777 results about "Candidate image" patented technology

The invention provides a subway tumble detection method and system based on cloud edge collaboration and an X-CLIP model, and relates to the technical field of subway tumble detection, and the method comprises the steps: obtaining a real-time video stream in a subway, and analyzing the video stream data into frame image data and video stream data; inputting the frame image data into the YOLOv8n-c3xghost neural network model to obtain all human body target area detection frames; performing identification detection on the human body target area detection frame through a BoT-SORT target tracking algorithm to obtain candidate images of abnormal behaviors; inputting the candidate image of the abnormal behavior into an X-CLIP model, outputting an abnormal region fragment classification result of the candidate image, identifying an abnormal region fragment again, determining the falling behavior according to the situation that the falling probability is greater than a set threshold value, outputting a position coordinate where the falling behavior occurs and a corresponding original image, and giving an alarm; wherein the model is deployed at the cloud end and trained, the model sinks to the edge side after being compressed and packaged, and real-time video frame detection is realized at the edge side.

The invention provides a subway pedestrian anomaly detection method and system based on an improved yov8 network and a clip model, and relates to the technical field of subway pedestrian detection, and the method comprises the steps: obtaining a video frame in a subway scene, and carrying out the preprocessing of the video frame; inputting the preprocessed frame data into an improved yolov8 neural network model to obtain an abnormal behavior target detection frame, and performing target candidate tracking on the target detection frame by using a target tracking algorithm to obtain a candidate image of a target area of a specific behavior; the method comprises the following steps: inputting a candidate image containing a target area into a clip model, inputting a set detection text into the clip model, mapping the detection text and the candidate image of the target area, performing text-image similarity matching, and optimizing a similarity matching threshold by adopting a Q-Learning reinforcement learning immune algorithm to obtain a similarity matching result; and the abnormal behavior detection accuracy is improved.

The invention provides a single-view image three-dimensional modeling method and system based on comparative learning, and the method comprises the steps: starting from a single input image, generating a multi-view candidate image through a pre-training two-dimensional diffusion model in combination with score distillation sampling; performing positive and negative sample division on the candidate image pairs based on texture fidelity and structure coherence; constructing a perceptual contrast learning mechanism, calculating perceptual loss between the images, and optimizing texture consistency in a three-dimensional content generation process; proposing a number-aware triple loss function, and dynamically adjusting the contribution ratio of positive and negative samples in training; then, a super-resolution reconstruction module is embedded to perform detail enhancement on the image, and the quality of a contrast gradient signal is improved; and finally, performing joint training on the perception comparison loss and a neural radiation field or volume rendering module to generate a high-quality three-dimensional grid or volume rendering result. The problems that in an existing diffusion generation method, multi-view textures are inconsistent, structure reconstruction is not coherent, and the semantic distinguishing capacity is insufficient are effectively solved.

The invention discloses a graph-driven intelligent medical image retrieval method and system, and relates to the field of image retrieval, and the method comprises the steps: firstly, coding a text query input by a user through constructing a medical knowledge graph, querying to generate a graph-driven enhanced embedded vector containing an advanced medical concept, and querying a related graph concept set; then, performing preliminary vector similarity retrieval by using the enhanced vector, and quickly screening out Top-K preliminary candidate images; and finally, on the basis of querying a related map concept set, carrying out map concept secondary sorting on the preliminary candidate images, and accurately matching deep medical semantics of the images by calculating a concept coincidence degree. The mechanism of two-stage retrieval and map deep fusion ensures that the retrieval result is not only similar in vision, but also highly related in medical concept, so that the accuracy and intelligent level of medical image retrieval are remarkably improved.

The invention relates to the technical field of image retrieval, in particular to an image retrieval method based on big data mining, which comprises the following steps: collecting images and metadata, screening metadata fields, determining a candidate image set and obtaining a metadata screening result, according to the method, in the image retrieval process, image metadata screening, geographic information clustering, visual feature fusion, index construction and matching strategy optimization are combined, and the retrieval accuracy and response efficiency are improved. In the metadata screening link, data is screened by using information such as time, places and equipment types, so that interference of irrelevant data is reduced, and a candidate image set is more targeted. Geographic information and timestamps are combined for classification, and clustering analysis is performed by utilizing a geographic information system, so that spatial distribution information plays a greater role in a retrieval process, and a more accurate result can be obtained for query of a specific region.

