196 results about "Good image" patented technology

The invention relates to the technical field of image processing, and discloses an image deblurring model and method based on double-domain feature fusion, and the method comprises the following steps: collecting data, and building a blurred image data set; building a DAF-UNet neural network based on spatial domain and frequency domain feature fusion; the DAF-UNet neural network is composed of n UNet sub-networks with different scales, and n is a positive integer; based on the blurred image data set, training the DAF-UNet neural network to obtain a trained image deblurring model; collecting a to-be-processed image; and processing a to-be-processed image by using the trained image deblurring model to obtain a clear image result. According to the method, the problems of low information mobility and key feature loss in the prior art are solved, and the method has the characteristics of good image recovery effect and model miniaturization.

The invention discloses a robust reversible watermarking method based on BFMs and adaptive normalization. The method comprises the steps of watermark embedding and watermark extraction. The watermark embedding specifically comprises the steps of generating watermark sequence information, performing integer wavelet transform on a carrier image, extracting a low-frequency region, and calculating BFMs of the low-frequency region; a proper watermark pair is selected through a private key, self-adaptive normalization is carried out according to the watermark pair, and the watermark is embedded in the normalized amplitude; reconstructing an image based on the amplitude of the embedded watermark; constructing a final reversible watermark image based on the reconstructed image; watermark extraction: judging whether the image is attacked or not; for an unattacked image, carrying out watermark extraction based on the auxiliary information; for an attacked image, watermark extraction is carried out by calculating the BFMs of the attacked image. The method has noise resistance and also has relatively good image reconstruction capability.

The disclosed technology provides for generating best images of a person in a retail environment. A method may include receiving, by an edge computing device from a camera, a continuous stream of image data of the retail environment, detecting, using object detection techniques, a person in the image data, the image data including a group of images that are part of a time series, generating bounding boxes for each of the group of images around the person based on detecting the person as they move in the images, identifying, based on applying a features model to each bounding box, at least one feature of the group of images depicting the person, selecting a subset of the bounding boxes having at least one feature that satisfies best images criteria, the subset having best images of the person, and returning the best images of the person.

The invention relates to the technical field of image processing, in particular to an in-orbit multi-stage real-time automatic dimming device and method for a spaceflight area-array camera. According to the device, an imaging module obtains an original scene image according to set integral time and transmits image data to an image coarse adjustment module; the image coarse adjustment module counts gray level distribution of an original image in real time, calculates first updating integral time, packages image data without overexposure or underexposure and sends the packaged image data to the image fine adjustment module; after the image fine adjustment module receives the image, scene recognition and gray level evaluation are carried out through a deep learning model, exposure parameters are dynamically adjusted according to the scene type, and image exposure control is achieved; and the image coarse adjustment module and the image fine adjustment module form a two-stage closed loop, so that the imaging module obtains integral time matched with a scene in real time. The method has the advantages that the integral time of the detector is automatically adjusted according to the detail content of the target scene, and the target image with the best image impression and detail is captured.

The invention provides a stereoscopic warehouse stacked goods checking method based on improved YOLO11-Seg. The stereoscopic warehouse stacked goods checking method comprises the steps of collecting goods images and performing labeling processing on the goods images; the method comprises the following steps: constructing an improved YOLO11-Seg model; carrying out segmentation identification on the cargo image by utilizing the trained improved YOLO11-Seg model; judging whether the condition is legal or not according to the segmentation mask and the mask category information; according to the geometric relationship of the convex hull points, judging whether the cargo top layer is in a full load condition, and if the cargo top layer is in the full load condition, directly calculating the cargo quantity; otherwise, calculating the number of the sub-top-layer boxes according to an estimation algorithm, and then calculating the total number of the cargo boxes. According to the method, the identification of stacked goods is realized by utilizing an improved YOLO11-Seg model, a goods segmentation mask and a mask type are generated, and the total goods quantity is calculated by identifying and distinguishing different layers of goods boxes; the method has the advantages of high generalization, low deployment cost, high checking efficiency and the like.

