Dr. P. Mathivanan - received his B.E degree in Electronics and Communication Engineering from St. Peter’s Engineering College, Affiliated to Anna University, Chennai, India in 2009. He received his M.E degree in Applied Electronics from Sriram Engineering College, Affiliated to Anna University, Chennai, India, 2012 and he completed his PhD in field of data hiding from the department of Information and Communication Engineering at Anna University, Chennai,2020. Currently he is working as Assistant Professor in the Department of Electronics and Communication Engineering at Velammal Engineering College, Chennai, India. His research interests are Data Hiding,Signal Processing, and Image Processing. Address: Tiruvallur, Tamil Nadu, India
The paper presents a colour image steganography, in which patient diagnose data is appended along... more The paper presents a colour image steganography, in which patient diagnose data is appended along with diagnosis information. The confidential information in the form of audio signal convolved with diagnoses data by using multi-bit embedding process. Each represented bit of both ECG and audio samples are convolved with each other and the resultant samples are given to modulo division process to separate quotient and remainder value. The quotient data is then embedded within selective locations of colour image component. Eventually, the reminder value will act as a key to retrieve original information. The performance of proposed steganography process is validated by estimating various quantitative and statistical metrics, such as peak signal to noise ratio (PSNR), normalized cross correlation (NCC), structural similarity index (SSIM) and signal to noise ratio (SNR). With increase in payload size the result shows better robustness and imperceptibility between cover and stego image. T...
Abstract This paper uses a novel color image stego-crypto technique to enhance the capacity, secu... more Abstract This paper uses a novel color image stego-crypto technique to enhance the capacity, security and robustness of data embedding process. The proposed scheme uses a base64 encoding technique and along with QR code generator to transform the secret data, i.e., watermark to a 2D binary matrix, i.e., QR code that considered as a first-level encryption process. Followed by, embedding of individual QR code into the color image components, namely red, green and blue using a dynamic bit replacement technique that guarantees a second level of security. Subsequently, scrambling of random block-pixels and the encryption of logistic chaos are performed. The experimental outcome of three levels of security scheme has proved to be with a much larger key space and a strong resistance to brute force, statistical and differential attacks. The correlation between adjacent pixels is observed to be uniformly distributed in the cipher image and the information entropy value is close to an ideal value of 8. From the simulation results, it is observed that the proposed security scheme has outperformed the existing standard encryption techniques within a single round of iteration in terms of security and robustness.
International Journal of Imaging Systems and Technology, 2020
The article presents a novel ECG steganography scheme based on the tunable Q‐factor wavelet trans... more The article presents a novel ECG steganography scheme based on the tunable Q‐factor wavelet transformation (TQWT) and also singular value decomposition (SVD) techniques that ensure better safety and confidentiality of patient information. Initial parameters such as Q, r, and J are used to decompose the cover signal into individual frequency sub‐bands with the tunable Q‐factor wavelet transform (TQWT). The singular value decomposition (SVD) technique is used to further decompose high‐frequency sub‐band coefficients into singular values. The watermark information is then embedded with high‐frequency sub‐band coefficients by involving the quantization process. The performance of this proposed system is successfully evaluated by considering various metrics, such as peak signal to noise ratio (PSNR), structural similarity index (SSIM), percentage residual difference (PRD), and bit error rate (BER). The simulation results of the proposed scheme are observed to be better than other traditional algorithms.
Journal of The Institution of Engineers (India): Series B, 2017
One of the challenging biometric authentication applications is gender identification and age cla... more One of the challenging biometric authentication applications is gender identification and age classification, which captures gait from far distance and analyze physical information of the subject such as gender, race and emotional state of the subject. It is found that most of the gender identification techniques have focused only with frontal pose of different human subject, image size and type of database used in the process. The study also classifies different feature extraction process such as, Principal Component Analysis (PCA) and Local Directional Pattern (LDP) that are used to extract the authentication features of a person. This paper aims to analyze different gender classification techniques that help in evaluating strength and weakness of existing gender identification algorithm. Therefore, it helps in developing a novel gender classification algorithm with less computation cost and more accuracy. In this paper, an overview and classification of different gender identification techniques are first presented and it is compared with other existing human identification system by means of their performance.
