34results about How to "Resistance to noise interference" patented technology

The invention discloses a time type network covert channel construction method based on discrete wavelet transform, and belongs to the technical field of network information security. According to the time type network covert channel construction method disclosed by the invention, a secret message is embedded in a wavelet domain coefficient of a normal IPD sequence, and the embedding position and intensity are controlled to reduce the time domain changes to the original carrier and the properties of the original carrier as much as possible; and the generated covert communication data streams have normal rules and features and better anti-detection property. Meanwhile, the secret message bits are embedded in a part of approximation coefficients of wavelet decomposition by adopting the odd-even quantification method, and the network noise interference can be resisted to a certain extent without importing additional synchronization cost, and the robustness is relatively good.

The invention discloses a printed circuit board image registration method based on the shape context of the mass center of a communicated region, and belongs to an image registration method based on characteristics. An interest point serves as the coordinate of the mass center of the communicated region of an image. The coordinate of the mass center is characterized in two aspects, firstly, with the coordinate of the mass center as a center, the shape context of the communicated region where the coordinate of the mass center is located describes the shape characteristics of the communicated region, and therefore the shape context is called as a shape descriptor; secondly, the coordinate of the mass center is opposite to the shape context of the coordinates of the mass centers of other communicated regions in the image where the communicated region is located, the shape context describes the relative positions of the coordinate of the mass center and the communicated region where the coordinate of the mass center is located in the whole image, and therefore the shape context is called as a relative position descriptor. Through the shape descriptor and the relative position descriptor of the coordinate of the mass center of the communicated region, accurate matching between a communicated region of reference images and a communicated region of images to be registered is achieved, transformation parameters between the reference images and the images to be registered are estimated according to the coordinates of the mass centers of the communicated regions which are accurately matched, and finally the images to be registered are accurately registered.

The invention relates to an automatic slant angle detecting method for a scanned document image, and belongs to the technical field of image processing. An existing automatic detecting technology is poor in adaptability and noise-jamming-resisting capability. The slant angle of the scanned document image is calculated by carrying out the initializing, iterative stroke processing, foreground area edge extracting, area increasing, area feature extracting, slant angle extracting and other methods on the scanned document image. By means of the method, the slant angle of the document image can be automatically detected through the image processing technology. The method has the advantages that a very high noise-jamming-resisting capability is achieved, the method is unrelated to text texture and is suitable for complex situations with various image-text mixed arranging and various writing directions existing at the same time, and the method is suitable for Chinese and English documents.

The invention discloses a digital watermarking method for communication modulation channel transmission. The method comprises three parts: watermark embedding, signal transmitting and watermark extracting. The method comprises the steps of adopting a communication digital watermark embedding method to embed watermark information into an original modulation signal, acquiring a watermark-embedded modulation signal through communication channel transmission, and adopting a communication digital watermark extracting method to extract a synchronization signal and to obtain original data and watermark information. According to the method, digital watermark information is embedded in a communication signal, and secret information can be transmitted in the communication signal by the use of digital watermarks without influencing normal communication. The methods of the invention can be used for correspondent identity verification and secret information transmission and provide a reliable guarantee for secret communication.

The invention belongs to the field of marine pollution treatment and particularly relates to a segmentation method of a sea surface oil spillage image. Particularly, a high-order Boolean energy function is established, wherein the high-order Boolean energy function comprises a first-order judging item and a high-order coupling item, the first-order judging item is obtained according to a single pixel energy function and an edge-preserving filter function through convolution so that a good effect of protecting the edge of an oil spillage zone can be achieved; and the high-order coupling item can be used for calculating the coupling performance of similar paired neighboring pixels in a pixel subset so that a long-strip-shaped oil spillage zone can be segmented precisely. The minimum value is obtained according to the established Boolean energy function by adopting a maximum flow / minimum segmentation algorithm, and each Boolean variable value corresponds to the optimum segmentation method of the oil spillage zone in the sea surface oil spillage image when the minimum energy value is adopted. By adopting the segmentation method, the influence of the edge diffusion phenomenon in the process of segmenting the long-strip shaped oil spillage zone can be overcome to a certain extent, and the segmentation accuracy can be improved.

