61 results about "Cellular neural network" patented technology

The system can be used for the automatic analysis of images (I), comprising a matrix of spots, such as images of DNA microarrays after hybridisation. The system can be associated-and preferably integrated in a single monolithic component implementing VLSI CMOS technology-to a sensor (10) for acquiring said images (I). The system comprises a circuit (20) for processing the signals corresponding to the images (I), configured according to a cellular neural network (CNN) architecture for the parallel analogue processing of signals.

The invention belongs to the technical field of information security, and particularly relates to a color image encryption method based on cellular neural network hyperchaos and a DNA sequence. The method provided by the invention comprises the steps of separating red, green and blue three primary color components of a color plaintext image; updating and generating parameters and initial values of a six-order cellular neural network hyperchaotic system and a Logistic-Sine mapping system through the plaintext image, respectively carrying out iterative operation on the two chaotic systems, and obtaining red, green and blue components of an encrypted image according to a DNA encoding rule and a DNA decoding rule; and finally, changing the pixel values of the encrypted image through key stream and a bitwise XOR operation to obtain a final ciphertext image. The decryption is a reverse operation of the encryption process. Compared with the existing image encryption method, the color image encryption method provided by the invention has the advantages of high security, good encryption effect, strong robustness, no information loss and the like, and thus being capable of being widely used in military, remote sensing, remote medical treatment, commerce and other fields.

A chaotic signal generator includes a set of elements connected together for generating chaotic signals. The connection scheme may correspond to the circuit generally referred to as Chua's circuit, particularly when implemented as a cellular neural network. Interposed in the connection scheme is at least one switch, such as a MOS transistor. Opening and closing of the switch causes variation in the chaotic dynamics of the generated signals. A command signal applied to the switch may correspond to a modulating signal for transmission on a channel, such as a high noise channel. The modulating signal may be a binary signal, and the command signal may be a switching signal having a frequency that increases or decreases depending on the logic level of the binary signal.

The invention provides a method and a device for carrying out classified management on clothes on the basis of picture processing. The method comprises the following steps: obtaining a picture which contains a clothes image; inputting the picture into a clothes detection model, and detecting to obtain a clothes image area in the picture; carrying out padding preprocessing on the clothes image area to obtain a standard-size picture which contains the clothes image area, and extracting the CNN (Cellular Neural Network) characteristics of the clothes image; inputting the CNN characteristics of the clothes image into an attribute identification model to identity to obtain an attribute sequence of the clothes image; and on the basis of the attribute sequence, carrying out classified storage on the picture which contains the clothes image to form an electronic wardrobe. The clothes image area in the picture is automatically detected, the attribute of the clothes image is automatically identified to carry out the classified storage on the clothe pictures so as to bring convenience for users to check, and a phenomenon that the user manually types in the attributes to carry out the classification storage is not required.

The invention relates to a hyper-chaos encryption method for a weak password based on a quantum cellular neural network, which belongs to the field of the information security technology. The method provided by the invention solves the problem that the existing password system is limited by limited key space so that the safety of the system is difficult to guarantee when running into brute-force attacks. The encryption method provided by the invention combines the hyper-chaos characteristic of the quantum cellular neural network with the advantages of human brain identification; the safe password is divided into two parts, wherein one part is translated into the picture form; in terms of an initial value and extremely sensitive characteristics of a control parameter, images are encrypted by using the high complexity of the hyper-chaos system; the other part of secret key is encrypted by using an AES (Advanced Encryption Standard) encryption method; and the two parts form a safe password for encrypting the data. The encryption method provided by the invention has great key space; the ability resisting to brute force attack is obviously strengthened; the calculated quantity in the encryption process is reduced when less password number is used; and the encryption method has the characteristics of high safety and convenience for remembering by a user.

The system can be used for the automatic analysis of images, including a matrix of spots, such as images of DNA microarrays after hybridization. The system can be associated—and preferably integrated in a single monolithic component implementing VLSI CMOS technology—to a sensor for acquiring the images. The system includes a circuit for processing the signals corresponding to the images, configured according to a cellular neural network (CNN) architecture for the parallel analogue processing of signals.

