Search results for "quantization"

showing 10 items of 253 documents

Deep Learning for Resource-Limited Devices

2020

In recent years, deep neural networks have revolutionized the development of intelligent systems and applications in many areas. Despite their numerous advantages and potentials, these intelligent models still suffer from several issues. Among them, the fact that they became very complex with millions of parameters. That is, requiring more resources and time, and being unsuitable for small restricted devices. To contribute in this direction, this paper presents (1) some state-of-the-art lightweight architectures that were specifically designed for small-sized devices, and (2) some recent solutions that have been proposed to optimize/compress classical deep neural networks to allow their dep…

business.industryComputer scienceDistributed computingDeep learningIntelligent decision support systemRedundancy (engineering)InitializationArtificial intelligencePruning (decision trees)businessAdaptation (computer science)Quantization (image processing)Convolutional neural networkProceedings of the 16th ACM Symposium on QoS and Security for Wireless and Mobile Networks
researchProduct

Blind Robust 3-D Mesh Watermarking Based on Mesh Saliency and QIM Quantization for Copyright Protection

2019

International audience; Due to the recent demand of 3-D models in several applications like medical imaging, video games, among others, the necessity of implementing 3-D mesh watermarking schemes aiming to protect copyright has increased considerably. The majority of robust 3-D watermark-ing techniques have essentially focused on the robustness against attacks while the imperceptibility of these techniques is still a real issue. In this context, a blind robust 3-D mesh watermarking method based on mesh saliency and Quantization Index Modulation (QIM) for Copyright protection is proposed. The watermark is embedded by quantifying the vertex norms of the 3-D mesh using QIM scheme since it offe…

business.industryComputer scienceWatermark robustness[INFO.INFO-TS] Computer Science [cs]/Signal and Image ProcessingData_MISCELLANEOUS[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]020207 software engineeringQuantization index modulationWatermark02 engineering and technologyVertex (geometry)[INFO.INFO-TS]Computer Science [cs]/Signal and Image Processing0202 electrical engineering electronic engineering information engineering020201 artificial intelligence & image processingComputer visionArtificial intelligencebusinessQuantization (image processing)Digital watermarkingSmoothingComputingMilieux_MISCELLANEOUS
researchProduct

Robustness of texture parameters for color texture analysis

2006

This article proposes to deal with noisy and variable size color textures. It also proposes to deal with quantization methods and to see how such methods change final results. The method we use to analyze the robustness of the textures consists of an auto-classification of modified textures. Texture parameters are computed for a set of original texture samples and stored into a database. Such a database is created for each quantization method. Textures from the set of original samples are then modified, eventually quantized and classified according to classes determined from a precomputed database. A classification is considered incorrect if the original texture is not retrieved. This metho…

business.industryCovariance matrixAutocorrelationComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONPattern recognitionMaxima and minimaQuantization (physics)Matrix (mathematics)Computer Science::GraphicsAutocorrelation matrixComputer Science::Computer Vision and Pattern RecognitionPrincipal component analysisRGB color modelComputer visionArtificial intelligencebusinessComputingMethodologies_COMPUTERGRAPHICSMathematicsSPIE Proceedings
researchProduct

Merging the transform step and the quantization step for Karhunen-Loeve transform based image compression

2000

Transform coding is one of the most important methods for lossy image compression. The optimum linear transform - known as Karhunen-Loeve transform (KLT) - was difficult to implement in the classic way. Now, due to continuous improvements in neural network's performance, the KLT method becomes more topical then ever. We propose a new scheme where the quantization step is merged together with the transform step during the learning phase. The new method is tested for different levels of quantization and for different types of quantizers. Experimental results presented in the paper prove that the new proposed scheme always gives better results than the state-of-the-art solution.

business.industryFractal transformVector quantizationTop-hat transformPattern recognitionArtificial intelligencebusinessQuantization (image processing)S transformTransform codingFractional Fourier transformData compressionMathematicsProceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium
researchProduct

Enhanced detection of contrast regions in echocardiograms by adaptive quantization

2002

The statistics of ultrasound echo images are governed by Rayleigh statistics. The authors derive some experimentally verifiable predictions of this theory, compare them with experimental results obtained from echocardiographic images, and derive a new color coding scheme, (adaptive quantization) that is adapted to the signal-dependent noise predicted by theory. This results in some technical advantages and in an improved discrimination of regions of the echo image that are enhanced by echo contrast material. >

business.industryQuantization (signal processing)Physics::Medical PhysicsEcho (computing)ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONContrast (statistics)Noisesymbols.namesakeSignal-to-noise ratioColors of noisesymbolsProbability distributionComputer visionArtificial intelligenceRayleigh scatteringbusinessAlgorithmMathematics[1991] Proceedings Computers in Cardiology
researchProduct

A Digital Watermarking Algorithm Based on Quantization of the DCT: Application on Medical Imaging

