397results about How to "Add nonlinearity" patented technology

The invention provides a deep learning-based traffic sign automatic identifying and marking method, which is used in the technical field of environmental perception of intelligent vehicles. A semantic segmentation structure is adopted for locating and detecting traffic signs, and candidate regions with the traffic signs are obtained, wherein the semantic segmentation structure comprises two parts: an encoding network; and a decoding network and a pixel-based classification layer. Then, the traffic signs in the candidate regions are subjected to classified identification and locating through a convolutional neural network of a quick region. Based on the traffic sign automatic identifying and marking method, the invention furthermore correspondingly provides an effective map navigation information updating method. According to the method, the candidate regions with the traffic signs are located by using the semantic segmentation method, so that a new idea is provided, a training parameter quantity is reduced, the storage space is saved, and the computing time is shortened; and the method is high in identifying accuracy, and can perform relatively accurate traffic sign information updating on map navigation information, thereby better serving drivers conveniently.

The invention relates to a recognition method of digital music emotion, belonging to the field of computer pattern recognition; the recognition method solves the problem that the existing recognition method of digital music emotion can not recognize sampling-based digital music format, the sorting technology based on a multi-class support vector machine is adopted, acoustic characteristic parameters and music theory characteristic parameters are combined, so as to carry out emotion recognition of digital music; the recognition method comprises the following steps: (1) pretreatment; (2) characteristic extraction; (3) training the multi-class support vector machine; (4) recognition. The music emotion is classified into happiness, impassion, sadness and relaxation, the emotion recognition is carried out based on a sampling-based digital music format file, the common acoustic characteristics in the speech recognition field are not only extracted, and a series of music theory characteristics are extracted according to the theory characteristics of music; meanwhile, the sorting method based on the support vector machine is adopted, the leaning speed is rapid, the sorting precision ratio is high and the recognition efficiency is improved.

The invention relates to synthesis of germanium sulphide glasses and optical devices formed therefrom. In a chemical vapour deposition process, germanium tetrachloride is reacted with hydrogen sulphide at temperatures in the range 450-700° C. to form germanium sulphide. Lower temperatures within this range of 450-550° C. directly produce a glass, whereas higher temperatures within the range of 600-700° C. produce a crystalline powder which can then be reduced to a glass by subsequent melting and annealing. The reaction is preferably carried out at atmospheric pressure or slightly higher. Thin films and bulk glasses suitable for optical waveguides can be formed directly in one processing step as can powders and microspheres. The materials synthesised are of a high purity with low oxide impurities and only trace levels of transition metal ions.

The invention relates to a single-wheel robot system and a control method thereof, belonging to the field of intelligent robot, in particular to a static unbalanced robot which can autonomically control movement balance further to autonomically ride a monocycle and a movement balance control method thereof. The single-wheel robot comprises a robot body and a balance detection and control system and is characterized in that the lower part of the robot body is provided with a single wheel which can rotate front and back, and the middle of the robot is provided with an inertial flywheel which can rotate left and right and is horizontally placed; the balance detection and control system comprises a movement controller, a posture detector and a servo drive controller, wherein the posture detector and the servo drive controller are connected with the movement controller; and the movement controller 10 receives a signal of a posture sensor and processes the received signal according to a control procedure so as to send a control order to control a motor through the servo drive controller to regulate the posture balance of the single-wheel robot. The invention provides a control platform which not only belongs to robotics but also belongs to the field of control science and intelligent control, thereby meeting the needs of multi-subject scientific research and teaching.

A Micro Electro-Mechanical System (MEMS) acceleration sensing device formed of a silicon substrate having a substantially planar surface; a pendulous sensing element having a substantially planar surface suspended in close proximity to the substrate planar surface; a flexure suspending the sensing element for motion relative to the substrate planar surface, the flexure having a both static geometric centerline and a dynamic centerline that is offset from the static geometric centerline; and a metal electrode positioned on the substrate surface for forming a capacitor with the pendulous sensing element, the metal electrode being positioned as a function of the dynamic centerline of the flexure.

An optical fiber, which has a zero-material dispersion wavelength equal to or greater than 2 μm, and a high nonlinear susceptibility χ³ equal to or greater than 1×10⁻¹² esu, and uses tellurite glass having sufficient thermal stability for processing into a low loss fiber, employs a PCF structure or HF structure having strong confinement into a core region. This enables light to propagate at a low loss. The size and geometry of air holes formed in the core region, and the spacing between adjacent air holes make it possible to control the zero dispersion wavelength within an optical telecommunication window (1.2-1.7 μm), and to achieve large nonlinearity with a nonlinear coefficient γ equal to or greater than 500 W⁻¹ km⁻¹.

A calibrated digital-to-analog converter (DAC) (15') has a main DAC (17) having a digital input and an analog output. An on-chip memory (21) stores measured INL values of the main DAC at a few selected input codes, and digital interpolation (50,17) is used to approximate INL error values at all input codes (14 . . . 14<VI>). To cancel the INL errors of the main DAC, outputs of this digital interpolation are sent to a calibration DAC (19), which has an analog output subtracted from the analog output of the main DAC. This subtraction can also be done in the digital domain, removing the need for a calibration-DAC when a main DAC with higher bit-count is designed.

A double-image encryption method based on chaotic and discrete fraction random transform comprises the steps of image expansion, chaotic scrambling, chaotic diffusion, image contraction, image reconstruction and discrete fraction random transform. According to the method, an expanded image is formed by a gray level image I1 and a gray level image I2 through the expansion step, the pixel positions and pixel intensity values are changed through the scrambling and diffusion process based on Logistic chaotic mapping, and then the expanded image is decomposed into a new image J1 and a new image J2 which are stable in white noise distribution through a contraction strategy; the image J2 is normalized and encrypted so as to form a pure phase matrix and a complex matrix J is obtained by multiplying the pure phase matrix and the image J1; finally, the complex matrix encrypted so as to form a temporary image through discrete fraction random transform based on Logistic chaos and the amplitude part is a final ciphertext image. Nonlinearity and randomness of a plaintext image in a spatial domain and a transform domain are enhanced, common attacks can be well resisted, and safety of an encryption system is improved.

