Adaptive Compression Technique Research Articles

UAV-Assisted Mobile Crowdsensing Systems Empowered by Wireless Power Transfer and Adaptive Compression Techniques

UAV-Assisted Mobile Crowdsensing Systems Empowered by Wireless Power Transfer and Adaptive Compression Techniques

Communication-efficient federated learning via knowledge distillation

Federated learning is a privacy-preserving machine learning technique to train intelligent models from decentralized data, which enables exploiting private data by communicating local model updates in each iteration of model learning rather than the raw data. However, model updates can be extremely large if they contain numerous parameters, and many rounds of communication are needed for model training. The huge communication cost in federated learning leads to heavy overheads on clients and high environmental burdens. Here, we present a federated learning method named FedKD that is both communication-efficient and effective, based on adaptive mutual knowledge distillation and dynamic gradient compression techniques. FedKD is validated on three different scenarios that need privacy protection, showing that it maximally can reduce 94.89% of communication cost and achieve competitive results with centralized model learning. FedKD provides a potential to efficiently deploy privacy-preserving intelligent systems in many scenarios, such as intelligent healthcare and personalization.

Deep reinforcement learning approach to predict head movement in 360° videos

The popularity of 360° videos has grown immensely in the last few years. One probable reason is the availability of low-cost devices and ease in capturing them. Additionally, users have shown interest in this particular type of media due to its inherent feature of being immersive, which is completely absent in traditional 2D videos. Nowadays such powerful 360° videos have many applications such as generating various content-specific videos (gaming, knowledge, travel, sports, educational, etc.), during surgeries by medical professionals, in autonomous vehicles, etc. A typical 360° video when seen through a Head Mounted Display (HMD) gives an immersive feeling, where the viewer perceives standing within the real environment in a virtual platform. Similar to real life, at any point in time, the viewer can view only a particular region and not the entire 360° content. Viewers adopts physical movement for exploring the total 360° content. However, due to the large volume of 360° media, it faces challenges during transmission. Adaptive compression techniques have been incorporated in this regard, which is in accordance with the viewing behaviour of a viewer. Therefore, with the growing popularity and usage of 360° media, the adaptive compression methodologies are in development. One important factor in adaptive compression is the estimation of the natural field-of-view (FOV) of a viewer watching 360° content using a HMD. The FOV estimation task becomes more challenging due to the spatial displacement of the viewer with respect to the dynamically changing video content. In this work, we propose a model to estimate the FOV of a user viewing a 360° video using an HMD. This task is popularly known as the Virtual Cinematography. The proposed FOVSelectionNet is primarily based on a reinforcement learning framework. In addition to this, saliency estimation is proved to be a very powerful indicator for attention modelling. Therefore, in this proposed network we utilise a saliency indicator for driving the reward function of the reinforcement learning framework. Experiments are performed on the benchmark Pano2Vid 360° dataset, and the results are observed to be similar to human exploration

Comprehensive Analysis of Flow Incorporated Neural Network based Lightweight Video Compression Architecture

The increasing video content over the internet motivated the exploration of novel approaches in the video compression domain. Though neural network based architectures have already emerge as de-facto in the field of image compression and analytics, their application in video compression also result in promising outputs. Adaptive and efficient compression techniques are required for video transmission over varying bandwidth. Several deep learning based techniques and enhancements were proposed and experimented but they didn’t exhibit full optimal behavior and are not end to end trained and optimized. In the zest of a pure and end to end trainable compression technique, a deep learning based video compression architecture has been proposed comprises of frame autoencoder, flow autoencoder and motion extension network for the reconstruction of predicted frames. The video compression network has been designed incrementally and trained with random emission steps strategy. The proposed work results in significant improvement in visual perception quality measured in SSIM and PSNR when compared to some state-of-art techniques but in trade-off with frame reconstruction time sheet.

