62 results about "High-dynamic-range imaging" patented technology

The invention discloses a high dynamic range imaging method for an ultrahigh-definition video, and the method comprises the following steps: collecting a low dynamic range frame of the ultrahigh-definition video, splicing images through exposure normalization and adaptive correction, and constructing multi-scale multi-channel data; extracting structure and texture features by using an improved multi-scale convolution and attention network, and calculating matching degree rearrangement support frame features; detecting a missing region in the image, and performing local recovery and sharpening on the missing region by using improved gating convolution in combination with residual errors and high-frequency enhancement branches; dynamically calculating a fusion weight, carrying out weighted fusion on the features, and outputting a final fusion feature; and inputting the fusion features into a global merging network for HDR reconstruction, and splicing the fusion features into a video sequence for coding and outputting by adopting adaptive mapping and time domain processing. The method has high robustness and noise immunity, video frames in a complex illumination environment can be effectively processed, HDR images with rich details and accurate colors are generated, and the visual effect of the video is remarkably improved.

A high dynamic range imaging method for an ultra-high-definition image comprises the following steps: acquiring the ultra-high-definition image, and preprocessing ultra-high-definition image data; constructing an ultra-high-definition and high-dynamic range imaging network; and inputting the preprocessed data into the constructed neural network to obtain an HDR image reconstruction result, calculating a loss function and training the model. Aiming at the problems of detail blurring, noise interference, limited dynamic range and the like in an ultra-high-definition image, the algorithm disclosed by the invention obviously improves detail richness and brightness balance of the image while keeping low calculation complexity, and supports real-time processing and low-resource hardware deployment; meanwhile, the method shows excellent robustness and generalization ability in underexposure image and high-exposure image reconstruction and noise suppression tasks, and can achieve a good effect in an actual scene.

According to the metasurface device for snapshot type high dynamic range imaging and the imaging system and method thereof provided by the invention, the dynamic range of a camera can be improved by integrating a plurality of superlenses with different transmittances on a single metasurface; the polarization-independent metasurface is designed by using a pure phase modulation principle, so that the system stability is greatly improved; compared with other backlight imaging methods, the method has the advantages of high integration level, single measurement and space and time cost saving, and has a certain reference effect on development of ultra-compact imaging equipment and improvement of dynamic range imaging capability.

The invention discloses a multi-frame fusion fixed image noise dynamic compensation system and method, particularly relates to the technical field of image processing, and is used for solving the problem that fusion boundary noise has artifacts due to existing independent noise compensation. The method comprises the following steps of: establishing a mapping relation between pixel-level imaging parameter difference degree and fusion weight by acquiring a multi-frame image sequence of different imaging parameters and identifying a transition region of brightness gradient dramatic change; extracting a frequency domain energy skewness feature and a chroma phase change boundary feature of the transition region, and generating a noise visibility correction factor through nonlinear fusion; dynamically constraining the compensation intensity range to realize multi-frame noise compensation collaboration; and finally, performing cooperative compensation and fusion output within the constraint range. Fusion boundary noise discontinuity is effectively eliminated, and the visual quality of high-dynamic-range imaging is remarkably improved.

The invention discloses an additive manufacturing process high dynamic range imaging method and system based on a multi-view light field. Firstly, an original light field image in the additive manufacturing process is collected through a light field camera; extracting R, G and B single-wavelength channel images from the original light field image by the established light field Bayer model; then, taking the R-channel image as a reference, and adopting a K-means clustering algorithm for segmentation to obtain a low-light area and a high-light area in the R-channel image; and finally, fusing the pixel value of the G channel image corresponding to the low-light area with the R channel image to enhance weak light details, and fusing the pixel value of the B channel image corresponding to the highlight area with the R channel image to suppress highlight, thereby reconstructing to obtain an HDR image. According to the method, the image quality and the information integrity monitored in the additive manufacturing process are effectively improved, and high-quality visual data are provided for additive manufacturing process optimization and defect recognition.

This disclosure provides a video production method, apparatus, electronic device, and storage medium. The method includes: when producing a video file using high dynamic range imaging technology, determining first metadata information, wherein the first metadata information is ambient light-related metadata information; and writing the first metadata information into the header information of the video file. This disclosure expands the metadata information of video files, improves the display effect of video files produced under different ambient light conditions, and enhances the usability of HDR technology.

