2929 results about "Dynamic range" patented technology

The invention relates to the technical field of organic display data processing, in particular to an organic display interface multichannel signal modulation method, which comprises the following steps of: generating a standardized data set through gamma correction and gamut mapping, and inputting the standardized data set into time sequence, amplitude and frequency sub-models deployed in a federal learning framework; and respectively analyzing the pulse width-refresh rate relevance, the driving voltage-brightness nonlinear relationship and the conduction period-mobility dynamic response, generating an optimized weight coefficient, and fusing the optimized weight coefficient into a boundary constraint condition of a cross-sub-model. A migration rate parameter index, amplitude gradient calculation and duty ratio correction service of chain calling is constructed based on a service grid, a reinforcement learning dynamic routing strategy is combined to avoid resource competition nodes, and brightness error characteristics are periodically fed back to reversely optimize gradient updating. According to the method, the conflict between time sequence deviation accumulation and local optimum is effectively suppressed, the color gradation continuity, the dynamic range and the edge sharpness are improved, and efficient cooperative modulation of multi-channel signals is realized.

The invention relates to the technical field of VR display control, in particular to an LBE enhanced multi-mode perception interaction VR display control method and system, and provides the following scheme that multi-mode perception data of a user is collected through VR head display equipment and a hand controller, position data and action data are fused by adopting a Kalman filtering algorithm, and the multi-mode perception interaction VR display control method and system are obtained. And calculating a continuous motion track of the user, dynamically adjusting the activity range based on the continuous motion track, and generating a directional dynamic range. And performing synchronization, feature extraction and weighted fusion processing on the multi-modal data to generate an interaction state vector. And analyzing real-time behaviors, action intentions and physiological feedback of the user according to the interaction state vector, judging the operation sensitivity and the emotional state of the user, and dynamically adjusting response parameters of the VR equipment. According to the method, the real-time performance, the flexibility and the immersion of large-space VR interaction are effectively improved.

The invention discloses a three-dimensional panoramic intelligent monitoring method and device for a coal conveying gallery, and relates to the technical field of industrial automation and intelligent monitoring, and the method comprises the steps: obtaining initial point cloud data and an initial multispectral image of the coal conveying gallery through a rotary laser radar, a high-dynamic-range camera and an infrared thermal imager; inertial measurement units are arranged in the devices to collect vibration data, and the vibration data are used for carrying out vibration compensation on point cloud and images to generate target point cloud data and target multispectral images; thirdly, performing three-dimensional modeling and multispectral texture mapping on the coal conveying gallery based on the data to obtain a target three-dimensional model; and finally, mapping equipment, personnel and environment data into the three-dimensional model in real time, displaying each piece of state information, and realizing comprehensive monitoring. According to the application, high-precision three-dimensional panoramic monitoring can be realized in a complex vibration environment of the coal conveying gallery, and the safety management level and the operation and maintenance efficiency of the coal conveying gallery are effectively improved.

The invention provides a multi-band radio frequency power amplifier dynamic range adaptive test method and a system thereof. The method comprises the following steps: acquiring an output signal frequency spectrum of a tested radio frequency power amplifier, extracting characteristic frequency points including a center frequency and a bandwidth, matching the characteristic frequency points with a preset frequency band database, and automatically loading a test scheme of a corresponding frequency band. According to the method, the frequency bands are automatically matched through spectrum sensing, the test scheme is loaded, the multi-frequency-band characteristics can be rapidly adapted without manual parameter configuration, index missing test is avoided, and the switching efficiency is improved; dynamically optimizing test parameters by using a genetic algorithm, and accurately capturing a nonlinear distortion critical point while balancing coverage and efficiency; by combining multi-dimensional signal analysis and a deep learning model, the problem of missing detection of traditional single-domain analysis is solved, and the abnormal signal recognition capability is remarkably enhanced; through linear model correction, impedance dynamic matching and index threshold iteration, full automation and high precision of the test are realized, the environmental interference influence is effectively reduced, and a self-adaptive test system is constructed.

The invention discloses a document layout analysis method based on a hybrid method, and belongs to the technical field of digital image processing. Named body recognition and relation extraction are carried out on each text node. The invention provides a novel channel shuffling dynamic range convolution algorithm to ensure that global channel features can learn features of similar pixels rather than adjacent pixels, and an adapter is used to learn text features with less parameter quantity. And finally, taking the predicted text block category and relationship as a result. Finally, a challenging document data set is used as a verification index (FUNSD), and the result shows that the performance of the method can be obviously improved.

