292 results about "Image transformation" patented technology

The invention discloses a multi-scale linear array camera splicing method and system based on point cloud. The method comprises the following steps: completing acquisition and preprocessing of point cloud data and image data of a target area; determining an overlapping region range between adjacent images; extracting spatial structure characteristics in the point cloud data, and performing multi-scale hierarchical decomposition on the point cloud through a multi-scale segmentation method; meanwhile, multi-scale image feature extraction is carried out on the images of the linear array camera; solving gradients in X and Y directions by adopting an optical flow method aiming at any pixel in the overlapping region, and calculating a motion vector between the pixels; fusing the optical flow information obtained under each scale, and constructing a globally consistent optical flow vector field; according to the fused optical flow vector, calculating to obtain a geometric transformation matrix of the whole overlapping region; and after image transformation and alignment are completed through the transformation matrix, fusion processing is carried out on overlapped areas. And the unification of the visual effect and the spatial integrity of the spliced image is ensured.

The invention provides an image defogging enhancement and intelligent perception collaborative optimization method based on stream matching, and the method comprises the steps: obtaining a foggy input image, and constructing an ordinary differential equation for defining image transformation; the input image is input into a fog perception vector field, the fog perception vector field comprises an atmospheric scattering purifier and a defogging perception color lookup table, and the atmospheric scattering purifier is used for extracting multi-scale features from the foggy input image to generate clearer output; the defogging perception color lookup table adaptively adjusts the image color through a nonlinear color transformation mechanism; and solving the ordinary differential equation through an RK4 solver, iteratively updating the foggy input image at each time step, and ensuring that the input image can be stably converted into a clear image from an initial foggy state. The technical problems that in the prior art, the existing method is forced to reduce the model capacity due to the calculation efficiency constraint in the ultra-high-definition image defogging, so that the fog concentration estimation is incomplete, and the perception quality and the calculation efficiency cannot be considered at the same time are solved.

Disclosed are systems and methods including software processes for developing test cases for testing robustness of AI-based image-recognition models-under-test (MUTs) with respect to types of image variation transformations. The system may generate various types of robustness metrics for the MUT and output user-readable reports about the MUT's performance. The system trains machine-learning architectures to generate test cases including augmented images according to the types of image transformations, applies the IR MUTs, and then evaluates the image feature vector embeddings and predicted classification produced by the IR MUTs to determine the accuracy of the MUT with respect to each type of transformation.

The invention discloses a panoramic image determination method and device, equipment, a medium and a product. The method comprises the following steps: acquiring a current scene image, shot by at least one camera device, of an environment to which a vehicle belongs, and kinematics data and attitude data of the vehicle from a current shooting moment to a previous shooting moment; determining a scene image transformation matrix based on the kinematics data and the attitude data; acquiring a historical scene image shot by the camera device at a previous shooting moment, and performing stereo matching on the current scene image and the historical scene image based on the scene image transformation matrix and shooting parameters of the camera device to obtain a scene stereo image corresponding to the camera device; and determining and displaying a panoramic image based on the scene stereo image of each camera device. The authenticity, accuracy and reliability of panoramic image determination are improved, and the driving safety is improved.

The invention relates to the technical field of image geographic positioning, in particular to a progressive cross-view-angle image geographic positioning method based on spatial feature aggregation and position awareness, which comprises the following steps: acquiring paired satellite images and ground panoramic images as sample images, and constructing a training sample set; a cross-view image geographic positioning model composed of a double-branch backbone network and a fine-grained prediction network is constructed, two branches of the double-branch backbone network are a satellite processing branch and a ground panorama processing branch, and the fine-grained prediction network is composed of a bird's-eye view image transformation module and a position sensing prediction module; inputting training samples in the training sample set into the cross-view image geographic positioning model, and performing iterative training on the cross-view image geographic positioning model until convergence; and inputting a to-be-positioned paired satellite image and ground panoramic image into the trained cross-view-angle image geographic positioning model to obtain a positioning result output by the trained cross-view-angle image geographic positioning model.

