This study proposed a metaheuristic pansharpening (MP) method, which performs in a Synthetic Variable Ratio (SVR)-like manner. The proposed method introduced the Symbiotic Organisms Search (SOS) algorithm, an advanced nature-based optimization algorithm, to estimate a weight for each multispectral (MS) band to achieve the optimum intensity. The SVR pansharpening formula was used as the objective function and the Root Mean Square Error (RMSE) metric was used as the fitness function of the SOS algorithm to optimize the intensity. The spectral and spatial quality of the results of the MP method were qualitatively and quantitatively compared against those of 15 widely-used pansharpening methods in 5 test sites in Turkey with different land cover features. The experiments aimed to spatially enhance WorldView-2 MS images by using WorldView-2 panchromatic (PAN) bands and a UAV-derived PAN orthophoto. It was also aimed to sharpen IKONOS MS images by using a QuickBird pansharpened image and an IKONOS PAN band. The MATLAB software was used to implement the proposed method and to compute the spatial and spectral quality metrics. The spatial quality of each pansharpened image was evaluated at full-scale, whereas the spectral quality of each pansharpened image was evaluated at both full-scale and reduced-scale. A scoring strategy based on giving a performance score with respect to the spatial and spectral quality metrics was used to ensure a fair comparison among the pansharpening methods used. The results demonstrated that, out of 16 pansharpening methods, the MP method achieved the highest overall spectral and spatial quality scores of 15.5 and 15.5, respectively. The proposed method was found to perform successfully with both singlesensor and multisensor input images. It was also concluded that the proposed method is able to deal with high spatial resolution ratio between the input images.

Metaheuristic pansharpening based on symbiotic organisms search optimization

Effective compression technique of on-board hyperspectral images has been an active topic in the field of hyperspectral remote sensintg. In order to solve the effective compression of on-board hyperspectral images, a new distributed near lossless compression algorithm based on multilevel coset codes is proposed. Due to the diverse importance of each band, a new adaptive rate allocation algorithm is proposed, which allocates rational rate for each band according to the size of weight factor defined for hyperspectral images subject to the target rate constraints. Multiband prediction is introduced for Slepian-Wolf lossless coding and an optimal quantization algorithm is presented under the correct reconstruction of Slepian-Wolf decoder, which minimizes the distortion of reconstructed hyperspectral images under the target rate. Then Slepian-Wolf encoder exploits the correlation of the quantized values to generate the final bit streams. Experimental results show that the proposed algorithm has both higher compression efficiency and lower encoder complexity than several existing classical algorithms.

Near lossless compression of hyperspectral images based on distributed source coding

An effective compression algorithm of hyperspectral image based on discrete wavelet transform (DWT) and improved Karhunen-Loeve transform(KLT) is proposed in this paper. It could achieve better performance with effective combination of the two methods which convert the energy of images to a small number of coefficients. In order to reduce the most redundancy of space, it deals with each spectrum with fast 5/3 2-D DWT firstly. Then the spectral 1-D KLT is applied on the 2-D DWT coefficients to reduce spectral redundancy and the other spatial redundancy. Finally, entropy coding is executed to obtain the compressed code stream. The experimental results show that the average peak signal to noise ratio (PSNR) of the proposed compression algorithm is 3.01 bit per pixel (bpp), which improves greatly through comparing with the other approaches, and reduces the operating time and improves the performance of hyperspectral image compression algorithm. It proposes the hardware implementation strategy at the same time, verifies the correctness and feasibility of our method. The work in this paper has a good reference value and significance for the similar case.

/pdf/a-compression-algorithm-for-hyperspectral-image-2aie459rpv.pdf

A Compression Algorithm for Hyperspectral Image

Abstract. We address the problem of the lossless compression of hyperspectral images and present two efficient algorithms inspired by the distributed source coding principle, which perform the compression by means of the blocked coset coding. In order to make full use of the intraband and interband correlation, the prediction error block scheme and the multiband prediction scheme are introduced in the proposed algorithms. In the proposed algorithms, the prediction error of each 16×16  pixel block is partitioned into prediction error blocks of size 4×4. The bit rate of the pixels corresponding to the 4×4 prediction error block is determined by its maximum prediction error. This processing takes advantage of the local correlation to reduce the bit rate efficiently and brings the negligible increase of additional information. In addition to that, the proposed algorithms can be easily parallelized by having different 4×4 blocks compressed at the same time. Their performances are evaluated on AVIRIS images and compared with several existing algorithms. The experimental results on hyperspectral images show that the proposed algorithms have a competitive compression performance with existing distributed compression algorithms. Moreover, the proposed algorithms can provide low-codec complexity and high parallelism, which are suitable for onboard compression.

