Adaptive synthesis of a wavelet transform using fast neural network

Stolarek, Jan

doi:10.2478/v10175-011-0002-5

Cited by 8 publications

(8 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…They are computationally efficient and powerful tools of digital signal processing which find wide application in, e.g., compression, filtering, and encryption of data [24][25][26][27][28][29][30][31][32]. FPOTs usually adopt the computational structures of known transforms with determined basis vectors.…”

Section: Fast Parametric Orthogonal Transformsmentioning

confidence: 99%

Joint compression and encryption of visual data using orthogonal parametric transforms

Puchala¹,

Yatsymirskyy²

2016

Bulletin of the Polish Academy of Sciences Technical Sciences

View full text Add to dashboard Cite

Abstract. In this paper, we introduce a novel method of joint compression and encryption of visual data. In the proposed approach the compression stage is based on block quantization while the encryption uses fast parametric orthogonal transforms of arbitrary forms in combination with a novel scheme of intra-block mixing of data vectors. Theoretical analysis of the method indicates no impact of encryption stage on the effectiveness of block quantization with an additional step of first order entropy coding. Moreover, a series of experimental studies involving natural images and JPEG lossy compression standard were performed. Here, the obtained results indicate a high level of visual content concealment with only a small reduction of compression performance. An additional analysis of security allows to state that the proposed method is resistant to cryptanalytic attacks known for visual data encryption schemes including the most efficient NZCA attack. The proposed method can be also characterized by high computational efficiency and feasibility of hardware realizations. In the view of the above motivation, we propose in this paper a novel method for joint compression and encryption of visual data. In the proposed approach the compression stage uses block quantization, which is the core of existing standards such as: JPEG, MJPEG and MPEG. It allowed to keep compliance with known standards and to obtain comparable performance of data compression. The encryption process is based on fast parametric orthogonal transforms (FPOTs) and a novel scheme of inter-block mixing of data vectors. The proposed method can be characterized by higher combinatorial complexity than existing methods based on linear transforms and, in addition, it is resistant to cryptanalytic attacks known for visual data encryption algorithms.In the research part of the paper it was proved theoretically that the proposed encryption scheme does not affect the compression performance based on block quantization with an additional step of first order entropy coding. Furthermore, a series of experiments concerning joint compression and encryption of natural images were performed by embedding the proposed encryption scheme into JPEG standard. Finally, we discuss in detail the aspects of computational complexity of the method and also provide a critical analysis of its resistance to the known types of cryptanalytic attacks. Existing visual data encryption methodsThe existing methods of visual data encryption can be divided into two groups: naive and selective. The naive approach

show abstract

Section: Fast Parametric Orthogonal Transformsmentioning

confidence: 99%

Joint compression and encryption of visual data using orthogonal parametric transforms

Puchala¹,

Yatsymirskyy²

2016

Bulletin of the Polish Academy of Sciences Technical Sciences

View full text Add to dashboard Cite

show abstract

“…Adding the watermark to the most prominent coefficients of phase of the DFT was proposed by Ruanaidh in [9]. The watermarking algorithm must take into consideration that the image to be real imposes the condition (11). So when watermarking, changes of phase must preserve negative symmetry:…”

Section: Dft Algorithm In the Same Article [2]mentioning

confidence: 99%

“…It has been shown that DWT performed best in all experiments. Further research should include adaptive DWT-based algorithms [11], complex domains, such as DWT-SVD and adaptive transforms, with particular emphasis on DWT adaptive transforms.…”

Section: On Domain Selection For Additive Blind Image Watermarkingmentioning

confidence: 99%

On domain selection for additive, blind image watermarking

Lipiński¹

2012

Bulletin of the Polish Academy of Sciences: Technical Sciences

View full text Add to dashboard Cite

Abstract. Recently, a variety of digital watermarking schemes have been developed for copyright protection of digital images. In robust watermarking, which is used in copyright protection, transform-based algorithms are used to ensure resilience of the watermark to common signal processing attacks. The most frequently used watermarking algorithms for additive watermark embedding involve DCT, DFT, SVD and DWT domains. In this article we verify which domain is optimal for robust, the additive watermark embedding scheme. We demonstrate that in additive watermark embedding the embedding domain plays more important role than the embedding formula.

show abstract

“…Moreover, special types of neural networks are used for adaptive synthesis of the wavelet transform [8], image segmentations [9] and systems identification and diagnosis [10,11], which proves that ANNs can be used in many technical areas. Based on this fact, for the prediction of corrections, the method based on artificial neural networks is examined in detail in this paper.…”

Section: Introductionmentioning

confidence: 99%

Prediction of corrections for the Polish time scale UTC(PL) using artificial neural networks

Luzar¹,

Sobolewski²,

Miczulski³

et al. 2013

Bulletin of the Polish Academy of Sciences: Technical Sciences

View full text Add to dashboard Cite

Abstract. In this paper, the effectiveness of using Artificial Neural Networks (ANNs) for predicting the corrections of the Polish time scale UTC(PL) (Universal Coordinated Time) is presented. In particular, prediction results for the different types of neural networks, i.e., the MLP (MultiLayer Perceprton), the RBF (Radial Basis Function) and the GMDH (Group Method of Data Handling) are shown. The main advantages and disadvantages of using such types of neural networks are discussed. The prediction of corrections is performed using two methods: the time series analysis method and the regression method. The input data were prepared suitable for the above mentioned methods, based on two time series, ts1 and ts2. The designation of prediction errors for specified days and the influence of data quantity for the prediction error are considered. The paper consists of five sections. After Introduction, in Sec. 2, the theoretical background for different types of neural networks is presented. Section 3 shows data preparation for the appropriate type of neural network. The experimental results are presented in Sec. 4. Finally, Sec. 5 concludes the paper.

show abstract

Adaptive synthesis of a wavelet transform using fast neural network

Cited by 8 publications

References 7 publications

Joint compression and encryption of visual data using orthogonal parametric transforms

Joint compression and encryption of visual data using orthogonal parametric transforms

On domain selection for additive, blind image watermarking

Prediction of corrections for the Polish time scale UTC(PL) using artificial neural networks

Contact Info

Product

Resources

About