42 results about "Optical depth" patented technology

The invention discloses a deep learning inversion AOD-assisted atmospheric correction-based method and device and a medium, and relates to the field of atmospheric remote sensing and deep learning cross technologies, and the method comprises the steps: constructing a multi-type training sample set fusing actual measurement and physical simulation data, and carrying out mass screening and layering processing to obtain a multi-type training sample set; obtaining a pre-training sample, a fine tuning sample and an extreme scene sample matched with the satellite remote sensing data; establishing a deep learning network model taking data loss and inverse operator constraint loss weighted fusion as a total loss function, and adopting a cross-satellite transfer learning strategy to sequentially complete pre-training, satellite exclusive fine tuning and extreme scene enhancement training to obtain a trained aerosol optical thickness inversion model; radiometric calibration, cloud detection, geometric correction preprocessing and input model reasoning calculation are carried out on satellite remote sensing original data, post-processing and quality grading are carried out on an output result, and an aerosol optical thickness inversion result used for assisting atmospheric correction is obtained.

The embodiment of the invention provides a remote sensing image atmospheric correction method and system for an ocean near-shore scene, and relates to the technical field of ocean remote sensing image atmospheric correction. According to the method, after remote sensing data is obtained, solar flare correction is executed by extracting a target mask and using a secondary simulation model of a satellite signal in a solar spectrum. And then a multi-band darkness index is calculated, and an optimal aerosol optical thickness value is inverted through a radiation transmission model. And calculating the atmospheric correction coefficient of each wave band according to the optimal aerosol optical thickness value so as to generate correction result data. According to the method, on the basis of robust flare correction of the SWIR proxy, scene-adaptive flare proxy is constructed by utilizing water body reflection and path radiation physical characteristics of the SWIR under a deepwater condition, and flare deduction is realized. Inversion is carried out through multiband combined darkness indexes, and the spectral characteristics of the deep water body reflected in the visible spectrum are utilized, so that single-band random noise is smoothed, and the accuracy of an AOT estimation result is improved.

In various examples, optical depth estimation for interior space monitoring systems and applications is disclosed. Absolute 3D depth estimates from monocular image data may be generated using a machine learning model using 3D geometry priors and a joint learning framework that combines depth estimation with object segmentation. Three-dimensional geometry priors provide surface-level information that enriches the model's understanding of the relevant spatial geometry and resolves scale ambiguity in monocular depth estimation within automotive in-cabin environments. The model may include a common (e.g., shared) encoder stage that outputs features extracted from an optical image sensor feed to separate decoder stages that include a depth estimation decoder and a segmentation decoder. Joint learning for depth estimation and segmentation tasks during training achieves a more nuanced understanding of the in-cabin environment, leading to significantly improved depth accuracy.

The application discloses a method for inverting marine aerosol optical depth by using spaceborne lidar noise, belongs to the technical field of laser remote sensing, and is used for quantitatively inverting aerosol optical depth. Under the premise that system hardware parameters of the spaceborne lidar and total background noise data detected by the spaceborne lidar, environmental parameters of a corresponding region and moment are given, a noise model of solar radiation items corresponding to water surface white froth diffuse reflection and water surface specular reflection, a water-leaving noise model (a noise model of solar radiation items of water body backscattering), and an atmospheric noise model (a noise model of solar radiation items of Rayleigh scattering of atmospheric molecules and Mie scattering of aerosol particles) are respectively constructed. Water surface noise, water-leaving noise and atmospheric Rayleigh scattering noise are removed from total noise rate of the atmosphere top detected by the lidar, and aerosol scattering noise results are obtained. Finally, marine aerosol optical depth of a satellite transit region and moment is quantitatively inverted by using the aerosol scattering noise results and the aerosol scattering noise model.

