116 results about "Spiral scanning" patented technology

The invention discloses a precision aspheric part profile measurement method. The method comprises the following steps: constructing a measurement system; point cloud data acquisition: controlling relative movement of a spectrum confocal sensor and a rotatable workpiece table in the measurement system, and acquiring three-dimensional point cloud data of the surface of the aspheric part in a spiral scanning mode or a layered annular scanning mode; performing fusion processing on the collected point cloud data, and preliminarily estimating the optical axis direction based on principal component analysis (PCA); and based on the fused point cloud data, calculating parameters such as an optical axis position and an attitude of the aspheric part, determining a curved surface equation of the optical element, calculating a transformation matrix, and performing point cloud coordinate transformation. According to the method, the spectrum confocal measurement technology is utilized, the high-precision position information aspheric part appearance measurement technology is fused, the three-dimensional appearance data of the whole surface of the aspheric part can be obtained in a non-contact and high-precision mode, and key parameters such as the optical axis of the aspheric part can be determined.

The invention relates to the technical field of transformer coil detection, and discloses a detection device for a transformer coil, which comprises a base frame, the driving mechanism is installed on the base frame and used for providing linear motion in the X direction, the Y direction and the Z direction; according to the scheme, by means of the detection mode that spiral scanning motion is combined with self-adaptive contact and laser ranging, the fatal defects that manual detection is low in efficiency, high in subjectivity and prone to making mistakes and missing detection are overcome, complete three-dimensional shape data of the surface of the coil can be rapidly and accurately obtained, and the detection accuracy is improved. According to the method, local fine defects which are difficult to discover manually can be detected, the reliability and the consistency of detection results are improved, objective and quantitative data support is provided for evaluating the winding quality, the mechanical strength and the long-term operation reliability of a transformer coil, and the manufacturing quality of a transformer and the operation safety of a power grid are guaranteed.

The invention discloses a multi-photon optical fiber array endoscopic imaging device and method based on hollow spiral scanning, and belongs to the technical field of biological imaging. The device comprises a laser collimation and beam expansion module, a pulse chirp module, a light beam switching module, an endoscopic detection module and a fluorescence acquisition module. The light beam switching module comprises an optical switch and an optical fiber array composed of at least two low-group-velocity dispersion single-mode optical fibers. The endoscopic detection module drives the optical fiber array to execute hollow spiral scanning through the resonance driver, the track starting point is not located at the view field original point, and central dense sampling is avoided. According to the method, the optical fibers are switched through the optical switch, so that the optical fibers execute spatial non-coincident hollow spiral scanning, multi-photon signals such as fluorescence, second harmonic and fluorescence lifetime are acquired, a central hole is complemented through image reconstruction and fusion, and a complete field imaging result is obtained. Light damage of the central area is remarkably reduced, the imaging resolution uniformity is improved, the fused image is true and reliable based on physical sampling, and the system stability is high.

The invention discloses a new energy automobile motor shaft damage detection method, and relates to the technical field of mechanical part tests.The new energy automobile motor shaft damage detection system comprises an eddy current detection module, an ultrasonic detection module and a track setting module, and the eddy current detection module sweeps a motor shaft through an eddy current detection method; the estimation module is used for estimating approximate distribution of distortion cracks and fatigue cracks, the ultrasonic detection module is used for scanning axial areas of the distortion cracks and the fatigue cracks by adopting a spiral scanning method and a circumferential stepping scanning method, and the track setting module is used for setting a scanning track according to the axial distribution of the fatigue cracks and the distortion cracks. The eddy current detection module comprises an eddy current sensor, a signal processing unit and a data acquisition module, the signal processing unit is electrically connected with the eddy current sensor, and the data acquisition module is electrically connected with the signal processing unit.