Disclosed is an imaging system with an image sensor and processor(s). The image sensor has a plurality of photo-sensitive cells arranged on a photo-sensitive surface of the image sensor; a colour filter array with colour filters of at least three different colours; and a polarisation mask with polarisation filters arranged on an optical path of at most a first predefined percent of the colour filters. The processor(s) is / are configured to: read out image data from the image sensor; and process the image data to generate at least one of: a full-resolution colour image, a full-resolution polarisation image, a high-dynamic range (HDR) image.

An image processing apparatus for forming an enhanced image is disclosed. The apparatus comprises one or more processors configured to: receive one or more input images; form, from each of the one or more input images, a respective feature representation, each feature representation representing features of the respective input image; and subject the one or more feature representations to a symmetric pooling operation to form an enhanced image from at least some of the features of the one or more feature representations identified by the symmetric pooling operation. The apparatus may generate images with increased photoreceptive dynamic range, increased bit depth and signal-to-noise ratio, with less quantization error and richer colour representation.

This invention relates to the field of high dynamic range imaging technology and discloses a deep learning-based high dynamic range imaging method. The method includes acquiring original images with different exposure types; extracting features from the original images; using a cross-frame entanglement mechanism to perform quantum motion compensation on abnormal exposure feature maps based on normal exposure feature maps, resulting in compensated abnormal exposure feature maps; fusing features between the normal exposure feature maps and the compensated abnormal exposure feature maps using a multi-scale feature pyramid and adaptive spatial attention, resulting in fused features; and reconstructing the image using an attention-guided residual reconstruction model to obtain a high dynamic range image. This invention solves the motion offset problem between multi-exposure images through quantum motion compensation, eliminating dynamic blur; achieves detail enhancement and color correction through a multi-scale feature pyramid and spatial attention; and reduces computational complexity and improves reconstruction efficiency through a lightweight reconstruction model.

The embodiment of the invention discloses a multi-lens shooting system. The multi-lens shooting system comprises a control assembly, a double-rate synchronous dynamic random access memory DDR, a first processing assembly comprising n image front-end processing units IFPU, a second processing assembly comprising m high dynamic range imaging HDR sensors, and an HDR pipeline arranged on any IFPU, and n and m are integers greater than 0; and the control assembly is used for performing opening and closing control on a connection circuit between the IFPU and the corresponding HDR sensor, controlling the IFPU to output corresponding exposure data to at least one of the DDR pipeline and the HDR pipeline, and controlling the HDR pipeline to perform hardware HDR processing on at least one path of exposure data. According to the invention, the purpose of HDR multi-camera video recording is realized by designing and multiplexing a set of HDR hardware based on time division, and on the premise of ensuring a certain rate, the chip area is effectively reduced, and the construction cost is reduced.

Image data from an image sensor is read out, wherein when reading out, processor(s) is / are configured to employ subsampling. The image data is processed, using neural network(s), to generate an image.

An image processing apparatus (700) comprising one or more processors (704) is described herein for: receiving (601) a plurality of input images (301, 302, 303); for each input image, forming (602) a decomposition dataset by decomposing the input image (301, 302, 303) or a filtered version thereof (307, 308, 309) into a plurality of frequency-specific components (313), each frequency-specific component representing occurrences of features of a respective frequency interval in the input image or filtered version thereof; processing (603) each decomposition dataset using one or more convolutional neural networks to form a combined image dataset (327); performing (604) a build operation on the combined image dataset (327), the build operation being adapted for image building from a plurality of frequency-specific components, thereby forming an output image (333) representing a combination of the input images. The generated HDR output image can have fewer artifacts and provide better quality results. Furthermore, the apparatus is computationally efficient, maintaining a good balance between accuracy and efficiency.

The invention discloses a high-dynamic-range imaging CMOS camera based on a K7 gigabit network. The high-dynamic-range imaging CMOS camera comprises a camera shell 9, and a high-dynamic-range imaging CMOS camera body is installed in the camera shell 9. The CMOS camera body comprises a CMOS sensor chip 1, an FPGA (Field Programmable Gate Array) programmable logic device 2, a DDR3 (Double Data Rate 3) high-speed memory 3, a FLASH memory chip 4, a gigabit Ethernet interface 5, a GPIO (General Purpose Input / Output) isolation interface 6 and a power supply module 7. The FPGA chip receives high-gain and low-gain 2048 * 2048 resolution image data from the CMOS sensor chip at the same time, data caching is carried out through the DDR3 high-speed memory, high-dynamic-range image synthesis is completed in real time, images with 46 frames per second (the resolution is 2048 * 2048) and the dynamic range reaching 70.3 dB are output through the gigabit Ethernet interface, and working parameters of the camera can be configured through a computer.

