Proceedings of the Institute for System Programming of RAS, 2021
Аннотация. Предлагается первый подход, основанный на применении вейвлет-анализа при обработке изо... more Аннотация. Предлагается первый подход, основанный на применении вейвлет-анализа при обработке изображений с целью обнаружения объектов c повторяющимися чертами и двоичной классификации в плоскости изображения, в частности, для навигации в симулируемых средах. На сегодняшний день стало привычным использовать алгоритмы на основе сверточных нейронных сетей (Convolutional Neural Networks, CNN) для обработки изображений, полученных с бортовых камер беспилотных летательных аппаратов (Unmanned Aerial Vehicles, UAV), в пространственной области, что способствует решению задач обнаружения и классификации. Архитектура CNN позволяет обучать сеть, используя в качестве входных данных изображения без предварительной обработки. Это позволяет извлекать характерные признаки изображения. Тем не менее, в этой работе мы утверждаем, что спектральные характеристики изображений на разных частотах, низких и высоких, также влияют на производительность CNN во время обучения. Мы предлагаем архитектуру CNN, дополненную двумерным дискретным вейвлет-преобразованием как методом выделения признаков. Такая информация улучшает способность сети к обучению, устраняет переобучение и обеспечивает более высокую эффективность при обнаружении цели.
The present investigation is proposing a new partial encryption algorithm for digital image, usin... more The present investigation is proposing a new partial encryption algorithm for digital image, using the synchronization of cellular automata based on the local rule 90. Unlike other partial encryption algorithm, which become vulnerable to attacks such as Replacement Attack or Reconstruction Attack, this system encodes different bit planes, in function of the secret key, that is, for each block of clear text, different bits are encrypted to prevent that with an elimination operation of the encrypted bits information can be revealed. The synchronization of cellular automata has proven to be a useful tool for data encryption because it is sensitivity to initial conditions and, in addition, rule 90 is considered a chaotic standard. Both characteristics ensure cryptographic and perceptive security. Based on the results of the security analysis, this research could be an attractive option for image encryption with less computer cost and without compromising information confidentiality.
Proceedings of the Institute for System Programming of the RAS, 2021
An acceptable level of quality in data is nowadays a paramount for any kind of organization or en... more An acceptable level of quality in data is nowadays a paramount for any kind of organization or enterprise that wishes its business processes to prosper. Thus, introducing activities focused in the data quality management is a crucial requirement for the analysts if the level of quality of data for the functionality or service at hand is to be ensured. Such specialized data quality management activities should be presented as early as possible during the software development process. So far and having done a search for proposals in this field, there is still a lack of either methodological or technological proposals with which a developer could be able to design data quality aware applications in the specific field of Web application development. Considering the benefits offered in the field of Model Driven Web Engineering, this work presents a partial outcome of our research in this novel field: a metamodel and a UML profile, both able to be used as data quality artefacts during the...
In this work is presented a numerical implementation of a system that combines a compression sche... more In this work is presented a numerical implementation of a system that combines a compression scheme with an improved encryption procedure, which is applied to digital images. For the compression stage is considered the two-dimensional Haar wavelet transform, where an energy criterion is contemplated. On the other hand, the encryption scheme regarded is based on the synchronization of the cellular automaton rule 90, this system presents a good performance to encrypt images and it is resistant to cryptanalysis attacks such as the Chosen/Known-plaintext attack. The numerical conjunction of these procedures could be an appealing option for real-time applications such as video communications, video-surveillance among others.
In this work is proposed a partial image encryption method based on the synchronization of the ce... more In this work is proposed a partial image encryption method based on the synchronization of the cellular automaton rule 90. The security analysis proves that this cryptosystem is resistant to different tests and attacks such as the Chosen/Knownplain image attack and bit-replacement. This algorithm is a variant from another encryption algorithm named ESCA, in this case we modified the ESCA system to a partial encryption version reducing the latency time up to a 50%. This version has two changes a)it only encrypts the three most significant bits, and b) the bits of the secret key are rotated instead of calculating a key for each block of plaintext. These changes allow to reduce the latency time and according to the tests do not compromise the security. This proposal could be a feasible and secure option for real-time applications.
In this paper, we study in detail the multifractal features of the main matrices of an encryption... more In this paper, we study in detail the multifractal features of the main matrices of an encryption system based on a rule-90 cellular automaton. For this purpose, we consider the scaling method known as the wavelet transform multifractal detrended fluctuation analysis (WT-MFDFA). In addition, we analyze the multifractal structure of the matrices of different dimensions, and find that there are minimal differences in all the examined multifractal quantities such as the multifractal support, the most frequent singularity exponent, and the generalized Hurst exponent.
