Showing papers on "Data compression published in 1985"

Vector quantization in speech coding

Abstract: Quantization, the process of approximating continuous-amplitude signals by digital (discrete-amplitude) signals, is an important aspect of data compression or coding, the field concerned with the reduction of the number of bits necessary to transmit or store analog data, subject to a distortion or fidelity criterion. The independent quantization of each signal value or parameter is termed scalar quantization, while the joint quantization of a block of parameters is termed block or vector quantization. This tutorial review presents the basic concepts employed in vector quantization and gives a realistic assessment of its benefits and costs when compared to scalar quantization. Vector quantization is presented as a process of redundancy removal that makes effective use of four interrelated properties of vector parameters: linear dependency (correlation), nonlinear dependency, shape of the probability density function (pdf), and vector dimensionality itself. In contrast, scalar quantization can utilize effectively only linear dependency and pdf shape. The basic concepts are illustrated by means of simple examples and the theoretical limits of vector quantizer performance are reviewed, based on results from rate-distortion theory. Practical issues relating to quantizer design, implementation, and performance in actual applications are explored. While many of the methods presented are quite general and can be used for the coding of arbitrary signals, this paper focuses primarily on the coding of speech signals and parameters.

Data compression using run length encoding and statistical encoding

Abstract: A compression device which uses both run length encoding and statistical encoding. The run length encoding scheme uses a flag byte symbol which is disposed between a character signal and a run length symbol. The statistical encoding process uses multiple statistical encoding tables which are selected based upon previously occurring data.

Matrix quantizer design for LPC speech using the generalized Llyod algorithm

Abstract: Rate-distortion theory provides the motivation for using data compression techniques on matrices of N LPC vectors. This leads to a simple extension of speech coding techniques using vector quantization. The effects of using the generalized Lloyd algorithm on such matrices using a summed Itakura-Saito distortion measure are studied, and an extension of the centroid computation used in vector quantization is presented. The matrix quantizers so obtained offer substantial reductions in bit rates relative to full-search vector quantizers. Bit rates as low as 150 bits/s for the LPC matrix information (inclusive of gain, but without pitch and voicing) have been achieved for a single speaker, having average test sequence and codebook distortions comparable to those in the equivalent full-search vector quantizer operating at 350 bits/s. Preliminary results indicate that higher quality or lower bit rates may be achieved with enough computational resources.

Approximate counting: a detailed analysis

Abstract: Approximate counting is an algorithm proposed by R. Morris which makes it possible to keep approximate counts of large numbers in small counters. The algorithm is useful for gathering statistics of a large number of events as well as for applications related to data compression (Todd et al.). We provide here a complete analysis of approximate counting which establishes good convergence properties of the algorithm and allows to quantify precisely complexity-accuracy tradeoffs.

Combinatorial Algorithms on Words

Abstract: Open Problems in Stringology- 1 - String Matching- Efficient String Matching with Don't-care Patterns- Optimal Factor Transducers- Relating the Average-case Cost of the Brute force and the Knuth-Morris-Pratt String Matching Algorithm- Algorithms for Factorizing and Testing Subsemigroups- 2 - Subword Trees- The Myriad Virtues of Subword Trees- Efficient and Elegant Subword Tree Construction- 3 - Data Compression- Textual Substitution Techniques for Data Compression- Variations on a Theme by Ziv and Lempel- Compression of Two-dimensional Images- Optimal Parsing of Strings- Novel Compression of Sparse Bit Strings- 4 - Counting- The Use and Usefulness of Numeration Systems- Enumeration of Strings- Two Counting Problems Solved via String Encodings- Some Uses of the Mellin integral Transform in the Analysis of Algorithms- 5 - Periods and Other Regularities- Periodicities in Strings- Linear Time Recognition of Square free Strings- Discovering Repetitions in Strings- Some Decision Results on Nonrepetitive Words- 6 - Miscellaneous- On the Complexity of some Word Problems Which Arise in Testing the Security of Protocols- Code Properties and Derivatives of DOL Systems- Words over a Partially Commutative Alphabet- The Complexity of Two-way Pushdown Automata and Recursive Programs- On Context Free Grammars and Random Number Generation

Encoding of Arbitrary Curves Based on the Chain Code Representation

Abstract: An efficient encoding scheme for arbitrary curves, based on the chain code representation, has been proposed. The encoding scheme takes advantage of the property that a curve with gentle curvature is divided into somewhat long curve sections (segments), each of which is represented by a sequence of two adjacent-direction chain codes. The coding efficiency of the scheme becomes higher as the segments become longer, while a variable-length coding technique makes it possible to encode short segments without an extreme increase in code amount. An experiment with complicated curves obtained from geographic maps has shown a high data compression rate of the proposed scheme, whose code amount is about 50-60 percent of that required for the conventional chain encoding scheme.

