266 results about "Product image" patented technology

The invention relates to the field of industrial defect detection, in particular to an industrial defect detection self-supervised segmentation method for iterative pseudo label refinement, which comprises the following steps of: constructing a system comprising a strategy model, a refinement model, a reward model and a meta-learning training module; inputting the multi-modal data into the strategy model, and outputting a rough defect mask; based on the rough defect mask, prompting refinement is carried out through a refinement model, and a refinement mask is output; based on the refinement mask, the to-be-detected product image, the standard template image, the depth image and the infrared image, calculating a comprehensive quality score through a reward model, and screening high-quality samples with qualified scores; and updating a preset training data set based on the high-quality sample, performing supervised training on the strategy model by using the updated training data set, and repeating the steps to form an iterative loop. By constructing a self-supervised closed loop, a system can be driven to autonomously learn defect features from an unlabeled production line multi-modal image only by a small amount of initial reference data.

The invention relates to an industrial quality inspection data cooperative transmission method driven by image recognition, in particular to the field of artificial intelligence, which analyzes video frames in real time through a lightweight neural network, accurately locates and extracts key areas and features in industrial product images, constructs a dynamic semantic scene graph to understand contents and divide priorities, and improves the quality inspection efficiency. The core of the method is that non-uniform intelligent coding is carried out on video streams according to semantic importance of contents, high-fidelity transmission of key information such as defects is ensured, non-key areas are greatly compressed to save bandwidth, scheduling is carried out in the transmission process according to priorities, confidence feedback based on a cloud recognition result is introduced, closed-loop control is formed, and high-fidelity transmission is realized. According to the method, the front-end analysis model, the coding strategy and the network path are adaptively optimized, so that low-delay and high-reliability visual quality inspection data transmission and accurate identification can be continuously and stably realized in a complex industrial network environment, and the overall efficiency and the intelligent level of online quality inspection are remarkably improved.

The invention provides a defect detection method and device for improving a YOLO model based on an attention mechanism. The method provided by the invention comprises the following steps: acquiring image data of a to-be-detected industrial product; the improved YOLOv10 model performs feature extraction, multi-scale feature fusion and defect positioning identification on the industrial product image data to obtain a defect detection result of the industrial product; wherein the SE module is used for strengthening a channel dependency relationship and detail representation of local detail features, and the CBAM module is used for modeling channel attention and space attention on global semantic features; the bridging layer controls middle and low layer features and high layer features to keep uniform channel dimension and spatial resolution by adjusting the size and stride of a convolution kernel; introducing a 1 * 1 convolutional layer after Concat operation of multi-scale feature fusion, and performing channel compression, linear fusion and semantic alignment on multi-scale fusion features; and constructing a loss function fusing the GHM loss and the dynamic IoU loss.

The invention discloses an automatic defect detection method and system based on cross-resolution and multi-level knowledge distillation. The method comprises the following steps: step 1, acquiring an image of an intelligent manufacturing product to be detected; 2, constructing a teacher-student network distillation model for intelligent manufacturing defect detection; 3, training the teacher network model in the step 2 by using the data set obtained in the step 1; 4, training a student network model in the teacher-student network distillation model in a cross-resolution and multi-level manner; and step 5, utilizing the trained student network model to carry out defect detection on a to-be-detected intelligent manufacturing product image. A cross-resolution and multi-level knowledge distillation framework is provided, and a collaborative distillation mechanism of local details and a spatial structure is designed; a morphological self-adaptive semantic distillation method is provided, and the method can be better suitable for dynamic and real-time defect detection of intelligent manufacturing products.

The invention discloses a product image recognition method, device and equipment, a medium and a product, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining a first multispectral image of a first product object, and carrying out the dynamic noise suppression and feature enhancement processing of the first multispectral image, and obtaining a second multispectral image; inputting the second multispectral image into a pre-trained image recognition model to obtain a defect probability thermodynamic diagram corresponding to pixel points in the first multispectral image and uncertainty confidence of defect positioning; determining a defect type and defect position information corresponding to the first multispectral image according to the defect probability thermodynamic diagram and the uncertainty confidence; wherein the image recognition model comprises a global feature extraction module based on a Vision Transform architecture and a local feature extraction module based on a dilated convolutional neural network. According to the technical scheme provided by the embodiment of the invention, the identification capability of sub-pixel-level microdefects can be improved, and the accuracy of a product image identification result is improved.