The invention provides an automatic image cutting method and system based on face key points, which are applied to the technical field of image processing, and the method comprises the steps: inputting a portrait image into a portrait segmentation model for foreground extraction, and obtaining a figure region mask; performing top positioning estimation based on the figure area mask to obtain a top position of the portrait image; inputting the portrait image into a face key point detection model to obtain a plurality of key points; determining a face midpoint of the portrait image based on the position relationship of the plurality of key points; performing affine transformation estimation based on the source point set and a preset target point set to obtain an affine transformation matrix; adjusting the vertical offset of the affine transformation matrix to obtain a longitudinal correction matrix; performing edge correction on the longitudinal correction matrix to obtain a cutting transformation matrix; and performing image transformation on the portrait image based on the cutting transformation matrix to obtain a standard composition image. According to the invention, the cut image with uniform size, standard composition and good detail retention can be generated.

Systems and methods for view-conditioned diffusion for real-world vehicle gaussian splatting. A single perspective image can be transformed using image transformation techniques to generate a training dataset that addresses a domain gap between synthetic data and real-world data in a traffic scene. A pre-trained diffusion model can be finetuned with the training dataset to obtain a fine-tuned diffusion model. Perspective-aware images having different perspective views of an entity from the single perspective image can be generated using the fine-tuned diffusion model. A large generative model (LGM) can be trained using the perspective-aware images to generate a gaussian splatting model for the entity. View-conditioned simulations from the single perspective image can be generated by using the gaussian splatting model for downstream tasks.

The application discloses a similarity optimal on-board image registration method based on inertial navigation data fast convergence, and comprises the following steps: constructing a projection transformation model between an image transformation matrix, a to-be-registered image and a reference image; obtaining the projection transformation matrix according to the field of view optical axis position and three-angle offset of the satellite at the imaging moment and coarse registration; performing global coarse matching by using the projection transformation model and the projection transformation matrix to obtain a coarse matching image; decoupling the image transformation matrix according to a mathematical model to obtain a to-be-solved transformation parameter, substituting the to-be-solved transformation parameter into the image transformation matrix to obtain an optimal transformation matrix, and applying the optimal transformation matrix to the to-be-registered image to obtain a fine registration image; and calculating the similarity of the fine registration image and the reference image according to a local normalized image similarity measurement algorithm to obtain an optimal matching result. The application has the advantages of meeting the on-board calculation capacity requirement, meeting the image processing precision requirement and meeting the on-board storage resource requirement.

A method for extracting at least one image feature from at least one image generated by at least one real camera, present in a motor vehicle, for monitoring the environment of the motor vehicle. The method includes, in a first step a), the image generated by the camera is transformed into a transformed image by means of a virtual camera which is assigned to the real camera and the optical axis of which extends along a horizontal direction with respect to the motor vehicle, according to which, in a second step b), at least one image feature is extracted from the transformed image.

The invention relates to the technical field of machine learning, in particular to a knowledge distillation method and system based on view alignment. The invention aims to solve the main limitations in the traditional distillation technology, including the problems of excessive confidence of a teacher model, confirmation deviation and the like. The invention provides a novel knowledge distillation framework based on logarithmic probability, which is called KDVA; specifically, z-score standardization is applied to the logarithmic probability of a model for smooth output, so that transmission of more teacher implicit knowledge is promoted. In addition, a same-view-angle and cross-view-angle alignment mechanism is introduced into the KDVA, and comparison information of weak and strong image transformation is utilized to enlighten a student model to obtain more knowledge. In addition, the student model is supervised by using the real label of the sample, so that the student model can obtain more information about each sample target category. According to the method, when different network architectures are applied to different data sets, high effectiveness and stability are shown.

In accordance with the described techniques, an image transformation system receives an input image and a text prompt, and leverages a generator network to edit the input image based on the text prompt. The generator network includes a plurality of layers configured to perform respective edits. A plurality of masks are generated based on the text prompt that define local edit regions, respectively, of the input image for respective layers of the generator network. Further, the generator network generates an edited image by editing the input image based on the plurality of masks, the respective edits of the respective layers, and the text prompt.

The invention provides a continuously constrained and difficulty guided passive domain adaptive three-dimensional medical image segmentation method. The method comprises the following steps: in a source domain pre-training stage, carrying out full-supervised training on a segmentation model by utilizing source domain annotation data; in the pseudo source domain image generation stage, a thought of combining coarse generation and fine generation is adopted, style migration is performed by using a frozen source domain pre-training segmentation model and target domain unlabeled data in coarse generation, and a target domain image is converted into a pseudo source domain image with a source domain style; and in the fine generation step, Fourier transform is utilized to remove artifacts and noise in the coarsely generated image. In the target domain adaptation stage, a pre-training segmentation model, a pseudo source domain image and a target domain image are utilized, and continuity constraint between slices and a difficult sample mining mechanism are fused to carry out an adaptation process from a source domain to a target domain. According to the method, under the condition that source domain data does not need to be accessed, the spatial context constraint and the difficult sample mining mechanism of the three-dimensional medical image are effectively fused.