The application discloses a kind of remote control step puncture robot systems, including front-end robot trolley, calibration mark module, optical depth navigation camera, remote control terminal module, patient fixation device, front-end robot trolley is connected with optical depth navigation camera by cable or wireless communication mode communication, the application also provides a kind of remote control step puncture robot system operation method.Compared with prior art, when using the puncture positioning of remote control needle insertion mechanism in the system of the application, the doctor can return to scanning control room without being in the X-ray radiation range all the time, operates needle insertion mechanism in control room and can view the actual image returned by CT machine, if there is deviation, puncture angle can be adjusted in time, the application can enable the doctor to operate outside CT room to avoid receiving the radiation of X-ray, while also conveniently realize accurate puncture, improve the operation efficiency, reduce the risk of secondary injury of patient during operation.

In some embodiments, a display system and method for generating a virtual or light field image includes a display and a field-emitting cavity. The field-emitting cavity modulates the optical depth of the virtual image. In some embodiments, the components of the display system are curved. The focal plane of the virtual image is curved and can approximate a portion of a human horopter. In some embodiments, the display system is integrated into a portion of a vehicle. The components of the display system can move relative to each other such that the display system is foldable or bendable when not in use.

The application discloses a vegetation dark background aerosol inversion method and device based on physical and machine learning fusion and a medium, relates to the field of aerosol optical depth inversion, and comprises the following steps: constructing a sample library based on AVHRR / MERSI observation data and MODIS surface reflectance product data; adopting a machine learning algorithm to establish a surface reflectance relationship model with medium / short-wave infrared apparent reflectance, NDVI and multiple angle data as input and label red band surface reflectance as output; determining the label red band surface reflectance by adopting the surface reflectance relationship model for satellite observation data to be processed, calculating the red band apparent reflectance, and then obtaining a set of dark background pixels to be inverted; for any dark background pixel to be inverted, a preset lookup table is called, the red band apparent reflectance is matched with the apparent reflectance in the preset lookup table, and the aerosol optical depth is determined. The application can realize high-precision AOD inversion.

The application discloses a prediction method for radiation refrigeration capacity by using aerosol optical thickness and precipitable water amount, comprising the following steps: S10, obtaining corresponding hourly weather data according to weather data; S20, judging whether the local latitude is greater than 33°, if yes, entering step S30, S30, calculating the local precipitable water amount (PWV H ) by using a high-latitude calculation method; if no, entering S40, S40, calculating the local precipitable water amount (PWV L ) by using a low-latitude calculation method; S50, calculating atmospheric radiation according to the calculated precipitable water amount (PWV H or PWV L ); S60, correcting the atmospheric radiation calculated in S50 according to the aerosol optical thickness in the weather data; and S70, calculating the radiation refrigeration capacity according to the atmospheric radiation corrected in S60.

The application provides a kind of improved remote sensing ecological index calculation method based on space-time data fusion, the method comprises the following steps: collecting remote sensing data of the study area;The collected remote sensing data is pretreated;The water in the processed remote sensing data is masked and treated;The remote sensing data after pretreatment is spatiotemporally fused;According to the remote sensing data after fusion processing, the index data for representing the improved remote sensing ecological index is calculated respectively;Get aerosol optical depth to represent air quality and nighttime light data to represent human activity intensity on the gee platform, and resample the two index data;The index data is parameter standardized, and the standardized data is obtained;Principal component analysis is used, and principal component extraction is carried out according to the standardized data, to obtain the evaluation index result.The application can improve the spatiotemporal resolution of data through space-time data fusion technology, and solve the problem of lack of comprehensiveness in index selection of remote sensing ecological index.

The invention relates to a display system for a vehicle, wherein the display system has a display field (4) with four outer edges and at least two mirror modules, wherein a reflection chamber (6a, 6b) is arranged on at least two mutually parallel outer edges of the display field (4), wherein front sides as front base sides of the two parallel reflection chambers (6a, 6b) are covered by at least one front mirror module and rear sides as rear base sides of the two parallel reflection chambers (6a, 6b) are covered by at least one rear mirror module, wherein an inner side of each reflection chamber (6a, 6b) is bounded by a wall designed as an outer edge of the display field (4), wherein an outer side of each reflection chamber (6a, 6b) is bounded by a further wall, wherein light sources (12) are arranged on at least one of the two walls of each reflection chamber (6a, 6b), wherein the light sources (12) are configured toto emit light into the respective reflection chamber (6a, 6b), wherein the two mirror modules are configured to reflect the light in the respective reflection chamber (6a, 6b), wherein reflections of the light through the at least one mirror module arranged on the front sides are visible and configured to create an optical depth effect for the display field (4).