The invention discloses a method for segmenting a fingerprint image based on a cellular neural network and morphology and belongs to the technical field of image processing. The method comprises the following steps of: firstly, determining an initial threshold value t in a curve trough between a fingerprint foreground peak value and a background peak value of a gray histogram h(k) of an initial fingerprint image, and computing the fuzziness mu(f(x, y)) of the initial fingerprint image; secondly, computing an entropy E(I) of the fuzziness by using a Shannon function S(mu(f(x, y))), and minimizing the entropy and computing an optimum fuzzy threshold value t*; thirdly, computing a threshold value z* of the cellular neural network by using the optimum fuzzy threshold value t*, and processing the fingerprint image by using two 3*3 square cellular neural network templates to obtain a substantial fingerprint foreground area image; and finally, performing morphological operation on the substantial fingerprint foreground area image by using a 9*9 morphological template to obtain a final fingerprint foreground area image. The method has the advantages that: computation quantity is relatively lower; and the contour of a fingerprint foreground subjected to the morphological operation is relatively smoother.

The invention provides a space network cross-domain anonymous identity authentication method based on hyper-chaos encryption, relates to the technology of space information network, and solves the problem that the security mechanism applied in the conventional network cannot be directly applied to space network, and space cross-domain identity authentication cannot be realized. Provided is a space network cross-domain anonymous identity authentication method based on hyper-chaos encryption. The method includes two parts: a register stage and an authentication stage, a chaos algorithm is sensitive to initial conditions and control parameters, the structure is complex, prediction and analysis are difficult, and pseudo random sequences with good randomness and complexity are provided so that the chaos is applicable to encryption. A quantum cellular neural network is a cellular neural network structure coupling via quantum cellular neural network automata, complex linear dynamic features are obtained from the polarizability and quantum phase of each quantum cellular automaton, a nano-scale hyper-chaos oscillator can be constructed, the power consumption is low, the integration level is high, and the application requirement of space network satellite nodes can be better met.

The invention discloses a simulation method for an adjustable weight module of a cellular neural network. A memristive synapsis bridge circuit is built mainly on the basis of a memristor to carry out simulation, and the magnitude and the time of the input pulse of the memristive bridge circuit can be regulated to accurately obtain a required cellular neural network weight template. By use of the simulation method, the adjustable weight module of the cellular neural network can be realized.

The invention discloses a cellular neural network hardware architecture optimization method. The method comprises the steps of constructing cellular neural network hardware architecture and realizingsystem-level optimal design, module-level optimal design and design space-level optimal design for a calculation acceleration unit, wherein the architecture consists of an external memory, a memory interface controller, an on-chip input cache, an on-chip output cache, the calculation acceleration unit and a bus; the calculation acceleration unit comprises a plurality of iteration units which are connected in sequence; each iteration unit comprises a plurality of parallel operation modules; data is operated in the calculation acceleration unit and the operation result is written into the on-chip output cache; and an operation of the whole cellular neural network is completed through the iteration units in an assembly line manner. According to the method, parallel calculation of the cellularneural network is realized through system-level optimization, memory bandwidth in hardware is sufficiently utilized and data transmission delay is decreased through module-level optimization, and optimal calculation performance of systems is achieved under the condition of limited hardware resources through design space-level optimization.

The invention discloses a quantum cellular neural network based video chaotic encryption method, relates to the technical field of video encryption and solves the problems such as relatively slow encryption speed, destruction of the coding format of a video, poor video transmission real-time performance and poor safety of the existing encryption method. The quantum cellular neural network based video chaotic encryption method comprises the steps of performing iteration solution on two cellular quantum cellular neural network hyper-chaotic systems to generate a matrix A; performing matrix transformation on the A to generate a chaotic sequence K and an index sequence Index; dividing the chaotic sequence K to generate an initial key pool; regarding elements in the index sequence Index as initial values of Logistic chaotic mapping, respectively and iterating to generate two chaotic index sequences and transforming to obtain two integer sequences; regarding the integer sequences as indexes respectively and carrying the indexes into the initial key pool to calculate and generate a Boolean key KeyB; dividing the Boole key KeyB into keyb1 and keyb2; encrypting the exponential-Golomb encoding information bit of H.264 by using the keyb1; and regarding the keyb2 as a key to encrypt the encoding data of the H.264 to realize video chaotic encryption of the quantum cellular neural network.