The invention discloses an infrared image registration method and an unsupervised learning image registration model training method, and belongs to the technical field of image processing. Aiming at the characteristics of weak texture and few features of a low-overlapping-rate infrared image, a weak texture feature extractor is firstly designed, and the feature extraction capability of a model on a weak texture image is improved through multi-scale feature fusion; then, by adopting multi-scale receptive field correlation calculation, the discrimination capability of the model on similar features is enhanced; furthermore, by adopting a homography transformation estimation method which is optimized gradually from coarse to fine, the estimation precision of a transformation matrix is improved; and finally, realizing registration and splicing of the infrared images through image transformation and overlapping region fusion processing. According to the method, the registration precision of the low-overlapping-rate infrared image can be effectively improved, and application scenes such as large-scene image splicing can be supported.

The invention provides a real-time video image stabilization method and system for performing intra-frame motion compensation based on an IMU (Inertial Measurement Unit). The method comprises the following steps: S1, preprocessing data; s2, track smoothing processing is carried out; s3, performing intra-frame motion compensation: S3.1, initializing line exposure delay time; s3.2, calculating the time difference between the current vth line and the middle line exposure in the current frame image; s3.3, acquiring the smoothed virtual camera pose of the current frame and the real camera pose of each layer; s3.4, acquiring a motion compensation matrix for compensating the line of image to the virtual pose of the frame of image; s3.5, acquiring an image frame perspective transformation matrix according to the camera internal reference matrix K and the motion compensation matrix; s3.6, performing intra-frame motion compensation on the image layers to obtain an initial motion compensation matrix queue; s3.7, carrying out black edge removal processing on the motion compensation matrix queue; obtaining a motion compensation matrix queue after black edge removal; and S4, image transformation output. The system comprises a data preprocessing module, a track smoothing module, an intra-frame motion compensation module and an image transformation output module. The operation efficiency is improved, and the anti-shake real-time performance is improved.

The present disclosure relates to systems, methods, and non-transitory computer readable media that utilize deep learning to identify regions of an image that have been editorially modified. For example, the image comparison system includes a deep image comparator model that compares a pair of images and localizes regions that have been editorially manipulated relative to an original or trusted image. More specifically, the deep image comparator model generates and surfaces visual indications of the location of such editorial changes on the modified image. The deep image comparator model is robust and ignores discrepancies due to benign image transformations that commonly occur during electronic image distribution. The image comparison system optionally includes an image retrieval model utilizes a visual search embedding that is robust to minor manipulations or benign modifications of images. The image retrieval model utilizes a visual search embedding for an image to robustly identify near duplicate images.

A method for generating aligned image data through an aligning parameter generated by an image transformation artificial intelligence (AI) model includes, through an encoder of the image transformation AI model, generating at least one or more aligning parameters from a first camera property and a second camera property related to a first camera and a second camera respectively. The method also includes, through an image transformer of the image transformation AI model, transforming, based on the at least one aligning parameter and a brightness parameter, first image data photographed by the first camera to be aligned with second image data photographed by the second camera. The method also includes training the encoder and a discriminator of the image transformation AI model by adversarial training. The image transformation AI model discriminates between the transformed first image data and the second image data.

Systems, methods, and computer-readable media are provided for using generative AI enriched with metadata about historical document characteristics to transform documents of various formats, including images, to the fields and values they represent. A prompt template may be selected in association with a type of document. The prompt template indicates field definition(s) of field(s) to be detected in the document and location(s) in which the field(s) have been detected in prior documents. A large language model is prompted with a prompt generated using the prompt template to generate a result that assigns value(s) to the field(s). Output from the language model is used for identifying the field to value mapping for the document, such that data detected from the document may be stored in appropriate database structures of a database. Metadata stored in association with the prompt template is updated based on location(s) in the document in which the field(s) were detected, and the value(s) of the field(s) are stored in a database. Outbound documents may be similarly translated to detect values of corresponding fields requested by third parties, even if those values are not stored in the database. In this scenario, values for fields may be detected in outbound documents using the prompt templates enriched with metadata as processed by the large language model before such information is prepared to be sent to a third party.

