297 results about "Intermediate image" patented technology

The invention discloses a machine vision-based automatic identification and calculation method for pore change of a rock section under an SEM (scanning electron microscope), and the method comprises the steps: obtaining original image data of the rock section through a scanning electron microscope, transmitting the original image data to a machine vision SEM pore intelligent quantification cloud platform, and dividing the original image data into a plurality of image sub-regions which are not overlapped with each other; calling a special threshold inversion model to analyze the gray information of each sub-region, and determining a dual-phase pore segmentation threshold; carrying out edge extraction on the sub-regions after threshold processing by adopting a pore edge gradient enhancement algorithm; performing topological structure analysis on the intermediate image data by using a biphase pore topological inversion model, constructing and optimizing a pore space topological network, and eliminating false pore structure data; and according to the optimized pore space topology network, integrating calculation results of all the sub-regions to obtain a pore change identification calculation result of the whole rock section. The method improves the pore identification precision and processing efficiency, and adapts to different types of rock samples.

The invention discloses a lightweight high-speed photographing system and method based on an event camera, and the system comprises a hybrid imaging module which is used for synchronously collecting the image data and event data of a scene, and comprises a visible light camera and an event camera; the hardware synchronization module is electrically connected with the visible light camera and the event camera, and is used for receiving a frame synchronization signal of the visible light camera and sending a trigger signal to the event camera, so that the event camera inserts a time stamp corresponding to a visible light image frame exposure moment in an event stream; the data processing and fusion unit is in communication connection with the hybrid imaging module and is used for receiving the image data and the event data and executing the following steps: performing time alignment on an event stream and a visible light image frame stream based on a time stamp; operating a high-frequency frame insertion reconstruction algorithm, and reconstructing a multi-frame intermediate image by using event stream data between two frames; synthesizing the original visible light image frame and the reconstructed intermediate image into a high-frame-rate video stream; and the power supply and interface module is used for supplying power to the system and outputting a high-frame-rate video stream.

The invention relates to the technical field of semantic segmentation, in particular to a night semantic segmentation method based on low illumination enhancement and edge optimization, and the method comprises the steps: inputting an image into a low-light enhancement repair network based on the Retinex theory, local contrast enhancement and adaptive feature fusion, obtaining a denoised and enhanced intermediate image, and carrying out the edge optimization of the intermediate image; inputting the intermediate image into a semantic segmentation network to obtain a category distribution diagram of each pixel; inputting a discriminator embedded with a channel attention module according to the category distribution diagram of each pixel, and optimizing a generator composed of a low light enhancement repair network and a semantic segmentation network through a multi-task joint optimization loss function; and inputting a night image to be detected and segmented into the optimized generator, and outputting a segmentation result. By adopting the method, the low-light enhancement repair network is combined with a local contrast enhancement and channel feature fusion mechanism, so that the overall brightness of the image is improved, the details and edge structures of the image are reserved, and the perception capability and robustness of the model are improved.

An optical system is configured to guide light from a light source to a first target and a second target, and to guide first reflected light from the first target and second reflected light from the second target to an image sensor. The optical system includes a first optical element having a conical surface. An intermediate image is formed between the first optical element and each of the first target and the second target. The first reflected light and the second reflected light are guided to an imaging plane of the image sensor such that at least part of the first reflected light and part of the second reflected light overlap each other on the imaging plane.

An optical assembly (17) has a primary optical rod (28) for guiding illumination light (4) along an illumination light beam path from a light source to a target field. An optical follower rod (29) of the optical assembly (17) is arranged in the illumination light beam path between the primary rod (28) and the target field. An intermediate image-generating optic (30) is arranged between the primary rod (28) and the follower rod (29). This generates an intermediate image between the primary rod (28) and the follower rod (29) along exactly one imaging dimension (y) perpendicular to the illumination light beam path. The result is an optical assembly in which the illumination light, after passing through the optical assembly, has a defined polarization state. The optical assembly can be part of an illumination optic and can, in particular, be part of an optical system of a projection exposure system containing such an illumination optic.

System and method for Generative Artificial Intelligence based image generation are disclosed. In an aspect, a user input is received requesting an image of a predetermined object to be generated. An enhanced prompt is then generated from the user input, the enhanced prompt includes one or more of a reference image of the predetermined object and a textual portion describing technical details for generating the image. Further, the enhanced prompt is input to an image generation model that preserves portions of the reference image including the predetermined object unchanged via masking the portions of the reference image, generates a depth map of the reference image, and generates an intermediate image by merging the reference image with the masked portions of the predetermined object and the depth map. Also, an enhanced version of the intermediate image is output as the image generated in response to the user input.