The application discloses a model rendering method and device, electronic equipment and storage medium. The method comprises the following steps: acquiring a fluid model comprising a plurality of fluid model parts surrounded by solid models and fluid density control information of the fluid model; determining a target fluid model part to be rendered on the fluid model according to a fluid model part through which a view ray of a virtual camera passes on the fluid model; setting a bounding box for a target solid model corresponding to the target fluid model part, and determining a ray path of the view ray in the bounding box; performing ray stepping analysis according to the fluid density control information and the ray path, determining optical depth information, and rendering the target fluid model part. Based on this, the fluid model is divided into a plurality of fluid model parts surrounded by solid models, so that the fluid model rendering mode is suitable for various fluid distribution scenes, and in the scene where the fluid distribution is uneven, the performance consumption is reduced, the rendering efficiency is improved, and the user experience is improved.

A method for measuring the equilibrium spin polarizability of alkali metal atoms based on the steady-state response of a triaxial magnetic field is proposed. First, the device is adjusted to its normal operating state. Then, DC magnetic fields are applied in the x, y, and y, z directions, with the magnitude of the magnetic field in the z direction varied. The ratio of the device outputs in the two applications is fitted with the square of the applied magnetic field in the z direction to obtain the total relaxation rate. Next, the pressure broadening of the alkali metal gas cell is obtained at low temperature using the optical depth of the detection light. Then, at the normal operating temperature, the power of the detection light is varied, and the transfer coefficient is established using the pressure broadening and the fitting slope. Finally, a DC magnetic field is applied along the y-axis with varying magnitude, and the device output is fitted with the B... y‑fit The relationship between these factors, combined with the transfer coefficient, yields the final equilibrium spin polarization. This method does not disrupt the SERF states required for normal system operation, is unaffected by temperature, and has small measurement errors. It provides a more accurate measurement method for further optimizing SERF magnetic field measurement devices to improve their sensitivity.

This invention discloses a method for measuring gas velocity in absorption spectra by constructing and matching absorption peak image features. The method includes: performing multiple phase-shift downsampling on a detected wavelength-modulated laser signal; reconstructing the multiple phase-shift sequences to obtain a two-dimensional grayscale image matrix containing the direct absorption spectrum profile; sharpening the two-dimensional grayscale image matrix extracted from the laser signal where the flow field velocity is zero to obtain an image template characterizing the absorption peak position; matching the position of the image template from the two-dimensional grayscale images obtained from laser detection signals in two directions within the flow field under test, based on the absorption peak position image template, to determine the offset of the absorption peak positions in the two directions; and calculating the airflow velocity using the absorption peak offsets in the two detection directions. This invention improves the accuracy of solving the absorption spectrum frequency shift at different optical depths, thereby improving the velocity measurement accuracy in complex flow fields with varying operating conditions.

The invention relates to the technical field of monitoring of the running state of a solar cell panel, and discloses a method and device for detecting the influence result of the solar cell panel, and the method comprises the steps: obtaining the related data of the solar cell panel during running; wherein the related data comprises aerosol optical thickness data, wind speed data and vibration root-mean-square data; determining a joint threshold based on the related data, and detecting whether a difference between the joint threshold and the actual power value is greater than a difference threshold; when the difference value between the combined threshold value and the actual power value is larger than a difference threshold value, the power attenuation rate is determined based on the aerosol optical thickness variable quantity and the vibration root-mean-square variable quantity; and detecting the influence result of the solar cell panel based on the power attenuation rate and the related data.