Described are systems and methods for training a machine-learning model to generate image of biological samples, and systems and methods for generating enhanced images of biological samples. The method for training a machine-learning model to generate images of biological samples may include obtaining a plurality of training images comprising a training image of a first type, and a training image of a second type. The method may also include generating, based on the training image of the first type, a plurality of wavelet coefficients using the machine-learning model; generating, based on the plurality of wavelet coefficients, a synthetic image of the second type; comparing the synthetic image of the second type with the training image of the second type; and updating the machine-learning model based on the comparison.

Presented herein are systems and methods for classifying features from biomedical images. A computing system may identify a first portion corresponding to an ROI in a first biomedical image derived from a sample. The ROI of the first biomedical image may correspond to a feature of the sample. The computing system may generate a first embedding vector using the first portion of the first biomedical image. The computing system may apply the first embedding vector to a clustering model. The clustering model may have a feature space to define a plurality of conditions. The clustering model may be trained using a second embedding vectors generated from a corresponding second portions with at least one of a plurality of image transformation. The computing system may determine a condition for the feature based on applying the first embedding vector to the clustering model.

A method for a driver assistant image customization of vehicle-related images, a data processing circuit, a computer program, a computer-readable medium, and a vehicle, can include, with at least one sensing device of the vehicle, obtaining an input image to be customized. With at least one human-machine-interface of the vehicle or connected thereto an input determining at least one customization scheme to be performed is received. Using at least one data processing circuit of the vehicle applying artificial intelligence the input image is transformed according to the at least one customization scheme into a transformed output image. With at least one smart mirror of the vehicle or a mobile device connected to the at least one data processing circuit the transformed output image is outputted. The at least one customization scheme includes a plurality of different types of adaptation modes.

The present application is a backdoor attack method based on image steganography, comprising: S1, constructing an image steganography network and an image transformation network; S2, performing spatial transformation on a poisonous image to obtain a first trigger; S3, inputting the spatially transformed poisonous image back into the attack network to restore it to obtain a second trigger; S4, constructing a steganalysis loss function of the image steganography network based on the distance loss between the first trigger and the second trigger; S5, iteratively training the image steganography network; S6, performing a backdoor attack using a poisonous image. When training the steganography network, a loss function is constructed based on the degree of deformation of the trigger, and the image steganography network generates a poisonous image that can adapt to image transformation through the back propagation of the loss function. The poisonous image generated by the present application can trigger an attack in the victim model even after image transformation processing, so that the victim model recognizes the corresponding target label.

The present invention discloses a method for detecting and removing stripe noise in satellite-borne spectral images. The method comprises an image classification and characteristic analysis module, a stripe detection and analysis module (including a stripe detection unit, a characteristic analysis unit, and a degradation modeling unit), an image transformation module, and a soft threshold signal decomposition module. Based on independent analysis of the original space-spectral domain and the transform domain, the present invention comprehensively integrates the space-spectral domain analysis method and the transform domain analysis method to simultaneously detect and remove stripe noise. The method also fully analyzes the image characteristics and wavelet distribution characteristics, deploys an image decomposition method in the wavelet domain, effectively analyzes the stripe components, and retains image detail information. The present invention has the advantages of being adaptable to the complexity of spectral image sources and the diversity of application environments, taking into account the differences in different stripe noise distributions, and having strong generalization and robustness.

Fisheye images that include objects at first, second, and third angles into rectilinear images are transformed with a first image transformation and the rectilinear images are transformed into bird's eye view images with a second image transformation. The bird's eye view images can be transformed into multiple images that include objects at multiple angles intermediate between the first, second, and third angles to generate a training dataset that includes ground truth regarding the multiple angles with a third image transformation. A machine learning model can be trained with the training dataset. A machine such as a vehicle can be operated with output from the machine learning model.

The application provides an image processing method, comprising: presenting an image transformation function item in a view interface, in response to a triggering operation on the image transformation function item, acquiring and presenting a to-be-processed image containing a first object; in response to a transformation determination operation triggered based on the to-be-processed image, generating and presenting a target image of a target image template. The application also provides an image processing device, an electronic device and a storage medium. The application can retain the characteristics of the to-be-processed image, avoid distortion of the replaced target image, realize batch processing of different to-be-processed images of the user, reduce the occupation of hardware resources in the image processing process, reduce the increase of the cost of hardware devices, and improve the user experience.