The invention provides a defect detection method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a template image and a product image of a to-be-detected product; performing feature matching on the template image and the product image, and performing non-rigid alignment on the product image based on feature point pairs obtained by feature matching to obtain an alignment image of which the geometric view angle is consistent with that of the template image; and defect detection is carried out based on the aligned image and the template image to obtain a detection result, so that the problem of image deformation caused by the fact that the surface of the to-be-detected product is a curved surface and the shooting angle is not correct in a traditional scheme is effectively solved, and local and global deformation of the product image can be corrected, thereby greatly improving the alignment precision of defect detection, and improving the detection accuracy of the to-be-detected product. Therefore, the real difference of the product surface can be more focused during defect detection, non-defect geometric deformation interference is effectively shielded, and the accuracy and robustness of defect detection of complex surfaces of household electrical appliances and the like are greatly improved.

Methods, computer systems, computer storage media, and graphical user interfaces are provided for facilitating generation of brand-aligned product images. In one implementation, a product-environment prompt including a text description of a reference image is obtained. Further, a set of image features extracted from the reference image is obtained. Thereafter, a brand-aligned product image is generated by performing outpainting from a product representation in accordance with the product-environment prompt and the set of image features extracted from the reference image. The brand-aligned product image can then be provided for display via a graphical user interface.

The invention provides an e-commerce commodity graph differentiation generation method and device, equipment and a medium, and the method comprises the steps: obtaining an original to-be-shelved commodity graph of e-commerce, and finally obtaining the disassembly information; the method comprises the following steps: adjusting an original commodity image into 980 * 980 pixels, performing Gaussian blur processing at the same time, and then converting image pixel data into a Base64 coded character string; constructing a composite cue word according to the image data, the disassembly information and a format required by generation; inputting the composite cue word into a model to generate a new scene description; inputting the new scene description, the original commodity graph and the synthesis requirement into a model to generate a synthesis cue word, and then inputting the original commodity and the synthesis cue word into an image-text model to generate a differentiated commodity graph; setting the output resolution, the number of iterations and the learning rate of the SuperIR model, and inputting the differentiated commodity graph into the SuperIR model to generate a differentiated commodity optimization graph; the defects of low efficiency, poor scene adaptation, main body distortion and difficulty in repeated avoidance of a traditional method are overcome.

A classification model generating system includes: an obtainer that obtains a good product image group including a plurality of good product images; a defective portion image generator that generates a plurality of defective portion images, based on a seed image group obtained by geometrically transforming a seed image simulating a defective portion, the seed image being an artificially drawn image; a combination processing unit that combines each of the plurality of defective portion images with the good product image group to generate a defective product image group; and a classification model generator that performs classification training by using a part of the good product image group and a part of the defective product image group as a training image group to generate a classification model.

The application discloses a one-way valve component missing loading detection system and method, and relates to the technical field of visual detection.The system comprises an image acquisition and positioning end, an angle and posture correction end and a hierarchical comparison and detection end.The image acquisition and positioning end acquires a component area image of a to-be-detected one-way valve, and preliminarily positions the image with a pre-stored qualified product image.The angle and posture correction end corrects the to-be-detected image to the same detection posture as the template image by calculating the rotational deviation of the to-be-detected image and the template image and the affine transformation technology.The hierarchical comparison and detection end divides the corrected image into a plurality of component corresponding areas, calculates the feature difference of each area, and judges whether there is missing loading.The application can eliminate light interference through image quality adaptive enhancement, overcome position deviation through posture correction, realize multi-component parallel detection through hierarchical comparison and detection, improve the precision and robustness of missing loading detection, and support online switching detection of different models of products.

The invention relates to a small sample defect detection method based on class awareness prototype migration, and the method comprises the steps: inputting a labeled image into a student network comprising a first CAM module, and outputting a corresponding first prediction result and first features corresponding to various truth value labels extracted by the first CAM module; calculating a first loss function based on the first prediction result to preliminarily optimize the student network, and updating prototypes of various truth value labels based on the first feature; inputting the label-free image into a teacher network comprising a second CAM module, calculating a comprehensive credibility score of a pseudo label based on a second feature output by the teacher network, a second prediction result and the updated prototype, and calculating a second loss function based on the comprehensive credibility score to secondarily optimize the student network; network parameters of the teacher network are updated based on the network parameters of the student network after secondary optimization; and the industrial product image to be detected is subjected to iterative optimization through the teacher network to obtain a final defect detection result.