Novel bit-level Adaptive and Asymmetric Data Compression Technique

Novel bit-level Adaptive and Asymmetric Data Compression Technique

Quality Verification of Audio and Image Modulation by the Simulation of PCM, DM and DPCM Systems

Modulation is a process through which a message has to pass in order to be effectively transmitted. However, there are some limitations to Pulse Code Modulation and Delta Modulation that can cause data redundancy, quantization error, slope overload distortion and granular noise which result in a bad communication process. Throughout the past few years, Pulse Code Modulation (PCM), Delta Modulation (DM) and Differential Pulse Code Modulation (DPCM), in digital communication systems, have proven to have unparalleled advantages over analog communication systems; this is in terms of error minimization and distances of transmission enhancements. Delta Modulation, a simplified version of Pulse Coded Modulation also pauses major problems in noise and quantization error. Consequently, and to combat the arising problems, communication engineers have developed newly adaptive compression and modulations techniques for better digital transmission. One of these innovative systems is the Differential Pulse Coded Modulation (DPCM) that can solve the aforementioned problems. Thus the focal point of this article is to explore the simulation of these systems using Simulink (The Math Works, Inc., USA). Eventually, the systems are tested on both image and audio inputs to prove the superiority of DPCM over DM and PCM systems in reducing noise and increasing the signal to quantization noise ratio, thus insuring a smooth and successful transfer of data.

Compact inverted index storage using general‐purpose compression libraries

SummaryEfficient storage of large inverted indexes is one of the key technologies that support current web search services. Here we re‐examine mechanisms for representing document‐level inverted indexes and within‐document term frequencies, including comparing specialized methods developed for this task against recent fast implementations of general‐purpose adaptive compression techniques. Experiments with the Gov2‐URL collection and a large collection of crawled news stories show that standard compression libraries can provide compression effectiveness as good as or better than previous methods, with decoding rates only moderately slower than reference implementations of those tailored approaches. This surprising outcome means that high‐performance index compression can be achieved without requiring the use of specialized implementations.

New method for Increasing watermarked image quality and security

Recently, information transmission is done electronically, information can be stored and manipulated easily using computer, information transferring should be done in secure way. Watermark is a technique that can be used to secure data.
 In this paper, a new watermark algorithm is proposed which depends on integer wavelet transform, arithmetic encoding and adaptive run length encoding. Adaptive run length encoding is used to compress watermark before embedding process. Arithmetic encoding can be used to compress data in original image that are needed to be retrieved from watermarked images to recover original image without any differences after watermark extraction process.
 To avoid the problem of traditional Run Length Encoding Technique that may produce file that is larger than original one. Adaptive run length encoding is suggested which is used to compress watermark in cover image. The adaptive compression technique can be done several times on watermark using different number of bits to represent counts of runs and produce several vectors for number of runs, compare several results and select smallest vector from them. Surely, the smallest vector means that watermarked image will be high quality image; capacity will be better and adaptively ensure embedding process will be more secure.

A Power and Area Efficient Ultra-Low Voltage Laplacian Pyramid Processing Engine With Adaptive Data Compression

This paper proposes a power and area efficient Laplacian Pyramid processing engine (LPPE) for real-time portable image/video processing applications. On the architecture level, an adaptive data compression technique is proposed to reduce the power and area of FIFOs in the LPPE while maintaining small mean square error (MSE). A new filtering extension method is proposed to reduce the output errors caused by image boundary pixels. On the circuit level, near-threshold operation is adopted to further reduce the power consumption. The proposed LPPE is fabricated in 0.18- $\mu\text{m}$ CMOS process technology and consumes only 452 $\mu\text{W/frame}$ with a clock frequency of 3.68 MHz and 112 frames per second at 0.5 V. The area is reduced by 18.98%–36.06% compared to conventional counterpart.

Adaptive searchless fractal image compression in DCT domain

In this paper, an adaptive searchless fractal image compression technique in the frequency domain is proposed. Proposed methods aim to improve image compression fidelity criteria in discrete cosine transform domain (DCT). In this work spatial domain, adaptive methods are exercised in discrete cosine domain for optimisation of image encoding. Variance-based selection of image blocks and adaptive threshold are two simple adaptive techniques administered on attaining objectives of image compression. First, range block exclusion adaptive method removes homogenous range blocks before FIC coding. This exclusion consequently recedes the number of range-domain blocks matching operations. Proposed hybrid compression algorithm localises domain block location about range block without searching entire domain pool. Second technique implemented is domain pool size reduction. Variances of range and domain blocks are compared. Domain blocks that are within threshold value are kept in the pool. Finally to further optimise the performance of encoding adaptive threshold in quadtree partitioning is applied in along with variance domain selection approach. Simulation results substantiate that adaptive techniques have higher coding performance in terms of encoding time, peak signal to noise ratio and compression ratio in contrast to the fractal image encoding without adaptive techniques.