Abstract The paper presents a novel and enhanced bio-signal (ECG) cryptographic technique in whic... more Abstract The paper presents a novel and enhanced bio-signal (ECG) cryptographic technique in which a QR code has been successfully utilized as a data container. The bio-signal encryption process considers ECG samples in the form of integers in order to create corresponding binary pair sequences that are converted as equivalent ciphertext information. Subsequently, the generated ciphertexts are transformed into QR code by using a QR code generator/reader in order to ensure further security. The encryption process generates a binary key for each binary pair that will be used during signal decryption. The imperceptibility of a reconstructed ECG signal is evaluated by analyzing performance metrics in terms of PSNR and PRD values. The average PSNR values are found to be above 42 dB, whereas PRD values observed were less than 3.43 percent. It is further proved that the proposed technique can encrypt a maximum of 226 kb of diagnose data by using merely seven QR codes.
International Journal of Imaging Systems and Technology, 2020
The article presents a novel ECG steganography scheme based on the tunable Q-factor wavelet trans... more The article presents a novel ECG steganography scheme based on the tunable Q-factor wavelet transformation (TQWT) and also singular value decomposition (SVD) techniques that ensure better safety and confidentiality of patient information. Initial parameters such as Q, r, and J are used to decompose the cover signal into individual frequency sub-bands with the tunable Q-factor wavelet transform (TQWT). The singular value decomposition (SVD) technique is used to further decompose high-frequency sub-band coefficients into singular values. The watermark information is then embedded with high-frequency sub-band coefficients by involving the quantization process. The performance of this proposed system is successfully evaluated by considering various metrics, such as peak signal to noise ratio (PSNR), structural similarity index (SSIM), percentage residual difference (PRD), and bit error rate (BER). The simulation results of the proposed scheme are observed to be better than other traditional algorithms. K E Y W O R D S ECG steganography, scaling factor and performance metrics, singular value decomposition (SVD), tunable Q-factor wavelet transform (TQWT)
This paper uses a novel color image stego-crypto technique to enhance the capacity, security and ... more This paper uses a novel color image stego-crypto technique to enhance the capacity, security and robustness of data embedding process. The proposed scheme uses a base64 encoding technique and along with QR code generator to transform the secret data, i.e., watermark to a 2D binary matrix, i.e., QR code that considered as a first-level encryption process. Followed by, embedding of individual QR code into the color image components, namely red, green and blue using a dynamic bit replacement technique that guarantees a second level of security. Subsequently, scrambling of random block-pixels and the encryption of logistic chaos are performed. The experimental outcome of three levels of security scheme has proved to be with a much larger key space and a strong resistance to brute force, statistical and differential attacks. The correlation between adjacent pixels is observed to be uniformly distributed in the cipher image and the information entropy value is close to an ideal value of 8. From the simulation results, it is observed that the proposed security scheme has outperformed the existing standard encryption techniques within a single round of iteration in terms of security and robustness.
The proposed process involves color image steganography using DWT-BRA (Bit Replacement Algorithm)... more The proposed process involves color image steganography using DWT-BRA (Bit Replacement Algorithm). The Discrete Wavelet Transform (DWT) is used to maintain image quality, imperceptibility, security and robustness. The cover image is decomposed into individual sub band for data embedding. The embedding process involves BRA technique to hide secret data. In the frequency subband coefficient only selective bit of the coefficient is replaced by the secret bit. The embedded data is extracted using bit extraction process and the performance analysis like PSNR and SSIM are evaluated. The proposed method obtained a 63.25 dB PSNR in high frequency subband (HH) and SSIM value of 1.6݁ ିହ .
The paper presents a colour image steganography, in which patient diagnose data is appended along... more The paper presents a colour image steganography, in which patient diagnose data is appended along with diagnosis information. The confidential information in the form of audio signal convolved with diagnoses data by using multi-bit embedding process. Each represented bit of both ECG and audio samples are convolved with each other and the resultant samples are given to modulo division process to separate quotient and remainder value. The quotient data is then embedded within selective locations of colour image component. Eventually, the reminder value will act as a key to retrieve original information. The performance of proposed steganography process is validated by estimating various quantitative and statistical metrics, such as peak signal to noise ratio (PSNR), normalized cross correlation (NCC), structural similarity index (SSIM) and signal to noise ratio (SNR). With increase in payload size the result shows better robustness and imperceptibility between cover and stego image. The absorbed SNR value of the extracted signal is absorbed to be 23 dB with minimum signal degradation.
In concepts such as connected health, patient data needs to be protected when they are transmitte... more In concepts such as connected health, patient data needs to be protected when they are transmitted over the internet to the caregiver. The current letter proposes embedding the QR code of patient data into discrete wavelet transform coefficients of the cover ECG Signal. The imperceptibility of hidden data is estimated by performance metrics such as peak signal-to-noise ratio, percentage residual difference, and Kullback-Leibler distance. The proposed approach is demonstrated on MIT-BIH database. Watermarking in the low coefficient values deteriorate the signal less and results in lower bit errors. It is shown that larger watermarks can be accommodated with a larger version of QR code for more or less the same imperceptibility. It is found that QR code version 40 allows embedding of maximum 2632 bytes of patient data with zero bit errors. Since QR code has an inbuilt error correction code, it provides additional layer of security to patient data and the proposed approach can be used for secure transfer of patient data.