1. Name of the product in this design: Graphical User Interface for Mobile Phones (Shopping "Gouwu" app). 2. Purpose of this design: This design is used for human-computer interaction to facilitate the purchase of goods. 3. The key design features of this product are its graphical user interface content. 4. The image or photograph that best illustrates the design's key points: the front view. 5. Due to the lack of design considerations, the rear view, left view, right view, top view, and bottom view are omitted. 6. Purpose of the graphical user interface: This design allows users to access the shopping system's purchase interface through operations on the interface, thus enabling the purchase of goods. 7. Description of the changing states of the graphical user interface: The main view and interface state change diagrams 1-3 show the content presented to the user during use; the main view displays the splash screen animation interface when entering the system. After clicking "Click to enter the world of 'Gouwu'" in the main view, the user enters the interface state change diagram 1. In the interface state change diagram 1, the user enters the mobile phone number and clicks "Send Verification Code". After filling in the verification code, the user clicks "OK" and enters the interface state change diagram 2. In the interface state change diagram 2, the user clicks "Buy" and enters the interface state change diagram 3. The logical changes between the above multiple interfaces are in a single direction. In the interface state change diagrams 2 and 3, the green blocks cover the content screen that is not part of the graphical user interface design itself, and is used to display product images. To enter interface change state diagram 3, you must pass through interface change state diagram 2; to enter interface change state diagram 2, you must pass through interface change state diagram 1; and to enter interface change state diagram 1, you must pass through the main view interface.

The application relates to the technical field of image data processing, in particular to a blister product defect intelligent detection method and system based on industrial vision. The method comprises the following steps: acquiring product images of to-be-detected blister products and calibration plate images of calibration plates respectively photographed on multiple detection stations; determining a relative offset amount of a finished product image relative to a finished calibration plate image based on the positions of physical marking points in the product images and the calibration plate images photographed on the multiple detection stations; evaluating a perspective distortion degree of the finished product image based on the morphological feature difference between a blister product region in the finished product image and a calibration plate region in the finished calibration plate image; and performing perspective transformation calibration on the finished product image based on the relative offset amount and the perspective distortion degree, wherein the calibrated finished product image is used for intelligent defect detection. The method can eliminate the influence of distortion on defect detection and improve the accuracy of defect detection.

The invention relates to a main and auxiliary screen control method for main and auxiliary screen linkage, computing equipment and a main and auxiliary screen linkage system.The main screen control method comprises the steps that in response to starting of a target program of a main screen, a program interface corresponding to the target program is displayed, and the auxiliary screen is controlled to execute operation of displaying a product image interface; the product image interface is used for displaying a product image obtained in the process of manufacturing the target product based on the target semiconductor technological process; and in response to determining that the display state of the product image interface in the secondary screen is abnormal display and the display state of the program interface is normal display, performing normal recovery adjustment aiming at the product image interface. The method and the device are used for solving the technical problem that the detection efficiency is reduced due to the fact that the production or detection process is forced to be interrupted and waiting for manual intervention because the image observation window of the secondary screen is abnormal.

The invention provides a dispensing method, electronic equipment and a storage medium. The dispensing method comprises the following steps: performing feature matching on an image of a product to be dispensed and a template image to obtain a plurality of matching point pairs; based on the multiple matching point pairs, the pose deviation of the product to be subjected to glue dispensing relative to the standard pose is solved through a rigid body transformation model; constructing a transformation matrix by using the pose deviation, and performing coordinate transformation on each path point in the preset dispensing path by using the transformation matrix to generate a corrected dispensing path; and controlling a dispensing execution mechanism to perform dispensing operation according to the corrected dispensing path. According to the method, the dispensing precision and the process stability can be improved.

An electronic device according to an embodiment of the present invention may include a processor that constructs an image generation model by applying a feature model generated based on a normal product test image and a defective product test image to a base model trained to generate a defective product image, the processor inputs defect information formed in at least one form of a text and an image to the image generation model, and the processor acquires a defective product image including a defect corresponding to the defect information by using the image generation model.