The invention discloses a cable strength detection method and system, relates to the technical field of cable detection, and solves the technical problems that the metal cable detection process is not standard, the defect positioning and quantification precision is low, the contradiction between the non-metal cable detection efficiency and precision is prominent, and microscopic verification is lacked. A high-frequency arc-shaped phased array probe, a bicrystal oblique TOFD probe and an oil-based high-viscosity coupling agent are adopted, a layered focusing technology is combined, laser ultrasonic non-contact detection and ultrasonic microscope high-resolution imaging are adopted for a non-metal cable, and spiral scanning and circumferential fixed-point scanning are combined, so that the defect omission ratio in the circumferential direction is reduced; the abnormal region judgment takes a calibration test block as a reference, the misjudgment rate is reduced, the authenticity is verified through coarse scanning and locking of a core region, fine scanning and acquisition of a high-resolution image and multi-angle imaging, the defect type judgment basis is clear, and the misjudgment rate is reduced.

The invention discloses a remote sensing image segmentation method based on double-sequence significance guidance and space gating, and belongs to the field of computer vision. According to the method, firstly, a training data set is constructed, features are extracted through a double-branch encoder, the features comprise a global sequence scanning branch and a saliency spiral scanning branch, the global sequence scanning branch adopts horizontal and vertical forward and backward scanning to reserve global layout information, and the saliency spiral scanning branch positions an initial anchor point through a saliency map to generate a bidirectional spiral path and focus key area features. In the encoder, deep fusion of the convolutional neural network and the Mama model is realized through a space gating state transition mechanism, and the Mama hidden state is guided to be updated by using space features. The decoder completes double-branch feature calibration and alignment through a multi-stage cross-scanning feature calibration module, reinforces semantic consensus and optimizes detail differences, and adopts a three-head supervision and consistency constraint strategy combined training model. According to the method, the segmentation precision and efficiency of the high-resolution remote sensing image complex ground feature are remarkably improved.

The invention relates to the technical field of steel roller surface treatment, and discloses a surface treatment method of a conductive steel roller for copper foil, which comprises the following steps: S1, pretreating the surface of the conductive steel roller, including ultrasonic cleaning and absolute ethyl alcohol wiping; s2, setting laser average power, pulse frequency, pulse width, light spot diameter, axial feeding speed, radial feeding step pitch and scanning overlapping rate based on the preprocessed conductive steel roller according to initial roughness and target roughness of the conductive steel roller; and S3, based on the set online laser grinding, when the steel roller rotates, a laser machining head spirally scans the surface of the steel roller at the set feeding speed and step pitch in the axial direction and the radial direction, and surface convex peaks are removed through laser. By means of the laser grinding and real-time feedback control technology, accurate adjustment of the surface roughness and shape of the steel roller is achieved, the machining efficiency and the consistency of the surface quality are improved, and high precision and uniformity are ensured.

The invention discloses an SLM (selective laser melting) forming method of an In718 high-temperature alloy material, which comprises the following steps: selecting 15-35 mu m spherical powder, and carrying out ultrasonic screening, plasma activation and 50-100nm aluminum oxide coating modification to improve wettability and oxidation resistance; in an inert gas cabin, a platform is preheated at the temperature of 100-150 DEG C, and a pre-tightening layer with the density larger than or equal to 98% is formed through vibration auxiliary powder laying; a pulse fiber laser is adopted to dynamically modulate parameters, a thin-wall area and a solid area are matched with a spiral scanning path, and the cooling rate of a molten pool is reduced by 30%-50%; a molten pool is monitored through dual-wavelength infrared temperature measurement and a high-speed CCD, and laser frequency compensation and argon blowing stable convection are triggered; constructing a three-layer temperature gradient; by means of the method, the density and the high-temperature performance of the formed part are improved, and the method is suitable for manufacturing of aerospace complex structural parts.