Distributed lossless compression algorithm for hyperspectral images based on the prediction error block and multiband prediction

The JPEG2000 image compression standard is the newest still image compression standard for digital image consumer electronics, such as digital cameras. However, Tier-1 encoding in the JPEG2000 standard algorithm requires a significant amount of memory and code time. In this study, a highly efficient and low-complexity rate control algorithm for JPEG2000 is proposed to reduce the memory requirements and the computation time. A revisable threshold is developed and is used to select the passes of each code block during Tier-1 encoding. Only the selected passes are encoded during Tier-1 and Tier-2 encoding. This routine avoids the encoding of all code passes and improves the coding efficiency. The experimental results based on standard test images show that the proposed algorithm can reduce the Tier- 1 coding time by up to 86.36% and the working memory by up to 95.50% at a bit rate of 0.125 bpp in comparison to the post-compression rate distortion algorithm. The peak-signal-to-noise ratio across the images remains almost unchanged. The proposed algorithm can not only significantly improve the encoding efficiency but also maintain the image quality. Therefore, it can be applied to consumer imaging devices such as digital cameras, cameras in mobile phones and compact consumer cameras in the form of either embedded or processing software to reduce the memory and hardware requirements.

A highly efficient rate control algorithm for JPEG2000 images

Distributed source coding(DSC) is applied to hyperspectral image compression due to its low complexity and error resilienceIn the framework of typical scalar coset coding based distributed compression method(s-DSC),it is pointed out in this paper that the infinity-norm minimization should be the best criterion for prediction,and a sub-optimal prediction method based on search for L∞ minimum is proposed to approach the criterionIn addition,the compression scheme is extended to near-lossless compressionThe experimental results show that the lossless compression bitrate of the proposed method is reduced by about 025bpp compared to s-DSC and the near-lossless compression outperforms JPEG-LS significantlyOwing to the advantages of low complexity,high performance and error resilience,the proposed method is quite suitable for onboard compression

Lossless and Near-Lossless Compression of Hyperspectral Images Based on Search for L_∞ Minimum and Coset Coding

Hyperspectral (HS) pansharpening aims at fusing a panchromatic (PAN) image with a hyperspectral image, generating an image with the high spatial resolution of the former and the high spectral resolution of the latter. Recently, in order to enhance this process, researches have combined hyperspectral unmixing with the HS fusion model, improving the fusion results . In this combined model , endmember subspace extraction is a crucial step. Traditionally, VCA and SVD are used for subspace extraction, but VCA extracts pixels in the HS image that contain impure material signatures. In this work, we use the sparse unmixing technique to extract the endmember subspace which contains pure material signatures, which may better represent the HS image. We combine sparse unmixing-based subspace extraction with the bayesian fusion model. Results indicate that the fusion algorithm using our subspace extraction method had better global performance.

Subspace selection for hyperspectral pansharpening using sparse unmixing

After the elaborate analysis for the rate allocation algorithm based on priority scanning, a fast and efficient rate control algorithm for JPEG2000 based on priority scanning and minimal slope discarding is presented to eliminate influence of priority scanning rate allocation algorithm to the reconstructed image quality. The coding passes of all coding blocks are encoded by T1 encoder with bit-plane priority. After the accumulate bit-rate exceeds the target bit-rate, minimal slope discarding scheme is used for all blocks to complete the rate-distortion optimization truncation, which can more efficiently reduce the redundancy of T1. The experimental results show that the proposed scheme can efficiently reduce both the computational cost and working memory size of the entropy coding process, while obtaining almost the same performance as the conventional rate control algorithm in JPEG2000 standard.

Wang Keyan

Papers

Lossless and Near-Lossless Compression of Hyperspectral Images Based on Search for L_∞ Minimum and Coset Coding

Subspace selection for hyperspectral pansharpening using sparse unmixing

Efficient rate control algorithm for JPEG2000 based on PSMSD