The application discloses an AI product image enhancement method based on fusion segmentation, and belongs to the technical field of image enhancement, comprising the following steps: constructing a data set of input images of products by using real shooting synthesis data; encoding the input images into latent noise vectors and illumination condition vectors; performing image feature extraction on the input images to obtain image feature vectors; simultaneously extracting panoramic illumination feature vectors; performing multi-modal attention fusion on the image feature vectors and the panoramic illumination feature vectors to obtain fused feature vectors; inputting the fused feature vectors into a segmentation layer to generate product masks and process background modes; and inputting the feature vectors with the masks into a generation layer to obtain generated images according to a speed predictor iteration refinement. The application realizes accurate matching of materials and illumination, restores the real texture of products, and enhances visual attraction by using the above method.

1. Name of the product in this design: Formal Dress (5L7166-1). 2. Intended use of this design: for wearing. 3. The key design feature of this product is its shape. 4. The image or photograph that best illustrates the design's key points: a 3D model. 5. The top surface of this design product is a part that is not easily seen or cannot be seen during use, so the top view is omitted; the bottom surface of this design product is a part that is not easily seen or cannot be seen during use, so the bottom view is omitted. 6. Other matters requiring explanation: The human model in the product images of this design is a supporting element and is not part of the design content for which protection is sought.

This invention provides a method, apparatus, device, and medium for generating product scene images based on image retrieval. The method includes: obtaining a style model and an image retrieval library based on a training set; receiving original product images uploaded by users to obtain user-uploaded product images with a white background; obtaining original scene prompt words associated with the target matching result; determining whether the retrieval result meets preset matching conditions; inputting the user-uploaded product images with a white background and the final prompt words into a generation network loaded with the style model to generate and output the final product display image with background, thereby improving the generation quality and accuracy and enhancing the usability of the generated image.

A system is provided.SOLUTION: A system comprising: means for inputting and transmitting body shape information of a user; server means for receiving and storing the body shape information in a database; means for generating a model of a product image based on the stored body shape information; and means for transmitting the generated model image to a terminal of the user.SELECTED DRAWING: Figure 1

The present application relates to a system and method whereby an enterprise system can identify items within an image and recommend similar enterprise products to a consumer. In some examples, such a system may include an object detection component trained on a known dataset, which may be used to identify objects / items in an image. Cropped items of interest may be extracted from the image. The system may include an image embedding and category prediction component, which may generate embedding vectors and predict category information for the cropped images. A visual similarity search component, which may reference embeddings for product images to identify products which are visually similar to the items identified from the image. An attribute-based ranking process may be applied to rank the top recommended items obtained from the visual similarity search component. A final ordering of recommended items may then be provided to a consumer.

The invention relates to an industrial visual defect detection system and method, and the system comprises an image collection module which is used for collecting industrial product images on a production line in real time; the detection reasoning module is used for performing multi-dimensional defect detection through a trained defect detection model according to the industrial product image to obtain multi-dimensional defect information of the industrial product, and the multi-dimensional defect information comprises defect type information, defect position information and defect shape information; and the output control module is used for executing different levels of early warning processing according to the multi-dimensional defect information and generating a corresponding quality inspection report. By means of the method and device, multi-dimensional defect detection of the industrial products on the assembly line is achieved, automatic recognition of various complex defects is completed based on the types, positions and shapes of the defects, early warning processing of different levels is executed, the refinement level of quality control is improved, and the problem of how to improve the defect detection reliability of the industrial products is solved.

Methods and systems are disclosed for generating an extended reality (XR) try-on experience based on an image produced by a diffusion model. The system receives an image depicting a real-world object and generates a prompt comprising a textual description of a fashion item. The system analyzes the image and the textual description of the fashion item using a generative machine learning model to generate an artificial image that depicts an artificial object that resembles the real-world object wearing an artificial fashion item matching the textual description of the fashion item. The system identifies an object comprising a real-world product image that matches visual attributes of the artificial fashion item and replaces the artificial fashion item in the artificial image with the object to generate an output image.

This invention provides a furniture product recognition method and system based on multimodal heterogeneous data processing, relating to the field of image recognition technology. The method includes: acquiring a target product image; preprocessing the target product image to obtain global text features, target text features, and visual features; processing the global text features, target text features, and visual features using a modal fusion model to obtain a classification result for the target product image; searching for a candidate image set in a pre-constructed database based on the classification result of the target product image; inputting the target product image and the candidate image set into a knowledge graph model, and using the knowledge graph model to perform entity alignment between the target product image and the candidate images; and outputting the recognition result for the product to be identified in the target product image based on the entity alignment result. This application can achieve accurate identification and rapid matching of furniture products.