The invention relates to the technical field of laser radars, and discloses a laser radar beam scanning path optimization and energy efficiency management method based on an FPGA, and the method comprises the following steps: S1, collecting the point cloud data and obstacle distribution information of a target region in real time through an environment sensing module; s2, generating a dynamic scanning path based on space partition, and generating a self-adaptive spiral scanning path for avoiding obstacles; s3, executing multi-parameter collaborative energy efficiency adjustment, and dynamically matching the laser transmitting power, the receiver gain and the MEMS driving voltage; and S4, synchronously outputting the optimized scanning path and the energy efficiency instruction to a laser driving circuit and a light beam deflection controller through a hierarchical scheduling module of the FPGA. A detection area is divided into priority grids through an FPGA, a self-adaptive spiral scanning path is generated, a hardware-level sudden obstacle avoidance algorithm is utilized, the response time is shortened to be smaller than or equal to 5 ms, and the integrity of point cloud data is ensured.

The invention relates to the technical field of medical image processing, in particular to an intestinal polyp segmentation method and system based on a spiral scanning state space model, and the method comprises the following steps: S1, obtaining and preprocessing an image; s2, performing multi-scale context feature coding; s3, self-adaptive frequency domain feature enhancement is carried out; s4, performing decoding and segmentation prediction; and S5, carrying out threshold processing to generate a binary segmentation mask image. According to the method, the boundary integrity of an irregular target is kept through a dynamic multi-focus spiral scanning strategy, and the adaptive frequency domain filtering module is introduced to enhance the discriminability of key features, so that the segmentation precision and robustness in complex medical image tasks such as intestinal polyp and the like are comprehensively improved.

The invention discloses an X-ray tomography in-situ testing device and method suitable for a seawater high-pressure environment. The X-ray tomography in-situ testing device comprises an X-ray source, a collimator, a flat panel detector, a seawater pressure cabin and a seawater pressure control device. During detection, the X-ray source is started, the sample is started to move spirally, the flat panel detector collects projection data, then a cross-sectional image is reconstructed through an existing method, and the X-ray tomography in-situ test of the sample in the seawater high-pressure environment is achieved. In order to solve the problems that a seawater in-situ detection device in the prior art is insufficient in penetrating power, low in resolution ratio, limited in sample movement freedom degree and poor in high-pressure environment adaptability, the invention discloses an X-ray tomography device and method, and in-situ dynamic imaging can be carried out on a large component in a seawater high-pressure environment. In-situ observation of micron-sized damage and defects (100-200 microns) is realized, and meanwhile, an imaging view field is expanded through a multi-degree-of-freedom sample table and a spiral scanning method, so that the detection efficiency is improved.

The invention relates to a high-rise building heat insulation layer hollowing detection method and system based on an unmanned aerial vehicle. The method comprises the following steps: converting point cloud data of a building into a three-dimensional network model by using a Poisson reconstruction algorithm, and generating a spiral scanning unmanned aerial vehicle route; the unmanned aerial vehicle flies according to the route of the unmanned aerial vehicle, collects flight attitude data and reflection echo signals of the laser radar, and suppresses flight jitter interference through an anti-jitter signal processing algorithm; performing time-frequency domain analysis on the radar signals after noise reduction, extracting double-interface reflection coefficient features, inputting the double-interface reflection coefficient features into a support vector machine classifier to identify a hollowing region, and calculating hollowing position coordinates; and generating a three-dimensional thermal image of the building, and outputting a hollowing detection report. Anti-interference processing is carried out through an anti-jitter signal processing algorithm, so that hollowing feature misjudgment caused by jitter can be avoided; the whole process is automatic, and labor time consumption can be saved; automatic, precise and quantitative evaluation of hollowing detection is realized.

The invention discloses an aircraft assembly drilling method and device based on a high-energy laser beam, and the method comprises the steps: adjusting the current posture of equipment through a normal alignment module in an end effector of the equipment, and enabling the irradiation direction of the high-energy laser beam to be the normal direction of a product; the product structure of the current hole forming position is locally pressed through a pressure foot module in an end effector of the hole forming equipment; a high-energy laser beam coupling module in an end effector of the drilling equipment is used for carrying out beveling annular spiral scanning on the product until a conical blind hole molded surface is formed; and the high-energy laser beam coupling module in the end effector of the hole making equipment is adjusted again to be the normal posture of the hole making position, the posture is kept, the center of the conical blind hole serves as the circle center, and circular cutting straight hole machining is conducted till a complete assembly through hole is formed. The problem that the machining precision is low due to local deformation of a weak-rigidity product can be solved, meanwhile, the link of disassembly, finish machining and secondary racking in the assembling and hole forming procedure can be omitted, and the comprehensive hole forming efficiency is improved.

