Abstract: An image processing apparatus in which image data is transformed to spatial frequencies to obtain multiple frequency components, the obtained frequency components are processed, the processed frequency components are inversely transformed to image data, and the number of gray levels of the inversely transformed image data is decreased, and the image processing apparatus comprising a frequency component judgment section for comparing the absolute value of each of specific multiple frequency components in the multiple frequency components obtained after transformation to spatial frequencies with a predetermined value in terms of magnitude, a frequency component changing section for changing each of specific frequency components on the basis of the result of the comparison, and a noise addition section for adding a specific value to each of the specific frequency components having been changed and other frequency components.

Abstract: An object input unit extracts the area of an object as an object area, and a mesh generation unit divides the object area into a plurality of meshes to obtain position information, gradient information, and color information of respective vertices that form the meshes. An optimization unit determines the position information, gradient information, and color information of the vertices by dividing each of the meshes into a plurality of small areas, and performing processing for changing the position information, gradient information, and color information of the vertices so as to locally minimize the differences between color information within the small areas and that within portions of the object area corresponding to the small areas. To divide each of the meshes into a plurality of small areas, the optimization unit changes the number or size of small areas in accordance with the number of times of changing processing.

Abstract: A low complexity visual masking method used as part of an image encoding process is described. The method is suitable for use in JPEG2000 image compression systems. Control weights used for rate allocation are generated based on integer order moments of wavelet transformed coefficients corresponding to a codeblock. The novel rate allocation weight generation method can, and in some embodiments is, combined with an apriori rate allocation algorithm, where allocation of bits to different portions of images is controlled as a function of one or more generated weights. The methods and apparatus of the present invention have the effect of increasing errors in busy areas of an image where they tend to be less noticeable and allocating a higher number of bits to less busy areas than some other systems, e.g., systems which attempt to minimize a mean squared error under a constraint of a user selected output rate.

Abstract: An image-encoding method and apparatus is provided for using a table statistically reflecting a selection frequency of a quantization parameter. The method includes the steps of preparing a table representing a relationship among a target bit quantity, a quantization parameter and a selection frequency as a table for statistically reflecting a frequency of selecting a quantization parameter according to a target bit quantity. The method further includes the steps of searching for a maximum selection frequency among the selection frequencies corresponding to an input target bit quantity by referring to the table, and selecting a quantization parameter corresponding to the input target bit quantity and the searched maximum selection frequency as an optimized quantization parameter to thereby prevent inadequate image-encoding due to large amounts of computations and provide a target bit rate and uniform image quality.

Abstract: A color image residue transform and/or inverse transform method and apparatus, and a color image encoding and/or decoding method and apparatus using the same are provided. The residue transform method includes: obtaining a residue corresponding to the difference of an original image and a predicted image; and transforming the residue by using a relation between residues of color image components. The residue inverse transform method includes: generating a residue for each component by performing residue inverse transform of the residue transformed original image; and restoring the original image by adding a predicted image to the residue of each component. Also, the color image lossless encoding method using the residue transform includes: obtaining a residue corresponding to the difference of an original image and a predicted image; and performing encoding by transforming the residue by using the relation between residues of predetermined components.

Abstract: Systems, methods and computer readable media are disclosed for improving compression efficiency and quality in a remote session via tile image classification and variable encoding. A server determines a set of codecs that are shared by both the server and a corresponding client. Then, when it receives an image, it determines whether classification of the image is required. Where classification of the image is not required, the server sends the client the image, either uncompressed or compressed with a default codec and default fidelity. Where classification of the image is required, the server classifies the image (e.g. the image comprises either text or photograph), and based on that classification determines a codec with which to encode the image, and a fidelity to use on the encoding. The server performs that encoding with the codec and the fidelity, and then sends this encoded image to the client.

Abstract: In a rate control system, an MMU temporarily stores in a mass storage an input bit stream including compressed and encoded data which is compressed and encoded according to the JPEG2000 standard. The MMU then reads out data from the mass storage and outputs it to a multiplexer according to readout control signals. The multiplexer multiplexes the data and outputs the result as an output bit stream. A bit truncation controller comprises an image quality controller for selecting a target to be coded according to target image quality; and a rate controller for controlling a rate according to a target rate. A layer splitting controller outputs to the MMU the readout control signal for generating an output bit stream which is divided into a plurality of layers.