The present invention discloses a continuously variable magnification prism telescope optical system, belonging to the technical field of optical telescope systems. The system comprises a first objective lens doublet, a third objective lens, a second objective lens doublet, a third objective lens doublet, a fourth objective lens doublet, a tenth objective lens, a first prism, a graticule, an eyepiece doublet, a fourth eyepiece lens, and a fifth eyepiece lens, arranged in sequence along a first optical path. Most of the objective lenses and eyepieces in the continuously variable magnification prism telescope optical system are doublets, which can effectively shorten the overall length of the optical system and reduce the number of lenses in the optical system, resulting in a smaller volume and weight of the entire optical system after assembly, while also achieving good image quality. The system achieves continuous changes in focal length and / or imaging magnification by moving the second objective lens doublet and / or the third objective lens doublet along the optical axis. The use of a variable magnification objective lens structure makes the eyepiece structure simple and highly reliable.

Systems and methods for implementing smart in-vehicle conferencing with user tracking and speaker prioritization. A controller system of a vehicle may determine, based at least in part on one or more cameras, a plurality of candidate best image sources, determine, from the plurality of candidate best image sources, a best image source for a speaker in the vehicle, determine, based at least in part on the best image source, a video feed, determine, based at least in part on one or more microphones, a plurality of candidate best audio sources, determine, from the plurality of candidate best audio sources, a best audio source for the speaker, determine, based at least in part on the best audio source, an audio feed, and provide, to a conference call line, the video feed and the audio feed.

The invention discloses a camera shooting control method and device using an optical fiber signal, equipment and a medium. After the system is powered on, the upper computer sends control commands such as communication and triggering, the control commands are converted into optical signals through the electro-optical conversion module, the optical signals are sent to controlled equipment (including a camera unit, a light source unit and an aperture unit) through optical fibers (plastic optical fibers and silica optical fibers), the optical signals are converted into electric signals through the photoelectric conversion module, and the system works based on the control commands. Operation such as camera exposure, light source flashing and aperture opening and closing is achieved, and ideal control states of good image sensitization, light source synchronization and moderate aperture are achieved. Compared with a traditional control mode that electric signals are transmitted through cable media, optical signals are transmitted through the optical fiber media, the electromagnetic interference influence can be reduced, the transmission distance is longer, the cable weight is lighter, the error control probability is reduced, and the overall system robustness is effectively improved.

The invention discloses a deep learning-based warehouse cargo image recognition method, and particularly relates to the technical field of warehouse cargo recognition, and the method comprises the steps: simulating the day and night illumination difference through an image enhancement module, and constructing a training data set covering a real scene; carrying out feature extraction by using an improved ResNet101 model, and reconstructing a 256-dimensional full-connection layer to output a high-discrimination cargo exclusive feature vector; a double-branch classification structure is innovatively adopted, the first branch judges the goods category, the second branch judges the packaging state, and according to a confidence coefficient threshold value, intelligent decision making of high-confidence-coefficient result automatic output and low-confidence-coefficient triggering rechecking is achieved; the system integrates an error analysis module, a performance evaluation module and a model dynamic optimization module, and supports scene-based directional incremental updating. The method overcomes the defect that the prior art is poor in adaptability to complex scenes, illumination changes and package anomalies, remarkably improves the recognition precision and the operation efficiency, and effectively reduces the inconsistency of inventory accounts.

An anti-peep light source module and an anti-peep display device having an anti-peep function and good image quality are provided. The anti-peep light source module includes a light guide plate, a first light emitting element, a second light emitting element, a plurality of optical microstructures and a first surface. The light guide plate has a first light incident surface and a second light incident surface. Each of the optical microstructures has a first optical surface facing the first light incident surface and a second optical surface facing the second light incident surface. A magnitude of a first included angle between the first optical surface and the first surface and a magnitude of a second included angle between the second optical surface and the first surface gradually change as getting farther away from the first surface.

