5442results about "Image data processing details" patented technology

A pattern which does not appear at a flat portion in normal binarization processing is set as a code pattern, and a code formed from this pattern is attached. At this time, code attachment with little degradation in image quality is implemented by selecting an unnoticeable pattern.

In a measurement processing apparatus for measuring a geometric feature of an object image: a measurement-reference-element setting unit automatically sets at least one first measurement reference element for use in measurement of the geometric feature of the object image, at at least one first position on the object image based on first image data representing the object image and position information indicating at least one second position of at least one second measurement reference element which is set on a measurement reference image corresponding to the object image; and a geometric-feature measurement unit measures the geometric feature of the object image based on the at least one first position of the at least one first measurement reference element.

The invention provides a bridge girder erection machine intelligent construction control method and system based on machine vision, and relates to the technical field of intelligent control. Image information of a construction site is collected through a plurality of high-definition cameras arranged on a bridge girder erection machine; inputting the image information into a pre-trained deep learning model, identifying and positioning a target in the construction scene, and constructing a digital twinborn model of the construction scene in combination with a graph structure of a construction scene topological structure subjected to spatial perception graph convolution operation; the intelligent decision agent generates an initial construction strategy according to the current construction scene state, an improved Monte Carlo tree is adopted to search and optimize the initial construction strategy, and an optimal construction strategy is generated through multi-objective solution; and after the construction is completed, based on the digital twin model and the whole process data, analyzing the construction quality and efficiency by using a graph neural network, and generating a multi-dimensional evaluation report and optimization suggestions.

Systems and methods for image compositing are provided. An aspect of the systems and methods includes obtaining a first image and a second image, wherein the first image includes a target location and the second image includes a target element; encoding the second image using an image encoder to obtain an image embedding; generating a descriptive embedding based on the image embedding using an adapter network; and generating a composite image based on the descriptive embedding and the first image using an image generation model, wherein the composite image depicts the target element from the second image at the target location of the first image.

The invention provides a digital image watermarking method and device based on a diffusion model and a spread spectrum technology, and relates to the technical field of image processing and artificial intelligence. The method comprises the steps of obtaining a carrier image and watermark information; performing down-sampling and forward diffusion on the carrier image by using a variational auto-encoder to obtain a first potential vector; performing channel dimension division on the first potential vector, selecting a one-dimensional channel, performing frequency domain transformation through a discrete cosine transform technology, extracting a high-frequency coefficient, and obtaining an embedded matrix; performing spread spectrum modulation on the watermark information by using an orthogonal code to obtain a spread spectrum watermark; embedding the spread spectrum watermark into the embedded matrix by combining a strength factor to obtain a watermark potential vector; and after carrying out reverse denoising on the watermark potential vector by utilizing a denoising network, restoring the watermark potential vector into a watermark image through a variational auto-encoder, and combining the watermark potential vector with a carrier image to obtain a final watermark image. According to the method, the imperceptibility and robustness of the watermark are greatly enhanced, and regenerative attacks can be effectively resisted.

An apparatus and method for static image of time series measured data to time series translation is disclosed. The apparatus comprises at least a processor configured to receive a static image of time series measured data, convert that static image from its initial domain to a usable time series within another user-selected domain, then to validate the conversion against a confidence threshold.

Techniques for latent space based steganographic image generation are described. A processing device, for instance, receives a digital image and a secret that includes a bit string. A pretrained encoder of an autoencoder generates an embedding of the digital image that includes latent code. A secret encoder is trained and utilized to generate an embedding of the secret to act as a latent offset to the latent code. The processing device leverages a pretrained decoder of the autoencoder to generate a steganographic image based on the embedding of the secret and the embedding of the digital image. The steganographic image includes the secret and is visually indiscernible from the digital image. Further, the processing device is configured to recover the secret from the steganographic image, such as by training and leveraging a secret decoder to extract the secret.