Image similarity search results can be improved by augmenting a query image with images that are generated based on the query image. The generated images can have different camera poses, different lighting conditions, etc. The generated images can be generated using machine learning models. Using an ensemble of images having the query image and the generated images when searching a library of candidate images can improve performance of image similarity search. Producing effective generated images is not trivial. Also, an image similarity search algorithm may be modified to identify and rank top matching images that are most similar to the ensemble of images.

The invention discloses a three-dimensional scene target blanking method and system based on 3DGS, and belongs to the field of computer vision processing. The method comprises the following steps: firstly, performing three-dimensional reconstruction through a multi-view image to generate an original Gaussian point cloud scene; after a scene is divided into foreground and background point clouds based on target text information, spatial expansion is performed on the foreground to generate a three-dimensional soft mask. And projecting the mask to a two-dimensional image space to generate a two-dimensional mask image of each view angle, and inputting the two-dimensional mask image and an original image into an image blanking model to generate a candidate blanking image set. After a user selects a reference image from an initial view angle, a candidate image most similar to a preorder view angle is selected as a final blanking result from a non-initial view angle, and fidelity enhancement is carried out. And finally, optimizing the original Gaussian point cloud parameters by using the blanking result of each view angle to obtain a target three-dimensional scene after blanking. Based on the invention, a user can carry out texture remodeling on the structure of the original three-dimensional scene, and the multi-view consistency of the three-dimensional scene is maintained.

The invention relates to the technical field of image processing, in particular to an image search precision optimization method and system based on the combination of rough arrangement and fine arrangement, and the method comprises the steps: a rough arrangement model stage: calculating the rough arrangement similarity of a query image and all images in a database, and sorting and screening out a preset number of candidate images according to the rough arrangement similarity; a model fine arrangement stage: performing multi-scale reconstruction on the query image to generate reconstructed images with different resolutions; extracting feature vectors of the reconstructed images through a fine arrangement model, performing expansion operation on the feature vectors, and calculating fine arrangement similarity between the query image and the candidate image; and a similarity weighting combination stage: performing normalization processing on the rough ranking similarity and the fine ranking similarity, dynamically adjusting the weight coefficients of the rough ranking similarity and the fine ranking similarity according to the difference of the rough ranking similarity, generating a final similarity through weighted average, and finally sequencing and outputting a retrieval result. And by dynamically combining the distance of rough arrangement and the distance of fine arrangement, the image retrieval precision is further improved.

Embodiments of the present invention provide systems, methods, and computer storage media for selection of the best image of a particular speaker's face in a video, and visualization in a diarized transcript. In an example embodiment, candidate images of a face of a detected speaker are extracted from frames of a video identified by a detected face track for the face, and a representative image of the detected speaker's face is selected from the candidate images based on image quality, facial emotion (e.g., using an emotion classifier that generates a happiness score), a size factor (e.g., favoring larger images), and / or penalizing images that appear towards the beginning or end of a face track. As such, each segment of the transcript is presented with the representative image of the speaker who spoke that segment and / or input is accepted changing the representative image associated with each speaker.

Systems and methods for image search result filtering can include obtaining a search query, determining a plurality of candidate image search results, processing the search query with a generative model to determine a plurality of search result criteria, and refining the plurality of candidate image search results based on determining whether the candidate results satisfy the plurality of search results criteria. The systems and methods can perform a plurality of determinations based on the output of the generative model.