The invention provides a roof leakage point intelligent positioning method and system based on machine vision, and relates to the technical field of constructional engineering detection.The method comprises the steps that 1, roof image data and surface texture, temperature field and environment illumination parameters are collected through a visible light and infrared thermal imaging two-channel sensing device, a multi-dimensional data set is constructed, and a multi-dimensional data set is established; generating visual data through dynamic range correction based on the multi-dimensional data set; step 2, based on visual data, performing detection area division on the acquired image, extracting a local temperature difference gradient, a surface texture abrupt change rate and a liquid flow track parameter, and establishing a multi-index characteristic matrix; according to the method, the features of the roof image are extracted through machine vision, intelligent positioning, risk grade division and dynamic evolution tracking of roof leakage points are realized, and the leakage positioning accuracy is improved.

A burst-mode optical receiver for an optical line terminal (OLT) of a passive optical network (PON) comprises a variable optical attenuator (VOA), a photodiode (PD) which may be a pin-PD or an APD, a transimpedance amplifier (TIA), and a feedback / control circuit for adjusting a bias voltage of the VOA in response to a burst-mode optical input signal level, to provide an attenuated optical output signal to the PD having a narrower dynamic range. Providing signal level adjustment in the optical domain mitigates the requirement for a burst-mode TIA with a large dynamic range and provides for fast switching. The burst-mode optical receiver may comprise a waveguide configuration, wherein a first electro-absorption modulator (EAM) is operable as the VOA and a second EAM is operable as the photodiode. A monolithically integrated electro-photonic circuit comprising the VOA, PD, TIA and feedback / control circuit may be provided using InP-based semiconductor materials.

The invention discloses a multi-mode speech enhancement system based on audio and video, and the system comprises a data collection module which collects target audio, background noise and interference sound through a microphone; a camera is used for synchronously collecting environment visual information, including facial expressions and mouth shapes of speakers, and noise and interference sounds are normalized to ensure that the noise and the interference sounds are matched with the dynamic range of a target audio; the data analysis module is used for preprocessing the collected audio, including noise suppression and speech enhancement, and improving the speech definition by applying a deep learning model; through the noise reduction and voice enhancement technology, the interference of background noise can be effectively reduced, and the voice is clearer. Thus, audiences can more easily understand speech contents, especially in noisy environments, such as public transport means or busy streets; and the voice recognition accuracy is improved, and the quality of the voice signal can be improved.

The invention relates to a target classification method based on dynamic multi-feature data fusion, and belongs to the technical field of radar signal processing. Comprising the following steps: 1) acquiring echo data of a target by a radar, preprocessing the echo data, and constructing a training data set by using time-frequency information, distance period information and distance period data of a signal; 2) constructing a dynamic multi-feature fusion network target classification model which comprises a global feature extraction module, a local feature extraction module, a dynamic feature extraction module and a feature fusion module; 3) inputting a training data set to carry out iterative optimization training on the dynamic multi-feature fusion network to obtain an optimal network parameter; and 4) preprocessing real-time radar echo signals, inputting the real-time radar echo signals into the trained dynamic multi-feature fusion network for testing, and completing target classification. According to the method, the Doppler information and the dynamic distance amplitude information of the radar signal can be subjected to feature extraction at the same time, the target classification accuracy is improved, and the actual application requirement is met.

Disclosed is a method for enhancement of a low-light image based on reinforcement learning and aesthetic evaluation. The method include: generating images of non-normal luminance under different lighting scenes, and constructing a training dataset for a reinforcement learning system based on the images; initializing the training dataset, a policy network, and a value network; updating, based on a no-reference reward score and an aesthetic assessment reward score, the policy network and the value network; completing model training and outputting an enhanced image result. By expanding the action space range defined in reinforcement learning, the enhancement operations for the input low-light image gain a greater dynamic range, offering higher flexibility for real-world scenarios and better meeting low-light image enhancement needs. Additionally, by incorporating the aesthetic quality assessment scores as part of the loss function, the enhanced image achieves better visual effects and higher subjective user evaluation scores.

