13 results about "Image reduction" patented technology

The application provides a differential-based white light lensless single-frame weakly conjugate reconstruction method, which can realize weakly conjugate image reduction and improve single-frame imaging resolution in a white light lensless imaging system without using additional elements and complex algorithms. The method comprises the following steps: step one, obtaining spectral distribution information of a system light source by using a spectrometer and performing differential processing to obtain a reconstruction wavelength; step two, performing preprocessing on a collected single-frame image by using a hologram average value reduction method; step three, generating light intensity diagrams of different defocusing surfaces by using an angular spectrum propagation formula, and determining a defocusing distance range based on a definition discrimination factor; and step four, calculating differential distribution on different focal planes in the distance range obtained in step three based on numerical differentiation, and obtaining a final result by selection. The application can reduce the requirement of spectral width of a traditional lensless microscope, effectively reduce the cost of the system, improve the imaging resolution of the white light lensless microscope, and reduce the weakly conjugate image in the case of single-frame image collection.

The application discloses a kind of based on visual device transmission line external damage hidden danger early warning method, and the method steps are as follows: S1 obtains the transmission line image containing large construction machinery and pre-processes;S2 is to the pre-processed image Shadow synthesis obtains enhanced image, reduces enhanced image, and obtains fusion feature image by fusing the features of three;S3 is constructed and is trained target recognition model with fusion feature image, and the optimal model is obtained by optimizing parameters using optimization algorithm;S4 visual device obtains real-time image, and optimal model identification, if mechanical is identified, then progress S5;S5 millimeter wave radar scans three-dimensional point cloud data, projection two-dimensional is compared with identification result, and if it coincides, then progress S6;S6 preset safety distance, calculate the height of machinery in combination with point cloud data and projection area, and if it exceeds safety distance, then alarm.The application can realize all-weather unattended, accurately identify machinery, filter invalid early warning, reduce missed detection, reduce the risk of early warning failure, avoid line trip-out and other accidents, and protect the safety of transmission line.

The application provides a large-view-field compact reflective projection objective lens, and belongs to the technical field of semiconductor manufacturing lithography. The projection objective lens system comprises six aspheric mirrors. After a light source irradiates a mask, the light source sequentially passes through a first mirror M1, a second mirror M2, an aperture, a third mirror M3, a fourth mirror M4, a fifth mirror M5 and a sixth mirror M6. The aperture is located between the second mirror M2 and the third mirror M3. Finally, the light source is imaged on a wafer surface, and the projection objective lens system realizes an image reduction ratio of 0.2. In the embodiment of the application, the light path does not have a light blocking phenomenon and the structure is compact. The main optical parameters are as follows: the numerical aperture on the image side is 0.2, the object field on the object side is 10mm, and the total length of the system is 1280mm.

The present application relates to remote sensing image processing technical field, especially to a kind of high bit depth remote sensing image reduction and enhancement method based on three levels optimization, comprising: cutting to remote sensing image, and screening out representative local sample block;Representative local sample block is decoupled and image enhancement in Lab space;With enhancement parameter as optimization variable, construct hierarchical objective function, and obtain optimal parameter by optimization algorithm optimization, and generate enhanced image by optimal parameter full-width processing original image;The whole scene enhanced image obtained is corrected.The present application is cooperated between representative local sample block construction, Lab decoupled enhancement and optimization learning, whole scene correction, realizes three levels closed loop operation of sample layer, parameter layer, whole scene layer, so that the last obtained high bit depth optical remote sensing image has good brightness level, detail distinguishability, display balance and color naturalness whole scene enhanced result.

The present application relates to the technical field of image processing, in particular to a grid-based image matching positioning method, system, device and medium; the method divides the input image into grids, performs template matching of the image according to the grid position, and reduces the input image matched to the smallest range, greatly reducing the number of matching comparisons, improving the running efficiency of the RPA process and the accuracy of image positioning; by setting a priority quadrant, the efficiency of positioning matching is improved, the input image is reduced to the smallest range, the number of matching is reduced under different resolutions, and the running efficiency and positioning accuracy are improved.

The embodiment of the invention discloses an image processing method and device, electronic equipment and a computer readable storage medium, and relates to the technical field of image processing. The method comprises the following steps: acquiring an original image, and shrinking the original image into a shrunk image; segmenting the original image into a plurality of block images; determining the size of a convolution kernel, and determining a plurality of edge pixel points in the plurality of block images according to the size of the convolution kernel; according to the coordinates of the edge pixel points and the size of the convolution kernel, determining a first local neighborhood of the edge pixel points in the block image and a second local neighborhood of the edge pixel points in the reduced image; performing convolution processing on the edge pixel points according to the first local neighborhood and the second local neighborhood, and performing lookup table LUT adjustment processing on the edge pixel points according to the reduced image to obtain target edge pixel points; and determining the target edge pixel points as the pixel points in the enhanced block image corresponding to the block image. According to the scheme, the problem of inconsistent boundary effects during image enhancement can be efficiently solved.