Abstract: A signal processing method include steps initializing a residual data signal representative of an acquired data signal, determining a significant coefficient corresponding to the residual data signal, updating the residual data signal using the significant coefficient to generate updated residual data signal, iteratively determining significant coefficients to generate a plurality of significant coefficients using the updated residual data signal, updating the plurality of significant coefficients by using a successive approximation technique, to improve the numerical accuracy of the significant coefficients and reconstructing a data signal using the updated plurality of significant coefficients.

Abstract: Provided are a method and an apparatus for selectively encoding/decoding point sequences to maximize bit efficiency of a lightweight application scene representation (LASeR) binary stream. The point sequence encoding method includes the steps of: for each point sequence, (a) selecting one of exponential-Golomb (EG) encoding and fixed length (FL) encoding schemes; (b) when the FL encoding scheme is selected, encoding the point sequence using the FL encoding scheme to generate a binary stream; and (c) when the EG encoding scheme is selected, encoding the point sequence using the EG encoding scheme to generate a binary stream. The binary stream includes a flag indicating which encoding scheme is selected and a parameter k, with which the EG encoding can be most effectively performed, when the EG encoding scheme is selected. According to the encoding method, LASeR point sequences can be efficiently encoded and, during a decoding process, a large overhead is not incurred to a decoder (terminal).

Abstract: An image coding method includes transforming a color space of a color image from a first color space to a second color space to generate a color space transformed color image (S12), removing part of samples included in the color space transformed color image to generate a subsampled color image (S13), coding the subsampled color image to generate a coded color image (S14), determining an upsampling coefficient used for upsampling (S16), determining a color space inverse transform coefficient for inversely transforming the color space from the second color space to the first color space (S17), and outputting the coded color image, the upsampling coefficient, and the color space inverse transform coefficient (S19).

Abstract: An apparatus for downloading a multimedia file including a file converter configured to convert a larger multimedia file into a smaller multimedia file by deleting prescribed medial information based on a media format contained in the larger file. Further, the file converter converts the larger file into the smaller file by deleting multi bit-rate media not corresponding to a selected bit rate or by deleting enhancement layer video.

Abstract: Low complexity (16 bit arithmetic) H.264 video compression replaces a single quantization table for all quantization parameters with multiple quantization tables and thereby equalizes quantization shifts and round-off additions; this eliminates the need for 32-bit accesses.

Abstract: The present invention relates to a process of bringing at least one subject data set into registration or conformity with a reference data set by electronic methods, each data set being a representation of a respective object.

Abstract: A decoding apparatus includes a random-number generating section and a decoding section. The random-number generating section generates random numbers according to distribution of original data corresponding to respective quantization indexes. The decoding section generates decoded data on a basis of the random numbers generated by the random-number generating section.

Abstract: A method and apparatus is disclosed herein for decoding data (e.g., video data) using transforms. In one embodiment, the decoding process comprises scaling a block of coefficients using a scaling factor determined for each coefficient by computing an index for said each coefficient and indexing a look-up table (LUT) using the index. The index is based on a quantization parameter, a size of the block of coefficients, and a position of said each coefficient within the block. The method also comprises applying a transform to the block of scaled coefficients.

Abstract: Methods and systems for watermarking of digital images are presented. In one aspect, a method of embedding information in a digital image includes transforming the digital image to a set of coefficient blocks having coefficients in a frequency domain, embedding a watermark-indicator in one or more of the coefficient blocks, and embedding a watermark in one or more watermark blocks that have a predetermined number of coefficient blocks. The embedded watermark is substantially invisible in the output watermarked digital image. Another aspect is a method of extracting a watermark from a digital image that includes transforming a digital image to a coefficient matrix, determining if the digital image is watermarked based on a predetermined watermark-indicator, and retrieving a watermark from a projected watermark block of a projected watermark block distribution.

Abstract: The disclosure relates to encoding and decoding of digital data, and in particular to lossless arithmetic encoding and decoding of digital data representing audio, image or video data. A probability density function used for lossless arithmetic encoding of digital data is controlled by employing one or more parameters that changes over the set of data to be encoded. A parametric model in the form of an envelope function describes the spread of quantization indices derived from the data in a transform domain. By transmitting the one or more parameters together with the arithmetically encoded data, a receiving decoder may decode the data by exploiting the same parametric model as used by the encoder.