The invention discloses a double-image encryption and decryption method which comprises: an image combination step of combing original images; a chaotic scrambling step of scrambling and expanding a combined image by using Logistic chaotic mapping; an image decomposition step of decomposing the chaotically-scrambled image; a modulation phase mask generating step of generating a phase mask based on a chaotic random sequence generated by the chaotic mapping and the decomposed image; a temporary image generating step of generating a temporary image according to the image obtained by decomposition and the phase mask; multi-parameter discrete fraction angle transformation steps of performing two-dimensional DMPFAT transformation on the temporary image and truncating a transformed phase part to obtain a final ciphertext image; an initial value of the chaotic mapping in chaotic scrambling and/or a parameter value of multi-parameter discrete fraction angle transformation are used as private secrete keys. The method disclosed by the invention is convenient for the storage, the transmission and the management of the secrete keys; the secret key space of a cryptographic system is enlarged, and the safety and the effectiveness of the method are improved.

The invention discloses a method for improving angular displacement measurement accuracy of an optical fiber gyroscope. The method comprises the following steps of: (1) fixing the optical fiber gyroscope on a high-accuracy rotary table to keep a measuring shaft of the optical fiber gyroscope in parallel to a rotating shaft of the high-accuracy rotary table; (2) marking scale factors of the optical fiber gyroscope at different rotary table input angular rates to acquire a correspondence table between the angular rates and the scale factors; (3) taking the optical fiber gyroscope off the high-accuracy rotary table and fixing on a measured rotating device to keep the measuring shaft of the optical fiber gyroscope in parallel to the rotating shaft of the measured rotating device; (4) solving an average value of output values of the optical fiber gyroscope in a testing time under the condition that the measured rotating device is static; (5) rotating the measured rotating device and continuously recording each output value of the optical fiber gyroscope; and (6) calculating a rotation angle, namely the angular displacement of the rotating device by using the correspondence table in thestep (2), the output average value in the step 4 and each output value in the step (5).

Provided is a dispersion-shifted fiber for an optical parametric amplifier. The dispersion-shifted fiber is a silica-based optical fiber and has a large nonlinear coefficient and a small effective area. The dispersion-shifted fiber also reduces a bending loss and a splice loss due to a mode field diameter difference between it and a conventional single mode fiber. For this, the dispersion-shifted fiber comprises a rectangular core with a very small radius and a high refractive index, a depressed inner clad, and an outer clad with a ring. The core and/or the inner clad are/is doped with a heavy metal.

The invention provides a multi-exposure high-dynamic range inverse tone mapping model construction method and device. The multi-exposure high-dynamic range inverse tone mapping model construction method comprises the following steps: intercepting a high dynamic range image from an original high dynamic range video, converting the standard dynamic range images into standard dynamic range images, adjusting exposure time to generate multi-exposure standard dynamic range images, and forming a supervised data set by the standard dynamic range images with different exposures and a high dynamic rangeimage with the same normal exposure to serve as a training data set; establishing a generative adversarial network based on a convolutional neural network and skip connection; and establishing a target loss function integrated by the image content features, the intrinsic features and the perception features for the generative adversarial network, and continuously training and optimizing by adopting the training data set to obtain a final model. According to the multi-exposure high-dynamic range inverse tone mapping model construction method, the brightness of the overexposure or underexposureimage can be adjusted, and the effect of the generated high-dynamic-range image is improved, and the brightness characteristic and the chrominance characteristic of the high-dynamic-range image are considered, and inverse tone mapping of the high dynamic range is better achieved.

A method and system for privacy-preserving ridge regression using masks is provided. The method includes the steps of requesting a garbled circuit from a crypto service provider, collecting data from multiple users that has been formatted and encrypted using homomorphic encryption, summing the data that has been formatted and encrypted using homomorphic encryption, applying prepared masks to the summed data, receiving garbled inputs corresponding to prepared mask from the crypto service provider using oblivious transfer, and evaluating the garbled circuit from the crypto service provider using the garbled inputs and masked data.

A two-point modulator using a voltage control oscillator includes: a modulation section including a feedback circuit for performing feedback control of a signal outputted from the voltage control oscillator based on an inputted modulated signal, and a feedforward circuit for calibrating the modulated signal and outputting the calibrated modulated signal to the voltage control oscillator; a signal output section for, upon calibration processing, outputting a predetermined reference signal in place of the modulated signal, to the modulation section; and a gain correction section for, in a state where the feedback circuit is in an open loop state, calculating a frequency transition amount of the reference signal outputted from the voltage control oscillator, and correcting a gain used for calibration of the modulated signal performed by the feedforward circuit, based on the calculated frequency transition amount.

An optical fiber includes a first core with a first refractive index located in a central portion of the optical fiber; a second core with a second refractive index located in an outer periphery of the first core; a third core with a third refractive index located in an outer periphery of the second core; and a cladding with a fourth refractive index located in an outer periphery of the third core, where among the refractive indices, the first one>the third one>the fourth one>the second one. The absolute value of dispersion at the wavelength of 1550 nm is not more than 20 ps/nm/km. The effective area at the wavelength is not more than 15 μm². The nonlinear constant n₂/Aeff at the wavelength of 1550 nm is equal to or more than 25×10⁻¹⁰/W.