The invention discloses a tunnel wall surface deformation monitoring method and system based on computer image recognition. The method comprises the following steps: acquiring a tunnel area image and a tunnel wall surface image of each image acquisition point; tunnel wall surface deformation identification and wall surface deformation prediction are carried out on the tunnel area image and each tunnel wall surface image, and deformation image identification data and a wall surface deformation prediction result are correspondingly obtained; obtaining deformation image identification data and spatial position data corresponding to a wall surface deformation prediction result, and generating a plurality of deformation monitoring units in a deformation monitoring area of the tunnel; taking each deformation monitoring unit as a basic unit, acquiring sensor monitoring data and a corresponding monitoring threshold interval, and comparing the sensor monitoring data and the corresponding monitoring threshold interval to obtain a sensor monitoring judgment result; based on all sensor monitoring judgment results, the corresponding deformation image recognition data and the wall surface deformation prediction result, space matching analysis is carried out, and a tunnel wall surface deformation monitoring result is correspondingly obtained; according to the invention, the monitoring effect of tunnel wall surface deformation monitoring is improved.

The invention relates to the technical field of titanium anodes, and provides an automatic coating method of a titanium anode and related equipment, and the method comprises the following steps: obtaining a target coating area of the titanium anode, coating, carrying out data acquisition on the target coating area in the coating process to obtain real-time coating data, and calculating coating uniformity and thickness change conditions; evaluating the coating uniformity, and generating a coating quality evaluation result and a coating defect distribution condition; based on the coating quality evaluation result and the coating defect distribution condition, the thickness change condition is analyzed, and thickness optimization parameters are obtained; and generating a feedback atlas according to the thickness optimization parameters, and adjusting coating parameters of the automatic coating equipment based on the feedback atlas. Through application of automatic coating equipment, a sensor array, real-time data acquisition and a coating analysis model, the problems that when uniformity and thickness in the coating process are controlled, the coating quality is unstable, the coating thickness is not uniform, and coating parameters are difficult to monitor and adjust in real time are solved.

A client device may be enabled to render a three-dimensional [3D] scene comprising one or more objects. At a server system, a video-based representation of an object may be streamed as one or more video streams to the client device. At the client device, the scene may be rendered from a viewing position within the scene to obtain a rendered view of the scene. Furthermore, a relative position between the viewing position and a position of the object in the scene may be determined. At the server system, the one or more video streams may be generated to show the object from a limited set of viewing angles, wherein the limited set of viewing angles is selected based on the relative position. At the client device, a viewing angle may be selected from the limited set of viewing angles and the video-based representation of the object may be placed at said selected viewing angle in the scene.

The invention provides a heavy truck battery compartment guiding method and system based on visual perception. The method and system are used for automatic battery replacement operation in a complex industrial environment. According to the system, a multi-camera fusion visual perception platform is constructed, a plurality of industrial cameras arranged on the ground or ceiling of a battery swap station are used for collecting local images of different visual angles of a battery compartment, and a complete visual field image is generated through feature matching and image splicing. A battery compartment is coarsely positioned by adopting a YOLO series model, a bounding box region is extracted, pixel-level contour segmentation is realized by introducing SAM, and the complex background and multi-interference environment recognition capability is enhanced. And after segmentation, calculating a minimum enclosing rectangle of the battery compartment, obtaining a center coordinate and a deviation angle, and transmitting a pose parameter to an upper computer control system. According to the method, the defects of the laser radar are avoided, image processing, the deep neural network and multi-view information are fused, the recognition precision and stability are improved, the battery replacement efficiency and the unmanned level of the electric heavy truck can be remarkably improved, and reliable support is provided for green traffic.

The present application provides a watermark information embedding method, device, storage medium and program product, wherein the method includes obtaining the target watermark information to be embedded and the original carrier information in response to a watermark embedding instruction; mapping the target watermark information to a latent variable space to obtain a watermark latent vector, and mapping the original carrier information to the latent variable space to obtain a carrier latent vector; fusing the watermark latent vector and the carrier latent vector to generate a noise latent vector with a watermark; denoising the noise latent vector to generate a target carrier information with the target watermark information. The present application realizes that when the target carrier information is subjected to various complex editing processes, the watermark information can still be effectively extracted, thereby improving the robustness of the watermark and providing a reliable data basis for subsequent digital information ownership protection.

The invention discloses a code printing quality control method and system of a laser vision code printing machine, relates to the technical field of code printing quality control, and aims to solve the problem that the laser vision code printing machine cannot perform code printing optimization according to a code printing process. According to the method, abnormal identification is carried out on the unqualified quality indexes in the code printing monitoring data, specific problems such as laser parameter abnormity, equipment faults and hardware problems can be accurately identified, the accuracy enables subsequent improvement measures to solve the problems in a more targeted manner, and the accuracy of code printing quality detection is improved by formulating clear code printing quality detection standards and targets. According to the method, the standardization and consistency of quality detection work are ensured, the division standard of quality evaluation is clarified, the quality evaluation result is more objective and accurate, the coding efficiency is improved by designing a reasonable path type, a starting point, an ending point and a path sequence, the accuracy and no intersection or overlapping of a coding path are ensured through the path simulation step, and the coding efficiency is improved. And repeated coding and waste are avoided.

