6 results about "Noise map" patented technology

The control device (20) of the present application includes: an image acquisition unit (203) that acquires a radiograph of a subject (F) irradiated with radiation and photographed through the subject (F); a noise map generation unit (204) that derives an evaluation value based on relationship data representing a relationship between a pixel value and an evaluation noise value, from a pixel value of each pixel of the radiograph, and generates a noise map that corresponds to the derived evaluation value and each pixel of the radiograph; and a processing unit (205) that inputs the radiograph and the noise map into a learning completed model (207) constructed by machine learning in advance, and performs image processing for removing noise from the radiograph.

The application provides a denoising method and device of an image and electronic equipment. The method comprises the following steps: obtaining image data to be denoised; performing noise component evaluation on the image data to obtain a noise component map corresponding to the image data; performing multi-frequency band decomposition on the noise component map to obtain noise maps of a plurality of frequency bands corresponding to the noise component map; performing different denoising intensity transformations on the noise maps of the plurality of frequency bands based on a plurality of different preset denoising intensities to obtain candidate noise maps of the plurality of different denoising intensities; reconstructing a spatial domain noise map corresponding to the image data based on the candidate noise maps of the plurality of different denoising intensities; and processing the image data according to the spatial domain noise map to obtain a target denoising map corresponding to the image data. Thus, the image data is adjusted in different frequency bands, and the image data is processed in combination with the adjusted spatial domain noise map, so that the target denoising map corresponding to the image data is accurately determined, and the quality of the image is improved.

A method, apparatus, non-transitory computer readable medium, and system for performing an accelerated denoising process includes obtaining a noise map from a noise distribution. Embodiments then compute, using an image generation model, a denoising vector based on an accelerated denoising trajectory, where the accelerated denoising trajectory accelerates a denoising rate based on a diffusion timestep. Then, embodiments generate, using the image generation model, a synthetic image by denoising the noise map based on the denoising vector.

An infrared image non-uniformity correction method, apparatus, device, and storage medium include: acquiring continuous frame infrared images and extracting gradients to obtain a gradient map; performing inter-frame differencing on the gradient map to obtain a gradient frame difference map; dividing the gradient frame difference map into sub-blocks and counting the number of moving sub-blocks to obtain motion statistics; estimating high-frequency noise in the continuous frame infrared images based on an adaptive learning rate and motion statistics to obtain a high-frequency fixed non-uniform noise map, wherein the adaptive learning rate is dynamically adjusted based on the gradient map; estimating low-frequency noise in the continuous frame infrared images based on image registration frame difference calculation and motion statistics, and performing time-domain iterative mean filtering to obtain a low-frequency fixed non-uniform noise map; performing multi-scale frequency division processing on the high-frequency fixed non-uniform noise map and the low-frequency fixed non-uniform noise map to obtain noise components in different frequency bands; weighting the noise components using different preset correction coefficients; and correcting the continuous frame infrared images to obtain an output corrected image.

This invention relates to the field of low-altitude airspace management technology, specifically to a method, system, and storage medium for generating noise maps and optimizing flight paths for low-altitude aircraft. The method includes: extracting deep environmental features through an environmental feature encoding network; calculating a real-time noise map centered on the aircraft's current position using a physical-data hybrid noise prediction network, combining a physical acoustic model with neural network correction terms; generating a sequence of optimized flight actions for the future time domain based on the current noise map, aircraft state, and target point using a silent path optimization network; executing the first action in the action sequence, updating the aircraft state, and returning to perform rolling optimization for the next moment until the target point is reached. This invention, through the deep integration of a physical model and a data-driven model, ensures the correctness of acoustic principles while utilizing neural networks to compensate for complex environmental factors, achieving high-precision, real-time noise distribution prediction.

The application discloses a kind of considering the noise exposure evaluation method and system of unmanned aerial vehicle, the method includes: obtaining city environment data and unmanned aerial vehicle related data, the sound source modeling of unmanned aerial vehicle noise is carried out and the propagation of unmanned aerial vehicle noise in city environment is simulated, generate unmanned aerial vehicle noise map;With present situation city noise map, the city noise map of unmanned aerial vehicle noise map and present situation city noise map is obtained by merging calculation;Identify crowd distribution in different time periods and map to geographical space, form crowd distribution map in different time periods, combine multi-source data to carry out feature identification and analysis to crowd, obtain crowd distribution map with crowd feature information;According to city noise map and crowd distribution map, calculate regional noise pollution total amount and per capita noise pollution index, and calculate the noise exposure level of different crowds.The application perfects the noise exposure evaluation method, improves the space-time accuracy of unmanned aerial vehicle noise exposure evaluation.