The invention provides a robot multi-mode touch sensing device based on electrical capacitance tomography, and the device comprises a flexible electrical capacitance tomography sensor which is disposed at an execution tail end of a robot and is used for sensing capacitance measurement values in a non-contact stage and a contact stage respectively; the multi-mode signal processing unit is used for acquiring the capacitance measurement value in real time, processing the capacitance measurement value to generate multi-mode touch information and feeding back the multi-mode touch information to the robot control system; wherein object category identification is carried out based on the capacitance measurement value in the non-contact stage, if the identified object is a contactable object, ECT image reconstruction is carried out by using the capacitance measurement value in the contact stage to obtain a contact image, and the contact image is converted into contact force distribution through a gray scale-pressure mapping relation. And carrying out object contour extraction by using the geometric features of the contact image. The adaptability and safety of the robot in a complex environment are improved, and the method is particularly suitable for the field of robots needing fine operation and environment interaction.

83 20 25 28 74 03 30 D ec 2 02 5 A B S T R A C T 2 0 2 5 2 8 7 4 0 3 3 0 D e c 2 0 2 5 8 3 IL LU M IN AT IO N PA TT ER N (S ) 11 0 A SE CO ND T YP E OF IM AG ES (E .G ., FL UO RE SC EN CE IM AG ES ) 12 4 A FI RS T TY PE O F IM AG ES (E .G ., BR IG HT -F IE LD IM AG ES 12 2 DI SC RI MI NA TO R 10 4 GE NE RA TO R 10 2 Tr ain in g Da ta 12 0 GA N Mo de l 1 00 FI G. 1 1 / 26 Training Data 120 GAN Model 100 ILLUMINATION A FIRST TYPE OF IMAGES (E.G., PATTERN(S) BRIGHT-FIELD IMAGES 110 122 GENERATOR 102 A SECOND TYPE OF IMAGES (E.G., DISCRIMINATOR FLUORESCENCE IMAGES) 104 124 FIG. 1 20 25 28 74 03 30 D ec 2 02 5 2 0 2 5 2 8 7 4 0 3 3 0 D e c 2 0 2 5 T r a i n i n g D a t a 1 2 0 ILLUMINATION A FIRST TYPE OF IMAGES (E.G., P A T T E R N ( S ) BRIGHT-FIELD IMAGES A SECOND TYPE OF IMAGES (E.G., FLUORESCENCE IMAGES)

The invention relates to the technical field of image processing, and discloses an image optimization method and device, equipment and a medium, and the method comprises the steps: recognizing the degradation information of a target image, carrying out the image transformation processing of the target image according to the image degradation information and a preset image transformation process, and obtaining an optimized image, performing multi-type visual task detection on the optimized image, generating a quantized image quality index, judging whether the image reaches a quality standard by comparing a detection parameter with a preset threshold value, if not, calculating a reward signal of a reinforcement learning agent model according to a detection result, and updating an image transformation process by using the signal, so as to improve the quality of the image. And then returning to the optimization step to process the image again, if the current strategy reaches the standard, indicating that the current strategy is effective, directly processing the subsequent to-be-processed image by the agent by using the updated strategy, and finally outputting a high-quality target optimized image. According to the invention, the efficiency and precision of image processing are improved.

An information processing device includes an image transformation unit, a left-right difference estimation unit, and an image generation unit. The image transformation unit performs warping to move positions of a feature point of a right-eye image and a feature point of a left-eye image based on right-eye and left-eye viewpoint information. The left-right difference estimation unit estimates a portion where a difference exceeding an allowable level occurs, due to the warping, between the right-eye image and the left-eye image as an inconsistent portion. The image generation unit makes a sharpness of the inconsistent portion different between the right-eye image and the left-eye image.

The application discloses an active interference image anti-counterfeiting method, system, device and storage medium, which are corresponding solutions, and in the solutions, the adversarial disturbance can be adaptively distributed in an area which is not sensitive to human eyes, so that the invisibility of the disturbance is significantly enhanced, the application further introduces Gaussian blur, and the ability of the disturbance to resist image transformation is improved through adversarial training, so that the application has high robustness; in addition, the application does not depend on prompt words, an attacker is difficult to bypass the protection by using multiple prompt words, and the application only fine-tunes the weight of a variational autoencoder, so that the training resources consumed are far less than those of a traditional active interference method for a U-Net.