The invention discloses a signaling type method of a mobile phone signal. The signaling type method of the mobile phone signal comprises the following steps: establishing a shielding processing system, and ensuring that the system frequency reference is synchronized with a base station; receiving frame signals of the base station time-divisionally by using the characteristic that the frame signalsof the base station are repeatedly transmitted according to a frame period of the base station; performing multi-frame averaging filtering processing on the frame signals of the base station to purify a frame reference signal of the base station; and shielding the base station by means of the purified frame reference signal. The multi-frame averaging filtering process on the frame signals causesthat modulated signals of the user data is cancelled out by most through vector addition filtering due to time-varying characteristics; but the reference signal given by the base station to the mobilephone is enhanced because the reference signal is always existed and the phase is unchanged. Synchronous emission is performed by using the frame reference signal of the base station obtained after purifying to disturb the base station signal, so that the mobile phone cannot distinguish the signal; and when the interference is large enough, the decoding fails and the network connection is lost. The signaling type method of the mobile phone signal is effective for a variety of mobile communications and has a good universality.

The invention discloses a license plate image correction method, a license plate image correction device, and a storage medium. The method comprises the following steps: based on an image segmentation network, processing a to-be-corrected image containing a to-be-processed license plate, and obtaining a first license plate mask image, wherein the image segmentation network comprises an encoder and a decoder, the encoder is used for encoding a to-be-corrected image to obtain a feature map, and the decoder is used for decoding the feature map according to a size of a corrected reference image corresponding to a to-be-processed license plate to obtain the first license plate mask image; based on the first license plate mask image, calculating angular points of a license plate in the to-be-corrected image and an external rectangular frame of the license plate; = correcting the image where the license plate is located based on the external rectangular frame and the angular points to obtain a corrected license plate image. In this way, the license plate correction effect can be improved.

The invention discloses a fire monitoring network positioning method based on RSSI distance measurement. The method comprises the following steps: establishing a logarithmic distance path loss model; according to the position of a known reference node, periodically and dynamically measuring to obtain the received signal strength between the known reference node and a neighbor reference node, and obtaining a path loss index n value in a monitoring area at the current moment by using a logarithmic distance path loss model; calculating distances between a reference node and other reference nodes in the network by measuring a PR (d) value by using an RF signal propagation attenuation model, comparing the measured distance with an actual distance, and obtaining a measurement error of the PR (d) value at the reference node; and estimating the coordinates of the monitoring nodes by using a forgetting adjustment factor-based RLS positioning algorithm. According to the method, the influence of factors such as RSSI ranging errors and the number of reference nodes on the positioning precision can be fully considered, the positioning effect is ideal, and the positioning requirements of a fire environment monitoring system with a complex and severe network environment and limited positioning cost can be met.

A coding method for information concealment is used for processing audio data of multiple channels at least including a first channel and a second channel. The coding method includes steps of detecting valid audio data segments in the multiple channels and judging audio sections having similarity among the valid audio data segments in the first channel and the second channel; segmenting the audiosections in the first channel and the second channel into multiple audio frames according to a preset number of sampling points; calculating the entropy values of the audio frames in the first channeland the second channel; adjusting the audio data in the audio sections of one channel of the first channel and the second channel so as to enable information to be concealed in a entropy value difference of the audio frames in the audio sections in the first channel and the second channel. The invention also provides a decoding method corresponding to the coding method and a coding and decoding system. The invention can prevent noise interference effectively and coding and decoding can be implemented conveniently and quickly.

The invention relates to a filter circuit and a structure thereof, wherein the filter circuit comprises an input end, an output end, a resonance circuit, a first coupling portion and a second coupling portion, wherein the resonance circuit is coupled between the input end and the output end, which comprises M resonators which are arrayed in sequence, and signals which are received by the input end can be transmitted to the output end by the resonance circuit through the mutual coupling of the adjacent resonators. The first coupling portion and the second coupling portion are respectively coupled on the non-adjacent resonators. Partial signals which are received by the input end are coupled to the second coupling portion through the first coupling portion in a crossed coupling mode. And therefore, the inhabitation to undesired signals near an operation frequency band is increased.

The invention relates to a quasi-continuum bound state magnetic field sensor based on photonic crystal panel support, and belongs to the technical field of photonic crystal sensors. According to the invention, a photonic crystal flat plate breaking a symmetrical structure is combined with a magnetic current film for the first time, and a symmetrical protection type quasi-continuum bound state mode is excited in a vertical incidence mode for magnetic field intensity sensing. The photonic crystal flat plate adopts the square nanopores, the symmetry of the nanopores is slightly broken, a non-suspension structure is adopted, the preparation difficulty is low, and the structure is stable. The bound state mode quality factor of the excited quasi-continuum is up to 2 * 10 < 4 >, meanwhile, the interaction between light and a magnetic current film is enhanced by a relatively large nanopore structure, and the magnetic field sensitivity of 88nm / T is realized. In addition, the resonance amplitude of the bound state of the quasi-continuum is up to 0.97, and interference of environmental noise can be effectively resisted. Compared with other types of photonic crystal magnetic field sensors, the photonic crystal magnetic field sensor has excellent sensing performance of high quality factor, high sensitivity and large resonance amplitude, and is suitable for high-precision magnetic field sensing in a complex environment.