The utility model relates to a magnetic resonance shoulder joint coil fixing device and a magnetic resonance imaging system. The fixing device comprises a waistcoat worn on a subject; the shoulder joint coils are detachably connected with the shoulders on the two sides of the waistcoat in an attached mode, each shoulder joint coil comprises a coil carrier, a coil and a cable base, the coil carrier is formed by compositing a lining layer, a memory sponge filling layer and an outer protection layer which are sequentially distributed from inside to outside, the coil is arranged in the memory sponge filling layer, and the cable base is arranged on the outer protection layer. And the cable base is positioned outside the coil carrier and is connected with the coil. The magnetic resonance imaging system comprises the magnetic resonance shoulder joint coil fixing device. Compared with the prior art, the magnetic resonance imaging device has the advantages that the fitting degree of corresponding positions of the shoulder joint can be effectively ensured, a good image acquisition signal-to-noise ratio is ensured, and the working efficiency of MR examination of related parts of the shoulder joint can be greatly improved.

This invention discloses a phase compensation method for ship targets using ISAR based on block phase gradient autofocus. The steps include: pulse compression; range cell migration correction; range-oriented block segmentation; block phase compensation; calculation of the offset between adjacent image blocks; and image block alignment and stitching. This invention performs range-oriented block segmentation of the target echo, overcoming the shortcomings of existing technologies that ignore the spatial variation of rotational phase error, leading to defocusing in ship imaging, thus improving the accuracy of phase compensation and achieving good image focusing effect. Because this invention uses a phase gradient autofocus algorithm to compensate for the phase error within each block, and then uses a cross-correlation method to align and stitch the images of each block region, it overcomes the shortcomings of parameter estimation methods, such as complexity, time consumption, and poor real-time performance, and has the advantages of simple implementation, high efficiency, and strong real-time performance.

The present application provides a corneal curvature calculation method and related device, which are applied to the field of image processing technology. By performing bright spot detection on each frame of eye image, multiple bright spots that meet a preset number range in each frame of eye image are quickly located, and the bright spot center point of each bright spot in each frame of eye image and the bright spot area of ​​each frame of eye image are further determined. At the same time, by performing curve fitting on the bright spot areas of multiple frames of eye images acquired during movement, the horizontal coordinate corresponding to the minimum bright spot area in the fitted curve is determined. Since the location with the minimum bright spot area during movement is the location with the best image quality, the corneal curvature is calculated using the ellipse parameters at this location. Even if the eye image is a low-quality, low-resolution image, high-precision angular curvature calculation can be achieved. At the same time, since the eye image used to calculate the corneal curvature is a low-quality, low-resolution image, the required computing power is relatively low.

The present application relates to the technical field of image segmentation, and provides a lung image segmentation method, which comprises the following steps: S1: inputting a lung CT dicom sequence and converting the original dicom sequence into a jpg format image as an original lung image; S2: enhancing the original lung image through an image enhancement Lap-CLAHE algorithm to obtain an enhanced image; and S3: inputting the enhanced image into a U-shaped symmetrical network segmentation model 3D-SDUnet composed of an encoder, a decoder, a DPA double-path attention and a skip connection, and outputting a final segmentation result of the lung image. According to the technical scheme, the lung image is first enhanced, the detail information is improved, and the good image contrast is maintained, so that high-quality input data is provided for subsequent model segmentation. The segmentation model is optimized, the skip connection composed of a dense Swin Transformer block (DATB) and the DPA attention module are added, the information connection between the encoder and the decoder is strengthened, the segmentation precision is improved, and a complete lung model is obtained.

The present invention provides a toner containing colored resin particles of a binder resin and a colorant and an external additive, the binder resin containing a carboxyl group-containing acrylate resin, and the external additive containing carboxyl group-containing acrylate polymer modified metal particles, and a method for preparing the same. By using the specific external additive and the adhesive resin, the prepared toner can still keep stable triboelectrification performance in a high-humidity environment, so that the problems of uneven image concentration and increase of bottom ash are avoided. In addition, the toner has good residual accumulation resistance, residual accumulation on the toner cartridge is reduced, the printing quality is improved, stable printing is still achieved in the high-speed printing process, and good image quality is obtained.