Data compression on a database system

Abstract: A general-purpose data-compression routine—implemented on the IMS database system—makes use of context to achieve better compression than Huffman's method applied character by character. It demonstrates that a wide variety of data can be compressed effectively using a single, fixed compression routine with almost no working storage.

Data compression system

Abstract: A data compression system for use in processing diagnostic image data which uses a predictor to predict future data. The actual data is subtracted from the predicted value to obtain data related difference values. The difference values are coded by coding the most prevalent difference by a code other than the least bit code but further coding the most prevalent number by string length coding, with the most common being the coded with a least bit code, and also coding the next most common difference with a least bit code and the subsequent most prevalent differences with sequential least bit codes to provide compression ratios of over 3:1. A procedure for finding the best predictor in noisy data is provided.

Variations on a theme by Ziv and Lempel

Abstract: The data compression methods of Ziv and Lempel are modified and augmented, in three ways in order to improve the compression ratio, and hold the size of the encoding tables to a fixed size. The improvements are in the area of dispensing with any uncompressed output, ability to use fixed size encoding tables by using a replacement strategy, and more rapid adaptation by widening the class of strings which may be added to the dictionary. Following Langdon, we show how these improvements also provide an adaptive probabilistic model for the input data. The issue of data structures for efficient implementation is also addressed.

Parallel algorithms for data compression

Abstract: Parallel algorithms for data compression by textual substitution that are suitable for VLSI implementation are studied. Both “static” and “dynamic” dictionary schemes are considered.

Data Compression in Scientific and Statistical Databases

Abstract: Scientific and statistical database systems heavily depend on data compression techniques to make possible the management and storage of their large databases. The efficiency of data compression methods has a signficant impact on the overall performance of these systems. The purpose of this paper is to show the importance of data compression to scientific/statistical databases, to discuss the pros and cons of data compression, and to survey data compression techniques relevant to scientific/statistical databases. The emphasis is on the basic idea, motivation, and tradeoffs of each approach. Both software and hardware methods are covered. The paper is concluded by a discussion of several points of research that seem worthy of further investigation.

Compression and ranking

Abstract: A complexity-theoretic approach to the classical data compression problem is to define a notion of language compression by a machine in a certain complexity class, and to study language classes compressible under the above definition. Languages that can be compressed efficiently (e.g. by a probabilistic polynomial time machine) are of special interest.We define the notion of language compressibility, and show that sufficiently sparse “easy” languages (e.g. polynomial time) can be compressed efficiently. We also define a notion of ranking (which is an optimal compression) and show that some “very easy” languages (e.g. unambiguous context-free languages) can be ranked efficiently. We exhibit languages which cannot be compressed or ranked efficiently.The notion of compressibility is closely related to Kolmogorov complexity and randomness. We discuss this relationship and the complexity-theoretic implications of our results.

Modified algorithm for block truncation coding of monochrome images

Abstract: Block truncation coding (BTC) is a recent technique used in the coding of image data. In the letter a modified technique for BTC coding of image data is presented which is algorithmically simple and hence easy to implement. This new technique uses only the first-order statistical information as `block overhead?. The new algorithm is shown to be optimum in the mean-square sense for a particular class of BTC algorithms. The letter presents the results of using the new algorithm for a typical image and compares the performance with that of the earlier algorithm for the same image.

Image data storing device

Abstract: An image data storing device according to the invention includes a data compression/expansion device between the main memory of a CPU and an auxiliary memory device, such that image data are transmitted between the main memory and the data compression/expansion device, while data which are obtained by compressing the image data are transmitted between the data compression/expansion device and the auxiliary memory device.