The application discloses a kind of based on deep learning inversion AOD auxiliary atmospheric correction method, device and medium, it is related to atmospheric remote sensing and deep learning cross technical field, including: the multiple type training sample set of fusion measured and physical simulation data is constructed, after quality screening and stratification processing, obtain the pre-training sample of adaptation satellite remote sensing data, fine-tuning sample and extreme scene sample;Create deep learning network model with data loss and anti-operator constraint loss weighted fusion as total loss function, adopt cross-satellite transfer learning strategy, in turn complete pre-training, satellite special fine-tuning and extreme scene enhancement training, obtain the training completed aerosol optical depth inversion model;Radiometric calibration is executed to satellite remote sensing original data, cloud detection, geometric correction preprocessing, input model inference calculation, to output result is post-processed and quality classification, obtain the aerosol optical depth inversion result used for auxiliary atmospheric correction.

To provide a display system for use in a vehicle.SOLUTION: A display system configured to be integrated into a vehicle is provided, comprising a light source and an optical subsystem optically coupled to the light source. The optical subsystem includes a reflector and at least one semi-reflective optical member, and is configured to generate light and direct light rays to provide a plurality of virtual images that appear to a viewer concurrently at each of a plurality of different optical depths. Each of the virtual images travels a different distance within the optical subsystem before exiting the optical subsystem on a path toward eyes of the viewer.SELECTED DRAWING: Figure 1

This application provides a method, apparatus, electronic device, and medium for aerosol optical thickness data completion, belonging to the field of remote sensing data processing and atmospheric environment monitoring. The method includes: acquiring multi-source data; encoding the multi-source data using a 3D encoding / decoding network to obtain an initial feature map; using auxiliary data as guiding weights for underlying surface geographic features, and combining this with a spatial attention module and the initial feature map to obtain a spatial feature map; obtaining a temporal feature map based on a temporal attention module based on the initial feature map; concatenating the spatial and temporal feature maps, and then adding them element-wise to the initial feature map to obtain a comprehensive feature map; decoding the comprehensive feature map to obtain an initial completion result; converting the initial completion result into corresponding completion values; determining missing regions based on each completion value; and completing the missing regions to obtain the completed aerosol optical thickness data. This application can improve the accuracy of aerosol optical thickness data completion.

Disclosed is computer-implemented method including marching first ray and second ray, along first gaze direction and second gaze direction that are estimated using gaze-tracking means, from given viewpoint into depth map, to determine first optical depth and second optical depth corresponding to first eye and second eye, respectively; calculating gaze convergence distance, based on first gaze direction and second gaze direction; detecting whether first optical depth lies within predefined threshold percent from second optical depth; and when it is detected that first optical depth lies within predefined threshold percent from second optical depth, selecting given focus distance as an average of at least two of: first optical depth, second optical depth, gaze convergence distance; and employing given focus distance for capturing given image using at least one variable-focus camera.

Light field images are displayed via a light field display unit to produce a synthetic light field presenting at least one virtual object to a given user. When it is detected that a difference between a focus depth at which the given user is gazing and a given optical depth at which the at least one virtual object is being presented is more than a predefined threshold difference, at least one of: dimming, blurring, is performed on a virtual image to be presented to a non-dominant eye of the given user. A current light field image is generated by utilising this virtual image and another virtual image to be presented to a dominant eye. The current light field image is displayed via the light field display unit to produce a current synthetic light field presenting the at least one virtual object.

The application discloses a FY-4B satellite aerosol observation data multiscale assimilation prediction method and system, relates to the meteorological analysis technical field, and the method comprises the following steps: obtaining land aerosol optical depth observation data inverted by an AGRI sensor carried by a FY-4B stationary meteorological satellite; constructing a double-layer nested based on a WRF-Chem regional weather-chemical coupling mode, and obtaining a collection background field based on a collection emission disturbance strategy; performing 4D-LETKF assimilation analysis based on the land aerosol optical depth observation data and the collection background field, so as to obtain a collection analysis field; and running a short-time WRF-Chem prediction based on the obtained collection analysis field, so as to obtain aerosol short-term prediction results. The application realizes analysis and short-term prediction of regional aerosols with high space-time resolution, and significantly improves the simulation capability for high-resolution spatial distribution characteristics and rapid change processes in the hour scale.

