scispace - formally typeset
Search or ask a question
Topic

Image conversion

About: Image conversion is a research topic. Over the lifetime, 2490 publications have been published within this topic receiving 19077 citations.


Papers
More filters
Patent
29 Apr 2020
TL;DR: In this article, a three-segment brightness correction recurrent curve was proposed to improve the quality of the output image by applying a linear segment for dark subrange (SR_d) of a range of brightness signals of colors of input images containing the darkest values of the input brightness signal.
Abstract: FIELD: transmission of video data.SUBSTANCE: invention is intended for image conversion. HDR video decoder has an image color processing device configured to display an input image (Im_in) input color (Y'UV_LDR), wherein input image has first dynamic range (DR_1) of brightness, in output color (R'o, G'o, B'o) of components of red, green and blue color of output image pixel (Im_res; REC_HDR), wherein output image has second dynamic range (DR_2) of brightness, due to which peak brightness of first dynamic range is at least 2 times lower than peak brightness of second dynamic range or vice versa, an approximate display unit configured to apply a three-segment brightness correction recurrent curve, which consists of a linear segment for dark subrange (SR_d) of a range of brightness signals of colors of input images containing the darkest values of the input brightness signal, which is determined by variable (InvBet) of inclination of second linear segment for lightest values of brightness input signal in bright subrange (SR_br), which is controlled by second inclination variable (InvAlph), and parabolic segment between two linear segments.EFFECT: technical result is higher image quality.15 cl, 5 dwg
Patent
26 Jun 2015
TL;DR: In this article, a panorama image processing unit generates a wide viewing angle image like panorama images which is seen from a given point of view, and an image conversion unit converts the generated wide view image into an image of a transmission format in which the angular resolution which is a number of pixels per unit angle centered at the view point is adjusted in response to a ray direction toward the viewpoint.
Abstract: A panorama image processing unit generates a wide viewing angle image like a panorama image which is seen from a given point of view. An image conversion unit converts the generated wide viewing angle image into an image of a transmission format in which the angular resolution which is a number of pixels per unit angle centered at the point of view is adjusted in response to a ray direction toward the point of view. An image provision unit transmits the image in the transmission format in which the angular resolution is adjusted to a head-mounted display unit. The head-mounted display unit converts the received image so as to conform the image to an optical characteristic of a device and displays the image on the device.
Patent
28 Jan 1987
TL;DR: In this paper, a buffer memory of a several line portion was used to save remarkably a memory by processing an image conversion without requiring a page memory before conversion, and the memory can be saved remarkably.
Abstract: PURPOSE:To save remarkably a memory by processing an image conversion by a buffer memory of a several line portion, without requiring a page memory before conversion. CONSTITUTION:For instance, it will suffice that the procedure of a figure (a) is executed successively in case of magnifying conversion, and the procedure of a figure (b) is executed successively in case of a reduction. When executing the magnification, first of all, when the magnification is executed in S1, a picture element having no data is generated, therefore, its interpolation is executed in S2, and subsequently, a parallel movement of S3 is executed. In such a case, omission of the data is not generated, therefore, only a coordinate conversion will suffice. Next, when a rotation is executed in S4, a picture element having no data is generated by the rotation, therefore, the data of this picture element is interpolated in S5, finally, a converted raster data is written in an output use memory 8. In such a way, with respect to the raster data, conversions such as magnification, reduction rotation, parallel movement, etc. can be processed at a high speed, and also, as for the data before a conversion, a full page memory is not required, and it will suffice that only a buffer memory of several lines is provided, therefore, the memory can be saved remarkably.
Patent
16 Dec 2010
TL;DR: In this paper, an image quality improvement method for improving the image quality of an inputted and magnified image (input image) consisting of reducing noise generated in the edge portions of the input image when magnifying the image to create an image with reduced noise was proposed.
Abstract: The present invention relates to an apparatus and method for improving image quality. The image quality improvement method for improving the image quality of an inputted and magnified image (input image) according to the present invention comprises the steps of: reducing noise generated in the edge portions of the input image when magnifying the image to create an image with reduced noise; creating a pseudo high-resolution image with respect to the image with reduced noise; converting the pseudo high-resolution image into an image (low-resolution image) having the same resolution as the image with reduced noise; and comparing the reduced image noise with the low-resolution image, calculating difference information on pixel data, and performing weight-added high-resolution conversion on the basis of the difference information. According to the present invention, data loss generated in a high-definition image conversion process of data can be compensated for, thereby improving image quality to enable the color of a high-resolution image to be naturally represented.
Patent
21 May 1996
TL;DR: In this paper, a registration flag of a RAM 13 is set on or OFF to detect the current date and time data as the data name through a time measuring instrument on condition that the flag is ON.
Abstract: PURPOSE: To eliminate the need to input a data name when image data are stored and to efficiently store the image data by automatically setting current date and time data (storage date and time) as the data name. CONSTITUTION: A CPU 11 sets a registration flag of a RAM 13 ON or OFF. When a PPC function is previously selected through initial setting, etc., the CPU 11 reads image data on a body to be processed by a scanner 22, stores the data in the RAM 13, and transfers the image data to a printer 18, which starts printing the data. During the printing, the CPU 11 checks the registration flag and detects the current date and time data as the data name through a time measuring instrument on condition that the flag is ON. The CPU 11 stores the image data by using the date and time data as the data name. When the image conversion of a code document is selected, the CPU 11 reads the code document and expands the code document on the RAM 13 by using a print image task stored in the RAM 13.

Network Information
Related Topics (5)
Image processing
229.9K papers, 3.5M citations
84% related
Feature (computer vision)
128.2K papers, 1.7M citations
83% related
Pixel
136.5K papers, 1.5M citations
82% related
Feature extraction
111.8K papers, 2.1M citations
80% related
Image segmentation
79.6K papers, 1.8M citations
79% related
Performance
Metrics
No. of papers in the topic in previous years
YearPapers
202132
202074
2019117
2018115
2017100
2016107