The application discloses a kind of based on deep learning and joint bilateral filtering's image dynamic range compression method, including steps: collecting original high bit image data, constructs data set, and preliminary processing is carried out to data set, obtains linear RGB graph and Y channel graph;Combined with the skip connection structure of UNet and the efficient light unit of ShuffleNet constructs deep learning network, utilizes deep learning network to linear RGB graph and Y channel graph processing, obtains luminance gain graph;With the help of joint bilateral filtering upsampling operation, luminance gain graph and original high bit image are processed, and original resolution luminance gain graph is obtained;Original high bit image is carried out inverse gamma operation, and then divided by original resolution luminance gain graph is carried out gamma operation to obtain compressed 8bit image.The application realizes a dynamic range adjustment algorithm, from high bit to low bit mapping process, beneficial to restore highlight details, brighten dark details, and maintain good image contrast.

To provide a method for manufacturing a regenerated conductive roller which contributes to good image formation by effectively removing contaminants sticking to the surface of a conductive roller which is used over a long period of time and on which the sticking of the contaminants remarkably occurs without impairing its conductivity.SOLUTION: The method for manufacturing a regenerated conductive roller includes a step of removing a contaminant fixed to the outer surface of the conductive roller having a support having a conductive outer surface and a conductive layer on the outer surface of the support, wherein the conductive layer includes a matrix containing a first rubber and a domain containing a second rubber and an electronic conductive agent, the matrix has a Young's modulus of G1 (MPa) and the domain has a Young's modulus of G2 (MPa), and G1 and G2 satisfy G1 <G2, and the contaminant removing step includes a step of blowing a gas to the surface of the conductive roller to which the contaminant is fixed.SELECTED DRAWING: Figure 1

The invention provides a K-TIG welding penetration state discrimination method and system based on deep learning, and the method comprises the steps: inputting an original molten pool image into a pre-trained image processing parameter prediction model, obtaining a plurality of groups of image processing parameters, generating a plurality of corresponding candidate images, and obtaining the quality confidence through an image qualification evaluation model; the optimal image processing parameters are selected to process the original molten pool image, and a target image is obtained; and comprehensively judging whether the current welding state meets a preset penetration condition or not by combining the geometric characteristic parameters of the molten pool and the current welding process parameters. The method comprises the following steps: automatically generating multiple groups of image processing parameters and candidate images for an original molten pool image by introducing an image processing parameter prediction model; the quality of each candidate image is evaluated through the image qualification evaluation model, and the optimal image processing parameter is screened according to the quality confidence of each candidate image, so that the accuracy of molten pool image extraction is greatly improved, and the improvement of the accuracy of penetration state recognition is facilitated.

The invention discloses an image scaling method and system, an intelligent terminal and a storage medium, and the method can calculate a scaling ratio, a partition parameter and a jitter address value according to a first size of an input image and a second size of an output image, and updates a head address of the current image scaling, so as to read partition image data in the input image, the method comprises the following steps of: performing gray conversion on read image data to obtain initial gray data, acquiring corresponding first gray energy and a maximum gradient value, performing scaling interpolation calculation to obtain first scaled data, acquiring second gray energy of the first scaled data, performing recovery processing on the first scaled data to obtain recovered data, and performing data processing on the recovered data to obtain the image data. And time-space domain filtering is carried out on the recovered data, so that scaling processing on the whole input image is realized, the high-frequency signal loss and the mosaic effect can be obviously improved in the scaling processing process, the calculation overhead is greatly optimized through channel combination and parameterization design, and the method has universal adaptability to various images and has good image visual effect and algorithm robustness.

This application provides an optical module and a head-mounted display device. The optical module includes a lens group, a beam splitter, a first phase retarder, and a polarizing reflector disposed between the optical paths of the lens group. The first phase retarder is located between the beam splitter and the polarizing reflector. The lens group consists of a first lens and a second lens bonded together, with the first and second lenses arranged adjacent to each other along the same optical axis. The ratio H / T2 of the height H of the second lens away from the surface of the first lens to the center thickness T2 of the second lens satisfies: 0.38 ≤ H / T2 ≤ 0.82. The optical module provided in this application has the characteristics of small overall length and good image quality.