Addressable subscription television system having multiple scrambling modes

Abstract: A multimode scrambling system for video signal transmission systems is described. The system provides for baseband video scrambling controlled by a central originating computer facility. The scrambling of each field of a video changes on a per field basis. The video signal is scrambled in several modes including vertical interval scrambling, alternate line inversion, bogus sync pulse generation, video compression and video offset techniques. A scrambling sequence is sent by a unique algorithm, sent during the vertical interval to each system decoder. The algorithm is reordered for each of a plurality of fields of the video signal and the reordering position is identified by a unique synchronization pulse transmitted during the vertical interval of the video signal. Additional security measures are provided to inhibit a subscriber from avoiding a transmitted disable command, or an attempt to invade the subscriber decoder mechanical packaging security.

Method of and apparatus for compressing image data

Abstract: The present invention relates to image data compression utilized either when an original image is displayed on a monitor or when an original is reproduced with a desired reduction ratio. Pixel data of an original image are sequentially inputted synchronously with a clock signal (CY), and an average value of pixel data inputted theretofore is renewed in an averaging circuit (40) each time pixel data are inputted. The renewal of the average value is repeated until an assignment signal (CX) is outputted from a DDA circuit (30). The assignment signal assigns pixels of the original image to the same number of blocks as the number of pixels of a desired compressed image, which is adapted for a desired image compression. An output obtained from the averaging circuit (40) represents an average value of pixel data with respect to each block aligned in the vertical direction of the original image. In a DDA circuit (50) an assignment for blocks in the horizontal direction of the original is carried out. Average values of pixel data with respect to each block adaptable for the desired image compression are obtained by an averaging circuit ( 60), which average values are used as pixel data of each pixel of the compression image.

SAPA-2 Is the Fan

Abstract: The algorithm SAPA-2, presented recently as a method for representing electrocardiographic wavefonns as a series of straight-line segments, appears to be equivalent to an older algorithm, the Fan.

The telesign project

Abstract: Telesign is designed to offer a means of visual communication over a 56- or 64-kbit/s data network. The purpose is to supply a means of visual telecommunication among the members of the deaf community using sign language or lip reading. The system consists of an edge detector followed by digital compression coding to meet channel requirements. Psychometric experiments have shown the need for 25 frames/s with a minimum definition of 128 × 128 points. Edge-detection techniques are discussed from the viewpoint of an unusual quality criterion: the subjective resemblance of the contoured image to the original and, closely related to this, the intelligibility of the image sequence. A classification of edge detectors is given, based upon the position of the contour with respect to the dark/light contrast boundaries in the picture. The relationship of this classification to the quality criterion yields the definition of a new gradient display, called the "shifted gradient," with improved visual quality. The pseudo-Laplacian, well-suited to the problem, is analyzed. Various binary coding techniques are reviewed, and the results of compression given. A psychovisual experiment, performed in conjunction with the INJS (National Institute for the Education of Young Deaf), Paris, France, is described; the results validate the proposed approach. Guidelines are proposed for the design of a real-time device. The application of Telesign to closed-caption TV programs for the deaf viewer is also suggested.

On-the-fly data compression system

Abstract: An on-the-fly data compression system for compressing the data transmitted between a data source and an ultimate utilization device. The system normally has two data compression modules implementing a conventional multi-buffering scheme and a decoder for reconstructing the compressed data to its original format. The data compression modules compresses the data using a word representation by associative processing buffer and a content induced transaction overlap transmission protocol which results in an interleaved transmission of data bits and bit position bits, the latter being indicative of the number of data bits that will be transmitted during the subsequent data transmission. The decoder reconstructs the compressed data to its original format in response to the transmitted data and bit position bits and stores them in a pair of buffers in the sequential order in which they were generated.

A Predictor Switching Scheme for DPCM Coding of Video Signals

Abstract: In DPCM coding of video signals, for improved data compression efficiency, the predictor is often switched adaptively between two or more different predictions. For low-bit-rate coders, the predictor switching is done on a pel-by-pel basis. In this paper, a new scheme for predictor switching on a pel-by-pel basis is proposed. This new scheme is a modification of an existing adaptive predictor switching scheme which has been implemented in pel-recursive motion compensated coding of video signals. To evaluate the efficiency of the new scheme, a theoretical expression for the output bit rate is derived. It is pointed out under what conditions it is advantageous to use the new scheme for predictor switching. Some details of implementation are discussed. Results of computer simulation on natural scenes are presented. These results validate the theoretical conclusions derived in the paper. The simulation results also indicate that the picture quality is improved significantly, for approximately the same output bit rate, with the new predictor switching scheme as compared to the existing scheme for predictor switching.