2013

International audience; The objective of this paper is to elaborate a new watermarking algorithm applied to the medical imaging. This algorithm must be invisible, robust and has a rate, relatively high, integrating data. The proposed method uses the standard JPEG compression for the integration of medical data. The insertion block is inserted just after the quantization phase. To control identification and eventually the correction (if possible) of the inserted data, we use a series of turbocodes to recover the inserted data, after application of several attacks. The simulation studies are applied on MRI medicals images.

business.industry[INFO.INFO-TI] Computer Science [cs]/Image Processing [eess.IV][INFO.INFO-TI]Computer Science [cs]/Image Processing [eess.IV][ INFO.INFO-TI ] Computer Science [cs]/Image ProcessingMedical imagingJpeg compressionDiscrete cosine transformTurbo codeComputer visionArtificial intelligencebusinessQuantization (image processing)Digital watermarkingAlgorithmTransform codingData compressionMathematics
researchProduct

Space-Frequency Quantization for Image Compression With Directionlets

2007

The standard separable 2-D wavelet transform (WT) has recently achieved a great success in image processing because it provides a sparse representation of smooth images. However, it fails to efficiently capture 1-D discontinuities, like edges or contours. These features, being elongated and characterized by geometrical regularity along different directions, intersect and generate many large magnitude wavelet coefficients. Since contours are very important elements in the visual perception of images, to provide a good visual quality of compressed images, it is fundamental to preserve good reconstruction of these directional features. In our previous work, we proposed a construction of critic…

image orientation analysisMultiresolution analysisVideo RecordingComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISIONImage processingnonseparable transformsmultiresolution analysisRate–distortion theoryWaveletDVMsImage Interpretation Computer-AssistedComputer GraphicsComputer visionQuantization (image processing)image codingimage segmentationMathematicsbusiness.industryWavelet transformNumerical Analysis Computer-AssistedSignal Processing Computer-AssistedWTsData CompressionImage EnhancementComputer Graphics and Computer-Aided Designwavelet transformsdirectional vanishing momentsdirectional transformsArtificial intelligencebusinessAlgorithmsSoftwareImage compressionData compressionIEEE Transactions on Image Processing
researchProduct

Analysis of Received Signal Strength Quantization in Fingerprinting Localization

2020

In recent times, Received Signal Strength (RSS)-based Wi-Fi fingerprinting localization has become one of the most promising techniques for indoor localization. The primary aim of RSS is to check the quality of the signal to determine the coverage and the quality of service. Therefore, fine-resolution RSS is needed, which is generally expressed by 1-dBm granularity. However, we found that, for fingerprinting localization, fine-granular RSS is unnecessary. A coarse-granular RSS can yield the same positioning accuracy. In this paper, we propose quantization for only the effective portion of the signal strength for fingerprinting localization. We found that, if a quantized RSS fingerprint can …

langattomat lähiverkotfingerprintingsisätilapaikannusindoor positioninglcsh:TP1-1185quantizationlcsh:Chemical technologyArticleSensors
researchProduct

Joint local quantization and linear cooperation in spectrum sensing for cognitive radio networks

2014

—In designing cognitive radio networks (CRNs), protecting the license holders from harmful interference while maintaining acceptable quality-of-service (QoS) levels for the secondary users is a challenge effectively mitigated by cooperative spectrum sensing schemes. In this paper, cooperative spectrum sensing in CRNs is studied as a three-phase process composed of local sensing, reporting, and decision/data fusion. Then, a significant tradeoff in designing the reporting phase, i.e., the effect of the number of bits used in local sensing quantization on the overall sensing performance is identified and formulated. In addition, a novel approach is proposed to jointly optimize the linear soft-…

non-ideal reporting channelcognitive radio (CR)quantizationdecision fusioncooperative spectrum sensing
researchProduct

Neural Classification of Compost Maturity by Means of the Self-Organising Feature Map Artificial Neural Network and Learning Vector Quantization Algo…

2019

Self-Organising Feature Map (SOFM) neural models and the Learning Vector Quantization (LVQ) algorithm were used to produce a classifier identifying the quality classes of compost, according to the degree of its maturation within a period of time recorded in digital images. Digital images of compost at different stages of maturation were taken in a laboratory. They were used to generate an SOFM neural topological map with centres of concentration of the classified cases. The radial neurons on the map were adequately labelled to represent five suggested quality classes describing the degree of maturation of the composted organic matter. This enabled the creation of a neural separator classify…

non-parametric classificationComputer science020209 energyHealth Toxicology and Mutagenesislcsh:Medicine02 engineering and technology010501 environmental sciencesengineering.material01 natural sciencesArticleDigital imageSoftwareArtificial Intelligence0202 electrical engineering electronic engineering information engineeringLearningTopological map0105 earth and related environmental sciencesLVQ algorithmLearning vector quantizationArtificial neural networkSOFM neural networkCompostbusiness.industryCompostinglcsh:RPublic Health Environmental and Occupational Health<i>LVQ</i> algorithmengineeringNeural Networks ComputerbusinessClassifier (UML)AlgorithmAlgorithmsSoftwareInternational Journal of Environmental Research and Public Health
researchProduct