The lens comprises a first lens, a second lens, a third lens, a fourth lens, a fifth lens, a sixth lens and a seventh lens which are sequentially arranged from an object plane to an image plane along an optical axis, the first lens has negative focal power, the second lens has negative focal power, the third lens has positive focal power, the fourth lens has positive focal power, the fifth lens has negative focal power, the sixth lens has positive focal power, and the seventh lens has negative focal power. The fifth lens has positive focal power, the sixth lens has negative focal power, and the seventh lens has positive focal power. According to the lens provided by the embodiment of the utility model, only seven lenses are adopted, and the focal power and the surface type of each lens are reasonably matched, so that the total optical length of the lens can reach 23.4 mm and below, the field angle FOV can reach 136 degrees, the image surface diameter can reach 7.5 mm, the aperture number F can reach 1.8, aberration can be well corrected, enough good image quality is ensured, and the lens is suitable for large-scale popularization and application. The lens which is high in resolution, compact in structure, large in field angle, large in target surface and large in aperture is realized.

For particularly good image quality, a method for correcting artifacts caused by extrafocal radiation (=OFR) in an X-ray image of an object is provided, comprising the following steps: X-ray image of the object is acquired by an X-ray detector with a large number of detector pixels, the object being irradiated by X-ray radiation emitted from an X-ray tube; provision of a primary X-ray spectrum of the X-ray radiation and provision of an estimated or determined extrafocal radiation spectrum for the primary X-ray spectrum; determination of an image X-ray spectrum from the X-ray image for each image pixel; division of the image X-ray spectrum into a high-energy component and a low-energy component for each image pixel.Determining a weighting value for each image pixel by calculating the quotient between high-energy irradiation and total energy irradiation, and determining a weighted image from the weighting values ​​of the image pixels; providing a high-energy image from the high-energy components; processing the low-energy components for each image pixel with respect to the extrafocal radiation component using at least one image processing algorithm, taking into account image features and / or structures of the high-energy image; and recombination of the high-energy component and the processed low-energy component for each image pixel and creating a corrected X-ray image from this.

The application relates to an optical imaging method for a large-view-field TOF camera. Incident light rays are converged by three pieces of spherical lens groups, the spherical lens groups in the light ray incident direction successively include a first piece of negative meniscus lens, a second piece of negative meniscus lens and a first piece of positive double-convex lens; the converged light rays are reduced in vignetting by an aperture diaphragm; an aspheric double-concave lens with an aspheric front surface and a spherical rear surface is used for aberration correction; then two pieces of spherical lens groups are used for twice refraction focusing light paths, correcting aberration and clearly imaging on an image plane; the two pieces of spherical lens groups successively include a positive double-concave lens and a second piece of positive double-convex lens. The imaging method provided by the application uses only one aspheric mirror, the rest of the lens surfaces are spherical, and all the lenses use the same glass material; the system uses an image far field and a reverse distance structure, and meets the requirements of the TOF camera on a large view field, miniaturization and good image quality.

The application discloses a multi-modal image data automatic recognition system and method based on machine learning, relates to the technical field of image data processing, and comprises the following steps: image data is called, pixel point feature information is extracted, and a pixel point feature sequence is constructed; feature matching degree analysis is performed on the pixel points, and hierarchical division is performed; spatial feature joint index analysis is performed on the pixel points in the same layer, and hierarchical pixel correlation index is obtained by integrating the analysis data; interlayer pixel classification analysis is performed on the pixel points between adjacent layers, pixel classification correlation analysis is performed on the adjacent layers, and pixel classification correlation index of the adjacent layers is determined; correlation evaluation of pixel layer correlation logic is performed on non-repeated image data, and the best image matching combination is output; the application realizes accurate recognition of multi-modal image data, reduces the misrecognition rate and interference rate caused by coarse recognition and coarse matching, and improves the accuracy of image recognition.

The application discloses a fixed focus lens, comprising: a first lens with negative refractive power, a second lens with negative refractive power, a third lens with negative refractive power, a diaphragm, a fourth lens with positive refractive power, a fifth lens with negative refractive power, a sixth lens with positive refractive power, a seventh lens with positive refractive power, and an eighth lens with negative refractive power arranged in sequence along an optical axis from an object side to an image side; wherein the fifth lens and the sixth lens are cemented; the first lens, the fourth lens, the fifth lens, and the sixth lens are all glass spherical lenses, and the second lens, the third lens, the seventh lens, and the eighth lens are all plastic aspherical lenses. The 4G4P combination can well correct aberration, reduce distortion, ensure good image quality, improve resolution, and the total optical length is less than or equal to 15 mm.