Techniques for performing a machine learning model based spatial-temporal adaptive shift for end-to-end text-video retrieval are described. According to some examples, a computer-implemented method includes receiving a video comprising a plurality of frames at a content delivery service; generating, by the content delivery service, a set of embeddings for each of a plurality of sections of each frame of the plurality of frames; determining, by a candidate selector machine learning model of the content delivery service, a proper subset of the plurality of sections of each frame of the plurality of frames for a time shift based on the set of embeddings; time shifting, by the content delivery service, the proper subset of the plurality of sections of each frame of the plurality of frames to generate time shifted frames; generating, by the content delivery service, an updated set of embeddings based on the time shifted frames; receiving a search request comprising input text from a user device; determining the video is a match for the search request based on the input text and the updated set of embeddings for the time shifted frames; and sending the video to the user device based on the match.

According to an aspect, a detection device includes: a plurality of optical sensors that are arranged in a detection area and each of which is configured to output a signal corresponding to light irradiating the optical sensor; a detection circuit configured to be supplied with the signals from the optical sensors; and a control circuit configured to output first data and second data for which detection conditions for acquiring the signals differ from each other. The control circuit is configured to output detection information data in which the detection conditions for the first data and the detection conditions for the second data are added as attribute information to a head of a data string in which the first data and the second data are sequentially arranged.

A medical image processing device of an embodiment includes first and second image acquirers, a generator, and a calculator. The first image acquirer acquires a first fluoroscopic image of a patient. The second image acquirer acquires a second fluoroscopic image according to radiation of the patient at a time point different from that of acquisition of the first image using a detector. The generator generates a reconstructed image by reproducing the second image from the first image virtually arranged in a three-dimensional space on the basis of a position of the detector in the three-dimensional space. The calculator obtains a suitable position on the first image in the three-dimensional space on the basis of a degree of similarity between the second image and the reconstructed image. The generator generates the reconstructed image for use in the calculator and has a range larger than a range corresponding to the second image.

This disclosure provides systems, methods and apparatus for synchronous channel access control of a wireless system. In some aspects, a device may use a TWT session to communicate with a second device during one or more TWT service periods. Uplink and downlink communications may be coordinated to both be in a TWT service period to allow a device to enter a low power mode outside of the TWT service period. The TWT session, including the service periods, may be configured and managed by the device or the second device to ensure the communications associated with an XR experience between the devices (such as pose data frames or tracking frames provided as uplink data and video data frames provided as downlink data) meet latency requirements or other requirements for the XR experience. Use of TWT service periods allows other devices to use the wireless medium outside of the TWT service periods.

The invention relates to the field of signal coding, and discloses a method for generating a 3D image watermark based on AIGC and a model framework, aiming at the problem that the watermark is easy to damage when being attacked by noise because the intensity of the embedded watermark cannot be enhanced at different scales in the existing 3D image watermark embedding method. According to the method, the NDOU module is embedded into the encoder based on the U-net, the multi-modal features of the middle viewpoint image and the depth map are extracted at the same time, watermarks are guided to be distributed on the middle viewpoint image at different scales and depths according to the internally-arranged WAM module, and the robustness of the embedded watermarks is improved; a discriminator and an encoder are arranged to adjust the parameters of each other in a confrontation manner, so that the visual effect and invisibility of the embedded watermark are ensured. A noise layer is arranged to improve the ability of the model to resist corresponding noise attacks, and a decoder is arranged to train the ability of an encoder to extract watermarks, so that the model can finally generate a watermarking-containing coded image with excellent robustness and invisibility.

The invention relates to the technical field of beef carcass segmentation, in particular to a beef carcass computer vision-assisted segmentation system, which comprises a vision acquisition and preprocessing module, a feature extraction module, a segmentation decision module, a dynamic target tracking module, an execution control module and a temperature monitoring and feedback module, the visual acquisition and preprocessing module is used for acquiring image data of beef carcasses; the feature extraction module extracts feature points and edge information; the segmentation decision module generates a segmentation path; the dynamic target tracking module corrects the generated segmentation path; the execution control module is used for cutting beef carcasses; and the temperature monitoring and feedback module monitors the temperature of the cutting equipment and the carcass cutting part in real time. According to the invention, the error and the labor intensity of manual operation are reduced, and the automation level and the overall benefit of the beef carcass segmentation process are improved.