The invention provides a low-light large dynamic image enhancement method based on dynamic weight and pyramid fusion. The low-light large dynamic image enhancement method is suitable for complex illumination imaging in security monitoring. According to the method, the global weight and the detail weight of an image are obtained according to the global average brightness, the local brightness and the gradient of an image sequence, and the total weight is obtained through weighting. And the halo phenomenon easily occurring in the fusion process is reduced through guide filtering. And combining Gaussian pyramid and ratio pyramid decomposition and weighting to obtain a final fusion image. According to the method, the problems of detail loss and limited dynamic range in the low-light scene are effectively solved. Experiments show that the method is superior to the prior art in information entropy, average gradient, edge strength, spatial frequency and other key indexes. The technology has an all-time adaptive capability, is suitable for the fields of intelligent transportation, urban security and protection, low-illumination monitoring and the like, and has a wide market prospect.

The invention discloses a roadway deformation monitoring device based on a millimeter wave radar. The roadway deformation monitoring device comprises a main shell, the millimeter wave radar, a laser sensor, a PCBA (Printed Circuit Board Assembly) and a bubble level, and an integrated multi-modal data fusion analysis module, a dynamic measurement range intelligent adjustment module, a real-time deformation prediction and control module and a self-calibration and fault diagnosis module. The multi-modal fusion system fuses radar and laser data through a Kalman filtering algorithm, and full-scale high-precision monitoring is realized by using median filtering, Gaussian fitting preprocessing and dynamic weight distribution; the dynamic range measurement module intelligently switches measurement and frequency sweeping ranges based on a decision tree algorithm according to mining progress, geological conditions and distance fluctuation coefficients; the real-time prediction module adopts an ARIMA model to predict the deformation in advance, and automatically adjusts the scanning angle and the acquisition frequency; the self-calibration module uses a high-precision reflecting plate to correct errors everyday; the system supports 5G communication and remote control, has the advantages of high precision, self-adaption and high reliability, and is suitable for long-term deformation monitoring of mine roadways.

The invention discloses a multimedia image optimization method and system, and relates to the technical field of image optimization, and the method comprises the steps: obtaining a to-be-optimized multimedia image, carrying out the HSI color space conversion, extracting a brightness component in an HSI color space, and carrying out the brightness enhancement; the contrast ratio of the enhanced brightness component is adjusted, and brightness area division and dynamic range optimization are carried out; and carrying out edge enhancement on the optimized brightness component, recombining to generate an enhanced HSI image, and converting the enhanced HSI image back to the RGB color space to obtain an optimized multimedia image. According to the method, the brightness, the contrast ratio and the details of the image are comprehensively optimized, the brightness layering sense is improved, the problem that the details of the image in highlight and dark areas are lost is solved, the marginal definition and the color fidelity of the image are improved, the visual quality of the multimedia image can be remarkably improved, and the actual application requirements of multiple scenes are met.

The invention relates to the technical field of robot control, and discloses a heterogeneous robot control system with intelligent and multi-modal perceptual driving, comprising: a quantum clock reference module used for generating a globally synchronized picosecond-level time reference signal and synchronizing a multi-modal sensor clock through a microwave pulse; the multi-mode sensing module comprises an event camera of which the dynamic range is greater than or equal to 120dB, a six-dimensional force sensor of which the noise density is less than or equal to 0.01 N / Hz and an inertial measurement unit of which the zero-bias stability is less than or equal to 0.8 degree / h; and the random differential manifold module adopts a four-layer full-connection neural network, and an input layer receives multi-modal data. According to the invention, a diamond NV color center system of the quantum clock reference module and a microwave pulse coupling technical scheme are adopted, so that a picosecond-level global signal synchronization effect is achieved. Compared with a scheme depending on a traditional crystal oscillator or a GPS clock in the prior art, the defects that electromagnetic interference is prone to occurring and signal jitter is large are overcome, and high-precision time sequence alignment of multi-module cooperation is ensured.

The invention discloses a laser radar test method based on a long-distance target distance measurement result, and the method comprises the following steps: obtaining a laser radar echo signal, and extracting the frequency spectrum information of the echo signal; doppler frequency shift characteristic parameters are extracted according to the frequency spectrum information, and the relative speed of the target is calculated based on a preset frequency shift calculation model; acquiring multi-source sensor data associated with the laser radar test platform, and fusing to construct a dynamic ranging reference coordinate system; performing dynamic motion compensation on a laser ranging result according to the relative speed of the target to obtain a target initial measurement distance after Doppler correction; mapping the target initial measurement distance after Doppler correction to the dynamic distance measurement reference coordinate system to obtain a remote target final distance measurement result aligned with a test platform coordinate system; and outputting the accurate position and moving state information of the target in the three-dimensional space according to the final distance measurement result. According to the invention, the ranging precision and stability can be improved.