Abstract: An encoder, a decoder, an encoding method and a decoding method are provided. The encoder includes a reversible color transform module, a difference pulse code modulation (DPCM) intra prediction module, a quantization module, a reversible frequency transform module and an entropy coding module. The reversible color transform module performs a reversible color transform to output a transformed video signal according to an input video signal. The DPCM intra prediction module performs a DPCM intra prediction to output a least residual according to the transformed video signal. The quantization module performs a quantization operation to output a quantization coefficient according to the least residual. The reversible frequency transform module performs a reversible frequency transform to output a frequency coefficient according to the quantization coefficient. The entropy coding module performs entropy coding to output a compression bit stream according to the frequency coefficient.

Abstract: This invention implements a fast lossless transform almost free from a delay with a small calculation amount. The lossless transform can be used to perform lossless coding and lossy coding quickly. A first calculation unit multiplies data D0, D1, D2, and D3 input to the input terminals by respective weighting coefficients {a0, a1, a2, a3} of {½, ?½, ?½, ?½}, and summates the products. A rounding unit in the first calculation unit rounds the sum into an integer and outputs the integer value E. A second calculation unit multiplies the value E by weighting coefficients {b0, b1, b2, b3} of {?1, 1, 1, 1} set for the respective input data, and adds the products to the respective input data. This invention sets, for the relationship between the first and second weighting coefficients, a condition that a0*b0+a1*b1+a2*b2+a3*b3=?2 or 0.

Abstract: A size limiter module used to limit a file size of an encoded file is disclosed. The size limiter module includes a size limiter set-up calculator and a size limiter processing engine. The size limiter calculator is configured to identify a maximum bit size for a set of data to be encoded based on a maximum value for the file size of the encoded file. The size limiter processing engine is configured to be in communication with the size limiter set-up calculator to process the set of data to be encoded and to monitor a cumulative bit size for the set of data according to an order. When the cumulative bit size is greater than a maximum bit size, the size limiter processing engine inserts end of block data into the set of data. The end of block data signals the substitution of a null value for all data units subsequent to the end of block data.

Abstract: An image encoding apparatus and method, and an image decoding apparatus and method. An image encoding apparatus includes: an image input unit receiving an image to be encoded; a transform encoding unit transform coding the image provided from the image input unit to generate a first bitstream; a pulse code modulation (PCM) encoding unit PCM coding the image provided from the image input unit to generate a second bitstream; and a selector selecting and outputting one of the first bitstream and the second bitstream depending on characteristics of the image.

Abstract: Methods and systems for compressing an image are described. A plurality of transformed and quantized values associated with a block of image data is accessed. The block corresponds to a position within the image. A count of the number of bits encoded during run-length encoding of the block is made. Run-length encoding of the block is concluded should the count reach a limit.

Abstract: The invention relates to the field of video compression and, more specifically, to a video encoding method applied to an input sequence of frames in which each frame is subdivided into blocks of arbitrary size.

Abstract: An image compression system and method is described, which makes use of the symmetry found in faces and heads to perform the compression. The image is divided along the line of symmetry, and pairs of corresponding pixels on the two divided sides are determined. A weighted average and a weighted variance of the pixel values of the pairs is computed, and is used to encode the image. A transform such as the Karhunen-Loeve transform is used to compute the weighted averages and variances.

Abstract: A coding apparatus for embedding second data into first data without deteriorating the first data includes a memory for storing at least partial data of the first data. An embedding coder embeds the second data into the first data by rearranging, according to the second data, at least partial data of the first data stored in the memory. A decoding apparatus for decoding coded data in to the original first data and the original second data without deteriorating the quality of the data includes a relation calculating unit for calculating a relation between first partial data and second partial data of the coded data. A decoder decodes the coded data into the original first data by moving the first partial data and the second partial data of the coded data based on the relation calculated by the relation calculating unit, and decodes the second data that is embedded in the coded data according to the movement.

Abstract: Analysis filter processing is recursively repeated on horizontal and vertical low frequency component coefficients obtained as a result of the analysis filter processing, until a predetermined decomposition level is reached. Coefficients obtained during a process of computation in the analysis filter processing and, except for a preset decomposition level, horizontal and vertical low frequency component coefficients obtained as a result of the computation in the analysis filter processing are stored in a first storage section independently for each decomposition level. The coefficients stored in the first storage section are read as appropriate and the read coefficients are supplied for the analysis filter processing. The horizontal and vertical low frequency component coefficients of the preset decomposition level are stored in a second storage section. The coefficients stored in the second storage section are read as appropriate. The read coefficients are supplied for the analysis filter processing.