The paper presents a colour image steganography, in which patient diagnose data is appended along... more The paper presents a colour image steganography, in which patient diagnose data is appended along with diagnosis information. The confidential information in the form of audio signal convolved with diagnoses data by using multi-bit embedding process. Each represented bit of both ECG and audio samples are convolved with each other and the resultant samples are given to modulo division process to separate quotient and remainder value. The quotient data is then embedded within selective locations of colour image component. Eventually, the reminder value will act as a key to retrieve original information. The performance of proposed steganography process is validated by estimating various quantitative and statistical metrics, such as peak signal to noise ratio (PSNR), normalized cross correlation (NCC), structural similarity index (SSIM) and signal to noise ratio (SNR). With increase in payload size the result shows better robustness and imperceptibility between cover and stego image. T...
Abstract This paper uses a novel color image stego-crypto technique to enhance the capacity, secu... more Abstract This paper uses a novel color image stego-crypto technique to enhance the capacity, security and robustness of data embedding process. The proposed scheme uses a base64 encoding technique and along with QR code generator to transform the secret data, i.e., watermark to a 2D binary matrix, i.e., QR code that considered as a first-level encryption process. Followed by, embedding of individual QR code into the color image components, namely red, green and blue using a dynamic bit replacement technique that guarantees a second level of security. Subsequently, scrambling of random block-pixels and the encryption of logistic chaos are performed. The experimental outcome of three levels of security scheme has proved to be with a much larger key space and a strong resistance to brute force, statistical and differential attacks. The correlation between adjacent pixels is observed to be uniformly distributed in the cipher image and the information entropy value is close to an ideal value of 8. From the simulation results, it is observed that the proposed security scheme has outperformed the existing standard encryption techniques within a single round of iteration in terms of security and robustness.
International Journal of Imaging Systems and Technology, 2020
The article presents a novel ECG steganography scheme based on the tunable Q‐factor wavelet trans... more The article presents a novel ECG steganography scheme based on the tunable Q‐factor wavelet transformation (TQWT) and also singular value decomposition (SVD) techniques that ensure better safety and confidentiality of patient information. Initial parameters such as Q, r, and J are used to decompose the cover signal into individual frequency sub‐bands with the tunable Q‐factor wavelet transform (TQWT). The singular value decomposition (SVD) technique is used to further decompose high‐frequency sub‐band coefficients into singular values. The watermark information is then embedded with high‐frequency sub‐band coefficients by involving the quantization process. The performance of this proposed system is successfully evaluated by considering various metrics, such as peak signal to noise ratio (PSNR), structural similarity index (SSIM), percentage residual difference (PRD), and bit error rate (BER). The simulation results of the proposed scheme are observed to be better than other traditional algorithms.
Journal of The Institution of Engineers (India): Series B, 2017
One of the challenging biometric authentication applications is gender identification and age cla... more One of the challenging biometric authentication applications is gender identification and age classification, which captures gait from far distance and analyze physical information of the subject such as gender, race and emotional state of the subject. It is found that most of the gender identification techniques have focused only with frontal pose of different human subject, image size and type of database used in the process. The study also classifies different feature extraction process such as, Principal Component Analysis (PCA) and Local Directional Pattern (LDP) that are used to extract the authentication features of a person. This paper aims to analyze different gender classification techniques that help in evaluating strength and weakness of existing gender identification algorithm. Therefore, it helps in developing a novel gender classification algorithm with less computation cost and more accuracy. In this paper, an overview and classification of different gender identification techniques are first presented and it is compared with other existing human identification system by means of their performance.
Abstract The paper presents a novel and enhanced bio-signal (ECG) cryptographic technique in whic... more Abstract The paper presents a novel and enhanced bio-signal (ECG) cryptographic technique in which a QR code has been successfully utilized as a data container. The bio-signal encryption process considers ECG samples in the form of integers in order to create corresponding binary pair sequences that are converted as equivalent ciphertext information. Subsequently, the generated ciphertexts are transformed into QR code by using a QR code generator/reader in order to ensure further security. The encryption process generates a binary key for each binary pair that will be used during signal decryption. The imperceptibility of a reconstructed ECG signal is evaluated by analyzing performance metrics in terms of PSNR and PRD values. The average PSNR values are found to be above 42 dB, whereas PRD values observed were less than 3.43 percent. It is further proved that the proposed technique can encrypt a maximum of 226 kb of diagnose data by using merely seven QR codes.
