Image similarity index based on moment invariants of approximation level of discrete wavelet transform

Compressed sensing (CS) has been applied to accelerate magnetic resonance imaging (MRI) for many years. Due to the lack of translation invariance of the wavelet basis, undersampled MRI reconstruction based on discrete wavelet transform may result in serious artifacts. In this paper, we propose a CS-based reconstruction scheme, which combines complex double-density dual-tree discrete wavelet transform (CDDDT-DWT) with fast iterative shrinkage/soft thresholding algorithm (FISTA) to efficiently reduce such visual artifacts. The CDDDT-DWT has the characteristics of shift invariance, high degree, and a good directional selectivity. In addition, FISTA has an excellent convergence rate, and the design of FISTA is simple. Compared with conventional CS-based reconstruction methods, the experimental results demonstrate that this novel approach achieves higher peak signal-to-noise ratio (PSNR), larger signal-to-noise ratio (SNR), better structural similarity index (SSIM), and lower relative error.

Download Full-text

Data Hiding Scheme for Video Watermarking Using Horizontal and Vertical Coefficients of Single Level Discrete Wavelet Transform Method

Journal of Circuits System and Computers ◽

10.1142/s021812662250044x ◽

2021 ◽

pp. 2250044

Author(s):

J. Jerisha Liby ◽

T. Jaya

Keyword(s):

Wavelet Transform ◽

Discrete Wavelet Transform ◽

Similarity Index ◽

Video Watermarking ◽

Data Matrix ◽

Experimental Result ◽

Video Frame ◽

Discrete Wavelet ◽

Transform Method ◽

Single Level

This paper proposes a new watermarking algorithm based on a single-level discrete wavelet transform (DWT). This method initially chooses ‘[Formula: see text]’ number of carrier frames to hide the data. After estimating the carrier frames, each frame is separated into RGB frames. Each R, G, and B frames are decomposed using a single-level DWT. The horizontal and vertical coefficients are selected to embed the watermark information since small changes in the horizontal and vertical coefficients do not highly affect the quality of the video frame. The watermark image pixels are shuffled using a predetermined key before embedding. The shuffled pixels are converted to binary, and they are grouped into three data matrices. Each data matrix is embedded in horizontal and vertical coefficients of the R, G and B frames of the video frame. After embedding the data, the watermarked video is reconstructed using the original approximation coefficients, the embed coefficients, and the original diagonal coefficients. During the extraction process, the watermark is extracted from the horizontal and vertical coefficients of the watermarked video. Experimental result reveals that the proposed method outperforms other related methods in terms of video quality and structural similarity index measurement.

Download Full-text

An Efficient Discrete Wavelet Transform Based Hybrid Image Watermarking Algorithm Using Human Visual Model

10.21203/rs.3.rs-961713/v1 ◽

2021 ◽

Author(s):

Sundhararaj V ◽

Meenakshipriya B ◽

Nirmala Devi P

Keyword(s):

Wavelet Transform ◽

Discrete Wavelet Transform ◽

Image Watermarking ◽

Similarity Index ◽

Input Image ◽

Digital Data ◽

Discrete Wavelet ◽

Weight Factor ◽

Visual Model ◽

Human Visual Model

Abstract More than ever with growing of multimedia technology, the digital data are exchanged in the internet, which can be duplicated by unauthorized users. To avoid this problem the watermarking technology has brought this paper. Performance improvement with compare to existing algorithm is obtained by proposing a new watermarking algorithm based on Human Visual Model (HVM) and Discrete Wavelet Transform (DWT) for securing the digital data and copyright protection. DWT is applied to the input image and at each level of DWT sub bands to embed a watermark image in selected coefficients of the sub band. HVM integrate the weight factor effect of human visualization by considering into the eye’s vision is less sensitivity area, depending on brightness, frequency band and texture areas of the image sub band. In the proposed approach robustly and imperceptibly, DWT and HVM are used for obtainable weight factor, according to human eye perceptual and to determine the optimal strength at which the threshold to embedding reaches the perceptual invisibility of watermarked image to various attacks. Performance is evaluated such as Peak Signal to Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM) and Quality Index (QI) is used to evaluate and achieve the imperceptible watermarked image. Results of PSNR values of hybrid image watermarking are between 49.73db to 36.24db. Experimental results show that our hybrid image watermarking process has enhanced robustness and displays the effectiveness of presenting images watermarking system.

Download Full-text

Integration of Discrete Wavelet Transform, DBSCAN, and Classifiers for Efficient Content Based Image Retrieval

Electronics ◽

10.3390/electronics9111886 ◽

2020 ◽

Vol 9 (11) ◽

pp. 1886

Author(s):

Muhammad Junaid Khalid ◽

Muhammad Irfan ◽

Tariq Ali ◽

Muqaddas Gull ◽

Umar Draz ◽

...

Keyword(s):

Wavelet Transform ◽

Image Retrieval ◽

Discrete Wavelet Transform ◽

Feature Vector ◽

Similarity Index ◽

Machine Learning Algorithms ◽

Image Feature ◽

Content Based Image Retrieval ◽

Support Vector ◽

Discrete Wavelet

In the domain of computer vision, the efficient representation of an image feature vector for the retrieval of images remains a significant problem. Extensive research has been undertaken on Content-Based Image Retrieval (CBIR) using various descriptors, and machine learning algorithms with certain descriptors have significantly improved the performance of these systems. In this proposed research, a new scheme for CBIR was implemented to address the semantic gap issue and to form an efficient feature vector. This technique was based on the histogram formation of query and dataset images. The auto-correlogram of the images was computed w.r.t RGB format, followed by a moment’s extraction. To form efficient feature vectors, Discrete Wavelet Transform (DWT) in a multi-resolution framework was applied. A codebook was formed using a density-based clustering approach known as Density-Based Spatial Clustering of Applications with Noise (DBSCAN). The similarity index was computed using the Euclidean distance between the feature vector of the query image and the dataset images. Different classifiers, like Support Vector (SVM), K-Nearest Neighbor (KNN), and Decision Tree, were used for the classification of images. The set experiment was performed on three publicly available datasets, and the performance of the proposed framework was compared with another state of the proposed frameworks which have had a positive performance in terms of accuracy.

Download Full-text