China 
Africa 
Explore chapters and articles related to this topic
Data Compression in Health Monitoring
Published in Rajarshi Gupta, Dwaipayan Biswas, Health Monitoring Systems, 2019
Sourav Kumar Mukhopadhyay, Rajarshi Gupta
Among the various compression techniques that are based on DWT decomposition of ECG, the tree-based methods and threshold-based methods are most popular. Among the tree-based methods, embedded zero tree wavelet (EZW) [31] and its extended version, set partitioning in hierarchical trees (SPIHT) [32], have demonstrated good performance for ECG compression. Here, the ECG dataset is successively decomposed to form a pyramidal structure. Some of the decomposed coefficients are considered as ‘significant’ when they are greater than a threshold and otherwise considered as non-significant. EZW is based on following principles: (a) the position and sign of the significant coefficients are considered; (b) non-significant coefficients are compactly encoded based on self-similarity across sub-bands; (c) successive approximation of the significant coefficients. The SPIHT algorithm sorts the coefficients transformed by wavelet-based decomposition in order of their significance or magnitude. This partial ordering is done by comparing them with an octavely decreasing threshold. For transmission applications, coefficients with low bit rate are ordered first, with gradually decreasing significant ones.
Compressive Sensing in Color Image Security
Published in S. Ramakrishnan, Cryptographic and Information Security, 2018
Rohit Thanki, Surekha Borra, Komal Borisagar, Nilanjan Dey
The techniques based on second approach are designed and implemented by applying cryptographic-based encryption followed by image compression to image to get encrypted + compressed image [20–42]. These techniques are designed by various encryption methods such as DES, AES, Bit XOR operation, selective encryption, elliptic curve, etc. These methods are based on symmetric key generation or asymmetric key generation. The image compression methods like DCT, lifting wavelet transform (LWT), SPIHT (Set Partitioning in Hierarchical Trees), JPEG, adaptive compression, etc., are used for compression of encrypted images in these techniques. These methods may be lossy or lossless compression methods. Table 8.2 shows the summary of compression + encryption techniques for security of images.
Foveated Image and Video Coding
Published in H.R. Wu, K.R. Rao, Digital Video Image Quality and Perceptual Coding, 2017
The embedded foveation image coding (EFIC) system [WB01b] is shown in Figure 14.13. First, the wavelet transform is applied to the original image. The foveated perceptual weighting mask calculated from given foveation points or regions is then used to weight the wavelet coefficients. Next, we encode the weighted wavelet coefficients using a modified set partitioning in hierarchical trees (SPIHT) encoder, which is adapted from the SPIHT coder proposed in [SP96]. Finally, the output bitstream of the modified SPIHT encoder, together with the foveation parameters, is transmitted to the communication network. At the receiver side, the weighted wavelet coefficients are obtained by applying the modified SPIHT decoding algorithm. The foveated weighting mask is then calculated in exactly the same way as at the encoder side. Finally, the inverse weighting and inverse wavelet transform are applied to obtain the reconstructed image. Between the sender, the communication network and the receiver, it is possible to exchange information about network conditions and user requirements. Such feedback information can be used to control the encoding bit-rate and foveation points. The decoder can also truncate (scale) the received bitstream to obtain any bit rate image below the encoder bit rate.
A novel resolution independent gradient edge predictor for lossless compression of medical image sequences
Published in International Journal of Computers and Applications, 2021
Urvashi Sharma, Meenakshi Sood, Emjee Puthooran
Avudaiappan compared Huffman coding, Arithmetic coding, Lossless predictive coding, and other various Lossless image compression techniques. Lossless JPEG has been found better compression technique among these entire techniques w.r.t time and compression ratio for images [20]. Sunil and Sharanabasaweshwar presented a new approach by introducing convex-smoothing problems. According to this approach, input image is divided into sub-problems which are solved by using compressed sensing method. Proposed model’s performance is achieved in terms of PSNR and the time taken to perform the reconstruction [21]. Kabir and Monda compared edge-based transformation and entropy coding (ETEC) and prediction-based transformation and entropy coding (PTEC) schemes with the existing lossless compression techniques: ‘joint photographic experts group lossless’ (JPEG-LS), ‘set partitioning in hierarchical trees’ (SPIHT), and ‘differential pulse code modulation’ (DPCM). The ETEC and PTEC algorithms provide better compression than other schemes. PTEC is more suitable than ETEC for compression when both CR and computation time are taken into consideration [22].
Multimedia transmission in MC-CDMA using adaptive subcarrier power allocation and CFO compensation
Published in International Journal of Electronics, 2018
The proposed method is applied in image compression to enhance the quality in terms of high PSNR for the retrieved image. The set partitioning in hierarchical trees (SPIHT), a wavelet-based image compression technique, is considered in this paper. The major limitation of SPIHT algorithm is the lack of error recovery. Moreover, an error in single bit leads to incorrect detection of successive bits which in turn outputs the distorted image at the receiver. Hence, the reduction of bit errors is an important issue for SPIHT-based image compression. Specifically in 4G, multicarrier modulation (MCM) is employed in physical layer for high data rate and spectral efficiency. The existence of CFO in MCM leads to high bit error and severely degraded the image quality. In order to reduce the bit errors in image transmission, the proposed residual CFO compensation is applied along with SPIHT image compression. Figure 11 illustrates (a) Original Lena image of size 512512 (b) Retrieved Lena image with CFO of 0.2 and (c) retrieved residual CFO compensated Lena image for SPIHT-based image compression. The obtained PSNR for Lena image without CFO compensation is 15 dB and with the proposed technique, the PSNR is 41 dB. Hence, the proposed method offered improved image quality in terms of high PSNR.