaluation Using Effective Loss Algorithm
An effective lossy algorithm for compressing hyperspectral images using singular value decomposition (SVD) and discrete cosine transform (DCT) has been proposed. A hyperspectral image consists of a number of bands where each band contains some specific in
- PDF / 474,032 Bytes
- 9 Pages / 439.37 x 666.142 pts Page_size
- 80 Downloads / 164 Views
Abstract An effective lossy algorithm for compressing hyperspectral images using singular value decomposition (SVD) and discrete cosine transform (DCT) has been proposed. A hyperspectral image consists of a number of bands where each band contains some specific information. This paper suggests compression algorithms that compress the hyperspectral images by considering image data, band by band and compress each band employing SVD and DCT. The compression performance of the resultant images is evaluated using various objective image quality metrics. Keywords Hyperspectral images decomposition DCT
Effective loss algorithm
Singular value
1 Introduction Hyperspectral images are obtained from airborne to space-borne sensors, thereby; transmit to base station for processing, while onboard systems exhibit limited storage and power. Compression algorithms with high performance and low
Srinivas Vadali (&) J.V.R. Murthy Department of Computer Science and Engineering, University College of Engineering, Jawaharlal Nehru Technological University Kakinada, Kakinada, Andhra Pradesh, India e-mail: [email protected] URL: http://www.springer.com/aisc J.V.R. Murthy e-mail: [email protected] G.V.S.R.Deekshitulu Department of Mathematics, University College of Engineering, Jawaharlal Nehru Technological University Kakinada, Kakinada, Andhra Pradesh, India e-mail: [email protected] © Springer Science+Business Media Singapore 2016 M. Pant et al. (eds.), Proceedings of Fifth International Conference on Soft Computing for Problem Solving, Advances in Intelligent Systems and Computing 436, DOI 10.1007/978-981-10-0448-3_77
919
920
Srinivas Vadali et al.
complexity are needed to compress image data [1]. The compression techniques are classified as lossless and lossy. The lossy compression achieves higher compression ratios when compared to lossless compression algorithms [2]. Various image compression algorithms are found in the literature for both lossy and lossless compression techniques. Cheng[] proposed an improved version of EZW (Embedded Image Coding using Zero trees of Wavelet Co-efficient) algorithm for compressing the AVIRIS images (Airborne Visible/Infrared Imaging Spectrometer) to perform compression ratio for both lossless and lossy compression [1]. A Joint KLT (Karhunen–Loeve Transform) Lasso algorithm for compressing hyperspectral images was proposed by Simplice et al. [2]. Cheng introduced a compression method that uses a hybrid transformation which includes integer Karhunrn–Loeve transformation (KLT) and integer discrete wavelet transformation (DWT) [3]. A low-complexity compression algorithm for hyperspectral images based on distributed source coding multi-linear was suggested by Nian and Wan [4]. A new lossy image compression technique which uses singular value decomposition (SVD) and wavelet difference reduction was discussed in [5]. SVD is numerical technique which is used to diagonalize matrices [6]. Medical images were compressed by discrete cosine transform (DCT) spectral similarity s
Data Loading...