The invention discloses a multistage video watermark embedding and extracting method and system, and the method comprises the steps: carrying out the perception analysis of an original video through a space-time attention neural network, and dynamically dividing a time domain stable region and a space domain sensitive region; a three-level watermark system is constructed, the three-level watermark system comprises a basic-level watermark, a first-level enhanced watermark and a second-level enhanced watermark, the basic-level watermark adopts an anti-rotation binary matrix generated by a Fourier descriptor, the first-level enhanced watermark is a DCT frequency domain Hash sequence, and the second-level enhanced watermark is a semantic fingerprint generated based on a motion vector. According to the method, through multi-dimensional collaborative optimization of the anti-attack capability, the dynamic security mechanism and the verification efficiency, the core defects that erasing is easy due to insufficient reversibility in a traditional video watermarking technology and counterfeit affiliation chaos is caused by a static key are overcome, and reliable technical guarantee is provided for digital content copyright protection.

Accurately identify a location based on images, even indoors SOLUTION: In this system, a control unit 22 of a server 20 that acquires an image captured by a terminal device 10 identifies the position of the terminal device. The server includes a database 23 that stores a large number of 3D maps showing each area and feature data related to the 3D maps. The feature data includes vectorized images of the 3D maps and related text information, context information of the 3D maps, etc. A map narrowing-down unit 222 of the control unit vectorizes the acquired images, etc., searches the vectors against feature data stored in the database, and also provides the acquired images and context information of the 3D maps to the LLM to narrow down the 3D map candidates (eliminates 3D maps with similar shapes). A map matching unit 221 matches the shapes of the narrowed-down 3D map candidates to accurately identify the position and orientation.

The invention relates to an image forgery active defense method based on multi-watermark fusion and cross-domain learning, and belongs to the technical field of computer vision. The method comprises the following steps: acquiring a to-be-processed image; an image to be processed is subjected to invisible watermark embedding and visible watermark embedding through a watermark encoder, and an image embedded with an invisible watermark and an image embedded with a visible watermark are obtained respectively; processing the image embedded with the invisible watermark through a noise layer to obtain a noise image; the image embedded with the visible watermark is processed through a noise layer, and an obvious false warning mark is generated at the position of the image embedded with the random noise through visible watermark joint optimization; the noise image is traced and detected through a watermark decoder, and the authenticity of the image is judged; and carrying out loss function supervision training. According to the method, whether the image is deeply counterfeited or not can be accurately judged, and the authenticity of an image source can be verified.

A system includes a plurality of satellites orbiting around a celestial body in a plurality of orbital planes. Each satellite includes an imaging device having a field of view (FOV) to capture an image of a sky that includes celestial image features of resident space object (RSO), stars, and / or planets. The satellite processor is configured to input the image data for each image into a pre-processing software pipeline to generate for each image a contrast-enhanced image data replica of each image, and an enhanced image data replica of each image, to receive from an output of a known-unknown RSO split data processing pipeline, a determination that a candidate RSO is a known RSO stored in an RSO catalog, or an unknown RSO, and to assign to the candidate RSO based on the determination, an RSO ID of the known RSO listed in the RSO catalog, or a new RSO ID.

The invention discloses an image steganography method and device, electronic equipment and a storage medium, a Res-SS2D module is designed, and through four-way global space modeling, the linear complexity is ensured, and meanwhile, the quality of a secret-containing image is improved; a high-frequency feature enhancement strategy is put forward, and secret information is guided to be preferentially embedded into a high-frequency area by extracting carrier image edges and fusing the carrier image edges to an encoder, so that the possibility of steganalysis and detection is reduced; and designing a multi-objective loss function, optimizing the generation quality by combining the PSNR and MS-SSIM, and ensuring that the secret information is preferentially hidden in the high-frequency component through the low-frequency consistency of the L1 norm loss constraint carrier and the secret-containing image of the low-frequency component. The invention provides an efficient and safe solution for the image steganography technology.