The application belongs to the technical field of laser precision machining, and particularly relates to a laser drilling method for heterogeneous composite materials based on partition parameter regulation. f In view of the problems that a multilayer heterogeneous structure such as a SiC / SiC composite material is prone to hole wall cracking, interface delamination and re-solidification blockage in through-hole machining, the workpiece is divided into a coating area and a matrix area, the scanning speed, the focal point increment and the scanning number matching the material characteristics are selected respectively, and linear variable pitch spiral scanning combined with workpiece self-rotation is adopted to realize the relative uniform distribution of the energy density in the hole. On the premise of ensuring the integrity of the coating / matrix interface, the method can effectively reduce the hole taper and obtain a high-quality through-hole with smooth hole wall and significantly reduced defects.

This disclosure provides a method, apparatus, and storage medium for on-orbit optical axis calibration of a spectrometer, relating to the field of astrophysical detector calibration technology. The method includes: performing a helical scan of the celestial body under test; during the scan, continuously acquiring the imaging position of the centroid of the celestial body on the imaging target surface of the imaging device, and acquiring the thermal radiation energy of its pointing position at a first angle based on a preset acquisition frequency using the spectrometer; inverting the continuously acquired imaging positions onto the first target surface, and mapping the acquired thermal radiation energy to the corresponding position on the first target surface according to the acquisition time, generating first thermal radiation data; and using the position of the point with the highest thermal radiation energy in the first thermal radiation data on the first target surface as the optical axis pointing position of the spectrometer at the first angle. This disclosure achieves joint observation by synchronously controlling the imaging device and spectrometer of the astrophysical detector, and determines the true pointing position of the spectrometer's optical axis through the observation results, which is simple to operate and has high accuracy.

The invention provides a lightweight semantic segmentation method and system for a remote sensing image of an urban scene, and relates to the technical field of semantic segmentation, and the method comprises the steps: carrying out the multi-scale transformation of the remote sensing image of the urban scene, carrying out the semantic feature extraction, obtaining a first semantic feature map under each reduced scale, adjusting the first semantic feature map, and obtaining a second semantic feature map under each reduced scale; using a second semantic feature map; performing forward, snake-direction and spiral scanning fusion operation on the second semantic feature map to obtain a third semantic feature map, and fusing the third semantic feature map with the third semantic feature map of the previous scale to obtain a fused semantic feature map; splicing the third semantic feature map and the up-sampling result of the fused semantic feature map of the previous reduced scale according to the channel dimension, and then performing two convolution operations and fusing; obtaining a fused feature image corresponding to a reduced scale; and taking an up-sampling result of the last scaled-down fusion feature image as a semantic segmentation result image.

The invention relates to a rapid scanning type laser active detection system based on laser angle on-line control, and the system comprises an adjustable beam expanding optical unit which can continuously adjust the beam expanding multiplying power on line through position closed-loop control; the scanning detection execution unit can perform large-range search and can compensate image motion caused by carrier motion and continuous scanning, and a variable-speed spiral scanning mode with high probability of finding a target is adopted as a scanning mode; the off-axis beam expanding unit is used for further compressing a laser divergence angle; the photoelectric imaging system can splice scanned images on line, so that the scanning process is clearer and more visual; and the time sequence control unit assists in image motion compensation control. The scanning process can be rapidly completed, imaging is clear in the scanning process, and the scanning result can be directly presented.