The invention discloses a method for resisting impulse noise interference and a device thereof, belonging to the field of communication. The method comprises the steps of: acquiring carrier wave original static line noise; calculating virtual static line noise according to the acquired original static line noise; and accessing a multiplexer device into a local side digital subscriber line, and sending the virtual static line noise. The device comprises an acquisition unit, a calculation unit and a transmission unit.The technical scheme of the invention can more effectively resist the noise interference.

The invention relates to a one-way valve fault feature extraction method based on improved permutation entropy, and belongs to the technical field of mechanical fault diagnosis and signal processing.The method comprises the following steps: firstly, aiming at the problem that the permutation entropy algorithm only considers the permutation sequence of signals and ignores the signal amplitude, adding a weighting coefficient to a reconstructed component to improve the permutation entropy, and further proposing an improved permutation entropy algorithm; directly adopting an improved permutationentropy algorithm to extract one-way valve fault features under a strong noise background, and completing fault diagnosis of the one-way valve. According to the method, the fault characteristics of the one-way valve are extracted by utilizing the permutation entropy, improvement is carried out aiming at the defects of the permutation entropy, the noise interference resistance is improved, the signal preprocessing step is omitted, the program running time is shortened, and the diagnosis efficiency is improved.

The invention discloses a hierarchical tracking method based on increment supervised gradient descent in the field of computer vision. The hierarchical tracking method comprises the following steps: initializing; collecting a plurality of samples in Gaussian distribution near a target position marked by a first frame, extracting features, and training a hierarchical regression model; positioning atarget; in a new frame, collecting a plurality of samples in Gaussian distribution near the tracking result of the previous frame, respectively regressing the samples to the tracking result accordingto a hierarchical regression model, and fusing the results to obtain a positioning target; collecting training samples online; collecting samples in Gaussian distribution near a tracking result, andperforming hierarchical regression on the samples to obtain samples of other layers for regular model updating; updating an incremental model; updating the hierarchical regression model online in an incremental learning mode; carrying out iterative tracking; repeating the step 2 to the step 4 until the complete video tracking is completed to obtain a more stable tracking result which can be used in artificial intelligence equipment.

The invention belongs to the technical field of seismic exploration, and particularly relates to a detector fully differential circuit based on charge amplification. The detector fully differential circuit comprises a charge amplification circuit consisting of a charge conversion circuit and a voltage amplification circuit, a movement unit, and a subsequent filtering and collecting circuit. The detector fully differential circuit can improve low frequency acceptance capability effectively without high input impedance, and can provide high common mode rejection ratio to resist noise interference regardless of the length of a transmission wire. The length of the transmission wire does not affect attenuation of voltage amplitude of the transmission signal, and distributed capacitance does notaffect a piezoelectric detector circuit of the frequency band width.

The invention provides a guiding system and a guiding method for non-excavating. The non-excavating guiding system comprises a detection probe (1) for acquiring information, a first processor (2) for pre-processing the information acquired by the detection probe, a second processor (3) for performing frequency domain processing on the information processed by the first processor, and a synchronous display (4) for displaying the information processed by the second processor. The second processor (3) comprises a spectrum analyzer (31); and the spectrum analyzer (31) is used for analyzing one or more predetermined signal frequency ranges, and processing the analyzed information to select one frequency or one frequency range of which the signal-to-noise ratio of signals is relatively larger.

The invention relates to a method for detecting luminous intensity and uniformity based on CCD perception, and the method comprises a calibration step and a detection step, and the detection step comprises the following steps: a1, taking a luminous surface to be detected, and setting parameters of a CCD to be the same as the parameters of the CCD in the calibration step; a2, in a darkroom environment, imaging a light-emitting surface to be measured to obtain an imaging image I; a3, extracting a brightness characteristic matrix M and a uniformity characteristic U of the imaging image I; a4, establishing and learning a brightness model according to the calibration step to obtain a learned brightness model; a5, calculating the brightness A of the light-emitting surface to be measured according to the learned brightness model and the brightness characteristic matrix; and calculating the uniformity u of the light-emitting surface to be measured according to the uniformity characteristic matrix U. According to the invention, brightness and uniformity detection of different light-emitting materials can be realized, detection of different light-emitting positions and different light-emitting sizes can be realized, noise interference can be resisted, and the method is simple and easy to implement.