This paper carries out the image encryption with a revised and modified cryptosystem based on a r... more This paper carries out the image encryption with a revised and modified cryptosystem based on a rule-90 cellular automaton. For this purpose, we used an appropriate and convenient matrix approach to implement the main components of the encryption system, two indexed families of permutations and a pseudo-random bit generator. Furthermore, we modify the encryption scheme for improving its security. This makes our encryption proposal resist some common attacks, e.g. chosen plain-text attack, and the encryption and decryption processes are performed in a flexible way. Also, some security aspects of the cryptosystem are evaluated with a series of statistical tests exhibiting good results. It is believed that this proposal fits naturally in the present digital technology and is capable of achieving high levels of performance.
The technique of detrended fluctuation analysis (DFA) has been widely used to unveil scaling prop... more The technique of detrended fluctuation analysis (DFA) has been widely used to unveil scaling properties of many different signals. In this paper, we determine scaling properties in the encrypted images by means of a two-dimensional DFA approach. To carry out the image encryption, we use an enhanced cryptosystem based on a rule-90 cellular automaton and we compare the results obtained with its unmodified version and the encryption system AES. The numerical results show that the encrypted images present a persistent behavior which is close to that of the 1/f-noise. These results point to the possibility that the DFA scaling exponent can be used to measure the quality of the encrypted image content.
In this work, we numerically implement an encryption system to be used for real-time encryption o... more In this work, we numerically implement an encryption system to be used for real-time encryption of audio and video signals. In such applications, the results obtained present in general a good performance: a low time processing that meets the demands of real time systems, and there is no correlation between the original signal and its encrypted version.
2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14), 2014
ABSTRACT This paper presents the hardware implementation of a cryptographic algorithm using a mat... more ABSTRACT This paper presents the hardware implementation of a cryptographic algorithm using a matrix approach, named ESCA (encryption by synchronization of cellular automata). The ESCA system is a symmetric key algorithm, which is based on the rule-90 cellular automaton to implement all their components. In the hardware implementation of the ESCA system was used the soft processor core Microblaze in a Virtex-5 VC5VLX110T FPGA. With this embedded system we can have a reconfigurable encryption system that allows to the user select among different configurations without the need to program the FPGA every time. With the different options to configure the ESCA system, we may increase the security or have a customized system for a faster performance. With this implementation we carry out the encryption of grayscale and RGB color images. In addition, a sparse matrix format was implemented to reduce the latency, and to prove the security of the algorithm, different tests were applied exhibiting good results.
Proceedings of the Institute for System Programming of RAS, 2021
Аннотация. Предлагается первый подход, основанный на применении вейвлет-анализа при обработке изо... more Аннотация. Предлагается первый подход, основанный на применении вейвлет-анализа при обработке изображений с целью обнаружения объектов c повторяющимися чертами и двоичной классификации в плоскости изображения, в частности, для навигации в симулируемых средах. На сегодняшний день стало привычным использовать алгоритмы на основе сверточных нейронных сетей (Convolutional Neural Networks, CNN) для обработки изображений, полученных с бортовых камер беспилотных летательных аппаратов (Unmanned Aerial Vehicles, UAV), в пространственной области, что способствует решению задач обнаружения и классификации. Архитектура CNN позволяет обучать сеть, используя в качестве входных данных изображения без предварительной обработки. Это позволяет извлекать характерные признаки изображения. Тем не менее, в этой работе мы утверждаем, что спектральные характеристики изображений на разных частотах, низких и высоких, также влияют на производительность CNN во время обучения. Мы предлагаем архитектуру CNN, дополненную двумерным дискретным вейвлет-преобразованием как методом выделения признаков. Такая информация улучшает способность сети к обучению, устраняет переобучение и обеспечивает более высокую эффективность при обнаружении цели.
The present investigation is proposing a new partial encryption algorithm for digital image, usin... more The present investigation is proposing a new partial encryption algorithm for digital image, using the synchronization of cellular automata based on the local rule 90. Unlike other partial encryption algorithm, which become vulnerable to attacks such as Replacement Attack or Reconstruction Attack, this system encodes different bit planes, in function of the secret key, that is, for each block of clear text, different bits are encrypted to prevent that with an elimination operation of the encrypted bits information can be revealed. The synchronization of cellular automata has proven to be a useful tool for data encryption because it is sensitivity to initial conditions and, in addition, rule 90 is considered a chaotic standard. Both characteristics ensure cryptographic and perceptive security. Based on the results of the security analysis, this research could be an attractive option for image encryption with less computer cost and without compromising information confidentiality.