The invention provides an injection molding part surface flaw detection method and device based on double-model collaboration, and the method comprises the steps: carrying out the local abnormal reflection feature analysis of a multi-view optical image collected on the surface of an injection molding part based on a first detection model, and obtaining a candidate flaw region; performing spatial positioning in the multi-view optical image based on the candidate flaw area to obtain a multi-view candidate image segment; performing surface geometric continuity analysis on the multi-view candidate image segments based on a second detection model to obtain a three-dimensional surface geometric consistency result; carrying out authenticity discrimination on the candidate flaw area based on a three-dimensional surface geometric consistency result to obtain a real flaw area; wherein the first detection model is used for capturing an optical response model of local abnormal reflection characteristics; the second detection model is a stereoscopic vision model for analyzing the geometric continuity of the multi-view lower surface. According to the invention, the false alarm rate of surface defect detection of the injection molded part under a complex surface condition is reduced.

The invention discloses a complex text image retrieval method and system based on a multi-modal large model thinking chain. The method comprises the following steps: S1, carrying out adaptive semantic disassembly; s2, reconstruction and optimization are carried out; s3, constructing a matching probability matrix; elements of the matching probability matrix represent probability scores of matching degrees between the candidate images and the matching texts; combining the matched text with the candidate image, and sending the combined image to a pre-training visual language model pair by pair to obtain a corresponding matching score; then a binary discriminant scoring mechanism is used to convert the matching score into a probability score that the cue word is' YES '; and S4, according to the matching probability matrix, calculating the matching degree between each candidate image and the original proposition, and selecting the candidate image with the highest matching degree as an image retrieval result. The invention aims to extract and optimize semantic features from complex text description by using a self-adaptive local deconstruction and global optimization method and a binary discriminant scoring mechanism guided by cue words, so as to improve the accuracy and generalization ability of image retrieval.

The invention discloses a machine vision-based enzyme freeze-dried powder quality detection method, and belongs to the technical field of machine vision, and the method specifically comprises the following steps: in a closed optical cavity, synchronously collecting a multispectral polarization sequence of a sample through annular multispectral polarization illumination and microscopic imaging; performing short-time airflow pulse disturbance and continuous imaging on the sample, and constructing a polarization time sequence texture map by means of inter-frame polarization retention difference; according to Stokes parameters, calculating a deskewness graph and a specular reflection component graph, and carrying out joint normalization on the deskewness graph and the specular reflection component graph and the polarization time sequence texture graph to form a multi-channel feature stack; constructing a double-branch encoder, and performing comparative learning on the feature stack to generate an impurity prototype dictionary; using the dictionary to match the feature stack pixel by pixel, and tracking and inhibiting powder agglomeration false detection through a topological constraint connected domain; and recovering a height map by using an oblique incidence fringe phase shift method, performing pixel-level fusion with the candidate image, retaining targets with height abrupt change and consistent polarization characteristics, and outputting an impurity mask.

The invention provides a method for distinguishing cell activity behaviors in a culture medium under the assistance of a computer, and relates to the technical field of cell image processing, and the method comprises the following steps: obtaining a target microscopic image of a target cell at the current moment, extracting the static characteristics of the target cell in the target microscopic image, and if the recognition confidence is lower than a first threshold value, determining that the target cell is a cell activity behavior; if yes, calling a historical image sequence of an area where the target cell is located; at least N candidate images are obtained through screening in the historical image sequence, time sequence dynamic characteristics of the target cells are obtained based on the candidate images, and the time sequence dynamic characteristics comprise the motion trail, the motion speed, the morphological change related to the activity behavior and the periodic change condition presented by the target cells within a certain time period; and combining the time sequence dynamic characteristics with the static characteristics, and carrying out activity behavior matching to obtain activity behaviors of the target cells. According to the method, the robustness and the accuracy of cell behavior recognition in a complex culture environment are remarkably improved through static-dynamic feature fusion.

The invention provides a pupil center positioning method and device and electronic equipment, and relates to the technical field of image processing, and the method comprises the steps: obtaining a target pupil image; performing gradient calculation on the target pupil image according to a pre-configured gradient direction to obtain a gradient filtering image; taking the MASK image of the target pupil image as a mask to execute mask operation to obtain a pupil position candidate image, filtering each circular arc contour in the pupil position candidate image to obtain a candidate contour, fitting the circle center of the candidate contour to obtain a circle center-containing set, filtering the circle centers in the circle center set to obtain a target circle center, and obtaining the target circle center. And determining the target circle center as the pupil center. According to the pupil center positioning method and device and the electronic equipment, multiple filtering modes are adopted in the whole positioning process, the pupil center can be accurately obtained, and meanwhile high accuracy can still be achieved under the conditions of poor image contrast, shielding, deformation and the like.