A CT device capable of easily and appropriately determining the start and end positions of imaging in helical scanning imaging is provided. [Solution] The CT apparatus is a helical scan type CT apparatus that continuously images an object being inspected while moving and rotating it, and includes an examination table 1 on which the object being inspected is placed and which moves and rotates the object being inspected, a radiation source 2 that irradiates the object being inspected with a radiation beam, a detector 3 that is provided opposite the radiation source 2 with the object being sandwiched between them, and an examination table position calculation unit 74 that calculates the position of the examination table 1 during imaging. The examination table position calculation unit 74 calculates a reconstruction start position Zs and a reconstruction end position Ze of the object being inspected, and calculates a data acquisition start position Zdas and a data acquisition end position Zdae based on the reconstruction start position Zs and the reconstruction end position Ze.

The invention discloses a method and system for treating hypertension through carotid artery blood vessel regulation and control based on low-frequency ultrasound and near-infrared light cooperation. The system consists of an integrated control system, a movable treatment head and a driving mechanism, wherein the treatment head adopts a cooperative structure of a circular low-frequency ultrasonic transducer (800kHz-2MHz) and a near-infrared weak laser transmitter (600-1000nm) surrounding the treatment head, and is driven by a planetary gear to realize spiral scanning along a neck blood vessel. During treatment, the common carotid artery and the carotid sinus are firstly scanned and positioned, the scanning path is calculated according to the blood vessel trend, ultrasonic waves (0.5-3 W / cm) and near-infrared light (50-200 mW / cm) are synchronously applied, single treatment lasts for 15-20 minutes, and the treatment time is more than or equal to 3 times per week.

The present invention relates to the technical field of digital X-ray imaging methods, and discloses an imaging device using slit scanning, comprising: a transmitting end for collimating X-rays into slit X-rays, continuously transmitting slit X-rays to an object to be imaged, and allowing the object to be imaged to be completely captured by the slit X-rays in a linear manner; and a receiving end for automatically activating a slit region of interest corresponding to the slit X-ray emission direction on the receiving end after each transmitted slit X-ray passes through the object to be imaged, acquiring the attenuated slit X-rays, converting them into slit images with the same range as the slit X-rays for storage, and splicing the slit images after they are completely captured. By collimating X-rays using slit scanning, the magnification effect of point light source imaging is avoided, and the device can be used to acquire images of a patient in true proportion. Furthermore, the radiation dose received is much lower than that of spiral scanning, being at the level of ordinary radiation.

The invention provides a railway line environment identification method and system based on artificial intelligence. According to the method, the laser point cloud, the track inspection vehicle state parameters and the millimeter wave radar data of the railway ballast area are synchronously collected, and the enhanced point cloud with the physical attribute is fused and constructed. Linear scanning laser and millimeter wave radar at the bottom of a rail inspection vehicle are used for spiral scanning and penetrating scanning, and a three-dimensional data matrix fusing the surface appearance and the internal dielectric characteristics is generated. The neural network analyzes the railway ballast contact surface curvature and gap distribution of the matrix, and enhances the weight of the loose area through space-time attention. And a model is generated based on adversarial training to complement the occlusion area, and a complete virtual point cloud is output. And finally, quantitatively analyzing the virtual point cloud and related characteristics, and outputting the railway ballast compactness and the coordinate range of the hardened area as key identification results. Through multi-source data fusion, deep learning and point cloud generation technologies, accurate, dynamic and non-contact quantitative evaluation of the railway ballast compactness and the hardened area is realized.

The process comprises the following steps: S1, crack detection: scanning the surface of a casting blank by adopting an infrared thermal imager, setting a temperature difference threshold value to be greater than or equal to 10 DEG C so as to identify the crack, generating a three-dimensional thermal field distribution diagram, and marking the depth, length and distribution density of the crack; s2, laser micro-melting repair is conducted, specifically, a mechanical arm is adopted to control a laser head, and according to path planning, spiral scanning or a Z-shaped track is adopted to conduct remelting on a crack area according to set parameters; s3, post-treatment and quality inspection are conducted, specifically, a repairing area is air-cooled to 200 DEG C or below, and residual stress accumulation is avoided; and then ultrasonic flaw detection and metallographic detection are conducted, the crack elimination rate and the structure uniformity are verified, it is ensured that the crack elimination rate is larger than or equal to 98%, and the depth of a heat affected zone is smaller than or equal to 0.2 mm. Subcutaneous cracks are accurately positioned through the thermal infrared imager, micro-area remelting is carried out on a defect area in combination with the movable optical fiber laser head, the cracks can be efficiently eliminated, meanwhile, the original structure performance of a casting blank is kept, and the yield of the continuous casting blank is remarkably increased.