The invention proposes a cellular neural network edge detection method based on relative variation regularization. The method includes image preprocessing and edge detection after the preprocessing. The image preprocessing mainly adopts the relative variation to smooth noises and textures; and the edge detection after the preprocessing mainly adopts a cellular neural network method to detect images. The method is capable of extracting effective and reliable and accurate effective edge information from pictures which are rich in noises and complex in textures. Compared with a traditional Canny method, the method restrains effects from the noises and the textures and compared with a standard cellular neural network method, the method prevents advance design of complex CNN template parameters.

A sensor having an array of photo sensitive elements for acquiring images of the passenger compartment in a motor vehicle and a circuit for processing the signals corresponding to the images generated by said photo sensitive elements. The processing circuit is configured according to a cellular neural network processing architecture of the image signals and can generate an output signal indicating the decision on whether to deploy an airbag to which the sensor is associated or to control the explosion of the airbag. Preferably, the photo sensitive array and the processing circuit are comprised on a single integrated component, preferably implementing CMOS technology.

An optical image encryption and decryption method based on compound chaos and quantum chaos is provided, which relates to the field of optical information security technology and solves the security defect of the non-linear shortage of the existing optical image encryption technology, the method comprising an encryption process and a decryption process, wherein the user encryption key and the decryption key are respectively set, which consist of the initial value, the control parameter and the order of the hyperchaotic system of a three-cell fractional quantum cellular neural network, the initial value and the control parameters of the composite chaotic map, and the series of the deformed fractional Fourier transform. The invention makes up for the security defect that the linear characteristic of the traditional optical image encryption technology is insufficient, and the fractional-order quantum chaotic system has a higher key dimension, bigger secret key space and greater sensitivity and can better resist all kinds of security attacks; the quantum chaotic system is a new type of nano-device, which is characterized in that quantum dots and quantum cellular automata transmit information mutually by the Coulomp action.

The invention discloses a method for generating pseudo-random numbers on the basis of cellular neural networks. By the aid of the method, problems of low pseudo-random number generation efficiency and poor statistical performance in the prior art mainly can be solved. The implementation scheme includes that the method comprises 1), generating random sequences P by the aid of the six-dimensional cellular neural networks and generating random sequences X by the aid of logic mapping so as to enhance the randomness of the random sequences; 2), respectively storing the generated random sequences P and X into two different matrixes and carrying out integer processing on data of the matrixes so as to extract the randomness of decimal portions of the data; 3), acquiring a novel matrix from the two different processed matrixes and generating 64 bits of the pseudo-random numbers by the aid of data in the novel matrix in each procedure. The method has the advantages that the pseudo-random number generation efficiency can be improved; requirements of international random number detection standards NIST SP800-22 can be met by the generated pseudo-random numbers, and the generated pseudo-random numbers can be used for secure communication.

A multi-diffusion image encryption and decryption method based on chaotic segmentation is disclosed, relating to the field of image encryption technology, and solving the problem of security risks inmultimedia data in the existing computer network environment. The multi-diffusion image encryption and decryption method based on chaotic segmentation removes the correlation between image blocks moreeffectively by a chaotic image segmentation method. A chaotic control table and a chaotic exchange table generated by a cellular neural network are used for intra-block and inter-block scrambling ona plain text image, and the correlation between image pixels is removed by multiple chaotic diffusion steps including static diffusion and dynamic diffusion. An image encryption and decryption methodwith large key space, good sensitivity and effective protection against security attacks is provided.