An electronic interface device is configured for use with associated optics for employing coded light for illuminating a liquid, gaseous, or other analyte or target, such as for measuring or determining a composition or other property of the analyte or target. The arrangement can help enable a far higher level of performance, such as for spectral analysis, imaging, or both. A shared common clock on both the illumination and receive sides of the system can be included, such as together with one or more of reference detection, per-channel wavelength-dependent modulation, multiple concurrent feedback loops at different levels, orthogonal and circulant codes, phase-scrambling to better utilize ADC dynamic range, oversampling, and other signal processing can be provided.

The invention discloses a wavefront reconstruction method for improving the dynamic range of a Shack-Hartmann wavefront sensor. Belongs to the technical field of wavefront detection. According to the method, a new Shack-Hartmann wavefront detection method is needed, the effective dynamic range of a Shack-Hartmann sensor can be expanded on the premise that the hardware complexity is not remarkably increased, local mismatching accumulation can be avoided through a reasonable matching strategy, and the detection accuracy is improved. Therefore, high-precision and high-reliability wavefront reconstruction can be realized in a large-amplitude wavefront distortion scene. Comprising the following steps: S1, detecting by a Shack-Hartmann wavefront sensor to obtain a light spot distribution diagram H (x, y); s2, extracting a light spot centroid pixel-level position through a matched filtering algorithm; s3, calculating the sub-pixel-level position of the mass center of the light spot through Gaussian interpolation; s4, a mass center-subaperture matching relation is established through a Kuhn-Munkres algorithm; and S5, calculating the wavefront slope according to the corresponding relationship between the centroid and the sub-aperture, and then recovering the wavefront according to the wavefront slope.

The present application provides a method and device for processing the dynamic range of an image, wherein the method includes: obtaining the first coordinate value of the first interpolation point and the first coordinate value of the third interpolation point related to the first cubic spline curve; determining the first coordinate value of the second interpolation point related to the first cubic spline curve according to the histogram information of the first brightness interval of the image to be processed, the first brightness interval being the interval between the first coordinate value of the first interpolation point and the first coordinate value of the third interpolation point, the first coordinate value of the first interpolation point, the first coordinate value of the second interpolation point and the first coordinate value of the third interpolation point being used to determine the function of the first cubic spline curve, the function of the first cubic spline curve being used to map the brightness value of the pixel whose brightness value is within the first brightness interval in the image to be processed to the first target brightness value, thereby correcting the dynamic range of the image to be processed. The present application improves the display effect of the image.

The invention discloses an optical fiber distributed sound wave sensing device based on double chirped pulses, which relates to the technical field of distributed optical fiber sensing and comprises a laser module (1), a chirped pulse modulation module (2), a sensing optical fiber module (3) and a Rayleigh scattering signal separation and receiving module (4), laser generated by the laser module passes through the chirped pulse modulation module to obtain double chirped pulse signals with different frequency bands, excites Rayleigh scattering signals in the sensing optical fiber module and inputs the Rayleigh scattering signals to the Rayleigh scattering signal separation and receiving module for signal separation and demodulation. The sensing optical fiber is inquired through double chirped pulses of different frequency bands, two Rayleigh scattering signals are excited to achieve a wide range and high sensitivity, the advantages of the two signals are combined through an adaptive filtering algorithm, the dynamic range of distributed sound wave detection is improved by one order of magnitude, and the detection accuracy is improved. The acquisition of large chirp bandwidth Rayleigh scattering signals is realized at a low sampling rate, and a distributed sound wave sensing technology with a large dynamic range is realized.

The invention belongs to the technical field of image processing, and particularly relates to an image processing method for high-voltage cable defects. The method comprises the following steps: firstly, performing multi-scale Gaussian smoothing and other processing on an acquired cable surface real-time image to solve the problem of uneven illumination, and enhancing a brightness dynamic range and local details; afterwards, the image resolution and details are improved through residual learning and frequency domain high-frequency enhancement; performing multi-scale edge detection on the to-be-detected image to obtain edge features; and finally, optimizing edge details through adaptive edge enhancement and noise suppression. The method effectively improves the accuracy and reliability of high-voltage cable defect detection, has remarkable advantages in image processing efficiency, precision and stability, and meets the requirements of high-voltage cable production on surface defect image processing.