A computer-implemented method for determining an estimation of a two-dimensional scattered radiation distribution in a two-dimensional X-ray image that is based on an imaging procedure performed by an X-ray detector is provided. The method includes receiving the X-ray image, applying an estimation algorithm to the X-ray image or to a two-dimensional intermediate image determined from the X-ray image. The estimation algorithm determines a respective two-dimensional partial scattered radiation image for a plurality of physical scatter processes such that image values of pixels of the respective partial scattered radiation image in each case describe an estimated value for a respective scattered radiation dose that was applied to a respective detector region of the X-ray detector assigned to the respective pixel during acquisition of the X-ray image by the respective physical scatter process. The estimation of the scattered radiation distribution is determined based on the plurality of partial scattered radiation images.

The invention discloses an optical cement surface defect detection method and system based on visual inspection. The method comprises the steps that an original image set is acquired and preprocessed to obtain an intermediate image sequence; performing dynamic noise estimation on the intermediate image sequence, and selecting a high-frequency noise region in the intermediate image sequence for optimization to obtain a stable image sequence; defect detection is carried out on the stable image sequence, edge texture enhancement processing is carried out after preliminary defect area distribution is obtained, coordinates are mapped, and defect positioning coordinates are obtained; performing real defect judgment on the image according to the defect positioning coordinates to obtain a defect aggregation degree, dividing a high-density aggregation region, extracting the defect size of the high-density aggregation region, and generating a defect distribution mapping graph in combination with the defect aggregation degree; and carrying out image superposition according to the mapping graph and the original image set, and carrying out risk assessment on the high-density defect area according to the superposed image to obtain a final defect detection report. According to the method, accurate detection and analysis can be ensured when tiny defects are processed.

The invention belongs to the technical field of computer vision and digital media, and discloses a dome-screen film production method and system based on a high-altitude aerial image. Projecting the multi-view aerial video sequence to a unified spherical coordinate system, and constructing a panoramic space with an overlapping region; performing non-rigid geometric deformation processing on the overlapping region of the panoramic space to generate an intermediate image with aligned pixels; calculating the fusion cost of each pixel point in the overlapping region in the intermediate image, and performing path search based on the fusion cost of each pixel point to generate a dynamic suture line; generating an image mask based on the dynamic suture line, performing multi-scale frequency domain adaptive fusion on the intermediate image by using the image mask to generate a fused panoramic image, and remapping the fused panoramic image into a projection format adaptive to dome screen playing to output; and the real reducibility and the manufacturing quality of the dome film are greatly improved.

An eyewear display device implements a curved lightguide to form an intermediate image between an incoupler and an outcoupler. The curved lightguide produces an approximately 20-degree diagonal field of view (FOV) full-color display having approximately 10% red, blue, green efficiency with uniform color and luminance for a micro-display at a temple of an eyewear display device. The curved lightguide incorporates freeform mirror incouplers and / or outcouplers and color-corrected relay optics. Using an intermediate image stage in the optical pathway allows for use of a compact incoupler that does not exceed a thickness of the curved lightguide.

The invention discloses a large-view-field space detection optical lens. The large-view-field space detection optical lens comprises a front lens group, a diaphragm, a rear lens group and an image surface which are sequentially arranged from an object side surface to an image side surface, the front lens group filters incident light waves and forms an intermediate image; the diaphragm controls the luminous flux by adjusting the relative aperture of the diaphragm; the rear lens group amplifies the intermediate image, corrects aberration and performs optical imaging through an image surface; the front lens group comprises a first optical filter, a first negative meniscus lens, a first biconvex lens and a second biconvex lens which are sequentially arranged from a position far away from the diaphragm to a position close to the diaphragm; and the rear lens group comprises a first biconcave lens, a third biconvex lens, a fourth biconvex lens, a first positive meniscus lens, a second negative meniscus lens, a second positive meniscus lens, a third negative meniscus lens and a second optical filter which are sequentially arranged from the diaphragm to the image plane. The optical lens provided by the invention has the characteristics of circular light spot, large view field, low chromatic aberration, low distortion, high resolution and small form deviation of light spots in each view field.