The invention relates to an image edge extraction method based on Canny detection operator optimization, and belongs to the technical field of image detection. The method comprises the following steps: acquiring a source image, converting the source image into a grey-scale image, performing down-sampling operation on the grey-scale image to obtain a down-sampling image, filtering the down-sampling image, and calculating a gradient magnitude and a direction of the filtered image to obtain a gradient magnitude image; removing non-edge points from the gradient magnitude image by optimizing a non-maximum suppression process, and storing candidate image edges; defining a real edge and a weak edge by setting a careful extraction interval, and setting a candidate image edge of which the gradient value is lower than the lower bound of the careful extraction interval as 0; lagging edge tracking is carried out on the weak edge to judge whether the weak edge and the strong edge are continuous or not, if yes, the weak edge is set as a real edge, and if not, the weak edge is removed. The accuracy and robustness of edge detection are remarkably improved in an image with more noise, and the phenomena of generation of false edges and edge breakage are reduced.

The invention relates to a video key frame obtaining method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining a plurality of candidate image frames corresponding to a to-be-processed video, carrying out the preprocessing of each candidate image frame, inputting the preprocessed candidate image frame into a pre-constructed image feature extraction model, obtaining a feature vector corresponding to each candidate image frame, and obtaining a feature vector corresponding to each candidate image frame; performing clustering processing on the feature vector corresponding to each candidate image frame through the target clustering parameter based on a preset clustering algorithm to obtain at least one target clustering result, and performing calculation based on the feature vector corresponding to each candidate image frame and the center vector corresponding to the target clustering result to obtain a target clustering result; and determining a target image frame of the to-be-processed video from the plurality of candidate image frames based on a calculation result. By adopting the technical scheme, the image frames with large scene and element differences can be screened from the continuous image frames, and the screened image frames are labeled and applied to the training of the automatic driving perception algorithm, so that the training efficiency can be greatly improved.

The invention provides a scheme detection method and device for a complex image-text mixed file, and the method comprises the steps: extracting standard review items and standard review contents in a standard manual, and carrying out the structural processing, and obtaining a standard review file; converting a to-be-detected file into a to-be-detected image, extracting a to-be-examined multi-scale candidate region image from the to-be-detected image by using an initial recognition model and a matching model obtained by OCR and pre-training, and extracting character information of the multi-scale candidate image to obtain character information of the candidate region; inputting the multi-scale candidate region image and the standard review file into a lightweight hybrid twin network obtained by pre-training for image feature matching, and outputting a matched feature image; inputting the matched feature image, the standard review file and the candidate area text information into a multi-modal large model obtained by pre-training for compliance analysis, and outputting a detection result; according to the method, the automation and intelligence degree of review of the complex image-text mixed file can be remarkably improved.

The invention provides a lens detection method and system, and the method comprises the steps: sequentially enabling transmission light to penetrate through a to-be-detected lens for imaging based on a plurality of different exposure values, and determining a plurality of candidate images; performing fusion according to the plurality of candidate images to determine a transmission image of the to-be-detected lens; identifying the transmission image to determine a plurality of abnormal pixel points of which the pixel values are smaller than a pixel threshold value; determining a defect area according to all the abnormal pixel points; and determining a detection result of the to-be-detected lens according to the defect area. Through the above detection method, the detection of the lens is more comprehensive, and the detection result is more reliable.

