33results about How to "Improve edge definition" patented technology

The inks provide excellent ink-jet prints having excellent imaging onto the media with sharp, smooth edge text print and overall print quality improvements. The aqueous inks each comprise an anionic colorant (CMYK), a short chain alcohol or diol and a cationic surfactant.

The invention discloses an image processing method and device, relates to the field of medical images, and can solve the problem that the contrast and resolution of medical images cannot be effectively enhanced in the prior art. The image processing method of the present invention includes: acquiring a first image and a second image, the first image and the second image being the same; performing a histogram equalization process on the first image; performing a histogram equalization process on the processed first image Two-dimensional discrete wavelet transform to obtain a first high-frequency sub-image and a first low-frequency sub-image; performing two-dimensional discrete wavelet transform on the second image to obtain a second high-frequency sub-image and a second low-frequency sub-image; Perform passivation and fuzzy processing on the second high-frequency sub-image; based on wavelet transform, the first high-frequency sub-image, the first low-frequency sub-image, the processed second high-frequency sub-image and the second low-frequency sub-image Perform image fusion. The present invention can be widely applied in the scene of processing medical images.

The present disclosure relates to a training method, image processing method, device, electronic equipment, and storage medium. The training method includes: acquiring a real image and a style image corresponding to the real image, wherein the real image and the style image The image styles of the images are different; the real image is input into the generating network to obtain a composite image; the edge in the style image is blurred to obtain a first image; through the discriminant network, according to the composite image and the The first difference between the first image and the second difference between the synthesized image and the style image is adjusted to obtain a generation network for generating a style image. The embodiments of the present disclosure can make the synthetic image generated by the trained generation network not only close to the style image, but also improve the edge definition.

The present application discloses an image segmentation method, an image segmentation device, electronic equipment, and a computer-readable storage medium, wherein the image segmentation method includes: acquiring an image to be processed; extracting a first image, wherein the preset region contains specific features in the image to be processed; the first image is enhanced to obtain a second image; the second image is input into an image segmentation network to obtain a segmentation result. The solution of the present application can reduce the occurrence of image segmentation errors, and improve the accuracy of image segmentation to a certain extent.

The invention discloses an analysis iteration fast frequency spectrum extrapolation method for single image restoration, comprising an iteration system initialization process and an analysis iteration process, wherein in the iteration system initialization process, a degraded image frequency spectrum is calculated, and a primary iteration solution and a system primary parameter are set; in the analysis iteration process, the level of images and gradient information along a vertical direction are estimated by continuously elaborate gradient information threshold values in a shrinking way, a forecasting frequency spectrum is forecasted and obtained by analyzing an iteration frequency spectrum and gradient information frequency spectrum, and a high-definition image is obtained by synthetically correcting the degraded image frequency spectrum and forecasting frequency spectrum analysis through image frequency spectrum synthesis correction. The analysis iteration fast frequency spectrum extrapolation method is based on a fast Fourier transform technique, achieves the fast image restoration with little complexity through frequency spectra below cut-off frequencies and by extrapolating frequency spectra above the cut-off frequencies and has obviously superior deblurring and noise restraining capability compared with a conventional method, thereby obviously enhancing the signal-to-noise ratio; and the analysis iteration fast frequency spectrum extrapolation method is convenient to realize by utilizing image processing hardware formed by fast Fourier transform DSP chips.

The invention provides a method for reducing artifacts in an image in real time. The method includes the following steps: collecting a reference image: moving an ultrasonic probe located on a treatment position to a position where the reference image can be obtained, collecting an image of a nidus on the reference image position, and using the collected image of the nidus as the reference image, wherein the reference position is one position which enables a portion of the image collected by the ultrasonic probe, which is corresponding to a treatment target section, to have few artifacts or no artifact; collecting a treatment image: moving the ultrasonic probe from the reference position back to the treatment position, and collecting a current treatment image of the nidus; and registering and fusing images: registering and fusing the collected treatment image and the reference image, and outputting an image after fusion which is an actual treatment image. Correspondingly, the invention provides a device for reducing the artifacts in the image in real time. The method and the device for reducing the artifacts in the image in real time can improve location accuracy of the ultrasonic probe to the treatment target section, and accordingly improve treatment efficiency and treatment effects of an operation guided by ultrasound.