A display device includes: a first sub-pixel configured to receive a first data voltage and drive a first light emitting element included in the first sub-pixel; a second sub-pixel configured to receive a second data voltage and drive a second light emitting element included in the second sub-pixel; a third sub-pixel configured to receive a third data voltage and drive a third light emitting element included in the third sub-pixel. At least one of dynamic ranges, maximum voltages, and minimum voltages of the first data voltage, the second data voltage, and the third data voltage are different from each other. An emission time of the first light emitting element, an emission time of the second light emitting element, and an emission time of the third light emitting element are different from each other.

Standard event-based cameras capture changes in luminance with high temporal resolution, but they do not measure static images. An event-based imaging system and methods that can capture static intensity is provided that eliminates the need for a dual-sensing system and significantly reduces the cost, footprint, and data bandwidth of the imaging system. The system is suited for applications where high temporal resolution and dynamic range are required, but where static scene information is also important. The system has a Dynamic Vision Sensor for capturing data and a processing unit for data aggregation, a fast approximate extraction of static information from noise, and scene reconstruction. The methods may use a model for reconstruction or may calibrate during system deployment by making measurements of known static luminance and recording average noise levels over a predetermined period to map between luminance and noise for reconstruction.

The invention relates to a design method of a star magnitude large dynamic range spectrum modulation system. The design method comprises the following steps: S1, adopting a cylindrical collimation beam expanding system, a star magnitude large dynamic range light splitting system and a DMD device; the star magnitude large dynamic range light splitting system is composed of a large dynamic light splitting optical system and an energy modulation optical system. The star isometric large dynamic light splitting optical system is composed of a steering reflecting mirror, a grating and an imaging reflecting mirror. The energy modulation optical system is composed of a steering mirror, a grating and a plane mirror. S2, selecting a DMD device and a grating; and S3, carrying out optimization design on the cylindrical collimation beam expanding system and the star magnitude large dynamic range beam splitting system. According to the invention, the cylindrical collimation and beam expanding system is adopted to replace a physical slit and a collimating mirror in the Czerny-Turner type plane diffraction grating optical structure, so that the imaging quality of the large-dynamic-range spectrum modulation optical system can be improved; and the problem of contradiction between the slit and the imaging quality in the initial structure can be solved.

The invention discloses a processor, a chip product, computer equipment and a tensor calculation method, and relates to the technical field of processors. A first tensor is stored in a first storage unit of the processor; a plurality of scaling factors corresponding to the first tensor are stored in a second storage unit of the processor, the scaling factor set comprises a plurality of scaling factors, each scaling factor corresponds to one sub-block in the first tensor, each sub-block comprises at least one element in the first tensor, and each sub-block comprises at least one element in the first tensor; the scaling factor is used for adjusting the dynamic range of each element in the same sub-block; a tensor calculation engine of the processor is used for reading a first tensor from the first storage unit and reading a scaling factor set from the second storage unit; and for any sub-block in the first tensor, executing logical operation on elements in the sub-block according to the scale factor corresponding to the sub-block to obtain an operation result. According to the method and the device, the tensor calculation engine can realize accurate operation of the low-precision tensor through the scale factor of the sub-block granularity.

The present disclosure relates to an active graded magnetic compensation system and method for an unshielded SERF magnetometer. The system comprises: a detection laser, a half-wave plate, a magnetic field compensation coil, a magnetic field modulation coil, a non-magnetic heating assembly, an alkali-metal atomic vapor cell, a polarizing beam splitter prism, a balanced photodetector, a graded feedback control module, a pump laser, a quarter-wave plate, a focusing lens and a photodetector. By means of the present disclosure, a graded compensation system is used to realize high-precision compensation for a magnetic field within a large dynamic range in a geomagnetic environment, thereby ensuring that an instrument carries out high-sensitivity magnetic field measurement under geomagnetic conditions.

The invention relates to a partial discharge detection system based on Rydberg atoms in the field of partial discharge detection of electrical equipment, the system comprises an optical path frequency stabilization module, an atom probe, a photoelectric detection module, a controller module and a power frequency synchronous trigger signal circuit, the optical path frequency stabilization module locks 852.35 nm probe light and pump light frequency less than 509.53 nm through SAS and EIT technologies, and the atomic probe is used for detecting the frequency of the probe light. Inducing transition from a cesium atom ground state to a first excited state to a Rydberg state; under the effect of partial discharge pulse, the atomic probe outputs a modulated EIT transmission spectrum signal, after photoelectric conversion and filtering, the controller module analyzes the amplitude and phase information of the pulse in combination with a power frequency synchronizing signal, the upper computer generates a PRPD diagram to analyze the discharge correlation, the detection sensitivity of the system reaches-50dBm, the frequency band of 300MHz-2GHz is covered, the dynamic range is greater than or equal to 50dB, the equivalent height is greater than or equal to 20mm, and the discharge correlation is greater than or equal to 20mm. Broadband and high-resolution detection is achieved, the problem of phase synchronization is solved, and the device is suitable for high-voltage environment on-line monitoring.