Abstract: An information processing apparatus includes a device for performing horizontal analysis filtering, a device for performing vertical analysis filtering, and a device for horizontal control. The information processing apparatus also includes a first control device. This device allows the device for horizontal control to repetitively perform the horizontal control until the horizontal analysis filtering is performed on all of columns of a processing object line. The information processing apparatus further includes a device for vertical control and a second control device that allows the device for vertical control to repetitively perform the vertical control until the vertical analysis filtering is performed on all of lines.

Abstract: An image information decoding method for decoding compressed image information which has been coded via a process including dividing an input image signal into blocks, performing an orthogonal transform on the blocks on a block-by-block basis, and quantizing resultant orthogonal transform coefficients. The decoding process includes performing dequantization such that a quantization parameter is weighted by an addition operation, and the dequantization is performed on each chroma components of the quantized coefficients using said weighted quantization parameter, and performing an inverse orthogonal transform.

Abstract: The invention relates to a method of encoding an image divided into non-overlapping macroblocks themselves divided into non-overlapping blocks of N by N pixels into a binary stream. It comprises the steps of: —transforming each of the block into a transformed block of coefficients comprising one low frequency coefficient and N2-1 coefficients, called high frequency coefficients, of higher frequencies than the low frequency; —quantizing each coefficient of each of the transformed blocks with a quantizing parameter; —encoding the quantized coefficients into a binary stream. According to the invention, the low frequency coefficients of the transformed blocks are quantized with a same quantizing parameter, called first quantizing parameter.

Abstract: An image processor includes a frequency transform unit performing frequency transform on a first pixel block as a target block, and a pre-filter performing prefiltering with a region which overlaps with plural unit regions for processing by the frequency transform unit as a unit region for processing, before frequency transform is performed. The pre-filter performs prefiltering on a second pixel block being a predetermined number of pixels each larger horizontally and vertically than the first pixel block as a target block. The pre-filter performs prefiltering sequentially on a plurality of second pixel blocks aligned horizontally. The number of pixel signals in a vertical direction within a group of pixel signals continuously inputted to the pre-filter for prefiltering is equal to the number of rows in the second pixel block.

Abstract: Methods for embedding digital watermarks in compressed video include perceptual adapting a digital watermark in predicted and non-predicted data based on block activity derived from the compressed video stream, embedding in predicted objects in a video stream having separately compressed video objects, and bit rate control of watermarked video.

Abstract: A method and apparatus compress projection data and store the compressed projection data in a rotatable part that is mounted for rotation within a stationary part. The data acquisition source, compressor and storage device are connected to the rotatable part. The compressor compresses projection data samples provided by the data acquisition source to form compressed packets. The compressed packets are stored in the storage device, for example one or more solid state drives mounted on the rotatable part. A data access array contains information related to the location of the stored compressed packets. Compressed packets are retrieved and transferred across the interface to the stationary part. A decompressor at the stationary part decompresses the received compressed packets to form decompressed samples of the corresponding projection data. This abstract does not limit the scope of the invention as described in the claims.

Abstract: This invention solves problems due to employing error degraded data in digital processing. It particularly solves multi-generation problems wherein transform data degrade during each inverse transform and forward transform cycle even without any processing due to the rounding and clipping errors. It provides methods, systems and devices for reduced-error processing of transform-coded data. After inverse transformation of transform data, high-precision numbers are manipulated. The converting to integers and clipping to an allowed range steps are executed at any stage in the manipulation to obtain integer representation of the inverse transformed data such as for displaying of the data. However, further processing including forward transforming back to the transform domain is executed on the high-precision numbers. Thus, the rounding and clipping errors are not present in the processed data.

Abstract: Design techniques and their circuit designs for versatile and scalable video coding are proposed, in which the inner product operation in the transform coding expression is taken apart into a series of add and shift operations, and the expression is partitioned into several sub-expressions. Taking each order of the add-and-shift series as a clock period, several adders/subtractors and a shift accumulator are used to carry out corresponding additions/subtractions and shift operations to finish the inner product operations. The calculating result is accumulated until all the orders are finished. The final accumulated value is the output of the transform coding. Data throughput rate can thus be enhanced to save the power consumption of the circuit system. Moreover, a dynamic guarded computation method and a switching power suppression technique are provided to further lower the power consumption.