Optimal Filter Design of Wavelet Transform to an Adaptive Medical Image Compression Technique Using Genetic Group Search Optimizer

Optimal Filter Design of Wavelet Transform to an Adaptive Medical Image Compression Technique Using Genetic Group Search Optimizer

A Novel Adaptive Compression Technique for Dealing with Corrupt Bands and High Levels of Band Correlations in Hyperspectral Images Based on Binary Hybrid GA-PSO for Big Data Compression

sensors generate useful information about cli- mate and the earth's surface in numerous contiguous narrow spectral bands, being widely used in resource management, agriculture, environmental monitoring, among others. The compression of hyperspectral data helps in long-term storage and transmission systems. This paper introduces a new adap- tive compression method for hyperspectral data. The method is based on separating the bands with different specifications by the histogram analysis and Binary Hybrid Genetic Algo- rithm-Particle Swarm Optimization (BHGAPSO). The new proposed method improves the compression ratio of the best- known JPEG standards, saves storage space, and speeds up the transmission system. The proposed method is applied on two different test cases, and the results are evaluated and compared with a few powerful compression techniques, such as lossless JPEG and JPEG2000. The results confirm that the proposed method is accurate, simple and fast, which can be useful for big data (i.e, a high volume of data) processing.

Gaze behavior data profiling and analysis system platform based on visual content representation

This paper presents a real-time system platform for profiling and analyzing the gaze behavior based on visual contents. The proposed system captures the gaze information from multiple users and provides the ability to measure the degree of visual content perception by the users through statistical analysis. Visual content representation scheme is presented for capturing and annotating the gaze behavior effectively. Information correlation property among multiple image frames is defined for providing the ability to analyze the pattern of perception of user based on complex visual contents. In order to monitor the real time gaze behavior of the users, the monitoring rules are incorporated in to the representation template. The number of users and the duration of observation time may significantly increase the profile data size. To alleviate the problem, an adaptive data compression techniques is incorporated. The capabilities and functionalities of the proposed system are verified for multiple users with different gaze behavior on a sequence of commercial image data.

Power Efficient Adaptive Compression Technique for Wireless Sensor Networks

Wireless sensor network consists of many tiny disposable low power devices called nodes, which are spatially distributed in order to perform various application global tasks. These nodes form a network by communicating with each other either directly or through other nodes. Sensor networks are fundamentally constrained by the difficulty and energy expense of delivery information from sensor to sink. Power saving is a critical issue in wireless sensor networks since sensor nodes are powered by batteries which cannot generally change or recharged. In this paper we have focused on radio communication which is often considered as the main cause of energy consumption. One of the solutions to this problem would be generally achieved by reducing the size of the transmitted/received data through data compression. A new adaptive compression technique will be considered for lossless data compression which reduces the amount of data that must be passed through the network and to the sink and thus have energy benefits that are multiplicative with the number of hops the data travel through the network. Possible extensions of this paper, use RUN LENGTH CODING to compress decompress the data in WSN.

Efficient DCT-based image compression technique

Although JPEG technique is considered as the most popular image compression standard, it behaves high visual degradation at low bit rates. In this paper, an efficient DCT–based image compression technique is proposed to achieve high Compression Ratio (CR) with high quality at both high and low bit rates. This technique uses switching between JPEG compression technique at high bit rates and a novel Adaptive Lossy Image Compression (ALIC) technique at low bit rates. ALIC is proposed to overcome the drawbacks of JPEG technique at low bitrates. The performance of the proposed technique is analysed at low and high bit rates on both grey and colour images. Performances of both JPEG and ALIC techniques are analysed and compared. The experimental results reveal that the proposed ALIC technique achieves better CR with acceptable SNR in comparison with JPEG technique. Also, the resultant CR of ALIC technique can be considerably increased with a slight decrease of its PSNR. This decrease in PSNR does not result in a noticeable visual degradation of the compressed image. On the other hand, increasing the CR of JPEG technique results in a noticeable visual degradation due to the appearance of blocking effect in the reconstructed image. Thus, it is greatly recommended to use ALIC technique in the applications that require high CR with stable PSNR. ALIC is a general purpose technique that can be applied, not only on images, but also on any data source which uses Huffman coding to achieve better CR. Therefore, it is suitable for compression of text, image and video.