The invention provides an image encryption and decryption method based on four-dimensional quantum Dicke mapping and relates to the field of image encryption technology. The invention aims to solve the shortage of key space and insufficient randomness, as well as inability to effectively resist known plaintext attacks and security flaws in selective plaintext attacks in the existing image encryption methods. The invention uses a quantum Dicke chaotic system as a key generator to generate a plurality of sets of scrambling keys and diffusion keys, which are randomly selected by a user. The invention ensures the key space and the key sensitivity, shortens the length of the key and reduces the cost of storing and transmitting the key by the user. The invention applies the hyperchaotic system of the fractional-order quantum cellular neural network to the image diffusion stage, The hyperchaotic characteristic of the fractional-order quantum cellular neural network system guarantees the randomness and ergodicity of image diffusion, and the encryption method of the invention can effectively resist the known plaintext attack and the selected plaintext attack because the diffusion key streamis related to plaintext.

The invention discloses an unsupervised classification and supervised correction fusion speech separation method related to spatial structural features. The method comprises the steps of performing speech segment feature extraction based on a time-delay cellular neural network and speech segment unsupervised adaptive classification based on a dynamic growth self-organizing mapping neural network;adaptively correcting a speech separation model based on a particle swarm optimization algorithm, and reconstructing speech based on binary masking. According to the method, unsupervised classification and supervised correction are combined, the generalization and accuracy of separation of mixed voices with unknown speaker number are improved, and a theoretically supported and practically feasiblescheme is provided for the urgent practical problem of single-channel multi-speaker voice separation.

The invention discloses an intelligent safe multi-license-plate positioning identification method based on a cellular neural network, which comprises the steps that after front-end camera-shooting equipment collects color monitoring images of a plurality of lanes, license plate areas are positioned based on cellular neural network edge detection, fusion color quantification division color code matrices and definite length-width ratio characteristics; the license plate areas are magnified by a bilinear interpolation method, and are binarized by a gradation adaptive method; characters are divided by a binary projection method; and the characters and color types of license plates are identified based on a template matching fusion color code matrix. The license plate areas that are as small as possible is positioned and identified from the images containing the license plates; the positioning speed is high; and the identification accuracy is high.

The invention discloses a dynamic reconfigurable logic gate realization method based on a cellular neural network (CNN) and a realization circuit thereof. An input signal, template parameters and offset parameters are utilized to construct a state equation, an equation of system state variables is achieved, and different logical functions can be realized dynamically only through changing of template parameter signals. The dynamic reconfigurable logic gate realization method based on the CNN and the realization circuit of the dynamic reconfigurable logic gate realization method based on the CNN have the advantages of being simple in circuit, high in system stability and suitable for realization of super-large-scale integrated circuits.

The invention provides a method for image restoration, which is used to restore a known degradation image of a degradation model. The process method of the invention is processing the degradation image to be processed with three-dimensional cell neural network image processing method, and the output is just the restored image. The method is simple and easy, and is suitable for being used in restoring a known degradation image of a degradation model. The method can have a very high processing speed when adopting a cellular neural network chip to realize the algorithm.

An ensemble learning based image classification system contains multiple cellular neural networks (CNN) based integrated circuits (ICs) operatively coupling together as a set of base learners of an ensemble for an image classification task. Each CNN based IC is configured with at least one distinct deep learning model in form of filter coefficients. The ensemble learning based image classificationsystem further contains a controller configured as a meta learner of the ensemble and a memory based data buffer for holding various data used in the ensemble by the controller and the CNN based ICs.Various data may include input imagery data to be classified. Various data may also include extracted feature vectors or image classification outputs out of the set of base learners. The extracted feature vectors or image classification outputs are then used by the meta learner to further perform the image classification task.

The invention discloses an authenticity self-identifying and tampering self-positioning method for a digital image, and belongs to the technical fields of image processing, computer security, computer application and the like. By the method, whether an authenticated image is substituted or tampered can be effectively judged, and a specific tampered position is pointed out according to an actual tampered condition. The digital image is processed through a hyper-chaotic cellular neural network and encrypted by a public key encryption system. The method can be widely applied to occasions with high image security requirements, such as medical treatment, military affairs, electronic commerce, electronic government affaires and the like.