The invention discloses a global enhancement method for a wide-area astronomical image. The method comprises the following steps: carrying out high-frame-frequency image acquisition on a starry sky area; converting the reference star from a star catalogue position to an observation position, and resolving a linear negative film model; geometric distortion correction is carried out based on the resolving result; based on the image after geometric distortion correction, matching and identifying a reference star in a full view field, and obtaining celestial coordinate information of all fixed star images in the view field; on the basis of celestial coordinate information, correcting a poorer astronomical effect of the astronomical image, establishing an ideal coordinate system taking the center of a view field as a tangency point, and projecting a time sequence observation image to the ideal coordinate system; translating the processed graph, accumulating all translated images, and averaging all accumulated pixel positions to realize image superposition enhancement, thereby effectively improving the dynamic range of a detector and the detection capability of a wide-area astronomical observation system; and meanwhile, the signal-to-noise ratio of the astronomical image can be effectively improved, and the astronomical image centering and light measuring precision can be further improved.

The invention relates to the technical field of production management, in particular to an equipment fault alarm method and system applied to production management, and the method comprises the steps: carrying out the signal collection of a vibration signal and a temperature of equipment under a normal operation condition; calculating dynamic characteristic parameters of the equipment operation state, and establishing a reference characteristic model; comparing the real-time measured value with the dynamic change range in the reference feature model one by one, calculating a real-time deviation value, and identifying the abnormality of the running state of the equipment; classifying the abnormal states of the equipment according to the characteristics and the change trend of the deviation value; and adjusting the dynamic change range in the reference feature model, and triggering different alarm levels according to the real-time deviation value. Through the fault detection method and device, the problems that in the prior art, dynamic changes of operation conditions cannot be adapted, deviation characteristics and trends are difficult to analyze, and an anomaly classification and grading alarm mechanism is lacked are effectively solved, the accuracy and adaptability of fault detection are improved, and efficient and reliable decision support is provided for production management.

The invention relates to the technical field of camera control, and discloses a megapixel camera control method based on multifunctional fusion. The method comprises the following steps: acquiring original image data of an image sensor, and preprocessing to generate an initial brightness distribution diagram; and calculating a scene dynamic range to determine a target exposure parameter set containing exposure time, a gain coefficient and an aperture value, synchronously analyzing color temperature information, matching a preset white balance model to output a correction coefficient, and combining the two to generate a first control instruction to be sent to the image sensor. Receiving a touch screen focusing signal, and extracting a focus area coordinate to adjust a stroke parameter of a focusing motor; ambient light intensity is monitored in real time, and when the ambient light intensity is lower than a preset threshold, a flash lamp compensation module is activated and a compensation light quantity value is calculated. In the photographing mode, the output power of the flash lamp is adjusted according to the compensation light quantity value, the target exposure parameter set is recalibrated based on the focus area coordinates, and the camera complex scene adaptation and imaging quality are improved.

The present invention belongs to the field of optical fiber measurement technology, and specifically relates to an optical fiber distributed polarization crosstalk fast measurement device based on optical frequency domain interference. The device comprises a tunable laser source module (1), a device to be measured module (2), an auxiliary interferometer module (3) and a signal acquisition and analysis module (4), and is characterized in that: a high-coherence tunable laser is used to perform fast wavelength scanning on an optical signal, and a linear frequency modulated continuous light (501) is output, which passes through a polarizer (202), a device to be measured (206) and an analyzer (210) in sequence, and a beat frequency signal generated by the interference of a coupling light (704) and a transmission light (701) is obtained in the optical frequency domain, and then an optical path difference and a normalized interference signal amplitude are calculated, thereby respectively determining the polarization crosstalk position and intensity. The present invention, in combination with a tunable laser, gets rid of the dependence on an optical path scanning delay device in polarization crosstalk measurement, greatly improves the speed and dynamic range of polarization crosstalk measurement, and has a compact structure, is stable and reliable, and is easy to implement.