Proceedings of the Institute for System Programming of the RAS, 2021
An acceptable level of quality in data is nowadays a paramount for any kind of organization or en... more An acceptable level of quality in data is nowadays a paramount for any kind of organization or enterprise that wishes its business processes to prosper. Thus, introducing activities focused in the data quality management is a crucial requirement for the analysts if the level of quality of data for the functionality or service at hand is to be ensured. Such specialized data quality management activities should be presented as early as possible during the software development process. So far and having done a search for proposals in this field, there is still a lack of either methodological or technological proposals with which a developer could be able to design data quality aware applications in the specific field of Web application development. Considering the benefits offered in the field of Model Driven Web Engineering, this work presents a partial outcome of our research in this novel field: a metamodel and a UML profile, both able to be used as data quality artefacts during the...
In this work is presented a numerical implementation of a system that combines a compression sche... more In this work is presented a numerical implementation of a system that combines a compression scheme with an improved encryption procedure, which is applied to digital images. For the compression stage is considered the two-dimensional Haar wavelet transform, where an energy criterion is contemplated. On the other hand, the encryption scheme regarded is based on the synchronization of the cellular automaton rule 90, this system presents a good performance to encrypt images and it is resistant to cryptanalysis attacks such as the Chosen/Known-plaintext attack. The numerical conjunction of these procedures could be an appealing option for real-time applications such as video communications, video-surveillance among others.
In this work is proposed a partial image encryption method based on the synchronization of the ce... more In this work is proposed a partial image encryption method based on the synchronization of the cellular automaton rule 90. The security analysis proves that this cryptosystem is resistant to different tests and attacks such as the Chosen/Knownplain image attack and bit-replacement. This algorithm is a variant from another encryption algorithm named ESCA, in this case we modified the ESCA system to a partial encryption version reducing the latency time up to a 50%. This version has two changes a)it only encrypts the three most significant bits, and b) the bits of the secret key are rotated instead of calculating a key for each block of plaintext. These changes allow to reduce the latency time and according to the tests do not compromise the security. This proposal could be a feasible and secure option for real-time applications.
In this paper, we study in detail the multifractal features of the main matrices of an encryption... more In this paper, we study in detail the multifractal features of the main matrices of an encryption system based on a rule-90 cellular automaton. For this purpose, we consider the scaling method known as the wavelet transform multifractal detrended fluctuation analysis (WT-MFDFA). In addition, we analyze the multifractal structure of the matrices of different dimensions, and find that there are minimal differences in all the examined multifractal quantities such as the multifractal support, the most frequent singularity exponent, and the generalized Hurst exponent.
This paper carries out the image encryption with a revised and modified cryptosystem based on a r... more This paper carries out the image encryption with a revised and modified cryptosystem based on a rule-90 cellular automaton. For this purpose, we used an appropriate and convenient matrix approach to implement the main components of the encryption system, two indexed families of permutations and a pseudo-random bit generator. Furthermore, we modify the encryption scheme for improving its security. This makes our encryption proposal resist some common attacks, e.g. chosen plain-text attack, and the encryption and decryption processes are performed in a flexible way. Also, some security aspects of the cryptosystem are evaluated with a series of statistical tests exhibiting good results. It is believed that this proposal fits naturally in the present digital technology and is capable of achieving high levels of performance.
The technique of detrended fluctuation analysis (DFA) has been widely used to unveil scaling prop... more The technique of detrended fluctuation analysis (DFA) has been widely used to unveil scaling properties of many different signals. In this paper, we determine scaling properties in the encrypted images by means of a two-dimensional DFA approach. To carry out the image encryption, we use an enhanced cryptosystem based on a rule-90 cellular automaton and we compare the results obtained with its unmodified version and the encryption system AES. The numerical results show that the encrypted images present a persistent behavior which is close to that of the 1/f-noise. These results point to the possibility that the DFA scaling exponent can be used to measure the quality of the encrypted image content.
In this work, we numerically implement an encryption system to be used for real-time encryption o... more In this work, we numerically implement an encryption system to be used for real-time encryption of audio and video signals. In such applications, the results obtained present in general a good performance: a low time processing that meets the demands of real time systems, and there is no correlation between the original signal and its encrypted version.
2014 International Conference on ReConFigurable Computing and FPGAs (ReConFig14), 2014
ABSTRACT This paper presents the hardware implementation of a cryptographic algorithm using a mat... more ABSTRACT This paper presents the hardware implementation of a cryptographic algorithm using a matrix approach, named ESCA (encryption by synchronization of cellular automata). The ESCA system is a symmetric key algorithm, which is based on the rule-90 cellular automaton to implement all their components. In the hardware implementation of the ESCA system was used the soft processor core Microblaze in a Virtex-5 VC5VLX110T FPGA. With this embedded system we can have a reconfigurable encryption system that allows to the user select among different configurations without the need to program the FPGA every time. With the different options to configure the ESCA system, we may increase the security or have a customized system for a faster performance. With this implementation we carry out the encryption of grayscale and RGB color images. In addition, a sparse matrix format was implemented to reduce the latency, and to prove the security of the algorithm, different tests were applied exhibiting good results.
Uploads
Papers by Tulio Torres