The application discloses a kind of multi-photon fiber array endoscopic imaging device and method based on hollow spiral scanning, belong to biological imaging technical field.The device includes laser collimation and beam expansion module, pulse chirp module, beam switching module, endoscopic detection module and fluorescence acquisition module.The beam switching module contains optical switch and at least two low group velocity dispersion single-mode fiber array;Endoscopic detection module drives fiber array to perform hollow spiral scanning by resonant driver, its track starting point is not located at the origin of field of view, to avoid central dense sampling.The method switches fiber by optical switch, so that each fiber performs spatial non-overlapping hollow spiral scanning, collects multi-photon signals such as fluorescence, second harmonic and fluorescence lifetime, and completes the central cavity by image reconstruction and fusion, to obtain complete field of view imaging result.The application significantly reduces the central region light damage, improves the uniformity of imaging resolution, and the fused image is based on physical sampling to ensure authenticity and reliability, with high system stability.

The application provides a weather radar spiral scanning mode radial velocity motion compensation and deblurring method, and relates to the technical field of weather radar signal processing, and the method comprises the following steps: calculating a line-of-sight unit vector and predicting a platform radial projection based on an attitude and scanning geometry; establishing a folding observation relationship according to the Nyquist velocity, constructing a finite candidate folding integer set by using the platform prediction projection, determining the folding integer in the candidate set by matching the consistency constraint, realizing the velocity blurring correction; applying the meteorological small velocity physical prior and the spatial continuity constraint to the blurring correction result to suppress the false correction; and adopting the sequence of deblurring first and then deducting the platform projection to obtain the compensated radial velocity. The application reduces the velocity folding jump under the spiral scanning condition, reduces the systematic deviation related to the scanning phase, improves the radial velocity consistency, and provides reliable input for the cloud and rain radial velocity product and subsequent cloud wind field inversion.

The invention provides a muscle mass evaluation method and system, and relates to the technical field of medical imaging, and the method comprises the steps: carrying out the ultrasonic scanning of human limbs, collecting an ultrasonic total scattering signal, and obtaining an ultrasonic image; acquiring a muscle structure image according to the ultrasonic image; performing segmentation processing on the muscle structure image to obtain contour information of a target muscle; and calculating the cross-sectional area and volume of the target muscle based on the contour information. According to the invention, the annular probe performs three-dimensional spiral scanning in the scanning process, and simultaneously obtains spatial information in the horizontal and vertical directions; echo signals are processed by adopting a delay superposition beam forming algorithm, and image segmentation is carried out in combination with a deep learning technology. According to the invention, radiation-free, high-precision, low-cost and portable muscle mass measurement is realized, and the method is especially suitable for evaluation, screening and monitoring of sarcopenia.

The invention discloses a wind power tower modal analysis method and system based on FMCW four-dimensional laser radar dual-stage scanning. According to the method, the unique capability of an FMCW laser radar for directly measuring the radial speed of a target based on the Doppler frequency shift principle is utilized, and three-dimensional space coordinates and speed information are synchronously obtained to form a four-dimensional point cloud. The method adopts a double-stage scanning strategy: in the first stage, the overall geometric morphology of the tower drum is quickly obtained through spiral scanning with an oscillation compensation item; and in the second stage, self-adaptive vertical line reciprocating scanning is performed based on the geometric model to obtain high-density time sequence speed data. A time sequence speed matrix is constructed through a weighted median algorithm, and modal parameters are calculated and extracted in combination with frequency domain analysis and fusion damping ratio. The contradiction between space coverage and time resolution in the prior art is solved, and compared with a traditional method, the scanning efficiency is improved by more than 80%, and the damping ratio recognition precision is improved by more than 50%.