A display system that is configured for integration into a vehicle comprises a light source and an optical subsystem optically coupled to the light source. The optical subsystem includes a reflector and at least one semi-reflective optic, wherein the optical subsystem is configured to direct light rays to provide a plurality of virtual images that appear to a viewer concurrently at a respective plurality of different optical depths, by causing light of each of the virtual images to travel a different distance within the optical subsystem before exiting the optical subsystem on a path toward eyes of the viewer.

The invention discloses an AI-based intelligent city patrol system and method, and the system comprises a multi-mode patrol unit which is composed of an acoustic sensor and an optical assembly, and is used for collecting the optical frame data and acoustic echo data of a foundation pit construction area. And the visual analysis unit is used for preprocessing the optical frame data so as to extract an optical depth map containing local deformation information and pixel depth values in the optical frame data. And the Beidou positioning unit is used for carrying out preprocessing and three-dimensional reconstruction on the acoustic echo data so as to determine three-dimensional point clouds of the edge and the internal structure of the foundation pit. And the multi-modal data fusion unit is used for performing point clouding processing on the optical depth map to obtain an optical point cloud, and fusing the optical point cloud with the three-dimensional point cloud to obtain a fused point cloud and a corresponding three-dimensional grid model. And the AI driving unit is used for extracting a feature tensor based on the three-dimensional grid model and carrying out depth reasoning according to the feature tensor so as to output a hidden danger prediction result of each slice, and foundation pit monitoring blind areas are reduced.

A structure light depth sensing system and a structure light depth sensing method using auxiliary time-of-flight sensing are provided. The structure light depth sensing system includes a time-of-flight depth sensing circuit and a structure light depth sensing circuit. The structure light depth sensing method includes: providing similarity values in a predetermined search range, in which the predetermined search range includes a largest similarity value; calculating a compensation range in the predetermined search range in accordance with a target depth value of depth data of the time-of-flight depth sensing circuit; calculating compensated similarity values in accordance with a weighting gain and the similarity values in the compensation range; and determining if a maximum value among the compensated similarity values is greater than the largest similarity value. When the determination result is yes, the maximum value among the compensated similarity values is determined to be a correct similarity value.

This invention discloses a high-resolution distributed acoustic-vibration demodulation method based on sweep-frequency optical deep anti-fading. First, continuous linear sweep-frequency probe light is injected into an optical fiber, and the original backscattered Rayleigh signal containing information distributed along the fiber is acquired through coherent beat frequency acquisition. Then, the original backscattered Rayleigh signal undergoes frequency domain demultiplexing and matched-filter pulse compression processing to reconstruct multiple high-resolution time-domain complex envelope signals from a single sweep. Next, the high-resolution time-domain complex envelope signals are input into a deep anti-fading demodulation network, where amplitude-phase joint modeling achieves adaptive suppression of random fading and noise in the time-domain complex envelope signal, outputting a de-fading phase sequence. Finally, the de-fading phase sequence is used for phase difference demodulation, and combined with an optical time-of-flight positioning model, to achieve highly robust demodulation and spatial positioning of vibration events. This method maintains high resolution and robustness even under strong noise, long distance, and deep fading conditions, making it suitable for distributed acoustic-vibration and ultrasonic monitoring scenarios.

An AI-based intelligent urban patrol system and method includes: a multimodal patrol unit, composed of acoustic sensors and optical components, for collecting optical frame data and acoustic echo data of the foundation pit construction area; a visual analysis unit, for preprocessing the optical frame data to extract an optical depth map containing local deformation information and pixel depth values; a BeiDou positioning unit, for preprocessing and 3D reconstruction of the acoustic echo data to determine the 3D point cloud of the foundation pit edge and internal structure; a multimodal data fusion unit, for point cloudification of the optical depth map to obtain an optical point cloud, and for fusing the optical point cloud with the 3D point cloud to obtain a fused point cloud and a corresponding 3D mesh model; and an AI-driven unit, for extracting feature tensors based on the 3D mesh model and performing depth inference based on the feature tensors to output the hazard prediction results for each slice, reducing blind spots in foundation pit monitoring.