Abstract: An image representation format for representing a digital image comprises: image information stored as a bitstream representing sequential image blocks, each block comprising one or more components, each component comprising one or more data units and each data unit being represented as a Huffman-coded stream of coefficients of basis functions, wherein a zeroth order coefficient is represented as a difference to the previous zeroth order coefficient of the corresponding component, and a block information table, which comprises: indicators to the first coefficient of a specified order of each image block in said bitstream, information indicating the number of bits in the bitstream between adjacent coefficients of said specified order of the image block, and the zeroth order coefficient of at least the first data unit of each component, said zeroth order coefficient being represented in a non-differential form.

Abstract: An image decoding apparatus includes an information acquisition section that acquires information of a bit-plane truncation position in an encoded stream from the encoded stream generated by performing bit-plane encoding on a transform coefficient, a decoding process section that performs a decoding process on the encoded stream and generates a transform coefficient, and a significant coefficient detection section that detects whether the transform coefficient generated in the decoding process section is significant.

Abstract: The invention concerns a method for processing images, texture pattern blocks representative of a typical texture of the images being stored in a memory, the method comprising the following steps: transforming the pixel grey levels of a current block into a current block of frequency coefficients using a predefined transform; building up a merger block, each frequency coefficient positioned at a defined place of the merger block being equal either to the frequency coefficient positioned at the same place of the current block, or to the frequency coefficient positioned at the same place of a texture pattern block among the texture pattern blocks stored in the memory; selecting a merger block according to a predefined function (J); and replacing in the image the current block by the selected merger block. The invention also concerns an electronic device for processing the corresponding images.

Abstract: Some representative embodiments are directed to systems and methods for compressing a data set. In one embodiment, a method comprises receiving a frame of data to be encoded, generating a residual frame that represents a difference between the received frame and one or several reference frames, performing a respective sum of absolute differences (SAD) calculation for each block within the residual frame, and applying a transform function to each data value within the residual frame, wherein the transform function is at least a function of a SAD value calculated for the block containing the respective data value.

Abstract: There is disclosed an image processing apparatus which implements an image process suitable for an effect for compositing or switching images, and an image process that allows smooth high-speed playback of even an image in motion. The apparatus has band segmentation means for segmenting an image signal into a plurality of frequency band components, and image composition means for, after the band segmentation means segments input first and second image signals, outputting a third image signal by replacing image data for respective segmented band components. When image data, which are recorded while being segmented into a plurality of frequency band components, are composited for respective band, and the composite image data is output, image data obtained by compositing a plurality of image frequency components is decoded and played back in a high-speed playback mode.

Abstract: An encoding method encodes, with an encoding device, second data made up of a plurality of first data representing predetermined numerical values. The encoding method includes taking a significant digit having, of the numerical values represented by each of the first data, the greatest absolute value, as a maximum significant digit of the first data. The encoding method also includes outputting, with the encoding device, code indicating the maximum significant digit and code indicating whether or not the maximum significant digit has changed. The encoding method additionally includes outputting code indicating the absolute value of a numerical value represented by the first data. Further, the encoding method includes outputting code indicating the sign of a numerical value represented by the first data.

Abstract: In certain embodiments, overlap operators are applied during encoding and/or decoding of digital media, where the overlap operators have reduced DC gain mismatch and/or DC leakage between interior overlap operators and overlap operators at the edge and/or corner. In other embodiments, information indicating a selected tile boundary option for overlap processing can be encoded and/or decoded. The selected tile boundary option indicates one of a hard tile boundary option and a soft tile boundary option for processing with overlap operators. Overlap transform processing can then be applied based at least in part on the selected tile boundary option.

Abstract: Image processing is carried out by transforming image data to image data having spatial frequency components, by judging whether the absolute values of the coefficients of the predetermined frequency components of the transformed image data are equal to or more than a predetermined value, by modifying the coefficients of the AC components of the image data depending on the result of the judgment, by inversely transforming the image data having the modified coefficients to image data having spatial coordinate components, by comparing the gray levels of the inversely transformed image data with predetermined values, and by transforming the gray levels to quantized levels corresponding to the result of the comparison. By carrying out proper image processing corresponding to the characteristics of the image, the number of gray levels of the image can be decreased while the characteristic portions of the original image are maintained properly.