A photon stream detector having high image acquisition performance and a CT device. The photon stream detector comprises a scintillator (100), a photon stream sensor (200), a main control board (400), and a substrate (300), wherein the scintillator (100) is coupled to the photon stream sensor (200), and an output end of the photon stream sensor (200) is connected to the main control board (400). The scintillator (100) is configured for converting an incident X-ray into visible light. The photon stream sensor (200) is configured for converting the visible light into an electrical signal, then amplifying, shaping, integrating and counting same to generate an integration signal and a counting signal, and outputting image data after data conversion and processing. The main control board (400) is configured for providing a clock signal, a working power supply, and data analysis and transmission. When the photon stream detector performs detection, the photon stream sensor (200) simultaneously generates an integration signal and a counting signal, forms image data after data conversion and processing, and uploads the image data to an image acquisition device, thereby achieving good image acquisition performance in the entire dose range.

The invention discloses an image processing device and method, a chip and equipment, and belongs to the technical field of image processing. A feature extraction module in the image processing device is configured to acquire a first image and extract a shallow feature map of the first image, and the shallow feature map is used for describing detail content of the first image; the mapping module is configured to perform feature extraction on the shallow feature map and determine a mapping image based on a feature extraction result, and the feature extraction result is obtained by fusing different level features of the shallow feature map; and an image reconstruction module configured to reconstruct a second image having a resolution greater than that of the first image based on the mapped image. In the process of extracting the features of the shallow feature map, the features of different levels are fused to enhance the deep features in the feature extraction result, the ability to restore image details can be improved in the process of reconstructing the image based on the enhanced deep features, and a good image reconstruction effect can be obtained while the lightweight level is kept.

The invention discloses a coal mine safety monitoring system and method based on big data analysis, and belongs to the technical field of mine safety monitoring. A video processing chip integrated in the explosion-proof distribution box is used for segmenting the monitoring video to obtain a monitoring image, the hardware core of the processing module is a server cluster deployed in a ground data center, and a processor in the server cluster is used for carrying out image enhancement processing on the collected monitoring image; and whether the image quality after the image enhancement processing accords with the preset image quality is judged through an image quality detection chip built in the server. The method has the advantages that the peak signal-to-noise ratio of the image is increased by combining weighted guide filtering and FPGA edge detection, dead-corner-free comprehensive supervision of underground coal mine operation is achieved through global comprehensive monitoring, and coal mine monitoring images with good image quality can be obtained under the condition that the actual environment is severe.

The present invention relates to the technical field of optical imaging, and disclosed are an optical lens (110), a camera module (100), and an electronic device (1000). The optical lens (110) comprises a first lens group (10), a second lens group (20), and a light guide module (50), wherein in a first imaging mode of the optical lens (110), the light guide module (50) is configured to reflect first light from the first lens group (10) to an image sensor (120), and in a second imaging mode of the optical lens (110), the light guide module (50) is configured to reflect second light from the second lens group (20) to the image sensor (120), the optical lens (110) has different effective focal lengths in the first imaging mode and the second imaging mode, and the light guide module (50) is further configured for jitter compensation to achieve optical image stabilization. The optical lens (110) achieves optical image stabilization while performing optical zoom, and offers good image quality and high image resolution, thereby improving the use experience of a user.

The invention relates to an image capturing device. The image capturing device sequentially comprises an optical modulation element from an object side to an image side along the optical axis of the image capturing device, at least part of the area of the optical modulation element is a modulation area, and the modulation area is configured to enable light to form diffuse reflection on the incident side of the modulation area after the light enters the modulation area, and keep wavefront information to be emitted on the transmission side of the modulation area; and the optical lens is configured to enable the light rays emitted by the optical modulation element to be converged on an imaging surface of the optical lens. According to the image capturing device, on one hand, the appearance with low transparency formed by diffuse reflection of the optical lens can be effectively camouflaged, so that the appearance is not easy to perceive, and on the other hand, external object light can normally penetrate through the optical modulation element and is imaged through the optical lens, so that the shooting effect is ensured.