International Journal of Imaging Systems and Technology, 2020
The article presents a novel ECG steganography scheme based on the tunable Q-factor wavelet trans... more The article presents a novel ECG steganography scheme based on the tunable Q-factor wavelet transformation (TQWT) and also singular value decomposition (SVD) techniques that ensure better safety and confidentiality of patient information. Initial parameters such as Q, r, and J are used to decompose the cover signal into individual frequency sub-bands with the tunable Q-factor wavelet transform (TQWT). The singular value decomposition (SVD) technique is used to further decompose high-frequency sub-band coefficients into singular values. The watermark information is then embedded with high-frequency sub-band coefficients by involving the quantization process. The performance of this proposed system is successfully evaluated by considering various metrics, such as peak signal to noise ratio (PSNR), structural similarity index (SSIM), percentage residual difference (PRD), and bit error rate (BER). The simulation results of the proposed scheme are observed to be better than other traditional algorithms. K E Y W O R D S ECG steganography, scaling factor and performance metrics, singular value decomposition (SVD), tunable Q-factor wavelet transform (TQWT)
This paper uses a novel color image stego-crypto technique to enhance the capacity, security and ... more This paper uses a novel color image stego-crypto technique to enhance the capacity, security and robustness of data embedding process. The proposed scheme uses a base64 encoding technique and along with QR code generator to transform the secret data, i.e., watermark to a 2D binary matrix, i.e., QR code that considered as a first-level encryption process. Followed by, embedding of individual QR code into the color image components, namely red, green and blue using a dynamic bit replacement technique that guarantees a second level of security. Subsequently, scrambling of random block-pixels and the encryption of logistic chaos are performed. The experimental outcome of three levels of security scheme has proved to be with a much larger key space and a strong resistance to brute force, statistical and differential attacks. The correlation between adjacent pixels is observed to be uniformly distributed in the cipher image and the information entropy value is close to an ideal value of 8. From the simulation results, it is observed that the proposed security scheme has outperformed the existing standard encryption techniques within a single round of iteration in terms of security and robustness.
The proposed process involves color image steganography using DWT-BRA (Bit Replacement Algorithm)... more The proposed process involves color image steganography using DWT-BRA (Bit Replacement Algorithm). The Discrete Wavelet Transform (DWT) is used to maintain image quality, imperceptibility, security and robustness. The cover image is decomposed into individual sub band for data embedding. The embedding process involves BRA technique to hide secret data. In the frequency subband coefficient only selective bit of the coefficient is replaced by the secret bit. The embedded data is extracted using bit extraction process and the performance analysis like PSNR and SSIM are evaluated. The proposed method obtained a 63.25 dB PSNR in high frequency subband (HH) and SSIM value of 1.6݁ ିହ .
The paper presents a colour image steganography, in which patient diagnose data is appended along... more The paper presents a colour image steganography, in which patient diagnose data is appended along with diagnosis information. The confidential information in the form of audio signal convolved with diagnoses data by using multi-bit embedding process. Each represented bit of both ECG and audio samples are convolved with each other and the resultant samples are given to modulo division process to separate quotient and remainder value. The quotient data is then embedded within selective locations of colour image component. Eventually, the reminder value will act as a key to retrieve original information. The performance of proposed steganography process is validated by estimating various quantitative and statistical metrics, such as peak signal to noise ratio (PSNR), normalized cross correlation (NCC), structural similarity index (SSIM) and signal to noise ratio (SNR). With increase in payload size the result shows better robustness and imperceptibility between cover and stego image. The absorbed SNR value of the extracted signal is absorbed to be 23 dB with minimum signal degradation.
In concepts such as connected health, patient data needs to be protected when they are transmitte... more In concepts such as connected health, patient data needs to be protected when they are transmitted over the internet to the caregiver. The current letter proposes embedding the QR code of patient data into discrete wavelet transform coefficients of the cover ECG Signal. The imperceptibility of hidden data is estimated by performance metrics such as peak signal-to-noise ratio, percentage residual difference, and Kullback-Leibler distance. The proposed approach is demonstrated on MIT-BIH database. Watermarking in the low coefficient values deteriorate the signal less and results in lower bit errors. It is shown that larger watermarks can be accommodated with a larger version of QR code for more or less the same imperceptibility. It is found that QR code version 40 allows embedding of maximum 2632 bytes of patient data with zero bit errors. Since QR code has an inbuilt error correction code, it provides additional layer of security to patient data and the proposed approach can be used for secure transfer of patient data.
Uploads