This invention discloses a method, system, device, and medium for quantitative calculation of brightness perception, relating to the field of brightness perception. It involves acquiring a high dynamic range (HDR) image of a target area; extracting brightness and chromaticity information from the HDR image; calculating a brightness index based on the brightness information using an image thresholding method; calculating a contrast index based on the brightness information using a hierarchical sampling method; calculating a color temperature index based on the chromaticity information using a color space transformation matrix; and calculating the perceived brightness value of the target area based on the brightness, contrast, and color temperature indices. This invention achieves quantitative calculation of brightness perception through multi-dimensional image indices.

The application discloses a panoramic depth high dynamic range imaging system and method based on compound eye structure, relates to the technical field of computational imaging and image processing, and synchronously collects and multi-scale motion estimation ensures the time sequence consistency between sub-images, effectively reduces the splicing dislocation and ghosting caused by the rapid movement or partial occlusion of an object, and improves the stability of a dynamic scene; an adaptive HDR fusion mechanism based on exposure difference and local features preferentially retains high-contrast unsaturated region details through intelligent weight distribution, overcomes the overexposure and underexposure problems under backlight or light mutation, and makes the output image naturally smooth in the bright-dark transition area; in addition, multi-scale depth synthesis and edge perception filtering work cooperatively, enhances the depth estimation accuracy of a low-texture area, avoids depth jumping and edge blurring, and guarantees the overall definition of the panoramic depth image; meanwhile, rolling shutter correction and space-time interpolation compensation further reduce dynamic ghosting and noise interference.

An imaging device includes a pixel-circuit and a readout circuit. The pixel-circuit includes a photodetector. The readout circuit is configured to perform a series of iterations, each iteration including (i) accumulating an electrical charge in the photodetector during an integration time, (ii) transferring a portion of the electrical charge accumulated in the photodetector to a floating diffusion, the portion being less than all the electrical charge accumulated in the photodetector, and (iii) coupling a voltage of the floating diffusion to a storage capacitance, to subsequently, in a final iteration, transfer a remaining electrical charge from the photodetector, and to derive an output digital value, indicative of a total energy of the light incident on the photodetector over the series of iterations and the final iteration, from the voltage on the storage capacitance and from the remaining electrical charge transferred in the final iteration.

The present invention discloses a method for selecting exposure time for high dynamic range imaging, which relates to the field of high dynamic range (HDR) imaging. First, the camera parameters required for calculating the brightness signal-to-noise ratio are calibrated; then, a scene brightness distribution histogram is obtained, and after selecting the brightness range of interest, the signal-to-noise ratio of the target after encoding is calculated according to the photoelectric conversion function; finally, the signal-to-noise ratio of the target after encoding and the calculation formula of the brightness signal-to-noise ratio are used to calculate the shortest exposure time sequence that meets the signal-to-noise ratio requirements, and this is output as the final selected exposure time sequence. The present invention combines HDR imaging exposure time selection with HDR encoding quantization interval, providing a practical optimization target for reducing noise and improving the efficiency of HDR image acquisition and reconstruction quality.

The application discloses a multi-frame fusion fixed image noise dynamic compensation system and method, and particularly relates to the technical field of image processing, and is used for solving the problem of the existing independent noise compensation leading to the noise band artifact of the fusion boundary; a mapping relationship between the pixel-level imaging parameter difference degree and the fusion weight is established by acquiring a multi-frame image sequence with different imaging parameters and identifying a transition area with a dramatic change in luminance gradient; a noise visibility correction factor is generated by extracting the frequency energy skewness feature and the chroma phase change boundary feature of the transition area through nonlinear fusion; the multi-frame noise compensation is cooperated by dynamically constraining the compensation intensity range; finally, the cooperative compensation is executed within the constraint range and fused and output; the discontinuity of the fusion boundary noise is effectively eliminated, and the visual quality of the high dynamic range imaging is significantly improved.