The invention provides a color image edge detection method. The method comprises the following steps of 1, inputting a color image and carrying out normalization processing on the input original image; 2, analyzing the monostable characteristic of the FHN, constructing an adjustment function to optimize a diffusion coefficient in a reaction diffusion equation according to the gray level change of a local region, and calculating a threshold value with gray level self-adaption; 3, constructing a cellular neural network structure based on a FitzHugh-Nagumo (FHN) reaction diffusion equation; 4, performing edge detection on a gray color image or a color image in an HSV color space by using the proposed cellular neural network model; and 5, outputting an edge detection result of the color image. According to the method, the dynamic property of the FHN reaction diffusion equation is analyzed, and feasibility of the FHN reaction diffusion equation in image edge extraction is proved; the CNN based on the FHN reaction diffusion equation is applied to edge detection of gray and color images; the diffusion coefficient is optimized by adopting the reaction diffusion equation, so the threshold value is more adaptive, and the adaptive threshold value can be adaptively selected according to brightness of the color image, so the provided color image edge detection method is effective.

Disclosed is a three-order mobius strip type cellular neural network chaotic circuit. Third, fourth and sixth operational amplifiers A3, A4 and A6 form linear inverting integrators, and output ends are chaotic signal output ends X1, X2 and X3 respectively; first and fifth operational amplifiers A1 and A5 form linear inverting amplifiers; a second operational amplifier A2 forms a nonlinear inverting limiting amplifier; the inverting integrator A3 is connected with the amplifiers A1 and A4; the inverting integrator A4 is connected with the integrator A3 and the amplifier A5; the inverting integrator A6 is connected with the integrator A4. By the chaotic circuit, various waveforms, phase diagrams and chaotic evolution curves of the three-order neural network chaotic circuit are outputted to form a chaotic privacy communication system.

The invention discloses a virtual machine security monitoring method based on a cellular neural network. The method includes: a privileged virtual machine dynamically acquires resources of a client virtual machine; the contents of the resources obtained are analyzed on the basis of the cellular neural network so as to judge whether the client virtual machine is under attack or not; finally operating state of the client virtual machine is displayed in real time. The virtual machine security monitoring method based on the cellular neural network has the advantages that the privileged virtual machine can quickly and dynamically monitor mass client virtual machines on a same physical machine and the security of the client virtual machines and whole virtual machine system is effectively improved.

The invention relates to a palmprint and face recognition method based on a cellular neural network hetero-associative memory model, and belongs to the technical field of intelligent recognition. Themethod comprises registration and identification. According to the method, the hetero-associative memory and the cellular neural network model are combined; palm print picture data and face picture data are converted into a series of parameters to be stored, identity information, namely the palm print picture data and the face picture data, is high in verification reliability, the picture storagemode is high in confidentiality and safety coefficient, and the identity information of people is effectively prevented from being leaked. The mode of converting the picture into the parameter throughthe model is adopted, and the method is simple, convenient, good in practicability, good in picture recognition effect and good in face picture data protection effect.

A handwriting recognition device using fuzzy logic and cellular neural network for unconstrained handwritten numeral classification is provided. The current mode VLSI classifier has a I / O circuit for inputting and outputting a plurality of membership functions. An extraction unit comprising a CCD extractor with a CNN structure and a compression unit receives a to-be-recognized character having a plurality of input features for generating a plurality of features values after compression. A membership function generator stores the plurality of membership functions and receives the plurality of features values to generate a plurality of current-type membership degrees. A plurality of switched-current integrators receives the plurality of current-type membership degrees for generating a plurality of synthesis membership degrees. A k-WTA circuit is provided for comparing the plurality of synthesis membership degrees and output the plurality of synthesis membership degrees as well as the corresponding characters in an order of magnitude.

The invention discloses an image enhancement method based on a bionic adaptive memristive cellular neural network, which integrates an adaptive three-Gaussian model simulating the human visual mechanism, a cellular neural network and a nano-memristor, in the typical adaptive three-Gaussian model Based on this, a linear adaptive three-Gaussian model that is easy to implement in hardware is proposed. Using the Gaussian kernel function and the image processing characteristics of the cellular neural network, it overcomes the limitations of fixed image enhancement templates and pure mathematical algorithms, and greatly enhances the advantages of image enhancement. Its enhancement effect is closer to that of the human eye; at the same time, the hardware design based on memristor is conducive to large-scale circuit integration, and it is convenient to realize hardware acceleration and chip-based end-side real-time image processing.