The invention provides an optical domain depth detection system and a method thereof. The optical domain depth detection system comprises a laser light source, a first lens, an all-optical Gaussian Laplacian calculator and a signal detection system, and the laser light source is used for emitting laser and irradiating the laser on a three-dimensional object to be detected; the first lens is used for imaging a three-dimensional object to obtain a first imaging light field; the all-optical Gaussian-like Laplacian calculator is used for completing Gaussian-like filtering and Laplacian operation on the first imaging light field in an optical domain so as to obtain a second imaging light field after Gaussian-like Laplacian operation; and the signal detection system is used for detecting the second imaging light field so as to obtain depth information of the three-dimensional object. According to the invention, the real-time performance of deep recognition can be greatly improved.

The embodiment of the invention provides a three-dimensional cloud layer generation method and device, equipment, a medium and a program product. The method comprises the steps of obtaining cloud layer density, determining shadow parameters of a cloud layer area according to optical depth and the cloud layer density based on a solar visual angle and a cloud layer visual angle, calculating color parameters of the cloud layer area according to the cloud layer density and the shadow parameters, and rendering in the cloud layer area according to the color parameters to obtain a three-dimensional cloud layer. In this way, the shadow attenuation effect of a cloud layer under sun irradiation can be simulated, cloud layer shadow details can be enriched, the sense of reality of volume clouds can be improved, the view angle of the camera can be switched to present the volume clouds at different angles, and the dynamic interaction ability of the volume clouds can be enhanced.

The invention relates to an aerosol optical thickness inversion model processing method and device, equipment and a medium. The method comprises the following steps: acquiring a training sample set; inputting the polarization remote sensing data of the sample pixels into an input unit of a to-be-trained aerosol optical thickness inversion model for feature initial extraction to obtain initial features; inputting the preliminary features into a deep backbone network of a to-be-trained aerosol optical thickness inversion model for feature depth extraction to obtain depth features; inputting the depth features into an output layer of a to-be-trained aerosol optical thickness inversion model to obtain aerosol optical thickness inversion data; according to the aerosol optical thickness inversion data and the sample aerosol optical thickness observation data, the to-be-trained aerosol optical thickness inversion model is optimized, the trained aerosol optical thickness inversion model is obtained, and the aerosol optical thickness inversion precision is improved.

An optical depth sensing system is provided including an active illumination source; and a colloidal quantum dot photodetector. The colloidal quantum dot photodetector having colloidal quantum dots as a light sensing elements and the system has rise times that are less than 5.0 ns and fall times less than 10.0 ns.

The invention relates to a high-energy ray Mars atmosphere transmittance modeling system and method, and belongs to the technical field of planet atmosphere detection. Comprising the following steps: acquiring Mars atmosphere mass density data under a specific space-time condition from a Mars atmosphere parameter database; converting the mass density of the Mars atmosphere into the total molecular number density of the Mars atmosphere by utilizing an Avogalide constant and the average molar mass of the Mars atmosphere; based on the volume mixing ratio of the carbon dioxide in the Mars atmosphere, calculating the molecular number density of the carbon dioxide; according to the carbon dioxide molecular number density and the high-energy photon mass absorption coefficient, the Mars atmospheric optical depth of the high-energy photons along the transmission path is calculated by combining a shell-separating integral method; according to the Mars atmosphere optical depth, the transmittance after the high-energy photons penetrate through the Mars atmosphere is calculated, multi-source climate data and the accurate physical effect can be fused, the method is suitable for a wide height range, and the problems that an existing model depends on simplified data, CO2 attenuation characteristic modeling is not accurate, and the applicable height is limited are solved.