The embodiment of the invention discloses a picture display method, electronic equipment and a computer readable storage medium, in the method, the electronic equipment responds to a target operation used for uploading a picture, and target information related to interface content of an application program is obtained; according to the first picture number and the target information, a target candidate picture set of each target element is obtained from the picture set, the target elements are determined according to the target information, the target candidate picture set comprises at least one target candidate picture, and the total number of the target candidate pictures of each target element is smaller than or equal to the first picture number; the first picture number is the number of pictures capable of being displayed on a first screen interface of the picture selection interface; and displaying the target candidate picture of each target element on the first screen interface. Thus, the user can obtain the pictures of the multiple target elements on the first screen interface, the pictures of the target elements do not need to be obtained through page turning, and the convenience of selecting the pictures by the user is improved.

A method for generating an image editing dataset is provided. The method may include obtaining a candidate image for inputting an AI model from among at least one candidate image. The method may include determining an editing operation to be performed by the AI model on the candidate image from among at least one editing operation. The method may include determining a base prompt and a control prompt based on inputting of the candidate image and the editing operation into the AI model, wherein the base prompt comprises base instructions for detection of at least one object within the candidate image, and the control prompt comprises control instructions relevant to the editing operation to be performed by the AI model on the candidate image. The method may include generating the image editing dataset based on the base prompt and the control prompt.

The embodiment of the invention provides an image retrieval method and device, equipment, a storage medium and a program product. The method comprises the following steps: on the basis of a local feature vector and a global feature vector corresponding to a to-be-retrieved image, matching a plurality of candidate images from a preset retrieval library; calculating a local feature overlap ratio between the to-be-retrieved image and each candidate image based on the local feature vector corresponding to each candidate image and the local feature vector corresponding to the to-be-retrieved image; for each candidate image, based on the corresponding local feature overlap ratio, fusing the global feature vector corresponding to the candidate image and the global feature vector of the to-be-retrieved image to obtain a fused global feature vector corresponding to the candidate image; and calculating the similarity between the global feature vector corresponding to the to-be-retrieved image and the fused global feature vector corresponding to each candidate image, and determining a retrieval result of the to-be-retrieved image based on the similarity. According to the method, the retrieval accuracy can be improved.

A robustness verification device uses the similarity between feature amounts obtained by extracting a feature amount to identify a similar image having a predetermined rank of similarity with respect to an input image within a candidate image group; counts the rank of the similar image with respect to the input image in the candidate image group in a case where adversarial perturbation is applied to the image; calculates the rank of the similar image with respect to the input image in the candidate image group in a case where adversarial perturbation is not applied to the image; and a verifies whether or not the counted rank of the similar image counted is within a predetermined range that includes the calculated rank of the similar.

The invention provides a sample generation method and device, a deep learning model training method and device, a video stream processing method and device, equipment and a medium, and relates to the technical field of artificial intelligence, in particular to the technical fields of computer vision, deep learning, large models, image processing and the like. The method comprises the steps of processing M candidate image sequences based on a first mode and a second mode respectively to obtain respective first detection information and second detection information of each candidate image sequence, the first detection information representing whether the candidate image sequence comprises at least one of a target object and a motion state of the target object, and the second detection information representing whether the candidate image sequence comprises at least one of the target object and the motion state of the target object; the second detection information represents whether the candidate image sequence comprises at least one of the target object and the motion state of the target object; determining N target image sequences according to the first detection information and the second detection information of each candidate image sequence; and generating a significant sample set according to the N target image sequences and the first detection information and the second detection information of each target image sequence.

The invention provides a zero-sample multi-modal combined image retrieval method and device based on context awareness, and relates to the technical field of image retrieval, and the method comprises the following steps: obtaining a training sample set; the training sample set is input into a target CLIP model, target image embedding and target text embedding of the training sample set are obtained, and learnable language prompt embedding and visual prompt embedding are introduced into a text encoder and an image encoder of the target CLIP model respectively; training the mapping network model according to the target image embedding, the target text embedding and the target loss function to obtain a target mapping network model; obtaining a target image retrieval model according to the target CLIP model and the target mapping network model, and inputting the reference image and the text description into the target image retrieval model to obtain a target query feature; and outputting a retrieval result according to the similarity between the target query feature and the image embedding of each candidate image. According to the image retrieval method, a multi-modal combined image retrieval task can be completed.