This invention provides a method for detecting adhesive buildup on the surface of a running track, comprising: acquiring an HSV image corresponding to the running track to be tested; performing a first image processing on the HSV image to obtain a first intermediate image identifying a feature region; the feature region consisting of undifferentiated adhesive buildup areas and interference areas; performing a second image processing on the HSV image to obtain a second intermediate image identifying interference areas; and determining the target information of the adhesive buildup on the surface of the running track to be tested based on the first and second intermediate images. A running track surface adhesive buildup detection device, a computer device, and a storage medium are also provided. Using the HSV color model, and through relevant image processing on the HSV image, the target information of the adhesive buildup on the surface of the running track to be tested is finally determined; compared with manual visual inspection, this method can greatly improve detection efficiency, reduce labor costs, and accurately determine the target information such as the area of ​​the adhesive buildup on the surface.

The embodiment of the invention provides an image processing method and device, equipment and a storage medium. The method relates to the field of digital image processing, and comprises the following steps: carrying out first zero setting judgment on non-zero pixels in an input image; and if the judgment result indicates that at least one first non-zero pixel needs to be zeroed, performing zeroing processing on the first non-zero pixel to obtain an intermediate image. Carrying out second zero setting judgment on non-zero pixels in the intermediate image; and if the judgment result indicates that at least one second non-zero pixel needs to be zeroed, performing zeroing processing on the second non-zero pixel to obtain a target image. And taking the target image as a new round of input image, repeating the process until all non-zero pixels do not need to be zeroed, and finally outputting the target image. According to the method, zero setting judgment is only carried out on the non-zero pixels in the image, repeated judgment on the pixels which are already zero and do not need to be processed is avoided, and therefore the image processing efficiency is remarkably improved.

The application relates to the technical field of carton detection, and provides a method and system for detecting express cartons based on an Internet of Things platform and electronic equipment. The method for detecting express cartons based on the Internet of Things platform comprises the following steps: acquiring a carton image; obtaining a rotation angle of the carton relative to a preset direction according to contour coordinate information of the carton in the carton image; processing the carton image by using an affine transformation algorithm according to the rotation angle to obtain an intermediate image containing a carton conforming to the preset direction; obtaining length data of each part of the carton according to the intermediate image; and judging whether the length data satisfies a preset length relationship and / or whether area data based on the length data satisfies a preset area relationship to obtain a detection result of the carton. The embodiment provided by the application has high matching degree for the feature detection result of a target model and obvious effect.

The invention relates to the technical field of photographing, and discloses a photographing preview method, electronic equipment, a readable storage medium and a program product. The method comprises the following steps: in the process of displaying a preview image, when the electronic equipment detects that the current zoom ratio is relatively high, performing preprocessing (including preliminary cutting) on an original image acquired by a camera, and outputting intermediate images with different cutting sizes corresponding to the requirements of different image processing algorithms in a customized manner. And then, processing the intermediate images based on image processing algorithms corresponding to the intermediate images of the cutting sizes to obtain a plurality of image processing results. And obtaining a preview image corresponding to the current zoom ratio based on a plurality of image processing results, and displaying the preview image in the shot preview image. Therefore, the requirements of different image processing algorithms on the image can be considered in the photographing preview process, and the processing effects of the different image processing algorithms can be ensured, so that the quality of the preview image can be ensured.

This invention relates to a high numerical aperture, large field-of-view total internal reflection microscope objective in the ultraviolet wide band. Along the optical axis from image to object, it comprises: a first objective consisting of a first and a second reflecting mirror arranged sequentially along the optical axis OA from image to object, used to receive parallel light and form an intermediate image plane; and a second objective consisting of a third and a fourth reflecting mirror arranged sequentially along the optical axis OA from image to object, used to image the intermediate image plane onto the object plane. The first reflecting mirror is a convex mirror with negative optical power; the second reflecting mirror is a concave mirror with positive optical power; the third reflecting mirror is a concave mirror with positive optical power; and the fourth reflecting mirror is a concave mirror with positive optical power. This invention employs a four-mirror total internal reflection structure, and by optimizing the focal length and aperture ratios of each reflecting mirror, it achieves a balance between high numerical aperture and a large field of view, overcoming the limitations of traditional total internal reflection objectives where numerical aperture and field of view are mutually constrained.

Provided is a display image generation apparatus including a captured image acquisition section that acquires data of an image captured by a camera, an object arrangement section that arranges a virtual object to be operated by a user in a virtual three-dimensional space, a display image generation section that generates a display image by drawing an image of the virtual object and synthesizing the image of the virtual object with the captured image, and an output section that outputs data of the display image. The display image generation section switches whether or not to use an intermediate image representing the image of the virtual object from a viewpoint of the camera when drawing the image of the virtual object, according to a state of a display world including the virtual object.