Abstract: An image processing apparatus includes a size acquiring unit that acquires a block size of an orthogonal transformation from encoded data of an image compressed by performing the orthogonal transformation and quantization for each block; a filter selecting unit that selects a filter for each block, the filter reducing an encoding distortion from each block of a decoded image; and an encoding-distortion reducing unit that reduces an encoding distortion by the filter on all pixels within the block. The filter selecting unit selects the filter so that a strength of the filter strengthens by increasing of the block size, or selects the filter so that the number of taps of the filter decreases by reduction of the block size.

Abstract: Close approximations of ideal linear transforms, such as the forward and inverse discrete cosine transformation (DCT), are formed with minimum complexity using fixed-point arithmetic. The transformation is decomposed into a smaller set of transformations (e.g., the LLM decomposition of the DCT). The multiplication factors of the decomposition are then approximated by a fixed-point representation. However, instead of simply applying scaling and rounding operations to produce fixed-point approximations closest to the multiplication factors themselves, fixed-point multiplication factors are chosen that have the effect (after the cascaded operations of the various stages of decomposition) of being the closest feasible approximations of the entries in the resulting complete ideal transformation matrix.

Abstract: In the present invention, a first pixel data group of several pixels existing in block into which an image is divided is divided, thereby generating several second pixel data groups. In the second pixel data group, an orthogonal transformation is performed in several layers using only the predetermined number of coefficients from the lowest frequency side from among the coefficients obtained by the orthogonal transformation. The predetermined number of coefficients from the lowest frequency side from among the coefficients obtained by the orthogonal transformation in the highest layer, and the coefficients that are obtained in a layer other than the highest layer not used for the orthogonal transformation are encoded, thereby generation the compressed data. Accordingly, encoding is performed while separating lower frequency components from higher frequency components.

Abstract: The invention describes a process to prevent counterfeiting or alteration of a printed or engraved surface, characterized by the incorporation of a signature of the form of a digital mark into parts or the entire document, and in particular a digital mark technology to hide information in an invisible way through over-printing by using a method called asymmetric amplitude modulation. This method can be applied to any type of printed material such paper, packaging, or any other surface. Visible information can also be printed over the digital mark. As an application example, applied to a paper document the digital mark can be used to guarantee the document authenticity, as it would be destroyed by a copy process.

Abstract: In a digital signal processing system, a method for selecting a transform function to apply to an input signal based on characteristics of the signal, and for self-adjusting criteria which are used in selecting a transform function to apply to a subsequent signal. Characteristics are obtained from the signal. The characteristics are compared to adjustable criteria which are used in selecting a transform function. Differing criteria are maintained for the different selectable transform functions. A record is maintained of transform functions selected and the particular characteristics that caused the selection. Based on the ability of a transform function to minimally define the coded signal, an inverse transform function is selected to decode the signal. The criteria used in selecting a transform function to apply to a subsequent signal are adjusted based on a quality measure of the decoded signal and the record of selected transform functions.

Abstract: Coding quantized transform coefficients as occur in image compression combines a position identifying method to identify the relative position of clusters of consecutive non-zero-valued coefficients, an amplitude event identifying method to identify amplitude events in the clusters of non-zero-valued coefficients, and a coding method to code the position events, amplitude events, and signs of the amplitudes in the clusters. The method is particularly applicable to a series of quantized transform coefficients where clusters are likely to occur.

Abstract: A method and apparatus is disclosed for efficiently encoding data representing a video image, thereby reducing the amount of data that must be transferred to a decoder. The method includes transforming data sets utilizing a tensor product wavelet transform which is capable of transmitting remainders from one subband to another. Collections of subbands, in macro-block form, are weighted, detected, and ranked enabling prioritization of the transformed data. A motion compensation technique is performed on the subband data producing motion vectors and prediction errors which are positionally encoded into bit stream packets for transmittal to the decoder. Subband macro-blocks and subband blocks which are equal to zero are identified as such in the bit stream packets to further reduce the amount of data that must be transferred to the decoder.

Abstract: A system and method for compressing digital images is provided. The method includes receiving an input image having a dimension d and a target dimension constraint dt, and determining a number of resolution levels R for generating an image having corresponding dimension dr. The difference between dr and dt is less than the difference between dr and the corresponding dimension of an image compressed at the next or the previous resolution level. The method additionally includes determining the scale factor S as a function of R and the dimension d of the input image, and scaling the input image in accordance with the scale factor S. The method further includes compressing the scaled input image with the R resolution levels for producing a compressed image including a lowest resolution image. The lowest resolution image has a corresponding dimension that is substantially the same as dt.