Locally Optimal Run-Length Compression Applied to CT Images

Abstract: Run-length data compression techniques are described that preserve image content. After decompression, images are restored to their original state without loss in image gray scale or resolution. The first technique introduces terminology and illustrates a simple method for efficient picture archiving. It demonstrates the principle of run-length techniques. A second more general approach encodes picture information in a manner that adapts to local variation in pixel standard deviation. Among several options of compression formats, the one that delivers the best local compression is selected. Results of our compression techniques are given for several hundred computed tomography (CT) pictures with comparison to image entropy measures. A general solution to the optimal run-length compression of digital data is outlined. Routine application of the locally optimal method is also described.

Data compression using static Huffman code-decode tables

Abstract: Both static and dynamic Huffman coding techniques are applied to test data consisting of 530 source programs in four different languages. The results indicate that, for small files, a savings of 22-91 percent in compression can be achieved by using the static instead of dynamic techniques.

Measurement of data compression in advanced group 4 facsimile systems

Abstract: The CCITT (Consultative Committee for International Telephone and Telegraph) has defined Group 4 facsimile apparatus as that which transmits an ISO A4 (210 × 297 mm) document over public data networks including Packet Switched Data Networks (PSDN) and Circuit Switched Data Networks (CSDN). Data compression is employed in Group 4 systems to minimize the number of bits required for transmission of a message. CCITT Recommendation T.6 defines the basic Group 4 coding scheme to be a variation of the modified READ code (MRC II) which is the Group 3 coding option. This paper measures the compression of the MRC II algorithm for three different documents at resolutions of 200, 240, 300, 400, and 480 picture elements per inch. The compression measurement was accomplished by means of computer simulation. It was determined that the bits per page measure increases linearly with resolution. Results are presented and conclusions drawn. The Group 4 CCITT Recommendations include a mode of operation known as Mixed Mode where a page is divided into two parts: character-coded information and facsimile-coded information. This paper describes four alternative mixed-mode algorithms and measures the compression of each coding technique for CCITT document number 1. The compression measurements were performed by means of computer simulation. Conclusions are drawn regarding the relative advantages of the four algorithms.

Apparatus and algorithm for compressing and decompressing data

Abstract: A compression and a decompression algorithm resides in a memory of a data processing system. The data processing system communicates to another data processing system via an input/output means and a telecommunications network. When a message, to be communicated to the other data processing system, is ready for transmission, the data processing system compresses the message data. A processor of the data processing system compresses the data using the compression algorithm stored in the memory of the data processing system. The compression algorithm: identifies a mask character in the data representing the most frequently occurring character in the data, develops a mask byte in association with the mask character, the mask byte identifying the locations of the mask character in the data and identifying the locations of each portion of a residual message associated with the data, and develops the residual message, the residual message representing the message data with the mask character extracted from the message data.

A naturalness-preserving transform for image coding and reconstruction

Abstract: We have developed a new image transform that possesses some unusual properties that may be useful for some special applications; especially for those instances in which the communications must pass through channels whose quality and characteristics are unpredictable due to nonstationary noise processes or jamming. We present an example of a use for the new transform that can be characterized as a man-in-the-loop error correction procedure. A stable and an efficient algorithm has been developed for the proposed correction scheme and the algorithm is amenable for an array processor. The algorithm is shown to be superior to the conventional least squares type algorithms and can operate in near real time.

Data compression control device

Abstract: A control device has a data compression control circuit for compressing data and a data buffer for temporarily storing the data to be written on a recording medium. The data transferred from a CPU is compressed and then recorded on the recording medium. Further provided is a buffer discriminating circuit for discriminating the quantity of data stored in the data buffer to control the data compression control circuit so that the data compressing operation and the operation of recording the data on the recording medium may be conducted in parallel while the amount of transferred data in the data buffer is prevented from becoming less than a predetermined quantity.