The application discloses a kind of human eye test conoscopy, comprising: front light barrier, the light aperture of the light barrier is not more than 4mm;Imaging lens group, close to object plane, with positive focal length, can be not less than 120 degree field of view incident light rays converge into, and form approximate telecentric intermediate image in rear;Relay lens group, close to image plane, with less than 1 magnification, for imaging intermediate image at image plane;The focal length GF1 of the imaging lens group is less than 16.2mm, the focal length GF1 of imaging lens group and the focal length GF2 of relay lens group satisfy the following relationship: 0.2<GF1 / GF2<0.3;Imaging circle diameter is D1, and the diameter of diaphragm is D2, satisfy: 8<D1 / D2<10.The above-mentioned human eye test conoscopy can realize similar test effect with direct observation of human eye, suitable for testing near-eye display device, and can meet the imaging effect of large image plane, high resolution and no vignetting.

The embodiment of the invention discloses a shooting method and device. After a shooting instruction is received, an image acquisition module is started to acquire continuous initial images, for the photo shooting instruction, a first initial image without exposure extreme is selected from the initial images as a reference image, and a preset number of adjacent initial images are selected after the reference image as target images; and obtaining a first intermediate image after noise reduction according to the target image and the reference image, determining a target brightness value, and performing brightness processing on the first intermediate image according to the target brightness value to obtain a shot photo. Therefore, the first initial image without exposure extreme is used as the reference image to carry out photo shooting processing, so that photo shooting does not need to wait for automatic exposure convergence, shooting delay is reduced, shooting timeliness is improved, meanwhile, additional devices do not need to be added, and equipment cost is reduced.

This application discloses a method, device, and storage medium for generating multi-action human body videos, comprising: inputting reference images and multi-action prompt text into a multimodal large language model; performing inference on the multi-action prompt text through the multimodal large language model to output a fine-grained action description sequence; generating intermediate image frames based on the reference images and the fine-grained action description sequence; extracting human skeleton key point data from the intermediate image frames to obtain a visual skeleton sequence; inputting the visual skeleton sequence, multi-action text prompts, and reference images into a video diffusion model; and integrating the visual skeleton sequence into the denoising generation process of the video diffusion model through a pose guidance adapter to output a coherent human body video matching the multi-action text prompts. This application achieves accurate, coherent, and efficient generation of complex multi-action videos by decoupling and fusing the chained inference of the multimodal language model, the intermediate representation of the skeleton sequence, and the pose guidance of the video diffusion model.

The invention relates to the technical field of data processing, in particular to a holographic projection method based on a compound eye array camera, and the method comprises the steps: collecting a multi-sub-eye image sequence of a target scene, and determining a mapping parameter set containing information entropy distribution and an error sensitive region; comparing the error sensitive area of any sub-eye among the multiple image frames, and determining the distortion area of each sub-eye; carrying out calibration processing on the distortion area of each sub-eye, and configuring an intermediate image block after preliminary calibration; configuring a view field range of a single sub-eye based on the coordinate of the middle image block, and calculating an overlapping region of adjacent sub-eyes through a coordinate overlapping algorithm; calling an optimal boundary line of each boundary point based on the update data of each boundary point in the overlapped region in the multi-image frame; and based on the optimal boundary line, carrying out pixel-by-pixel splicing fusion on the view field range of the single sub-eye to obtain a fused global view field set. And the precision and efficiency of image processing are realized.

An apparatus for generating a defined laser illumination (12) on a work plane (14) has a laser light source (20) that generates a raw laser beam (22). An optical assembly (24) receives the raw laser beam (22) and shapes it into an illumination beam (26) along an optical axis (40). The illumination beam (26) defines a beam direction (28) that intersects the work plane (14). The illumination beam (26) has a beam profile (42; 42') in a region of the work plane (14) that has a major axis (44) perpendicular to the beam direction (28) with a major axis beam width and a minor axis (46) with a minor axis beam width. The optical assembly (24) includes a beam transformer (30) with an exit aperture, a first set of optical elements (56, 60, 62, 64) for beam shaping on the major axis, and a second set of optical elements (34, 36, 38) for beam shaping on the minor axis. The beam transformer (30) widens the raw laser beam (22) on the major axis to generate a widened raw laser beam. The first set of optical elements (56, 60, 62, 64) includes a homogenizer (56) that homogenizes the widened raw laser beam on the major axis. The second set of optical elements (34, 36, 38) includes at least one lens (38) that images the exit aperture of the beam transformer (30) into the work plane. The first set of optical elements (56, 60, 62, 64) generates an intermediate image (66) behind the homogenizer (56) and further implements an imaging optical unit that images the intermediate image (66) into the work plane (14).