The present invention discloses a mapping and positioning method based on multi-sensor fusion. This method uses three sensors, namely LocalSense, 2D lidar, and RGB camera. When constructing a map in real time, LocalSense is used to track the rough pose of the mobile robot; candidate areas are screened around the rough pose, and the pose is corrected through the result of 2D laser information matching; the cartographer algorithm is used to establish a two-dimensional grid map; and the RGB image of the current pose is segmented, and each piece of the image is compressed into a one-dimensional vector through a convolutional autoencoder and stored, corresponding to the current pose. In the positioning stage, candidate RGB images are screened by feature matching, and then candidate positioning areas are screened according to the poses corresponding to the candidate images. Laser information matching is performed in the candidate positioning areas to obtain the accurate pose. The method of the present invention does not depend on the content and features of the images, can be applied to different types of images and different application scenarios, and can better adapt to some scenarios with fewer corners and edges.

A method for (3D) object detection includes: receiving an input image with respect to a 3D space, an input point cloud with respect to the 3D space, and an input language with respect to a target object in the 3D space; using an encoding model to generate candidate image features of partial areas of the input image, a point cloud feature of the input point cloud, and a linguistic feature of the input language; selecting a target image feature corresponding to the linguistic feature from among the candidate image features based on similarity scores of similarities between the candidate image features and the linguistic feature; generating a decoding output by executing a multi-modal decoding model based on the target image feature and the point cloud feature; and detecting a 3D bounding box corresponding to the target object by executing an object detection model based on the decoding output.

The invention relates to the technical field of automobile part detection, in particular to an automobile part detection method, which comprises the following steps of: dividing a region range of a to-be-detected surface of an automobile part to obtain a plurality of image acquisition regions, and setting acquisition points for the plurality of image acquisition regions; acquiring basic feature data of a plurality of image acquisition areas, setting corresponding acquisition environment light sources, and performing image acquisition by an image acquisition camera corresponding to the acquisition environment light sources to obtain a plurality of detection images; performing feature analysis comprehensive processing on the plurality of detection images to obtain a defect area candidate image, and performing defect feature enhancement processing on the defect area candidate image to obtain a defect positioning image; and inputting the defect positioning image into the pre-training model, and outputting a detection result according to the defect parameters. According to the invention, the acquisition effect of the detection image is guaranteed, the visibility of the detection image is improved, and the accuracy of defect identification is guaranteed while the detection efficiency is improved.

The invention relates to the technical field of visual position recognition, and particularly discloses a multi-modal visual position recognition reordering method and system based on guidance, and the method comprises the steps: obtaining a query image, and retrieving a plurality of candidate images based on a pre-trained visual basic model and the query image; constructing a composite multi-modal prompt object, wherein the composite multi-modal prompt object comprises an image pair formed by the query image and the current candidate image, and an instruction text used for guiding a multi-modal large language model to perform visual comparison; outputting a structured similarity judgment result, wherein the result comprises a quantitative similarity score; and sorting based on the similarity scores corresponding to all the candidate images, and determining the candidate image with the highest score as an optimal matching result. Through combination of guiding type prompt engineering and structured output, an intermediate text generation link is avoided fundamentally, and the calculation efficiency is improved while the fidelity of all original visual information is reserved.

The invention discloses a task execution method and device, equipment and a medium, and relates to the technical field of artificial intelligence. The method comprises the steps that a task prompt word and an interface image are obtained, the task prompt word is used for describing a to-be-executed interaction task, and the interface image is divided into at least two image areas; acquiring image feature representations corresponding to the at least two image areas and text feature representations corresponding to the task cue words; identifying the relevance between the image feature representation and the text feature representation corresponding to the at least two image regions, and determining at least one candidate image region of which the relevance meets the requirement; and generating a task instruction based on the at least one candidate image area, wherein the task instruction is used for instructing execution of an interaction task on the first interface. In combination with the text feature representation corresponding to the task prompt word, redundant feature input is reduced, and the correlation between the candidate image area and the interaction task is improved, so that the accuracy of the task instruction and the execution efficiency of the interaction task are improved.