A system is provided.SOLUTION: A system comprising: means for a user to upload an item image and an item description; means for a server to receive and store the item image and the item description; means for the server to analyze the item image and the item description using a generative AI model to generate an assessment result; and means for the server to provide the assessment result to the user.SELECTED DRAWING: Figure 1

The invention relates to the technical field of visual identification, in particular to a part product flexibility collinear diagnosis method and system, and the method comprises the steps: collecting an image; identifying a product; correcting the attitude; inspecting product characteristics; and manually rechecking and outputting an inspection result. According to the scheme, multiple types of products can be inspected, the images can be further processed, the products in the images can meet the requirement for feature inspection, then feature inspection is conducted on the products in the images, the method adopts processing of the product images to replace the requirement for the posture and position of the products, and the accuracy of feature inspection is improved. And therefore, the degree of freedom is higher when the products are placed, and the products can be placed more easily.

The application relates to the technical field of industrial visual inspection and edge computing, and provides a distributed storage and hardware cooperative deployment method and system of an industrial quality inspection multi-task visual model, which comprises the following steps: constructing a multi-task visual model; storing model weight files of a shared backbone network in a first storage partition in a read-only mode; storing task head weight files of a plurality of lightweight task heads in the form of independent files in a second storage partition in a distributed mode; determining a target task identifier corresponding to a current product to be inspected; loading the shared backbone network and a task head corresponding to the target task identifier by edge computing hardware; inputting a product image into the shared backbone network and extracting general visual features of the product image; inputting the general visual features into the task head corresponding to the target task identifier in parallel to obtain classification results of the task heads; and determining the quality state of the current product to be inspected according to the classification results. Thus, systematic cost reduction is achieved.

The application discloses a visual-based contact type measuring robot adaptive measuring point relative guiding method, which comprises a 2D visual-based product image contour matching method, realizes 4-degree-of-freedom measuring point guiding of plane translation, rotation and height scaling, and is suitable for contact type six-axis cooperative measuring robot measuring point guiding measurement of products with local features; the method comprises the following steps: teaching contact type six-axis cooperative measuring robot measuring points of a product to be measured; determining a teaching 2D measuring pose, obtaining a product multi-scale contour image template under the teaching 2D measuring pose and a multi-scale image of the current product, and performing contour matching of corresponding scale levels; based on the contour matching result, calculating a first pose deviation under the teaching 2D measuring pose, transforming the first pose deviation to all teaching point positions of the teaching 2D, performing pose transformation, performing iterative updating on all measuring point positions, and obtaining all updated first measuring point poses. The application better realizes adaptive contact type measuring point measurement of products.

The invention discloses a plastic product surface defect intelligent detection method based on computer vision, and relates to the technical field of defect intelligent detection, and the method comprises the following steps: synchronously obtaining three plastic product images with different exposures in the same scene, extracting pixel values, calculating the basic credibility weight of the pixel values, and calculating the credibility weight of the plastic product; a saturation punishment mechanism is introduced to correct the basic credibility weights of the pixel values to obtain a second credibility weight of each pixel value, and the three plastic product images are fused to generate an extended dynamic range image; extracting brightness values from the extended dynamic range image, marking candidate seeds according to the brightness values, performing condition screening on each candidate seed to obtain seeds, arranging the seeds in a descending order according to the brightness values to establish a seed point queue, performing coarse growth and fine growth in sequence, and determining a highlight area; the confidence score of each highlight area is analyzed, corresponding defect detection operation is carried out, and after detection results of all the highlight areas are fused, a complete defect marking graph is generated.

The invention discloses an abnormal perception-based industrial defect grouping refining method. The method comprises the following steps of: taking a batch of industrial product images as input, dividing the batch of images into a plurality of groups with a fixed image quantity, and carrying out preliminary detection on the current group of images through an existing zero sample defect detection algorithm; screening pseudo normal samples and synthesizing pseudo abnormal samples and masks thereof; training a defect refining decoding network provided by the invention by using the synthesized pseudo abnormal sample and the mask thereof; after preliminary detection is carried out on the images in the next group, a refined output result is obtained by using the trained defect refining decoding network, and a pseudo-abnormal sample is synthesized based on the result to be used for continuously training the defect refining decoding network; and repeating the operation in the previous step until all the images are detected. According to the invention, the accuracy of a zero sample defect detection algorithm in defect positioning can be improved to a great extent. The invention further provides a corresponding industrial defect grouping and refining device based on anomaly perception.