The invention discloses a face deep counterfeiting-oriented robust countermeasure watermark generation method, device and equipment, and relates to the technical field of deep counterfeiting defense. The method comprises the following steps: inputting an original face image into a forward diffusion process of a denoising diffusion implicit model to obtain a noise potential variable, and executing a reverse denoising process; in each denoising step, a current intermediate image is generated, then the loss of counterfeit output is calculated, gradient information is obtained according to the loss and added to the current intermediate image, and an intermediate noise image is obtained. And a denoising step is executed until a complete denoising process is completed, an intermediate step adversarial image is obtained, a post-processing image is obtained through noise layer processing, then mean square error loss of counterfeit output and structural consistency loss of the intermediate step adversarial image and the original face image are calculated, and a total loss function is obtained through weighted fusion. And based on the total loss function, gradient information is obtained and added to the noise potential variable, and a new round of denoising is carried out. And finally, outputting an intermediate step confrontation image of the last iteration.

The present disclosure relates to systems and methods for image processing. The systems may acquire imaging data. The systems may generate multiple intermediate images based on the imaging data by performing image optimization of multiple image optimization dimensions using a plurality of trained machine learning models. The systems may generate a target image based on the intermediate images.

Embodiments of this disclosure provide methods, apparatus, and computer program products for processing images. The method includes: performing style transfer on an original image based on a background image to obtain an intermediate image, wherein the intermediate image contains the same content as the original image and has the same style as the background image; cropping an object image from the intermediate image, wherein the object image contains a target object; and compositing the object image with the background image to obtain a composite image.

A microscope having a light source, an illumination beam path for guiding excitation light into a sample region, a detector for detecting emission light emitted by a sample, a detection beam path having a microscope objective for guiding the emission light onto the detector, and a control unit for evaluating emission light detected by the detector. The illumination beam path, for providing an intermediate image plane, comprises at least one relay optical unit, the illumination beam path comprises a multimode light guide, and an output end of the multimode light guide is arranged in the intermediate image plane or in the vicinity of the intermediate image plane.

The invention provides an image processing method and a computer program product, and relates to the technical field of image processing, and the method comprises the steps: obtaining a target image and a plurality of reference images, each reference image comprising a to-be-inserted main body object; performing main body identification on each reference image to determine a main body object, and generating a separated main body image corresponding to the main body object; the separated main body images are fused into a single embedded image through a reference image embedding mechanism, the single embedded image is embedded into the target image to obtain a composite image, and the reference image embedding mechanism is used for avoiding semantic conflicts caused by multi-image input; adjusting details of the composite image to obtain an intermediate image, and optimizing the intermediate image according to a trained fine adjustment model until a loss function meets an iteration end condition to obtain a generated image; the loss function comprises semantic similarity used for restraining consistency of the reference image and the generated image. The image synthesis quality can be improved.

A beamlet is generated from an electron beam using an array of apertures downstream of a single global collimating lens. The beamlets are directed through an image lens array downstream of the aperture array in the path of the beamlets, the beamlets being individually focused onto an intermediate image plane using the image lens array. The beamlet is then directed at a workpiece on a stage using a transfer lens array downstream of the image lens array. The paths of the beamlets do not include intersections.

The present application relates to the technical field of intelligent model, and discloses an image generation method and device, computer equipment, storage medium and program product, the present application determines similar preset prompt words, potential classification list of to-be-generated images and confusion category pairs based on original prompt words, so as to utilize the confusion category pairs for classification correction, avoid misjudging the theme category of the to-be-generated images, and improve the accuracy of the theme category. If the theme category is a person theme, the original prompt word is optimized through the scene type of the to-be-generated images to obtain a target prompt word, so that the image generation model generates images by utilizing the target prompt word rich in details and the scene type. When the image generation model has not ended iteration, the intermediate image generated by the image generation model is subjected to face repair, so as to avoid image repair after the iteration ends, thereby prolonging the generation time, improving the image generation efficiency, avoiding the phenomenon of face collapse, and improving the image generation quality.