The invention relates to the technical field of digital image processing, in particular to a light energy fusion-based high dynamic range image acquisition method, which comprises the following steps of: constructing an imaging system comprising a low-sensitivity sensor, a medium-sensitivity sensor and a high-sensitivity sensor; the method comprises the following steps: carrying out high dynamic range imaging on a shooting scene by using an imaging system, and respectively obtaining three corresponding images in three sensors: respectively carrying out corresponding calculation on the three images according to respective inverse response functions of the three sensors to obtain light energy corresponding to the three sensors; corresponding fusion weights are set for the three kinds of light energy; fusing the three kinds of light energy by using the fusion weight to obtain shooting scene light energy of the shooting scene; and according to the three inverse response functions and the corresponding light energy, carrying out step conversion on the scene light energy to obtain a high dynamic range image of the shooting scene.

The present invention provides a color restoration method suitable for a multiple-exposure mechanism, comprising: obtaining Bayer images at different exposure levels; obtaining Bayer images at different exposure levels after applying white balance gain; obtaining a first matrix at different exposure levels; obtaining tristimulus value images under equal-energy light sources at different exposure levels; obtaining fused tristimulus value images under equal-energy light sources; obtaining a second matrix; obtaining an image after applying a color correction matrix; and obtaining an image after applying a gamma curve. The present invention can optimize the problem of color cast in the final image caused by inconsistent white balance gains corresponding to different regions in a scene under mixed light sources. Furthermore, the present invention simultaneously considers the gamma curve and target saturation during the calculation process, and the calculated color correction matrix matches the gamma curve and target saturation, effectively resolving the problem of abnormal color restoration under high dynamic range imaging conditions.

The high dynamic range imaging network method based on pyramid structure spatial feature transformation is based on three sub-networks: a deformable convolutional alignment network, a PSFT conditional network, and a fusion network of deformable convolutional residual dense blocks. The process is as follows: 1) The exposure of the input image is aligned using gamma correction, and then all images with and without exposure alignment are input into two convolutional layers of the deformable convolutional alignment network with shared offsets to obtain preliminary image features; 2) The exposure-aligned image features are geometrically aligned in the deformable convolutional alignment network with shared offsets, and the offset obtained in the geometric alignment process is also applied to the LDR image features without exposure alignment for geometric alignment; only gamma correction is used to align the exposure; 3) The geometrically aligned LDR image features without exposure alignment are input into the PSFT-based conditional network to obtain optimized features and obtain a high dynamic range image.

High dynamic range video frames may be generated from images captured using different exposure times. The merging of the images may combine pixel values of the images using a parametric exposure time ratio to form a merged video frame. The quality of the merged video frames may depend on the accuracy of the exposure time ratio. In some scenarios, the exposure time ratio is unknown, or reported information from an automatic exposure controller about the exposure time ratio is inaccurate. Using an inaccurate exposure time ratio to merge images will result in undesired artifacts in the merged video frames. To solve this problem, a self-calibration technique may be implemented to derive an exposure time ratio based on the image itself.

In accordance with some embodiments, systems, methods, and media for high dynamic range imaging using single-photon and conventional image sensor data are provided. In some embodiments, the system comprises: first detectors configured to detect a level of photons proportional to incident photon flux; second detectors configured to detect arrival of individual photons; a processor programmed to: receive, from the first detectors, first values indicative of photon flux from a scene with a first resolution; receive, from the second detectors, second values indicative of photon flux from the scene with a lower resolution; provide a first encoder of a trained machine learning model first flux values based on the first values, provide the second encoder of the model second flux values; receive, as output, values indicative of photon flux from the scene; and generate a high dynamic range image based on the third plurality of values.

The invention discloses a compressor valve plate welding positioning method and system based on machine vision, and relates to the field of image welding positioning, polarization structured light is actively projected, the interference of mirror reflection on imaging is inhibited from a physical source by using polarization characteristics, and meanwhile, the structured light technology is used for extracting fragile gray-scale edges from a positioning problem; and more stable light stripes which are deformed or interrupted on the surface of the object are analyzed. And by combining a high dynamic range imaging technology and fusing a multi-exposure frame sequence, it is ensured that feature details of highlight and dark areas are completely reserved. On the basis of the high-quality image, a group of stable and reliable feature point sets can be obtained by decoding the structured light and accurately extracting discontinuous points generated at the outline of the valve plate. And subsequently, contour fitting and pose calculation are carried out based on the point set, so that high-precision and high-robustness positioning of the valve plate is realized.