According to implementations of the present disclosure, a scheme for image scaling is proposed. According to the scheme, an input image having a first resolution is obtained. An output image having a second resolution and high-frequency information subject to a predetermined distribution are generated based on the input image by using a trained invertible neural network, where the first resolution is higher than the second resolution, and the input image and the output image have the same semantics. Furthermore, another input image having the second resolution is obtained. Another output image having the first resolution is generated based on the other input image and the high-frequency information subject to the predetermined distribution by using an inverse network of the invertible neural network, where the other input image and the other output image have the same semantics. The scheme is capable of scaling down original images into low-resolution images with the same semantics and visually pleasing, and is capable of reconstructing high-quality high-resolution images from the low-resolution images.

To realize a new technology for performing multiple image processing operations using a support program in an information processing device that incorporates a general-purpose printing program. [Solution] When a print command is output from the editing application 43, the general-purpose printing program 41 generates intermediate data based on the image to be printed, and the auxiliary program 42 obtains the intermediate data and the print settings corresponding to the print command from the general-purpose printing program 41. The auxiliary program 42 performs a synthesis process to synthesize composite information such as watermarks into the intermediate data according to the print settings obtained from the general-purpose printing program 41, and after the synthesis process, it performs a layout process to reduce the size of images from multiple pages and place them on a single sheet of paper, or to divide an image from one page so that it is printed on multiple sheets of paper. Furthermore, the auxiliary program 42 controls the printer 2 to perform printing based on the edited image.

A method and device (1600, 1700) for image reduction. The method for image reduction comprises: obtaining a phase coefficient corresponding to a position of a first pixel point in an image comprising N*N pixel points, and performing demosaicing on the pixel points in the image to obtain N*N first pixel points, and then performing reduction processing on the N*N first pixel points according to the phase coefficient. Through the demosaicing processing, a plurality of pixel points can be converted into the same pixel point, so that the first pixel points in a region of the image are increased, which helps to maintain more high-frequency performance when performing the reduction operation, thereby improving the definition of the reduced image.

The present disclosure relates to adaptive neural network selection methods and systems, the method comprising: obtaining one or more images from a camera; obtaining a zoom factor of the camera, the zoom factor representing a measure of a current zoom level applied by the camera when capturing the one or more images; selecting a neural network from a plurality of neural networks configured to segment the one or more images and / or detect one or more objects in the one or more images based on the zoom factor, the plurality of neural networks being configured to operate at different image resolutions, a higher zoom factor corresponding to selecting a neural network configured to operate at a lower image resolution, and a lower zoom factor corresponding to selecting a neural network configured to operate at a higher image resolution; downscaling the one or more images to an image resolution required by the selected neural network; applying the selected neural network to segment the downscaled one or more images and / or detect the one or more objects in the down scaled one or more images.

The application discloses an unmanned aerial vehicle horizon real-time detection method and system based on improved YOLO segmentation, and relates to the technical field of horizon detection, and comprises the following steps: S1, acquiring a front view image or a video frame sequence of an unmanned aerial vehicle; S2, performing sky region segmentation by using a lightweight sky segmentation network; S3, extracting a low sky boundary point set from a sky mask; S4, fitting a current frame horizon parameter based on the low sky boundary point set; S5, performing adaptive time sequence smoothing processing based on boundary integrity; and S6, outputting the horizon parameter after time sequence smoothing. The application adopts the unmanned aerial vehicle horizon real-time detection method and system based on improved YOLO segmentation, extracts a sky region by using a lightweight sky segmentation network, reduces the horizon search range from a complex original image to a real demarcation region between the sky and the non-sky, and effectively reduces the interference of buildings, ground textures, cloud changes and complex background edges on the detection result.

The invention provides an image data processing method and device, and belongs to the technical field of image data processing, and the method comprises the steps: obtaining a bit width reduced image of an original image, and original image data of the original image in a first preset region; the bit width reduction image is an image generated by performing bit width reduction processing on the original image; performing blurring processing and bit width recovery processing on the bit width reduced image in sequence to generate an intermediate image with the same bit width as the original image; and replacing the image data corresponding to the first preset area in the intermediate image with the original image data to generate a target image. The finally generated target image is the image data of the original image in the first preset area, all features of the original image are reserved, the data transmission amount is reduced compared with the original image for the part outside the first preset area, the immersion of a user is improved during display through fuzzy processing, and the user experience is improved. And the use experience of the user is improved.