The invention discloses a model watermark robustness enhancement method based on antagonistic immunity, which mainly solves the problem that the existing model watermark is difficult to resist various watermark removal attacks, and adopts the scheme that a public data set and a convolutional neural network suitable for an image classification task are respectively used as original data and a target model; training a target model by using the original data; taking the generation model as a watermark generation model and training the watermark generation model by using the original data; inputting the original data into the trained generated watermark model for forward propagation to output watermark data; splicing the original data and the watermark data, and inputting the spliced original data and watermark data into the trained target model for forward propagation to obtain a model watermark; tiny disturbance is added to original data to generate an adversarial sample, the adversarial sample is spliced with the original data and then input into the model watermark for robustness training, and the model watermark with enhanced robustness is obtained. The method can effectively enhance the generalization ability of the model watermark under different data distributions, improves the robustness of the model, and can be used in the fields of information security, financial services, medical care and the like.

A system and method for mapping colors of a color calibrated image or video are provided. An image or video may first be color calibrated. The colors appearing in different areas of the image or video may then be compared to one or more selected target colors to determine if color values of the colors appearing in the image or video are within a selected color value tolerance of the target color. If areas of the image or video are within the tolerance, those areas may be visually delineated from other areas to display one or more mapped color areas within the image or video.

Techniques for utilizing invisible and / or inaudible watermarking techniques in media streams (e.g., video and / or audio streams) as a form of in-band steganography to authenticate the source capture camera and / or microphone to prevent deepfakes in network-based communication systems. User device(s) may establish cloud-based connections to a network-based communication system hosting media stream(s). An encoder of a watermarking component of a user device may receive video / audio data directly from a camera / microphone thereof and encode the video / audio data with watermark(s) representing a signed signature. A decoder of a watermarking component of a recipient user device and / or another device associated with the communication system may extract the watermarks and authenticate the signed signature using public certificate(s) associated with the sending user device to validate authenticity of the capture source(s) of the sending user device and that the video / audio data has not been modified since the time of capture.

The invention discloses an intelligent robot construction technology and system for building surface treatment, and relates to the technical field of building construction.The technology comprises the following specific processes that three-dimensional contour data collection is conducted, specifically, a building surface is comprehensively scanned through high-precision three-dimensional laser scanning equipment, and space coordinate information of all points on the surface of the building surface is obtained; by integrating three-dimensional laser scanning, image processing and spray gun and polishing control, full-process automatic operation from putty spraying, polishing to paint spraying is achieved, the construction period is greatly shortened, the efficiency advantage is particularly remarkable for large-area or complex-shape building surface treatment, and meanwhile, the construction efficiency is greatly improved. Due to the fact that the robot can work strictly according to the preset program and path and conduct accurate adjustment in combination with real-time feedback data, high consistency and uniformity of construction quality are ensured, and the problems of spraying omission, repeated spraying or uneven grinding and the like possibly occurring in manual construction are effectively avoided.

The invention is suitable for the field of digital watermarking, and provides an anti-compression coding robust video watermark generation method based on an adversarial neural network, and the method comprises the steps: constructing an MSCA-GAN model; the model comprises an encoder, a decoder, a distortion layer, a discriminator network and an opponent network, and the specific steps are as follows: step S1, the encoder extracts different scale features of a video frame by using a multi-scale convolution attention mechanism, calculates attention weights, efficiently hides and embeds binary watermark information into the video, generates a watermark-containing video, and transmits the watermark-containing video to the decoder; performing confrontation optimization on an embedding strategy with a discriminator in training; according to the method, for H.264 compression layer special training, the multi-scale convolution attention mechanism and the depth separable convolution are combined, and the anti-compression robustness of the watermark under the H.264 standard is effectively improved; through common attack training such as noise layer simulation cutting and zooming, the watermark can still keep high extraction accuracy and robustness in a complex environment.

In various examples, visual differences between video data and an encoded version of the video data are used to determine updates to quantization parameters (QPs) used to encode the video data to store or upload video clips that corresponds to notable events associated with a machine. The video data may correspond to images applied to a machine learning model to perform control operations for the machine. A metric may be used to quantify the visual differences. To evaluate the visual differences, samples of the video data and encoded video data may be determined and analyzed, rather than entire images or frames. To selectively enable updates to QPs, the system may detect that a deviation between a bitrate corresponding to the encoded data and a reference bitrate has exceeded a threshold.