A method for obtaining the average acquisition probability of a helical scan is presented, solving the problem that existing methods for obtaining acquisition probability cannot accurately measure the satellite's space acquisition capability, and belonging to the field of communications. This invention includes: uniformly generating N target points (x...) in an uncertain region of the helical scan. i y i Based on the spiral scan parameters and the coordinates of the dwell point, the spiral scan is used to calculate the helical scan for each target point (x). i y i The capture probability p(x) i y i , σ v ), for the capture probability p(x) i y i , σ v By fitting the data, the envelope function is obtained. Rotating the envelope function around the y-axis yields the probability distribution function P(x, y, σ). v );σ v Represents the vibration noise of the transmitting satellite; obtains the probability density function P(x, y, σ) of the receiving satellite appearing at (x, y). p ), σ p Standard deviation of satellite position change at the receiver; average acquisition probability

The invention belongs to the field of part flaw detection, and particularly relates to a pipe fitting inner wall defect profiling TMR array eddy current imaging detection device and method. Conventional eddy current detection channels are small in number and low in detection efficiency, the length of a defect cannot be judged through an impedance diagram, and the conventional eddy current detection channels are difficult to adapt to defect detection of aluminum pipes with different inner diameters within the range of 120-150 mm due to the influence of a probe structure. According to the invention, the defect detection of the inner wall of the aluminum pipe with the inner diameter of not less than 120mm can be realized, the 100% full coverage detection of the machining area of the inner wall can be realized through spiral scanning, the three-dimensional C-scanning imaging of the machining area of the inner wall can be realized through a positioner-portal frame-array eddy current linkage mode, the shape and size of the defect can be intuitively judged, and the detection accuracy is improved. And meanwhile, the high sensitivity characteristic of the TMR sensor can ensure that closed cracks can be detected.

The invention discloses a laser film removing method and system for curved-surface coated glass. The method comprises the following steps: fixing a curved-surface glass workpiece subjected to hot bending forming; multi-angle images are collected through the CCD camera, and the control system generates a three-dimensional digital model matched with an actual curved surface through three-dimensional reconstruction; a three-dimensional scanning path containing X-axis and Y-axis tracks and Z-axis height is planned according to a preset film removing area and a three-dimensional digital model, spiral scanning is adopted, and the interval is smaller than or equal to 0.05 mm; and synchronously driving the 2D galvanometer and the dynamic focusing shaft to enable the laser beam to scan along the path and adjust the focal length in real time, so as to ensure that the laser focus is always focused on the surface of the film layer to remove the target film layer. By means of dynamic real-time tracking of the laser focus, laser energy acts on the whole film removing path accurately and evenly, film layer residues, base material damage and optical distortion caused by defocusing are eliminated, the diopter of a film removing area is smaller than or equal to 100 mdpt, the light transmittance is larger than or equal to 92%, and the requirement for high optical performance of an intelligent driving camera window is met.

The application provides a kind of urban scene remote sensing image light weight semantic segmentation method and system, it is related to semantic segmentation technical field, the remote sensing image of urban scene is carried out multi-scale transformation, and semantic feature extraction is carried out, to obtain the first semantic feature map under each reduced scale, the first semantic feature map is adjusted, to second semantic feature map;Second semantic feature map is carried out forward, snake direction and spiral scanning fusion operation, to obtain third semantic feature map, third semantic feature map and the third semantic feature map of last scale are fused, to obtain fusion semantic feature map;Third semantic feature map and the up-sampling result of the fusion semantic feature map of last reduced scale are spliced according to channel dimension after, two convolution operations are carried out and are fused;Obtain the fusion feature image corresponding to reduced scale;The up-sampling result of the fusion feature image of last reduced scale is used as semantic segmentation result image.