The application relates to an ultrahigh dynamic range imaging system and method based on a diffusion model image prior, the system comprising a pre-alignment module, a variational autoencoder, a decomposition fusion control module, a fidelity control module and a variational self-decoder, and the method comprising the following steps: adopting an optical flow estimation algorithm to transform and align a short-exposure image onto a long-exposure image, and masking a blocked area to obtain a pre-alignment result; decomposing information of the short-exposure image and the long-exposure image, and fusing and tone mapping the pre-alignment result by controlling a fusion process of a diffusion model to obtain a fusion result; controlling fidelity of the fusion result according to the long-exposure image and the pre-alignment result, and obtaining a reconstructed image through decoding. Compared with the prior art, the application can improve alignment and fusion robustness under the condition that exposure levels are greatly different, and can generate natural and realistic tone mapping effects in an ultrahigh dynamic range scene.

The invention relates to the technical field of image processing, and provides a high dynamic range imaging exposure interpolation method and device, equipment and a medium. The method comprises the following steps: acquiring a high-exposure image and a low-exposure image for the same scene, wherein the exposure time of the high-exposure image is longer than that of the low-exposure image; respectively mapping the high-exposure image and the low-exposure image to a target exposure time to obtain a first intermediate image and a second intermediate image; performing weighted fusion processing according to the first intermediate image and the second intermediate image to generate an initial intermediate exposure image; and inputting the initial intermediate exposure image into a pre-trained residual diffusion model, and performing residual detail recovery on the initial intermediate exposure image through the pre-trained residual diffusion model so as to generate a high dynamic range image based on the recovered residual details. According to the method, through a technical path of combining physical exposure interpolation and residual diffusion optimization, high-quality and high-dynamic range image reconstruction with natural vision and rich details is realized.

Described herein is an image processing apparatus (1200) for forming an enhanced image, the apparatus comprising one or more processors (1204) configured to: receive (1101) one or more input images (301, 302, 404, 405, 406, 501, 502, 503, 601, 602, 603); form (1102) a respective feature representation from each of the one or more input images, each feature representation representing features of the respective input image; perform (1103) a symmetric pooling operation (305) on the one or more feature representations to form an enhanced image (408, 505, 608) from at least some of the features of the one or more feature representations identified by the symmetric pooling operation. The apparatus can generate images with greater dynamic range of sensitivity, greater bit depth and signal-to-noise ratio, and with less quantization error and richer color representation.

The present disclosure provides a video production method and apparatus, an electronic device, and a storage medium. The method comprises: when a video file is produced using high dynamic range (HDR) imaging technology, determining first metadata information, the first metadata information being metadata information related to ambient light; and writing the first metadata information into header file information of the video file. The present disclosure expands metadata information of the video file, improves the display effect of video files produced under different ambient light conditions, and improves the availability of HDR technology.

The invention particularly relates to a semantic knowledge migration-based high dynamic range imaging method, which comprises the following steps of: acquiring a low dynamic range image, and generating an initial high dynamic range image by utilizing an initial high dynamic range reconstruction model; converting the initial high dynamic range image into a reconstructed image of a standard dynamic range domain through domain migration operation; processing the reconstructed image of the standard dynamic range domain by adopting a semantic segmentation network SAM to obtain semantic priori; processing the reconstructed image of the standard dynamic range domain by combining semantic prior with a semantic prior guidance reconstruction model to obtain a refined reconstructed image; based on a knowledge distillation method, cross-domain semantic priori migration is realized at three levels of content, color and features, and an initial high dynamic range reconstruction model is optimized through a combined loss function; in the reasoning stage, only the optimized initial high dynamic range reconstruction model is adopted to output a final high dynamic range image. According to the method, high-dynamic-range image reconstruction is carried out by effectively utilizing the advanced capability of semantic priori.

This invention provides a solid-state imaging device, a driving method for a solid-state imaging device, and electronic equipment that enable sufficient high dynamic range and image quality, and that offer an AD conversion unit with a small circuit area and low power consumption for high dynamic range imaging. [Solution] The pixel signal processing unit 400 includes an AD conversion unit 430 capable of performing AD conversion processing to convert the first conversion gain signal and the second conversion gain signal of the input pixel signal from an analog signal to a digital signal, and a multiple bit memory (SRAM) unit 440. The AD conversion unit 430 determines which of the first conversion gain signal and the second conversion gain signal is required for image generation, and selects information to be stored in the memory unit 440 based on the determination information.