Query Evaluation for XML Databases using Partial Decompression

This research work demonstrates the Extraction, compression and query processing of XML documents for Adaptive Compression Techniques and Efficient Query Evaluation. Proposed here are the algorithms for xml compression and Efficient Query Evaluation as - Feasible XML compression using data compression algorithm. Qurey Processor using Sax parsing and Interfaces. It is shown that using the proposed techniques for xml data compression will pave a way for better compression and improve the compression ratio and performance of compressor system

A CFAR detector for MIMO array radar based on adaptive pulse compression-Capon filter

Under the condition of strong clutter and finite training samples, a novel CFAR detector based on the proposed adaptive pulse compression-Capon filter (APC-Capon filter) is presented for MIMO array radar which transmits intrapulse coded waveforms. Both theoretical analysis and numerical simulations show that the proposed APC-Capon filter can utilize the non-reiterative adaptive pulse compression technique to alleviate both range sidelobes and mutual interference; at the same time it also can borrows ideas from the coherent signals cancellation of adaptive beamforming to perform clutter suppression. The detection performance simulations show that, with the same size parameters (including code length and the numbers of transmitter elements, receiver elements as well as pulses), the proposed CFAR detector is inferior to the adaptive matched filter (AMF) detector. But as the size parameters increase, the proposed CFAR detector will gradually outperform the AMF detector with smaller size parameters, and more importantly, unlike the AMF detector which has a lot of difficulties in performing due to the inordinate demands for the computational complexity and training sample size, this demands of the proposed CFAR detector will not significantly increase. The simulations also show that, the correlation properties of these intrapulse coded waveforms have certain effect on the APC-Capon filter and its corresponding CFAR detector. But fortunately, the effect will decrease as the number of intrapulse coded waveforms increases.

Image compression using Least Probable Coefficients Approximation technique

A novel Adaptive Lossy Compression technique, namely Least Probable Coefficients Approximation (LPCA) technique, is proposed to achieve better performance and simpler implementation than JPEG at low bitrate. Compression using LPCA based on reducing the number of source symbols by applying efficient processing on the Discrete Cosine Transform (DCT) coefficients of the image in addition to efficient quantisation to reduce the transmitted values. It is a general compression technique that can be applied to any digital data not just images. Experimental comparisons are carried out to compare the performance of the proposed technique with that of JPEG. The experimental results show that: The proposed compression technique achieves high Compression Ratio (CR) with higher Signal to Noise Ratio (SNR) than that of JPEG at low bitrate without the great visual degradation that appears in case of JPEG.

Comparison of the performance of different radar pulse compression techniques in an incoherent scatter radar measurement

Abstract. Improving an estimate of an incoherent scatter radar signal is vital to provide reliable and unbiased information about the Earth's ionosphere. Thus optimizing the measurement spatial and temporal resolutions has attracted considerable attention. The optimization usually relies on employing different kinds of pulse compression filters in the analysis and a matched filter is perhaps the most widely used one. A mismatched filter has also been used in order to suppress the undesirable sidelobes that appear in the case of matched filtering. Moreover, recently an adaptive pulse compression method, which can be derived based on the minimum mean-square error estimate, has been proposed. In this paper we have investigated the performance of matched, mismatched and adaptive pulse compression methods in terms of the output signal-to-noise ratio (SNR) and the variance and bias of the estimator. This is done by using different types of optimal radar waveforms. It is shown that for the case of low SNR the signal degradation associated to an adaptive filtering is less than that of the mismatched filtering. The SNR loss of both matched and adaptive pulse compression techniques was found to be nearly the same for most of the investigated codes for the case of high SNR. We have shown that the adaptive filtering technique is a compromise between matched and mismatched filtering method when one evaluates its performance in terms of the variance and the bias of the estimator. All the three analysis methods were found to have the same performance when a sidelobe-free matched filter code is employed.

General-purpose compression for efficient retrieval

Compression of databases not only reduces space requirements but can also reduce overall retrieval times. In text databases, compression of documents based on semistatic modeling with words has been shown to be both practical and fast. Similarly, for specific applications—such as databases of integers or scientific databases—specially designed semistatic compression schemes work well. We propose a scheme for general-purpose compression that can be applied to all types of data stored in large collections. We describe our approach—which we call RAY—in detail, and show experimentally the compression available, compression and decompression costs, and performance as a stream and random-access technique. We show that, in many cases, RAY achieves better compression than an efficient Huffman scheme and popular adaptive compression techniques, and that it can be used as an efficient general-purpose compression scheme.